Omega-Liutex Method

Prediction of the Vortex Evolution and Influence Analysis of Rough Bed in a Hydraulic Jump with the Omega-Liutex Method

Omega-Luitex법을 이용한 수력점프 발생시 러프 베드의 와류 진화 예측 및 영향 분석

Cong Trieu Tran, Cong Ty Trinh

Abstract

The dissipation of energy downstream of hydropower projects is a significant issue. The hydraulic jump is exciting and widely applied in practice to dissipate energy. Many hydraulic jump characteristics have been studied, such as length of jump Lj and sequent flow depth y2. However, understanding the evolution of the vortex structure in the hydraulic jump shows a significant challenge. This study uses the RNG k-e turbulence model to simulate hydraulic jumps on the rough bed. The Omega-Liutex method is compared with Q-criterion for capturing vortex structure in the hydraulic jump. The formation, development, and shedding of the vortex structure at the rough bed in the hydraulic jumper are analyzed. The vortex forms and rapidly reduces strength on the rough bed, resulting in fast dissipation of energy. At the rough block rows 2nd and 3rd, the vortex forms a vortex rope that moves downstream and then breaks. The vortex-shedding region represents a significant energy attenuation of the flow. Therefore, the rough bed dissipates kinetic energy well. Adding reliability to the vortex determined by the Liutex method, the vorticity transport equation is used to compare the vorticity distribution with the Liutex distribution. The results show a further comprehension of the hydraulic jump phenomenon and its energy dissipation.

Keywords

flow-3D; hydraulic Jump; omega-liutex method; vortex breakdown

References

[1] Viti, N., Valero, D., & Gualtieri, C. (2019). Numerical Simulation of Hydraulic Jumps. Part 2: Recent Results and Future Outlook. Water, 11(1), 28. https://doi.org/10.3390/w11010028
[2] Peterka, A. J. (1978.) Hydraulic Design of Stilling Basins and Energy Dissipators. Department of the Interior, Bureau of Reclamation.
[3] Bejestan, M. S. & Neisi, K. (2009). A new roughened bed hydraulic jump stilling basin. Asian journal of applied sciences, 2(5), 436-445. https://doi.org/10.3923/ajaps.2009.436.445
[4] Tokyay, N. D. (2005). Effect of channel bed corrugations on hydraulic jumps. Impacts of Global Climate Change, 1-9. https://doi.org/10.1061/40792(173)408
[5] Nikmehr, S. & Aminpour, Y. (2020). Numerical Simulation of Hydraulic Jump over Rough Beds. Periodica Polytechnica Civil Engineering, 64(2), 396-407. https://doi.org/10.3311/PPci.15292
[6] Hunt, J. C., Wray, A. A., & Moin, P. (1988). Eddies, streams, and convergence zones in turbulent flows. Studying turbulence using numerical simulation databases. 2. Proceedings of the 1988 summer program.
[7] Gao, Y. & Liu, C. (2018). Rortex and comparison with eigenvalue-based vortex identification criteria. Physics of Fluids, 30(8), 085107. https://doi.org/10.1063/1.5040112
[8] Liu, C., Gao, Y., Tian, S., & Dong, X. (2018). Rortex – A new vortex vector definition and vorticity tensor and vector decompositions. Physics of Fluids, 30(3), 035103. https://doi.org/10.1063/1.5023001
[9] Liu, C. et al. (2019). Third generation of vortex identification methods: Omega and Liutex/Rortex based systems. Journal of Hydrodynamics, 31(2), 205-223. https://doi.org/10.1007/s42241-019-0022-4
[10] Liu, C., Wang, Y., Yang, Y. et al (2016). New omega vortex identification method. Science China Physics, Mechanics & Astronomy, (8), 56-64. https://doi.org/10.1007/s11433-016-0022-6
[11] Tran, C. T. & Pham, D. C. (2022). Application of Liutex and Entropy Production to Analyze the Influence of Vortex Rope in the Francis-99 Turbine Draft Tube. Tehnički vjesnik, 29(4), 1177-1183. https://doi.org/10.17559/TV-20210821070801
[12] Dong, X., Gao, Y., & Liu, C. (2019). New normalized Rortex/vortex identification method. Physics of Fluids, 31(1), 011701. https://doi.org/10.1063/1.5066016
[13] Wang, L., Zheng, Z., Cai, W. et al. (2019). Extension Omega and Omega-Liutex methods applied to identify vortex structures in viscoelastic turbulent flow. Journal of Hydrodynamics, 31(5), 911-921. https://doi.org/10.1007/s42241-019-0045-x
[14] Xu, H., Cai, X., & Liu, C. (2019). Liutex (vortex) core definition and automatic identification for turbulence vortex structures. Journal of Hydrodynamics, 31(5), 857-863. https://doi.org/10.1007/s42241-019-0066-5
[15] Tran, C. T. et al. (2020). Prediction of the precessing vortex core in the Francis-99 draft tube under off-design conditions by using Liutex/Rortex method. Journal of Hydrodynamics, 32, 623-628. https://doi.org/10.1007/s42241-020-0031-3
[16] Liu, C. et al. (2019). A Liutex based definition of vortex axis line. arXiv preprint arXiv:1904.10094. https://doi.org/10.48550/arXiv.1904.10094
[17] Samadi-Boroujeni, H. et al. (2013). Effect of triangular corrugated beds on the hydraulic jump characteristics. Canadian Journal of Civil Engineering, 40(9), 841-847. https://doi.org/10.1139/cjce-2012-0019
[18] Ghaderi, A. et al. (2020). Characteristics of free and submerged hydraulic jumps over different macroroughnesses. Journal of Hydroinformatics, 22(6), 1554-1572. https://doi.org/10.2166/hydro.2020.298
[19] Wu, Z. et al. (2021). Analysis of the influence of transverse groove structure on the flow of a flat-plate surface based on Liutex parameters. Engineering Applications of Computational Fluid Mechanics, 15(1), 1282-1297. https://doi.org/10.1080/19942060.2021.1968955
[20] Ji, B., et al. (2014). Numerical simulation of threedimensional cavitation shedding dynamics with special emphasis on cavitation – vortex interaction. Ocean Engineering, 87, 64-77. https://doi.org/10.1016/j.oceaneng.2014.05.005
[21] Tran, C., Bin, J., & Long, X. (2019). Simulation and Analysis of Cavitating Flow in the Draft Tube of the Francis Turbine with Splitter Blades at Off-Design Condition. Tehnicki vjesnik – Technical Gazette, 26(6). https://doi.org/10.17559/TV-20190316042929
Download PDF
Numerical Investigation of the Local Scour for Tripod Pile Foundation

Numerical Investigation of the Local Scour for Tripod Pile Foundation

Waqed H. Hassan | Zahraa Mohammad Fadhe* | Rifqa F. Thiab | Karrar Mahdi
Civil Engineering Department, Faculty of Engineering, University of Warith Al-Anbiyaa, Kerbala 56001, Iraq
Civil Engineering Department, Faculty of Engineering, University of Kerbala, Kerbala 56001, Iraq
Corresponding Author Email: Waqed.hammed@uowa.edu.iq

OPEN ACCESS

Abstract: 

This work investigates numerically a local scour moves in irregular waves around tripods. It is constructed and proven to use the numerical model of the seabed-tripod-fluid with an RNG k turbulence model. The present numerical model then examines the flow velocity distribution and scour characteristics. After that, the suggested computational model Flow-3D is a useful tool for analyzing and forecasting the maximum scour development and the flow field in random waves around tripods. The scour values affecting the foundations of the tripod must be studied and calculated, as this phenomenon directly and negatively affects the structure of the structure and its design life. The lower diagonal braces and the main column act as blockages, increasing the flow accelerations underneath them.  This increases the number of particles that are moved, which in turn creates strong scouring in the area. The numerical model has a good agreement with the experimental model, with a maximum percentage of error of 10% between the experimental and numerical models. In addition, Based on dimensional analysis parameters, an empirical equation has been devised to forecast scour depth with flow depth, median size ratio, Keulegan-Carpenter (Kc), Froud number flow, and wave velocity that the results obtained in this research at various flow velocities and flow depths demonstrated that the maximum scour depth rate depended on wave height with rising velocities and decreasing particle sizes (d50) and the scour depth attains its steady-current value for Vw < 0.75. As the Froude number rises, the maximum scour depth will be large.

Keywords: 

local scour, tripod foundation, Flow-3D​, waves

1. Introduction

New energy sources have been used by mankind since they become industrialized. The main energy sources have traditionally been timber, coal, oil, and gas, but advances in the science of new energies, such as nuclear energy, have emerged [1, 2]. Clean and renewable energy such as offshore wind has grown significantly during the past few decades. There are numerous different types of foundations regarding offshore wind turbines (OWTs), comprising the tripod, jacket, gravity foundation, suction anchor (or bucket), and monopile [3, 4]. When the water depth is less than 30 meters, Offshore wind farms usually employ the monopile type [4]. Engineers must deal with the wind’s scouring phenomenon turbine foundations when planning and designing wind turbines for an offshore environment [5]. Waves and currents generate scour, this is the erosion of soil near a submerged foundation and at its location [6]. To predict the regional scour depth at a bridge pier, Jalal et al. [7-10] developed an original gene expression algorithm using artificial neural networks. Three monopiles, one main column, and several diagonal braces connecting the monopiles to the main column make up the tripod foundation, which has more complicated shapes than a single pile. The design of the foundation may have an impact on scour depth and scour development since the foundation’s form affects the flow field [11, 12]. Stahlmann [4] conducted several field investigations. He discovered that the main column is where the greatest scour depth occurred. Under the main column is where the maximum scour depth occurs in all experiments. The estimated findings show that higher wave heights correspond to higher flow velocities, indicating that a deeper scour depth is correlated with finer silt granularity [13] recommends as the design value for a single pile. These findings support the assertion that a tripod may cause the seabed to scour more severely than a single pile. The geography of the scour is significantly more influenced by the KC value (Keulegan–Carpenter number)

The capability of computer hardware and software has made computational fluid dynamics (CFD) quite popular to predict the behavior of fluid flow in industrial and environmental applications has increased significantly in recent years [14].

Finding an acceptable piece of land for the turbine’s construction and designing the turbine pile precisely for the local conditions are the biggest challenges. Another concern related to working in a marine environment is the effect of sea waves and currents on turbine piles and foundations. The earth surrounding the turbine’s pile is scoured by the waves, which also render the pile unstable.

In this research, the main objective is to investigate numerically a local scour around tripods in random waves. It is constructed and proven to use the tripod numerical model. The present numerical model is then used to examine the flow velocity distribution and scour characteristics.

2. Numerical Model

To simulate the scouring process around the tripod foundation, the CFD code Flow-3D was employed. By using the fractional area/volume method, it may highlight the intricate boundaries of the solution domain (FAVOR).

This model was tested and validated utilizing data derived experimentally from Schendel et al. [15] and Sumer and Fredsøe [6]. 200 runs were performed at different values of parameters.

2.1 Momentum equations

The incompressible viscous fluid motion is described by the three RANS equations listed below [16]:

(1)

\frac{\partial u}{\partial t}+\frac{1}{{{V}_{F}}}\left( u{{A}_{x}}\frac{\partial u}{\partial x}+v{{A}_{y}}\frac{\partial u}{\partial y}+w{{A}_{z}}\frac{\partial u}{\partial z} \right)=-\frac{1}{\rho }\frac{\partial p}{\partial x}+{{G}_{x}}+fx

(2)

\frac{\partial v}{\partial t}+\frac{1}{{{V}_{F}}}\left( u{{A}_{x}}\frac{\partial v}{\partial x}+v{{A}_{y}}\frac{\partial v}{\partial y}+w{{A}_{z}}\frac{\partial v}{\partial z} \right)=-\frac{1}{\rho }\frac{\partial p}{\partial y}+{{G}_{y}}+\text{f}y

 (3)

\frac{\partial w}{\partial t}+\frac{1}{{{V}_{F}}}\left( u{{A}_{x}}\frac{\partial w}{\partial x}+v{{A}_{y}}\frac{\partial w}{\partial y}+w{{A}_{z}}\frac{\partial w}{\partial z} \right)=-\frac{1}{\rho }\frac{\partial p}{\partial z}+{{G}_{z}}+\text{fz}

where, respectively, u, v, and w represent the x, y, and z flow velocity components; volume fraction (VF), area fraction (Ai; I=x, y, z), water density (f), viscous force (fi), and body force (Gi) are all used in the formula.

2.2 Model of turbulence

Several turbulence models would be combined to solve the momentum equations. A two-equation model of turbulence is the RNG k-model, which has a high efficiency and accuracy in computing the near-wall flow field. Therefore, the flow field surrounding tripods was captured using the RNG k-model.

2.3 Model of sediment scour

2.3.1 Induction and deposition

Eq. (4) can be used to determine the particle entrainment lift velocity [17].

(4)

{{u}_{lift,i}}={{\alpha }_{i}}{{n}_{s}}d_{*}^{0.3}{{\left( \theta -{{\theta }_{cr}} \right)}^{1.5}}\sqrt{\frac{\parallel g\parallel {{d}_{i}}\left( {{\rho }_{i}}-{{\rho }_{f}} \right)}{{{\rho }_{f}}}}

α𝛼i   is the Induction parameter, ns the normal vector is parallel to the seafloor, and for the present numerical model, ns=(0,0,1), θ𝜃cr is the essential Shields variable, g is the accelerated by gravity, di is the size of the particles, ρi is species density in beds, and d∗ The diameter of particles without dimensions; these values can be obtained in Eq. (5).

(5)

{{d}_{*}}={{d}_{i}}{{\left( \frac{\parallel g\parallel {{\rho }_{f}}\left( {{\rho }_{i}}-{{\rho }_{f}} \right)}{\mu _{f}^{2}} \right)}^{1/3}}

μ𝜇f is this equation a dynamic viscosity of the fluid. cr was determined from an equation based on Soulsby [18].

(6)

{{\theta }_{cr}}=\frac{0.3}{1+1.2{{d}_{*}}}+0.055\left[ 1-\text{exp}\left( -0.02{{d}_{*}} \right) \right]

The equation was used to determine how quickly sand particles set Eq. (7):

(7)

{{\mathbf{u}}_{\text{nsettling},i}}=\frac{{{v}_{f}}}{{{d}_{i}}}\left[ {{\left( {{10.36}^{2}}+1.049d_{*}^{3} \right)}^{0.5}}-10.36 \right]

vf  stands for fluid kinematic viscosity.

2.3.2 Transportation for bed loads

Van Rijn [19] states that the speed of bed load conveyance was determined as:

(8)

{{~}_{\text{bedload},i}}=\frac{{{q}_{b,i}}}{{{\delta }_{i}}{{c}_{b,i}}{{f}_{b}}}

fb  is the essential particle packing percentage, qb, i is the bed load transportation rate, and cb, I the percentage of sand by volume i. These variables can be found in Eq. (9), Eq. (10), fb, δ𝛿i the bed load thickness.

(9)

{{q}_{b,i}}=8{{\left[ \parallel g\parallel \left( \frac{{{\rho }_{i}}-{{\rho }_{f}}}{{{\rho }_{f}}} \right)d_{i}^{3} \right]}^{\frac{1}{2}}}

(10)

{{\delta }_{i}}=0.3d_{*}^{0.7}{{\left( \frac{\theta }{{{\theta }_{cr}}}-1 \right)}^{0.5}}{{d}_{i}}

In this paper, after the calibration of numerous trials, the selection of parameters for sediment scour is crucial. Maximum packing fraction is 0.64 with a shields number of 0.05, entrainment coefficient of 0.018, the mass density of 2650, bed load coefficient of 12, and entrainment coefficient of 0.01.

3. Model Setup

To investigate the scour characteristics near tripods in random waves, the seabed-tripod-fluid numerical model was created as shown in Figure 1. The tripod basis, a seabed, and fluid and porous medium were all components of the model. The seabed was 240 meters long, 40 meters wide, and three meters high. It had a median diameter of d50 and was composed of uniformly fine sand. The 2.5-meter main column diameter D. The base of the main column was three dimensions above the original seabed. The center of the seafloor was where the tripod was, 130 meters from the offshore and 110 meters from the onshore. To prevent wave reflection, the porous media were positioned above the seabed on the onshore side.

image013.png

Figure 1. An illustration of the numerical model for the seabed-tripod-fluid

3.1 Generation of meshes

Figure 2 displays the model’s mesh for the Flow-3D software grid. The current model made use of two different mesh types: global mesh grid and nested mesh grid. A mesh grid with the following measurements was created by the global hexahedra mesh grid: 240m length, 40m width, and 32m height. Around the tripod, a finer nested mesh grid was made, with dimensions of 0 to 32m on the z-axis, 10 to 30 m on the x-axis, and 25 to 15 m on the y-axis. This improved the calculation’s precision and mesh quality.

image014.png

Figure 2. The mesh block sketch

3.2 Conditional boundaries

To increase calculation efficiency, the top side, The model’s two x-z plane sides, as well as the symmetry boundaries, were all specified. For u, v, w=0, the bottom boundary wall was picked. The offshore end of the wave boundary was put upstream. For the wave border, random waves were generated using the wave spectrum from the Joint North Sea Wave Project (JONSWAP). Boundary conditions are shown in Figure 3.

image015.png

Figure 3. Boundary conditions of the typical problem

The wave spectrum peak enhancement factor (=3.3 for this work) and can be used to express the unidirectional JONSWAP frequency spectrum.

3.3 Mesh sensitivity

Before doing additional research into scour traits and scour depth forecasting, mesh sensitivity analysis is essential. Three different mesh grid sizes were selected for this section: Mesh 1 has a 0.45 by 0.45 nested fine mesh and a 0.6 by 0.6 global mesh size. Mesh 2 has a 0.4 global mesh size and a 0.35 nested fine mesh size, while Mesh 3 has a 0.25 global mesh size and a nested fine mesh size of 0.15. Comparing the relative fine mesh size (such as Mesh 2 or Mesh 3) to the relatively coarse mesh size (such as Mesh 1), a larger scour depth was seen; this shows that a finer mesh size can more precisely represent the scouring and flow field action around a tripod. Significantly, a lower mesh size necessitates a time commitment and a more difficult computer configuration. Depending on the sensitivity of the mesh guideline utilized by Pang et al., when Mesh 2 is applied, the findings converge and the mesh size is independent [20]. In the next sections, scouring the area surrounding the tripod was calculated using Mesh 2 to ensure accuracy and reduce computation time. The working segment generates a total of 14, 800,324 cells.

3.4 Model validation

Comparisons between the predicted outcomes from the current model and to confirm that the current numerical model is accurate and suitably modified, experimental data from Sumer and Fredsøe [6] and Schendel et al. [15] were used. For the experimental results of Run 05, Run 15, and Run 22 from Sumer and Fredsøe [6], the experimental A9, A13, A17, A25, A26, and A27 results from Schendel et al. [15], and the numerical results from the current model are shown in Figure 4. The present model had d50=0.051cm, the height of the water wave(h)=10m, and wave velocity=0.854 m.s-1.

image016.png

Figure 4. Cell size effect

image017.png

Figure 5. Comparison of the present study’s maximum scour depth with that authored by Sumer and Fredsøe [6] and Schendel et al. [15]

According to Figure 5, the highest discrepancy between the numerical results and experimental data is about 10%, showing that overall, there is good agreement between them. The ability of the current numerical model to accurately depict the scour process and forecast the maximum scour depth (S) near foundations is demonstrated by this. Errors in the simulation were reduced by using the calibrated values of the parameter. Considering these results, a suggested simulated scouring utilizing a Flow-3D numerical model is confirmed as a superior way for precisely forecasting the maximum scour depth near a tripod foundation in random waves.

3.5 Dimensional analysis

The variables found in this study as having the greatest impacts, variables related to flow, fluid, bed sediment, flume shape, and duration all had an impact on local scouring depth (t). Hence, scour depth (S) can be seen as a function of these factors, shown as:

(11)

S=f\left(\rho, v, V, h, g, \rho s, d_{50}, \sigma g, V_w, D, d, T_v, t\right)

With the aid of dimensional analysis, the 14-dimensional parameters in Eq. (11) were reduced to 6 dimensionless variables using Buckingham’s -theorem. D, V, and were therefore set as repetition parameters and others as constants, allowing for the ignoring of their influence. Eq. (12) thus illustrates the relationship between the effect of the non-dimensional components on the depth of scour surrounding a tripod base.

(12)

\frac{S}{D}=f\left(\frac{h}{D}, \frac{d 50}{D}, \frac{V}{V W}, F r, K c\right)

where, SD𝑆𝐷 are scoured depth ratio, VVw𝑉𝑉𝑤 is flow wave velocity, d50D𝑑50𝐷 median size ratio, $Fr representstheFroudnumber,and𝑟𝑒𝑝𝑟𝑒𝑠𝑒𝑛𝑡𝑠𝑡ℎ𝑒𝐹𝑟𝑜𝑢𝑑𝑛𝑢𝑚𝑏𝑒𝑟,𝑎𝑛𝑑Kc$ is the Keulegan-Carpenter.

4. Result and Discussion

4.1 Development of scour

Similar to how the physical model was used, this numerical model was also used. The numerical model’s boundary conditions and other crucial variables that directly influence the outcomes were applied (flow depth, median particle size (d50), and wave velocity). After the initial 0-300 s, the scour rate reduced as the scour holes grew quickly. The scour depths steadied for about 1800 seconds before reaching an asymptotic value. The findings of scour depth with time are displayed in Figure 6.

4.2 Features of scour

Early on (t=400s), the scour hole began to appear beneath the main column and then began to extend along the diagonal bracing connecting to the wall-facing pile. Gradually, the geography of the scour; of these results is similar to the experimental observations of Stahlmann [4] and Aminoroayaie Yamini et al. [1]. As the waves reached the tripod, there was an enhanced flow acceleration underneath the main column and the lower diagonal braces as a result of the obstructing effects of the structural elements. More particles are mobilized and transported due to the enhanced near-bed flow velocity, it also increases bed shear stress, turbulence, and scour at the site. In comparison to a single pile, the main column and structural components of the tripod have a significant impact on the flow velocity distribution and, consequently, the scour process and morphology. The main column and seabed are separated by a gap, therefore the flow across the gap may aid in scouring. The scour hole first emerged beneath the main column and subsequently expanded along the lower structural components, both Aminoroayaie Yamini et al. [1] and Stahlmann [4] made this claim. Around the tripod, there are several different scour morphologies and the flow velocity distribution as shown in Figures 7 and 8.

image023.png

Figure 6. Results of scour depth with time

image024.png

image025.png

image026.png

image027.png

Figure 7. The sequence results of scour depth around tripod development (reached to steady state) simulation time

image028.png

image029.png

image030.png

image031.png

Figure 8. Random waves of flow velocity distribution around a tripod

4.3 Wave velocity’s (Vw) impact on scour depth

In this study’s section, we looked at how variations in wave current velocity affected the scouring depth. Bed scour pattern modification could result from an increase or decrease in waves. As a result, the backflow area produced within the pile would become stronger, which would increase the depth of the sediment scour. The quantity of current turbulence is the primary cause of the relationship between wave height and bed scour value. The current velocity has increased the extent to which the turbulence energy has changed and increased in strength now present. It should be mentioned that in this instance, the Jon swap spectrum random waves are chosen. The scour depth attains its steady-current value for Vw<0.75, Figure 9 (a) shows that effect. When (V) represents the mean velocity=0.5 m.s-1.

image032.png

(a)

image033.png

(b)

image034.png

(c)

image035.png

(d)

Figure 9. Main effects on maximum scour depth (Smax) as a function of column diameter (D)

4.4 Impact of a median particle (d50) on scour depth

In this section of the study, we looked into how variations in particle size affected how the bed profile changed. The values of various particle diameters are defined in the numerical model for each run numerical modeling, and the conditions under which changes in particle diameter have an impact on the bed scour profile are derived. Based on Figure 9 (b), the findings of the numerical modeling show that as particle diameter increases the maximum scour depth caused by wave contact decreases. When (d50) is the diameter of Sediment (d50). The Shatt Al-Arab soil near Basra, Iraq, was used to produce a variety of varied diameters.

4.5 Impact of wave height and flow depth (h) on scour depth

One of the main elements affecting the scour profile brought on by the interaction of the wave and current with the piles of the wind turbines is the height of the wave surrounding the turbine pile causing more turbulence to develop there. The velocity towards the bottom and the bed both vary as the turbulence around the pile is increased, modifying the scour profile close to the pile. According to the results of the numerical modeling, the depth of scour will increase as water depth and wave height in random waves increase as shown in Figure 9 (c).

4.6 Froude number’s (Fr) impact on scour depth

No matter what the spacing ratio, the Figure 9 shows that the Froude number rises, and the maximum scour depth often rises as well increases in Figure 9 (d). Additionally, it is crucial to keep in mind that only a small portion of the findings regarding the spacing ratios with the smallest values. Due to the velocity acceleration in the presence of a larger Froude number, the range of edge scour downstream is greater than that of upstream. Moreover, the scouring phenomena occur in the region farthest from the tripod, perhaps as a result of the turbulence brought on by the collision of the tripod’s pile. Generally, as the Froude number rises, so does the deposition height and scour depth.

4.7 Keulegan-Carpenter (KC) number

The geography of the scour is significantly more influenced by the KC value. Greater KC causes a deeper equilibrium scour because an increase in KC lengthens the horseshoe vortex’s duration and intensifies it as shown in Figure 10.

The result can be attributed to the fact that wave superposition reduced the crucial KC for the initiation of the scour, particularly under small KC conditions. The primary variable in the equation used to calculate This is the depth of the scouring hole at the bed. The following expression is used to calculate the Keulegan-Carpenter number:

Kc=Vw∗TpD𝐾𝑐=𝑉𝑤∗𝑇𝑝𝐷                          (13)

where, the wave period is Tp and the wave velocity is shown by Vw.

image037.png

Figure 10. Relationship between the relative maximum scour depth and KC

5. Conclusion

(1) The existing seabed-tripod-fluid numerical model is capable of faithfully reproducing the scour process and the flow field around tripods, suggesting that it may be used to predict the scour around tripods in random waves.

(2) Their results obtained in this research at various flow velocities and flow depths demonstrated that the maximum scour depth rate depended on wave height with rising velocities and decreasing particle sizes (d50).

(3) A diagonal brace and the main column act as blockages, increasing the flow accelerations underneath them. This raises the magnitude of the disturbance and the shear stress on the seafloor, which in turn causes a greater number of particles to be mobilized and conveyed, as a result, causes more severe scour at the location.

(4) The Froude number and the scouring process are closely related. In general, as the Froude number rises, so does the maximum scour depth and scour range. The highest maximum scour depth always coincides with the bigger Froude number with the shortest spacing ratio.

Since the issue is that there aren’t many experiments or studies that are relevant to this subject, therefore we had to rely on the monopile criteria. Therefore, to gain a deeper knowledge of the scouring effect surrounding the tripod in random waves, further numerical research exploring numerous soil, foundation, and construction elements as well as upcoming physical model tests will be beneficial.

Nomenclature

CFDComputational fluid dynamics
FAVORFractional Area/Volume Obstacle Representation
VOFVolume of Fluid
RNGRenormalized Group
OWTsOffshore wind turbines
Greek Symbols
ε, ωDissipation rate of the turbulent kinetic energy, m2s-3
Subscripts
d50Median particle size
VfVolume fraction
GTTurbulent energy of buoyancy
KTTurbulent velocity
PTKinetic energy of the turbulence
ΑiInduction parameter
nsInduction parameter
ΘΘcrThe essential Shields variable
DiDiameter of sediment
d∗The diameter of particles without dimensions
µfDynamic viscosity of the fluid
qb,iThe bed load transportation rate
Cs,iSand particle’s concentration of mass
DDiameter of pile
DfDiffusivity
DDiameter of main column
FrFroud number
KcKeulegan–Carpenter number
GAcceleration of gravity g
HFlow depth
VwWave Velocity
VMean Velocity
TpWave Period
SScour depth

  References

[1] Aminoroayaie Yamini, O., Mousavi, S.H., Kavianpour, M.R., Movahedi, A. (2018). Numerical modeling of sediment scouring phenomenon around the offshore wind turbine pile in marine environment. Environmental Earth Sciences, 77: 1-15. https://doi.org/10.1007/s12665-018-7967-4

[2] Hassan, W.H., Hashim, F.S. (2020). The effect of climate change on the maximum temperature in Southwest Iraq using HadCM3 and CanESM2 modelling. SN Applied Sciences, 2(9): 1494. https://doi.org/10.1007/s42452-020-03302-z

[3] Fazeres-Ferradosa, T., Rosa-Santos, P., Taveira-Pinto, F., Pavlou, D., Gao, F.P., Carvalho, H., Oliveira-Pinto, S. (2020). Preface: Advanced research on offshore structures and foundation design part 2. In Proceedings of the Institution of Civil Engineers-Maritime Engineering. Thomas Telford Ltd, 173(4): 96-99. https://doi.org/10.1680/jmaen.2020.173.4.96

[4] Stahlmann, A. (2013). Numerical and experimental modeling of scour at foundation structures for offshore wind turbines. In ISOPE International Ocean and Polar Engineering Conference. ISOPE, pp. ISOPE-I.

[5] Petersen, T.U., Sumer, B.M., Fredsøe, J. (2014). Edge scour at scour protections around offshore wind turbine foundations. In 7th International Conference on Scour and Erosion. CRC Press, pp. 587-592.

[6] Sumer, B.M., Fredsøe, J. (2001). Scour around pile in combined waves and current. Journal of Hydraulic Engineering, 127(5): 403-411. https://doi.org/10.1061/(ASCE)0733-9429(2001)127:5(403)

[7] Jalal, H.K., Hassan, W.H. (2020). Effect of bridge pier shape on depth of scour. In IOP Conference Series: Materials Science and Engineering. IOP Publishing, 671(1): 012001. https://doi.org/10.1088/1757-899X/671/1/012001

[8] Hassan, W.H., Jalal, H.K. (2021). Prediction of the depth of local scouring at a bridge pier using a gene expression programming method. SN Applied Sciences, 3(2): 159. https://doi.org/10.1007/s42452-020-04124-9

[9] Jalal, H.K., Hassan, W.H. (2020). Three-dimensional numerical simulation of local scour around circular bridge pier using Flow-3D software. In IOP Conference Series: Materials Science and Engineering. IOP Publishing, 745(1): 012150. https://doi.org/10.1088/1757-899X/745/1/012150

[10] Hassan, W.H., Attea, Z.H., Mohammed, S.S. (2020). Optimum layout design of sewer networks by hybrid genetic algorithm. Journal of Applied Water Engineering and Research, 8(2): 108-124. https://doi.org/10.1080/23249676.2020.1761897

[11] Hassan, W.H., Hussein, H.H., Alshammari, M.H., Jalal, H.K., Rasheed, S.E. (2022). Evaluation of gene expression programming and artificial neural networks in PyTorch for the prediction of local scour depth around a bridge pier. Results in Engineering, 13: 100353. https://doi.org/10.1016/j.rineng.2022.100353

[12] Hassan, W.H., Hh, H., Mohammed, S.S., Jalal, H.K., Nile, B.K. (2021). Evaluation of gene expression programming to predict the local scour depth around a bridge pier. Journal of Engineering Science and Technology, 16(2): 1232-1243. https://doi.org/10.1016/j.rineng.2022.100353

[13] Nerland, C. (2010). Offshore wind energy: Balancing risk and reward. In Proceedings of the Canadian Wind Energy Association’s 2010 Annual Conference and Exhibition, Canada, p. 2000. 

[14] Hassan, W.H., Nile, B.K., Mahdi, K., Wesseling, J., Ritsema, C. (2021). A feasibility assessment of potential artificial recharge for increasing agricultural areas in the kerbala desert in Iraq using numerical groundwater modeling. Water, 13(22): 3167. https://doi.org/10.3390/w13223167

[15] Schendel, A., Welzel, M., Schlurmann, T., Hsu, T.W. (2020). Scour around a monopile induced by directionally spread irregular waves in combination with oblique currents. Coastal Engineering, 161: 103751. https://doi.org/10.1016/j.coastaleng.2020.103751

[16] Yakhot, V., Orszag, S.A. (1986). Renormalization group analysis of turbulence. I. Basic theory. Journal of Scientific Computing, 1(1): 3-51. https://doi.org/10.1007/BF01061452

[17] Mastbergen, D.R., Van Den Berg, J.H. (2003). Breaching in fine sands and the generation of sustained turbidity currents in submarine canyons. Sedimentology, 50(4): 625-637. https://doi.org/10.1046/j.1365-3091.2003.00554.x

[18] Soulsby, R. (1997). Dynamics of marine sands. https://doi.org/10.1680/doms.25844

[19] Van Rijn, L.C. (1984). Sediment transport, part I: Bed load transport. Journal of Hydraulic Engineering, 110(10): 1431-1456. https://doi.org/10.1061/(ASCE)0733-9429(1984)110:10(1431)

[20] Pang, A.L.J., Skote, M., Lim, S.Y., Gullman-Strand, J., Morgan, N. (2016). A numerical approach for determining equilibrium scour depth around a mono-pile due to steady currents. Applied Ocean Research, 57: 114-124. https://doi.org/10.1016/j.apor.2016.02.010

Investigating effects of lateral inflow characteristics on main flow using numerical modeling

Investigating effects of lateral inflow characteristics on main flow using numerical modeling

수치모델링을 이용한 측면 유입특성이 본류에 미치는 영향 조사

Mohammad Raze Raeisi Dehkordi1*, Amir Hossein Yeganeh Mazhar1
, Farzaneh Kheradzare2
1– PhD. Student in the Department of Construction and Water Management, Science and Research Unit, Islamic Azad
University, Tehran, Iran
2– M.Sc. Graduate Water resource management, Department of Civil Engineering and Mechanics, Ghiaseddin Jamshid
Kashani University, Qazvin, Iran

  • Corresponding author: mohamadreza.raeisi.d@gmail.com

Keywords

Channel Confluence, Channel cross, sectional area, Cross channel angles, Modelling, Flow-3D

Abstract

Introduction

One of the key issues in river engineering is analyzing the flow properties at the intersection of natural rivers and canals. The flow of the side channel moves away from the intersection of the two channels as a result of the exchange of input force from the side channel with the main flow after coming into contact with it. One of the most evident properties of the flow in these sections is the development of a revolving region with low pressure and even negative pressure close to the inner wall of the side channel. One advantage of the whirling flow in this low-pressure region is that it gives the flow enough space to sediment, but it also increases flow speed near the channel’s bottom and outside wall by lowering the intersectional area of the flow. One of the most crucial considerations in the design of these intersections is minimizing sedimentation in the rotating region and scouring in the area above the shear plane.

Materials and methods:

The channel (flume) created in the laboratory based on Weber et al., (2001) model, was employed in the current investigation to confirm the validity and examine other study objectives. The main channel is 21. 95 meters long, while the side channel, which is at a 90-degree angle to the main channel, is 3. 66 meters long. The total downstream discharge is approximately 0. 17 m3/s, with the upstream velocities of the main channel being 0. 166 m/s and the side channel being 0. 5 m/s. In both channels, the flow depth and width are 0. 91 meters and 0. 296 meters, respectively. In this study, 6 various models’ angles of intersection between the main and side channels, inlet flow velocity, intersectional area, and side channel length have been examined. Models 2 and 3 have intersection angles of 60 and 30 degrees, respectively, and share the rest of their attributes with the fundamental model, or model number 1. Model 1 is the same as Weber’s experimental model. The length of the side channel in model 4 is different from model 1. The only difference between model 6 and the basic model is the side channel intake speed.

Results and Discussion

Analyzing the intersection angle The angle between the main channel and the side channel is investigated in this section of the findings. Models 1, 2, and 3 are assessed using the intersection angles of 90, 60, and 30 degrees, respectively. In some studies, the impact of the intersection angle has been examined, but in this study, three-dimensional investigation in transverse and longitudinal sections as well as the plan of the intersection is discussed, as can be observed from the literature review. Considering three models with intersection angles of 90, 60, and 30 degrees, the kinetic energy contours at the channel’s middle height can be obtained for each model. The channel with a 30-degree intersection angle (model 3) has the maximum kinetic energy in the flow. The channel with a 60-degree intersection has the minimum kinetic energy. As a result of the maximum deviation of the flow in the main channel caused by the flow of the side channel, the channel with a 90-degree intersection also has the maximum kinetic energy near the wall in front of the side channel.

Examining the side channel length In model 1, the side channel is 3. 66 meters long, whereas in model 4, it is 5. 52 meters long. This study aims to determine how changing the side channel’s length affects the flow pattern where two channels intersect. The kinetic energy contours were obtained for two states of the channel length, which are known to extend the lateral channel, increase the energy of the flow after the intersection, and shorten the length of the high-kinetic energy zone. When compared to model 1 with a shorter length of the side channel, the width of the flow separation zone is reduced by approximately 20%, which results in less flow sedimentation. Figure 12 illustrates the rotating zones in the flow separation area. The flow separation region’s length is essentially unchanged. Studying the intersection of the lateral channel After determining the lateral channel’s length, its width and, consequently, its intersectional area should be evaluated.

This section compares model 1 width of 0. 91 meters to model 5 width of 1. 40 meters. One of the most recent topics related to the intersection of the main and side channels is examining the intersection of the side channel. In model 5, the side channel’s flow rate has also increased due to an increase in the width or intersection of the channel. The flow rate through the intersection and the momentum of the flow from the side channel and the main channel increase when the side channel flow rate rises. The findings indicate that when flow width and side channel flow rise, energy increases after the inlet.

Investigating the value of inlet speed in the side channel Unlike the preceding sections, which were all concerned with the channel geometry, the inlet velocity in the side channel is one of the hydraulic parameters of the flow. In this section, models 1 and 6 with inlet velocities of the side channel of 0. 5 and 0. 75 m/s are evaluated. According to the modeling, the flow is somewhat horst before and immediately on the intersection of the flow level, but it undergoes a substantial prolapse just after the intersection. Model 6 has a larger volume and height of flow, but a smaller and softer prolapse after the intersection.

Conclusion

Some hydraulic and geometric properties of the intersection of channels have been examined using Flow-3D software. The RNG turbulence model was used for three-dimensional modeling. Some of the results are listed below. The flow is uniform upstream of the main and minor channels and only slightly becomes horst at the intersection. The analysis of the lengthening of the side channel revealed a 20% reduction in the separation zone’s width and a considerable reduction in the kinetic energy at the intersection. The input flow rate of this channel to the intersection increases with the speed and width of the side channel, which accounts for the local drop in the width of the main channel flow.

References

  • Azhdari, K., Talebi, Z. & Hosseini, S. H. (2020). Simulation of Subcritical Flow Distribution and Water Surface Fluctuations in Fourbranch Open Channel Junction with FLOW 3D. Irrigation and Drainage, 14(3), 1018- 1031. (In persian).
  • Behdarvandi, M., Hajipour, M., Parsi, E. & Ansari ghojghar, M. (2022). Investigation of Velocity Changes in a Straight Asymmetric pattern at river bend. Water and Soil Conservation, 22(6), 81-89. (In Persian).
  • Ghobadian, R. & Seyedi tabar, Z. (2016). Numerical investigating of the effect of lateral channel junction position on flow Rectangular Composite Channel Using Flow3D Software. Irrigation and Water Engineering, 13(1), 1-16. Doi: 10.22125/iwe.2022.158503 (In Persian).
  • Burqaʻi, S. M. & Nazari, A. (2003). Laboratory investigation of sediment pattern at the intersection of channels. 6th International Civil Engineering Conference, Amirkabir University of Technology, Tehran, Iran (In Persian).
  • Hemmati, M. & Aghazade-Soureh, T. (2018). Simulation of the Effect of Bed Discordance on Flow Pattern at the River Confluence by Flow-3D Model. Irrigation and Drainage, 11(5), 785-797.
  • Hosseini, S, M. & Abrishami, J. (2018). OpenChannel Hydraulics. 35th Edition: Imam Reza International University, 613 pages (In Persian).
  • Karami moghadam, M., Keshavarz, A. & Sabzevar, T. (2019). The Effect of Diversion Flow, Intake Inlet Shape, Topography and Bed Roughness on the Flow Separation Dimensions and Shear Stress at the Lateral Intake. Irrigation and Drainage Structures Engineering Research, 73(19), 113-126. (In Persian).
  • Khosravinia, P., Hosseini, S.H. & Hosseinzadeh Dalir, A. (2018). Numerical analyzing of flow in open channel junction with effect of side slope of channel. Irrigation and Water Engineering, 10(1), 1-16. Doi: 10.22125/iwe.2019.95871 (In Persian).
  • Kwanza, J.K., Kinyanjui, M. & Nkoroi, J.M. (2007). Modelling fluid flow in rectangular and trapezoidal open channels. Advances and Applications in Fluid Mechanics, 2(2), 149- 158.
  • Masjedi, A. & Taeedi, A. (2011). Experimental Investigations of Effect Intake Angle on Discharge in Lateral Intakes in 180 Degree Bend. World Applied Sciences Journal, 15(10), 1442-1444
  • Musavi Jahromi, S.M., & Goudarzizadeh, R. (2011). Numerical Simulation of 3D Flow Pattern at Open-Channel Junctions. Irrigation Sciences and Engineering, 34(2), 61-70 (In Persian).
  • Nikpour, M. & Khosravinia, P. (2018). Numerical Simulation of Side Slope Effect of Main Channel Wall on Flow Behavior in Open Channels Junction. Irrigation and Drainage, 11(6), 1024-1037. (In persian).
  • Raeisi Dehkordi, M. (2022). Description of types of pollution in water resources and protection of water resources, New Approaches in Civil Engineering, 6(1), 42- 52. Doi: 10.30469/jnace.2022.154373 (In Persian).
  • Ramamurthy, A.S., Carballada, L.B. & Tran, D.M. (1988). Combining Open Channel Flow at Right Angled Junctions. Journal of hydraulic engineering, 114(12), 1449-1460.
  • Tabesh, M. (2018). Advanced Modeling of Water Distribution Networks. 4th Edition: University of Tehran Press, 585 pages.
  • Taylor, E. (1944). Flow Characteristics at Rectangular Open-Channel Junctions. Journal of hydraulic engineering, 10(6), 893- 902.
  • Thiong’o, J.W. (2011). Investigations of fluid flows in open rectangular and triangular channels. Master’s thesis, Jomo Kenyatta University of Agriculture and Technology, Juja, Kenya.
  • Weber, L.J., Schumate, E.D. & Mawer, N. (2001). Experiments on Flow at a 90° Open-Channel Junction. Journal of hydraulic engineering, 127(5), 340-350.

PDF 원문 보기
Estimating maximum initial wave amplitude of subaerial landslide tsunamis: A three-dimensional modelling approach

Estimating maximum initial wave amplitude of subaerial landslide tsunamis: A three-dimensional modelling approach

해저 산사태 쓰나미의 최대 초기 파동 진폭 추정: 3차원 모델링 접근법

Ramtin Sabeti a, Mohammad Heidarzadeh ab

aDepartment of Architecture and Civil Engineering, University of Bath, Bath BA27AY, UK
bHydroCoast Consulting Engineers Ltd, Bath, UK

https://doi.org/10.1016/j.ocemod.2024.102360

Highlights

  • •Landslide travel distance is considered for the first time in a predictive equation.
  • •Predictive equation derived from databases using 3D physical and numerical modeling.
  • •The equation was successfully tested on the 2018 Anak Krakatau tsunami event.
  • •The developed equation using three-dimensional data exhibits a 91 % fitting quality.

Abstract

Landslide tsunamis, responsible for thousands of deaths and significant damage in recent years, necessitate the allocation of sufficient time and resources for studying these extreme natural hazards. This study offers a step change in the field by conducting a large number of three-dimensional numerical experiments, validated by physical tests, to develop a predictive equation for the maximum initial amplitude of tsunamis generated by subaerial landslides. We first conducted a few 3D physical experiments in a wave basin which were then applied for the validation of a 3D numerical model based on the Flow3D-HYDRO package. Consequently, we delivered 100 simulations using the validated model by varying parameters such as landslide volume, water depth, slope angle and travel distance. This large database was subsequently employed to develop a predictive equation for the maximum initial tsunami amplitude. For the first time, we considered travel distance as an independent parameter for developing the predictive equation, which can significantly improve the predication accuracy. The predictive equation was tested for the case of the 2018 Anak Krakatau subaerial landslide tsunami and produced satisfactory results.

Keywords

Tsunami, Subaerial landslide, Physical modelling, Numerical simulation, FLOW-3D HYDRO

1. Introduction and literature review

The Anak Krakatau landslide tsunami on 22nd December 2018 was a stark reminder of the dangers posed by subaerial landslide tsunamis (Ren et al., 2020; Mulia et al. 2020a; Borrero et al., 2020; Heidarzadeh et al., 2020; Grilli et al., 2021). The collapse of the volcano’s southwest side into the ocean triggered a tsunami that struck the Sunda Strait, leading to approximately 450 fatalities (Syamsidik et al., 2020; Mulia et al., 2020b) (Fig. 1). As shown in Fig. 1, landslide tsunamis (both submarine and subaerial) have been responsible for thousands of deaths and significant damage to coastal communities worldwide. These incidents underscored the critical need for advanced research into landslide-generated waves to aid in hazard prediction and mitigation. This is further emphasized by recent events such as the 28th of November 2020 landslide tsunami in the southern coast mountains of British Columbia (Canada), where an 18 million m3 rockslide generated a massive tsunami, with over 100 m wave run-up, causing significant environmental and infrastructural damage (Geertsema et al., 2022).

Fig 1

Physical modelling and numerical simulation are crucial tools in the study of landslide-induced waves due to their ability to replicate and analyse the complex dynamics of landslide events (Kim et al., 2020). In two-dimensional (2D) modelling, the discrepancy between dimensions can lead to an artificial overestimation of wave amplification (e.g., Heller and Spinneken, 2015). This limitation is overcome with 3D modelling, which enables the scaled-down representation of landslide-generated waves while avoiding the simplifications inherent in 2D approaches (Erosi et al., 2019). Another advantage of 3D modelling in studying landslide-generated waves is its ability to accurately depict the complex dynamics of wave propagation, including lateral and radial spreading from the slide impact zone, a feature unattainable with 2D models (Heller and Spinneken, 2015).

Physical experiments in tsunami research, as presented by authors such as Romano et al. (2020), McFall and Fritz (2016), and Heller and Spinneken (2015), have supported 3D modelling works through validation and calibration of the numerical models to capture the complexities of wave generation and propagation. Numerical modelling has increasingly complemented experimental approach in tsunami research due to the latter’s time and resource-intensive nature, particularly for 3D models (Li et al., 2019; Kim et al., 2021). Various numerical approaches have been employed, from Eulerian and Lagrangian frameworks to depth-averaged and Navier–Stokes models, enhancing our understanding of tsunami dynamics (Si et al., 2018; Grilli et al., 2019; Heidarzadeh et al., 2017, 2020; Iorio et al., 2021; Zhang et al., 2021; Kirby et al., 2022; Wang et al., 2021, 2022; Hu et al., 2022). The sophisticated numerical techniques, including the Particle Finite Element Method and the Immersed Boundary Method, have also shown promising results in modelling highly dynamic landslide scenarios (Mulligan et al., 2020; Chen et al., 2020). Among these methods and techniques, FLOW-3D HYDRO stands out in simulating landslide-generated tsunami waves due to its sophisticated technical features such as offering Tru Volume of Fluid (VOF) method for precise free surface tracking (e.g., Sabeti and Heidarzadeh 2022a). TruVOF distinguishes itself through a split Lagrangian approach, adeptly reducing cumulative volume errors in wave simulations by dynamically updating cell volume fractions and areas with each time step. Its intelligent adaptation of time step size ensures precise capture of evolving free surfaces, offering unparalleled accuracy in modelling complex fluid interfaces and behaviour (Flow Science, 2023).

Predictive equations play a crucial role in assessing the potential hazards associated with landslide-generated tsunami waves due to their ability to provide risk assessment and warnings. These equations can offer swift and reasonable evaluations of potential tsunami impacts in the absence of detailed numerical simulations, which can be time-consuming and expensive to produce. Among multiple factors and parameters within a landslide tsunami generation, the initial maximum wave amplitude (Fig. 1) stands out due to its critical role. While it is most likely that the initial wave generated by a landslide will have the highest amplitude, it is crucial to clarify that the term “initial maximum wave amplitude” refers to the highest amplitude within the first set of impulse waves. This parameter is essential in determining the tsunami’s impact severity, with higher amplitudes signalling a greater destructive potential (Sabeti and Heidarzadeh 2022a). Additionally, it plays a significant role in tsunami modelling, aiding in the prediction of wave propagation and the assessment of potential impacts.

In this study, we initially validate the FLOW-3D HYDRO model through a series of physical experiments conducted in a 3D wave tank at University of Bath (UK). Upon confirmation of the model’s accuracy, we use it to systematically vary parameters namely landslide volume, water depth, slope angle, and travel distance, creating an extensive database. Alongside this, we perform a sensitivity analysis on these variables to discern their impacts on the initial maximum wave amplitude. The generated database was consequently applied to derive a non-dimensional predictive equation aimed at estimating the initial maximum wave amplitude in real-world landslide tsunami events.

Two innovations of this study are: (i) The predictive equation of this study is based on a large number of 3D experiments whereas most of the previous equations were based on 2D results, and (ii) For the first time, the travel distance is included in the predictive equation as an independent parameter. To evaluate the performance of our predictive equation, we applied it to a previous real-world subaerial landslide tsunami, i.e., the Anak Krakatau 2018 event. Furthermore, we compare the performance of our predictive equation with other existing equations.

2. Data and methods

The methodology applied in this research is a combination of physical and numerical modelling. Limited physical modelling was performed in a 3D wave basin at the University of Bath (UK) to provide data for calibration and validation of the numerical model. After calibration and validation, the numerical model was employed to model a large number of landslide tsunami scenarios which allowed us to develop a database for deriving a predictive equation.

2.1. Physical experiments

To validate our numerical model, we conducted a series of physical experiments including two sets in a 3D wave basin at University of Bath, measuring 2.50 m in length (WL), 2.60 m in width (WW), and 0.60 m in height (WH) (Fig. 2a). Conducting two distinct sets of experiments (Table 1), each with different setups (travel distance, location, and water depth), provided a robust framework for validation of the numerical model. For wave measurement, we employed a twin wire wave gauge from HR Wallingford (https://equipit.hrwallingford.com). In these experiments, we used a concrete prism solid block, the dimensions of which are outlined in Table 2. In our experiments, we employed a concrete prism solid block with a density of 2600 kg/m3, chosen for its similarity to the natural density of landslides, akin to those observed with the 2018 Anak Krakatau tsunami, where the landslide composition is predominantly solid rather than granular. The block’s form has also been endorsed in prior studies (Watts, 1998; Najafi-Jilani and Ataie-Ashtiani, 2008) as a suitable surrogate for modelling landslide-induced waves. A key aspect of our methodology was addressing scale effects, following the guidelines proposed by Heller et al. (2008) as it is described in Table 1. To enhance the reliability and accuracy of our experimental data, we conducted each physical experiment three times which revealed all three experimental waveforms were identical. This repetition was aimed at minimizing potential errors and inconsistencies in laboratory measurements.

Fig 2

Table 1. The locations and other information of the laboratory setups for making landslide-generated waves in the physical wave basin. This table details the specific parameters for each setup, including slope range (α), slide volume (V), kinematic viscosity (ν), water depth (h), travel distance (D), surface tension coefficient of water (σ), Reynolds number (R), Weber number (W), and the precise coordinates of the wave gauges (WG).

Labα(°)V (m³)h (m)D (m)WG’s Location(ν) (m²/s)(σ) (N/m)Acceptable range for avoiding scale effects*Observed values of W and R ⁎⁎
Lab 1452.60 × 10−30.2470.070X1=1.090 m1.01 × 10−60.073R > 3.0 × 105R1 = 3.80 × 105
Y1=1.210 m
W1 = 8.19 × 105
Z1=0.050mW >5.0 × 103
Lab 2452.60 × 10−30.2460.045X2=1.030 m1.01 × 10−60.073R2 = 3.78 × 105
Y2=1.210 mW2 = 8.13 × 105
Z2=0.050 m

⁎

The acceptable ranges for avoiding scale effects are based on the study by Heller et al. (2008).⁎⁎

The Reynolds number (R) is given by g0.5h1.5/ν, with ν denoting the kinematic viscosity. The Weber number (W) is W = ρgh2/σ, where σ represents surface tension coefficient and ρ = 1000kg/m3 is the density of water. In our experiments, conducted at a water temperature of approximately 20 °C, the kinematic viscosity (ν) and the surface tension coefficient of water (σ) are 1.01 × 10−6 m²/s and 0.073 N/m, respectively (Kestin et al., 1978).

Table 2. Specifications of the solid block used in physical experiments for generating subaerial landslides in the laboratory.

Solid-block attributesProperty metricsGeometric shape
Slide width (bs)0.26 mImage, table 2
Slide length (ls)0.20 m
Slide thickness (s)0.10 m
Slide volume (V)2.60 × 10−3 m3
Specific gravity, (γs)2.60
Slide weight (ms)6.86 kg

2.2. Numerical simulations applying FLOW-3D hydro

The detailed theoretical framework encompassing the governing equations, the computational methodologies employed, and the specific techniques used for tracking the water surface in these simulations are thoroughly detailed in the study by Sabeti et al. (2024). Here, we briefly explain some of the numerical details. We defined a uniform mesh for our flow domain, carefully crafted with a fine spatial resolution of 0.005 m (i.e., grid size). The dimensions of the numerical model directly matched those of our wave basin used in the physical experiment, being 2.60 m wide, 0.60 m deep, and 2.50 m long (Fig. 2). This design ensures comprehensive coverage of the study area. The output intervals of the numerical model are set at 0.02 s. This timing is consistent with the sampling rates of wave gauges used in laboratory settings. The friction coefficient in the FLOW-3D HYDRO is designated as 0.45. This value corresponds to the Coulombic friction measurements obtained in the laboratory, ensuring that the simulation accurately reflects real-world physical interactions.

In order to simulate the landslide motion, we applied coupled motion objects in FLOW-3D-HYDRO where the dynamics are predominantly driven by gravity and surface friction. This methodology stands in contrast to other models that necessitate explicit inputs of force and torque. This approach ensures that the simulation more accurately reflects the natural movement of landslides, which is heavily reliant on gravitational force and the interaction between sliding surfaces. The stability of the numerical simulations is governed by the Courant Number criterion (Courant et al., 1928), which dictates the maximum time step (Δt) for a given mesh size (Δx) and flow speed (U). According to Courant et al. (1928), this number is required to stay below one to ensure stability of numerical simulations. In our simulations, the Courant number is always maintained below one.

In alignment with the parameters of physical experiments, we set the fluid within the mesh to water, characterized by a density of 1000 kg/m³ at a temperature of 20 °C. Furthermore, we defined the top, front, and back surfaces of the mesh as symmetry planes. The remaining surfaces are designated as wall types, incorporating no-slip conditions to accurately simulate the interaction between the fluid and the boundaries. In terms of selection of an appropriate turbulence model, we selected the k–ω model that showed a better performance than other turbulence methods (e.g., Renormalization-Group) in a previous study (Sabeti et al., 2024). The simulations are conducted using a PC Intel® Core™ i7-10510U CPU with a frequency of 1.80 GHz, and a 16 GB RAM. On this PC, completion of a 3-s simulation required approximately 12.5 h.

2.3. Validation

The FLOW-3D HYDRO numerical model was validated using the two physical experiments (Fig. 3) outlined in Table 1. The level of agreement between observations (Oi) and simulations (Si) is examined using the following equation:(1)�=|��−����|×100where ε represents the mismatch error, Oi denotes the observed laboratory values, and Si represents the simulated values from the FLOW-3D HYDRO model. The results of this validation process revealed that our model could replicate the waves generated in the physical experiments with a reasonable degree of mismatch (ε): 14 % for Lab 1 and 8 % for Lab 2 experiments, respectively (Fig. 3). These values indicate that while the model is not perfect, it provides a sufficiently close approximation of the real-world phenomena.

Fig 3

In terms of mesh efficiency, we varied the mesh size to study sensitivity of the numerical results to mesh size. First, by halving the mesh size and then by doubling it, we repeated the modelling by keeping other parameters unchanged. This analysis guided that a mesh size of ∆x = 0.005 m is the most effective for the setup of this study. The total number of computational cells applying mesh size of 0.005 m is 9.269 × 106.

2.4. The dataset

The validated numerical model was employed to conduct 100 simulations, incorporating variations in four key landslide parameters namely water depth, slope angle, slide volume, and travel distance. This methodical approach was essential for a thorough sensitivity analysis of these variables, and for the creation of a detailed database to develop a predictive equation for maximum initial tsunami amplitude. Within the model, 15 distinct slide volumes were established, ranging from 0.10 × 10−3 m3 to 6.25 × 10−3 m3 (Table 3). The slope angle varied between 35° and 55°, and water depth ranged from 0.24 m to 0.27 m. The travel distance of the landslides was varied, spanning from 0.04 m to 0.07 m. Detailed configurations of each simulation, along with the maximum initial wave amplitudes and dominant wave periods are provided in Table 4.

Table 3. Geometrical information of the 15 solid blocks used in numerical modelling for generating landslide tsunamis. Parameters are: ls, slide length; bs, slide width; s, slide thickness; γs, specific gravity; and V, slide volume.

Solid blockls (m)bs (m)s (m)V (m3)γs
Block-10.3100.2600.1556.25 × 10−32.60
Block-20.3000.2600.1505.85 × 10−32.60
Block-30.2800.2600.1405.10 × 10−32.60
Block-40.2600.2600.1304.39 × 10−32.60
Block-50.2400.2600.1203.74 × 10−32.60
Block-60.2200.2600.1103.15 × 10−32.60
Block-70.2000.2600.1002.60 × 10−32.60
Block-80.1800.2600.0902.11 × 10−32.60
Block-90.1600.2600.0801.66 × 10−32.60
Block-100.1400.2600.0701.27 × 10−32.60
Block-110.1200.2600.0600.93 × 10−32.60
Block-120.1000.2600.0500.65 × 10−32.60
Block-130.0800.2600.0400.41 × 10−32.60
Block-140.0600.2600.0300.23 × 10−32.60
Block-150.0400.2600.0200.10 × 10−32.60

Table 4. The numerical simulation for the 100 tests performed in this study for subaerial solid-block landslide-generated waves. Parameters are aM, maximum wave amplitude; α, slope angle; h, water depth; D, travel distance; and T, dominant wave period. The location of the wave gauge is X=1.030 m, Y=1.210 m, and Z=0.050 m. The properties of various solid blocks are presented in Table 3.

Test-Block Noα (°)h (m)D (m)T(s)aM (m)
1Block-7450.2460.0290.5100.0153
2Block-7450.2460.0300.5050.0154
3Block-7450.2460.0310.5050.0156
4Block-7450.2460.0320.5050.0158
5Block-7450.2460.0330.5050.0159
6Block-7450.2460.0340.5050.0160
7Block-7450.2460.0350.5050.0162
8Block-7450.2460.0360.5050.0166
9Block-7450.2460.0370.5050.0167
10Block-7450.2460.0380.5050.0172
11Block-7450.2460.0390.5050.0178
12Block-7450.2460.0400.5050.0179
13Block-7450.2460.0410.5050.0181
14Block-7450.2460.0420.5050.0183
15Block-7450.2460.0430.5050.0190
16Block-7450.2460.0440.5050.0197
17Block-7450.2460.0450.5050.0199
18Block-7450.2460.0460.5050.0201
19Block-7450.2460.0470.5050.0191
20Block-7450.2460.0480.5050.0217
21Block-7450.2460.0490.5050.0220
22Block-7450.2460.0500.5050.0226
23Block-7450.2460.0510.5050.0236
24Block-7450.2460.0520.5050.0239
25Block-7450.2460.0530.5100.0240
26Block-7450.2460.0540.5050.0241
27Block-7450.2460.0550.5050.0246
28Block-7450.2460.0560.5050.0247
29Block-7450.2460.0570.5050.0248
30Block-7450.2460.0580.5050.0249
31Block-7450.2460.0590.5050.0251
32Block-7450.2460.0600.5050.0257
33Block-1450.2460.0450.5050.0319
34Block-2450.2460.0450.5050.0294
35Block-3450.2460.0450.5050.0282
36Block-4450.2460.0450.5050.0262
37Block-5450.2460.0450.5050.0243
38Block-6450.2460.0450.5050.0223
39Block-7450.2460.0450.5050.0196
40Block-8450.2460.0450.5050.0197
41Block-9450.2460.0450.5050.0198
42Block-10450.2460.0450.5050.0184
43Block-11450.2460.0450.5050.0173
44Block-12450.2460.0450.5050.0165
45Block-13450.2460.0450.4040.0153
46Block-14450.2460.0450.4040.0124
47Block-15450.2460.0450.5050.0066
48Block-7450.2020.0450.4040.0220
49Block-7450.2040.0450.4040.0219
50Block-7450.2060.0450.4040.0218
51Block-7450.2080.0450.4040.0217
52Block-7450.2100.0450.4040.0216
53Block-7450.2120.0450.4040.0215
54Block-7450.2140.0450.5050.0214
55Block-7450.2160.0450.5050.0214
56Block-7450.2180.0450.5050.0213
57Block-7450.2200.0450.5050.0212
58Block-7450.2220.0450.5050.0211
59Block-7450.2240.0450.5050.0208
60Block-7450.2260.0450.5050.0203
61Block-7450.2280.0450.5050.0202
62Block-7450.2300.0450.5050.0201
63Block-7450.2320.0450.5050.0201
64Block-7450.2340.0450.5050.0200
65Block-7450.2360.0450.5050.0199
66Block-7450.2380.0450.4040.0196
67Block-7450.2400.0450.4040.0194
68Block-7450.2420.0450.4040.0193
69Block-7450.2440.0450.4040.0192
70Block-7450.2460.0450.5050.0190
71Block-7450.2480.0450.5050.0189
72Block-7450.2500.0450.5050.0187
73Block-7450.2520.0450.5050.0187
74Block-7450.2540.0450.5050.0186
75Block-7450.2560.0450.5050.0184
76Block-7450.2580.0450.5050.0182
77Block-7450.2590.0450.5050.0183
78Block-7450.2600.0450.5050.0191
79Block-7450.2610.0450.5050.0192
80Block-7450.2620.0450.5050.0194
81Block-7450.2630.0450.5050.0195
82Block-7450.2640.0450.5050.0195
83Block-7450.2650.0450.5050.0197
84Block-7450.2660.0450.5050.0197
85Block-7450.2670.0450.5050.0198
86Block-7450.2700.0450.5050.0199
87Block-7300.2460.0450.5050.0101
88Block-7350.2460.0450.5050.0107
89Block-7360.2460.0450.5050.0111
90Block-7370.2460.0450.5050.0116
91Block-7380.2460.0450.5050.0117
92Block-7390.2460.0450.5050.0119
93Block-7400.2460.0450.5050.0121
94Block-7410.2460.0450.5050.0127
95Block-7420.2460.0450.4040.0154
96Block-7430.2460.0450.4040.0157
97Block-7440.2460.0450.4040.0162
98Block-7450.2460.0450.5050.0197
99Block-7500.2460.0450.5050.0221
100Block-7550.2460.0450.5050.0233

In all these 100 simulations, the wave gauge was consistently positioned at coordinates X=1.09 m, Y=1.21 m, and Z=0.05 m. The dominant wave period for each simulation was determined using the Fast Fourier Transform (FFT) function in MATLAB (MathWorks, 2023). Furthermore, the classification of wave types was carried out using a wave categorization graph according to Sorensen (2010), as shown in Fig. 4a. The results indicate that the majority of the simulated waves are on the border between intermediate and deep-water waves, and they are categorized as Stokes waves (Fig. 4a). Four sample waveforms from our 100 numerical experiments are provided in Fig. 4b.

Fig 4

The dataset in Table 4 was used to derive a new predictive equation that incorporates travel distance for the first time to estimate the initial maximum tsunami amplitude. In developing this equation, a genetic algorithm optimization technique was implemented using MATLAB (MathWorks 2023). This advanced approach entailed the use of genetic algorithms (GAs), an evolutionary algorithm type inspired by natural selection processes (MathWorks, 2023). This technique is iterative, involving selection, crossover, and mutation processes to evolve solutions over several generations. The goal was to identify the optimal coefficients and powers for each landslide parameter in the predictive equation, ensuring a robust and reliable model for estimating maximum wave amplitudes. Genetic Algorithms excel at optimizing complex models by navigating through extensive combinations of coefficients and exponents. GAs effectively identify highly suitable solutions for the non-linear and complex relationships between inputs (e.g., slide volume, slope angle, travel distance, water depth) and the output (i.e., maximum initial wave amplitude, aM). MATLAB’s computational environment enhances this process, providing robust tools for GA to adapt and evolve solutions iteratively, ensuring the precision of the predictive model (Onnen et al., 1997). This approach leverages MATLAB’s capabilities to fine-tune parameters dynamically, achieving an optimal equation that accurately estimates aM. It is important to highlight that the nondimensionalized version of this dataset is employed to develop a predictive equation which enables the equation to reproduce the maximum initial wave amplitude (aM) for various subaerial landslide cases, independent of their dimensional differences (e.g., Heler and Hager 2014; Heller and Spinneken 2015; Sabeti and Heidarzadeh 2022b). For this nondimensionalization, we employed the water depth (h) to nondimensionalize the slide volume (V/h3) and travel distance (D/h). The slide thickness (s) was applied to nondimensionalize the water depth (h/s).

2.5. Landslide velocity

In discussing the critical role of landslide velocity for simulating landslide-generated waves, we focus on the mechanisms of landslide motion and the techniques used to record landslide velocity in our simulations (Fig. 5). Also, we examine how these methods were applied in two distinct scenarios: Lab 1 and Lab 2 (see Table 1 for their details). Regarding the process of landslide movement, a slide starts from a stationary state, gaining momentum under the influence of gravity and this acceleration continues until the landslide collides with water, leading to a significant reduction in its speed before eventually coming to a stop (Fig. 5) (e.g., Panizzo et al. 2005).

Fig 5

To measure the landslide’s velocity in our simulations, we attached a probe at the centre of the slide, which supplied a time series of the velocity data. The slide’s velocity (vs) peaks at the moment it enters the water (Fig. 5), a point referred to as the impact time (tImp). Following this initial impact, the slides continue their underwater movement, eventually coming to a complete halt (tStop). Given the results in Fig. 5, it can be seen that Lab 1, with its longer travel distance (0.070 m), exhibits a higher peak velocity of 1.89 m/s. This increase in velocity is attributed to the extended travel distance allowing more time for the slide to accelerate under gravity. Whereas Lab 2, featuring a shorter travel distance (0.045 m), records a lower peak velocity of 1.78 m/s. This difference underscores how travel distance significantly influences the dynamics of landslide motion. After reaching the peak, both profiles show a sharp decrease in velocity, marking the transition to submarine motion until the slides come to a complete stop (tStop). There are noticeable differences observable in Fig. 5 between the Lab-1 and Lab-2 simulations, including the peaks at 0.3 s . These variations might stem from the placement of the wave gauge, which differs slightly in each scenario, as well as the water depth’s minor discrepancies and, the travel distance.

2.6. Effect of air entrainment

In this section we examine whether it is required to consider air entrainment for our modelling or not as the FLOW-3D HYDRO package is capable of modelling air entrainment. The process of air entrainment in water during a landslide tsunami and its subsequent transport involve two key components: the quantification of air entrainment at the water surface, and the simulation of the air’s transport within the fluid (Hirt, 2003). FLOW-3D HYDRO employs the air entrainment model to compute the volume of air entrained at the water’s surface utilizing three approaches: a constant density model, a variable density model accounting for bulking, and a buoyancy model that adds the Drift-FLUX mechanism to variable density conditions (Flow Science, 2023). The calculation of the entrainment rate is based on the following equation:(2)�������=������[2(��−�����−2�/���)]1/2where parameters are: Vair, volume of air; Cair, entrainment rate coefficient; As, surface area of fluid; ρ, fluid density; k, turbulent kinetic energy; gn, gravity normal to surface; Lt, turbulent length scale; and σ, surface tension coefficient. The value of k is directly computed from the Reynolds-averaged Navier-Stokes (RANS) (k–w) calculations in our model.

In this study, we selected the variable density + Drift-FLUX model, which effectively captures the dynamics of phase separation and automatically activates the constant density and variable density models. This method simplifies the air-water mixture, treating it as a single, homogeneous fluid within each computational cell. For the phase volume fractions f1and f2​, the velocities are expressed in terms of the mixture and relative velocities, denoted as u and ur, respectively, as follows:(3)��1��+�.(�1�)=��1��+�.(�1�)−�.(�1�2��)=0(4)��2��+�.(�2�)=��2��+�.(�2�)−�.(�1�2��)=0

The outcomes from this simulation are displayed in Fig. 6, which indicates that the influence of air entrainment on the generated wave amplitude is approximately 2 %. A value of 0.02 for the entrained air volume fraction means that, in the simulated fluid, approximately 2 % of the volume is composed of entrained air. In other words, for every unit volume of the fluid-air mixture at that location, 2 % is air and the remaining 98 % is water. The configuration of Test-17 (Table 4) was employed for this simulation. While the effect of air entrainment is anticipated to be more significant in models of granular landslide-generated waves (Fritz, 2002), in our simulations we opted not to incorporate this module due to its negligible impact on the results.

Fig 6

3. Results

In this section, we begin by presenting a sequence of our 3D simulations capturing different time steps to illustrate the generation process of landslide-generated waves. Subsequently, we derive a new predictive equation to estimate the maximum initial wave amplitude of landslide-generated waves and assess its performance.

3.1. Wave generation and propagation

To demonstrate the wave generation process in our simulation, we reference Test-17 from Table 4, where we employed Block-7 (Tables 3, 4). In this configuration, the slope angle was set to 45°, with a water depth of 0.246 m and a travel distance at 0.045 m (Fig. 7). At 0.220 s, the initial impact of the moving slide on the water is depicted, marking the onset of the wave generation process (Fig. 7a). Disturbances are localized to the immediate area of impact, with the rest of the water surface remaining undisturbed. At this time, a maximum water particle velocity of 1.0 m/s – 1.2 m/s is seen around the impact zone (Fig. 7d). Moving to 0.320 s, the development of the wave becomes apparent as energy transfer from the landslide to the water creates outwardly radiating waves with maximum water particle velocity of up to around 1.6 m/s – 1.8 m/s (Fig. 7b, e). By the time 0.670 s, the wave has fully developed and is propagating away from the impact point exhibiting maximum water particle velocity of up to 2.0 m/s – 2.1 m/s. Concentric wave fronts are visible, moving outwards in all directions, with a colour gradient signifying the highest wave amplitude near the point of landslide entry, diminishing with distance (Fig. 7c, f).

Fig 7

3.2. Influence of landslide parameters on tsunami amplitude

In this section, we investigate the effects of various landslide parameters namely slide volume (V), water depth (h), slipe angle (α) and travel distance (D) on the maximum initial wave amplitude (aM). Fig. 8 presents the outcome of these analyses. According to Fig. 8, the slide volume, slope angle, and travel distance exhibit a direct relationship with the wave amplitude, meaning that as these parameters increase, so does the amplitude. Conversely, water depth is inversely related to the maximum initial wave amplitude, suggesting that the deeper the water depth, the smaller the maximum wave amplitude will be (Fig. 8b).

Fig 8

Fig. 8a highlights the pronounced impact of slide volume on the aM, demonstrating a direct correlation between the two variables. For instance, in the range of slide volumes we modelled (Fig. 8a), The smallest slide volume tested, measuring 0.10 × 10−3 m3, generated a low initial wave amplitude (aM= 0.0066 m) (Table 4). In contrast, the largest volume tested, 6.25 × 10−3 m3, resulted in a significantly higher initial wave amplitude (aM= 0.0319 m) (Table 4). The extremities of these results emphasize the slide volume’s paramount impact on wave amplitude, further elucidated by their positions as the smallest and largest aM values across all conducted tests (Table 4). This is corroborated by findings from the literature (e.g., Murty, 2003), which align with the observed trend in our simulations.

The slope angle’s influence on aM was smooth. A steady increase of wave amplitude was observed as the slope angle increased (Fig. 8c). In examining travel distance, an anomaly was identified. At a travel distance of 0.047 m, there was an unexpected dip in aM, which deviates from the general increasing trend associated with longer travel distances. This singular instance could potentially be attributed to a numerical error. Beyond this point, the expected pattern of increasing aM with longer travel distances resumes, suggesting that the anomaly at 0.047 m is an outlier in an otherwise consistent trend, and thus this single data point was overlooked while deriving the predictive equation. Regarding the inverse relationship between water depth and wave amplitude, our result (Fig. 8b) is consistent with previous reports by Fritz et al. (2003), (2004), and Watts et al. (2005).

The insights from Fig. 8 informed the architecture of the predictive equation in the next Section, with slide volume, travel distance, and slope angle being multiplicatively linked to wave amplitude underscoring their direct correlations with wave amplitude. Conversely, water depth is incorporated as a divisor, representing its inverse relationship with wave amplitude. This structure encapsulates the dynamics between the landslide parameters and their influence on the maximum initial wave amplitude as discussed in more detail in the next Section.

3.3. Predictive equation

Building on our sensitivity analysis of landslide parameters, as detailed in Section 3.2, and utilizing our nondimensional dataset, we have derived a new predictive equation as follows:(5)��/ℎ=0.015(tan�)0.10(�ℎ3)0.90(�ℎ)0.10(ℎ�)−0.11where, V is sliding volume, h is water depth, α is slope angle, and s is landslide thickness. It is important to note that this equation is valid only for subaerial solid-block landslide tsunamis as all our experiments were for this type of waves. The performance of this equation in predicting simulation data is demonstrated by the satisfactory alignment of data points around a 45° line, indicating its accuracy and reliability with regard to the experimental dataset (Fig. 9). The quality of fit between the dataset and Eq. (5) is 91 % indicating that Eq. (5) represents the dataset very well. Table 5 presents Eq. (5) alongside four other similar equations previously published. Two significant distinctions between our Eq. (5) and these others are: (i) Eq. (5) is derived from 3D experiments, whereas the other four equations are based on 2D experiments. (ii) Unlike the other equations, our Eq. (5) incorporates travel distance as an independent parameter.

Fig 9

Table 5. Performance comparison among our newly-developed equation and existing equations for estimating the maximum initial amplitude (aM) of the 2018 Anak Krakatau subaerial landslide tsunami. Parameters: aM, initial maximum wave amplitude; h, water depth; vs, landslide velocity; V, slide volume; bs, slide width; ls, slide length; s, slide thickness; α, slope angle; and ����, volume of the final immersed landslide. We considered ����= V as the slide volume.

EventPredictive equationsAuthor (year)Observed aM (m) ⁎⁎Calculated aM (m)Error, ε (%) ⁎⁎⁎⁎
2018 Anak Krakatau tsunami (Subaerial landslide) *��/ℎ=1.32���ℎNoda (1970)1341340
��/ℎ=0.667(0.5(���ℎ)2)0.334(���)0.754(���)0.506(�ℎ)1.631Bolin et al. (2014) ⁎⁎⁎13459424334
��/ℎ=0.25(������ℎ2)0.8Robbe-Saule et al. (2021)1343177
��/ℎ=0.4545(tan�)0.062(�ℎ3)0.296(ℎ�)−0.235Sabeti and Heidarzadeh (2022b)1341266
��/ℎ=0.015(tan�)0.10(�ℎ3)0.911(�ℎ)0.10(ℎ�)−0.11This study1341302.9

⁎

Geometrical and kinematic parameters of the 2018 Anak Krakatau subaerial landslide based on Heidarzadeh et al. (2020), Grilli et al. (2019) and Grilli et al. (2021): V=2.11 × 107 m3, h= 50 m; s= 114 m; α= 45°; ls=1250 m; bs= 2700 m; vs=44.9 m/s; D= 2500 m; aM= 100 m −150 m.⁎⁎

aM= An average value of aM = 134 m is considered in this study.⁎⁎⁎

The equation of Bolin et al. (2014) is based on the reformatted one reported by Lindstrøm (2016).⁎⁎⁎⁎

Error is calculated using Eq. (1), where the calculated aM is assumed as the simulated value.

Additionally, we evaluated the performance of this equation using the real-world data from the 2018 Anak Krakatau subaerial landslide tsunami. Based on previous studies (Heidarzadeh et al., 2020; Grilli et al., 2019, 2021), we were able to provide a list of parameters for the subaerial landslide and associated tsunami for the 2018 Anak Krakatau event (see footnote of Table 5). We note that the data of the 2018 Anak Krakatau event was not used while deriving Eq. (5). The results indicate that Eq. (5) predicts the initial amplitude of the 2018 Anak Krakatau tsunami as being 130 m indicating an error of 2.9 % compared to the reported average amplitude of 134 m for this event. This performance indicates an improvement compared to the previous equation reported by Sabeti and Heidarzadeh (2022a) (Table 5). In contrast, the equations from Robbe-Saule et al. (2021) and Bolin et al. (2014) demonstrate higher discrepancies of 4200 % and 77 %, respectively (Table 5). Although Noda’s (1970) equation reproduces the tsunami amplitude of 134 m accurately (Table 5), it is crucial to consider its limitations, notably not accounting for parameters such as slope angle and travel distance.

It is essential to recognize that both travel distance and slope angle significantly affect wave amplitude. In our model, captured in Eq. (5), we integrate the slope angle (α) through the tangent function, i.e., tan α. This choice diverges from traditional physical interpretations that often employ the cosine or sine function (e.g., Heller and Hager, 2014; Watts et al., 2003). We opted for the tangent function because it more effectively reflects the direct impact of slope steepness on wave generation, yielding superior estimations compared to conventional methods.

The significance of this study lies in its application of both physical and numerical 3D experiments and the derivation of a predictive equation based on 3D results. Prior research, e.g. Heller et al. (2016), has reported notable discrepancies between 2D and 3D wave amplitudes, highlighting the important role of 3D experiments. It is worth noting that the suitability of applying an equation derived from either 2D or 3D data depends on the specific geometry and characteristics inherent in the problem being addressed. For instance, in the case of a long, narrow dam reservoir, an equation derived from 2D data would likely be more suitable. In such contexts, the primary dynamics of interest such as flow patterns and potential wave propagation are predominantly two-dimensional, occurring along the length and depth of the reservoir. This simplification to 2D for narrow dam reservoirs allows for more accurate modelling of these dynamics.

This study specifically investigates waves initiated by landslides, focusing on those characterized as solid blocks instead of granular flows, with slope angles confined to a range of 25° to 60°. We acknowledge the additional complexities encountered in real-world scenarios, such as dynamic density and velocity of landslides, which could affect the estimations. The developed equation in this study is specifically designed to predict the maximum initial amplitude of tsunamis for the aforementioned specified ranges and types of landslides.

4. Conclusions

Both physical and numerical experiments were undertaken in a 3D wave basin to study solid-block landslide-generated waves and to formulate a predictive equation for their maximum initial wave amplitude. At the beginning, two physical experiments were performed to validate and calibrate a 3D numerical model, which was subsequently utilized to generate 100 experiments by varying different landslide parameters. The generated database was then used to derive a predictive equation for the maximum initial wave amplitude of landslide tsunamis. The main features and outcomes are:

  • •The predictive equation of this study is exclusively derived from 3D data and exhibits a fitting quality of 91 % when applied to the database.
  • •For the first time, landslide travel distance was considered in the predictive equation. This inclusion provides more accuracy and flexibility for applying the equation.
  • •To further evaluate the performance of the predictive equation, it was applied to a real-world subaerial landslide tsunami (i.e., the 2018 Anak Krakatau event) and delivered satisfactory performance.

CRediT authorship contribution statement

Ramtin Sabeti: Conceptualization, Methodology, Validation, Software, Visualization, Writing – review & editing. Mohammad Heidarzadeh: Methodology, Data curation, Software, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding

RS is supported by the Leverhulme Trust Grant No. RPG-2022-306. MH is funded by open funding of State Key Lab of Hydraulics and Mountain River Engineering, Sichuan University, grant number SKHL2101. We acknowledge University of Bath Institutional Open Access Fund. MH is also funded by the Great Britain Sasakawa Foundation grant no. 6217 (awarded in 2023).

Acknowledgements

Authors are sincerely grateful to the laboratory technician team, particularly Mr William Bazeley, at the Faculty of Engineering, University of Bath for their support during the laboratory physical modelling of this research. We appreciate the valuable insights provided by Mr. Brian Fox (Senior CFD Engineer at Flow Science, Inc.) regarding air entrainment modelling in FLOW-3D HYDRO. We acknowledge University of Bath Institutional Open Access Fund.

Data availability

  • All data used in this study are given in the body of the article.

References

PDF 원문 보기
Figure 3 – Free surface views. Bottom left: k-ε RNG model. Bottom right: LES.

Physical Modeling and CFD Comparison: Case Study of a HydroCombined Power Station in Spillway Mode

물리적 모델링 및 CFD 비교: 방수로 모드의 HydroCombined 발전소 사례 연구

Gonzalo Duró, Mariano De Dios, Alfredo López, Sergio O. Liscia

ABSTRACT

This study presents comparisons between the results of a commercial CFD code and physical model measurements. The case study is a hydro-combined power station operating in spillway mode for a given scenario. Two turbulence models and two scales are implemented to identify the capabilities and limitations of each approach and to determine the selection criteria for CFD modeling for this kind of structure. The main flow characteristics are considered for analysis, but the focus is on a fluctuating frequency phenomenon for accurate quantitative comparisons. Acceptable representations of the general hydraulic functioning are found in all approaches, according to physical modeling. The k-ε RNG, and LES models give good representation of the discharge flow, mean water depths, and mean pressures for engineering purposes. The k-ε RNG is not able to characterize fluctuating phenomena at a model scale but does at a prototype scale. The LES is capable of identifying the dominant frequency at both prototype and model scales. A prototype-scale approach is recommended for the numerical modeling to obtain a better representation of fluctuating pressures for both turbulence models, with the complement of physical modeling for the ultimate design of the hydraulic structures.

본 연구에서는 상용 CFD 코드 결과와 물리적 모델 측정 결과를 비교합니다. 사례 연구는 주어진 시나리오에 대해 배수로 모드에서 작동하는 수력 복합 발전소입니다.

각 접근 방식의 기능과 한계를 식별하고 이러한 종류의 구조에 대한 CFD 모델링의 선택 기준을 결정하기 위해 두 개의 난류 모델과 두 개의 스케일이 구현되었습니다. 주요 흐름 특성을 고려하여 분석하지만 정확한 정량적 비교를 위해 변동하는 주파수 현상에 중점을 둡니다.

일반적인 수리학적 기능에 대한 허용 가능한 표현은 물리적 모델링에 따라 모든 접근 방식에서 발견됩니다. k-ε RNG 및 LES 모델은 엔지니어링 목적을 위한 배출 유량, 평균 수심 및 평균 압력을 잘 표현합니다.

k-ε RNG는 모델 규모에서는 변동 현상을 특성화할 수 없지만 프로토타입 규모에서는 특성을 파악합니다. LES는 프로토타입과 모델 규모 모두에서 주요 주파수를 식별할 수 있습니다.

수력학적 구조의 궁극적인 설계를 위한 물리적 모델링을 보완하여 두 난류 모델에 대한 변동하는 압력을 더 잘 표현하기 위해 수치 모델링에 프로토타입 규모 접근 방식이 권장됩니다.

Figure 1 – Physical scale model (left). Upstream flume and point gauge (right)
Figure 1 – Physical scale model (left). Upstream flume and point gauge (right)
Figure 3 – Free surface views. Bottom left: k-ε RNG model. Bottom right: LES.
Figure 3 – Free surface views. Bottom left: k-ε RNG model. Bottom right: LES.
Figure 4 – Water levels: physical model (maximum values) and CFD results (mean values)
Figure 4 – Water levels: physical model (maximum values) and CFD results (mean values)
Figure 5 – Instantaneous pressures [Pa] and velocities [m/s] at model scale (bay center)
Figure 5 – Instantaneous pressures [Pa] and velocities [m/s] at model scale (bay center)

Keywords

CFD validation, hydro-combined, k-ε RNG, LES, pressure spectrum

PDF Download

REFERENCES

ADRIAN R. J. (2007). “Hairpin vortex organization in wall turbulence.” Phys. Fluids 19(4), 041301.
DEWALS B., ARCHAMBEAU P., RULOT F., PIROTTON M. and ERPICUM S. (2013). “Physical and
Numerical Modelling in Low-Head Structures Design.” Proc. International Workshop on Hydraulic
Design of Low-Head Structures, Aachen, Germany, Bundesanstalt für Wasserbau Publ., D.B. BUNG
and S. PAGLIARA Editors, pp.11-30.
GRENANDER, U. (1959). Probability and Statistics: The Harald Cramér Volume. Wiley.
HIRT, C. W. and NICHOLS B. D. (1981). “Volume of fluid (VOF) method for the dynamics of free
boundaries.” Journal of Computational Physics 39(1): 201-225.
JOHNSON M. C. and SAVAGE B. M. (2006). “Physical and numerical comparison of flow over ogee
spillway in the presence of tailwater.” J. Hydraulic Eng. 132(12): 1353–1357.
KHAN L.A., WICKLEIN E.A., RASHID M., EBNER L.L. and RICHARDS N.A. (2004).
“Computational fluid dynamics modeling of turbine intake hydraulics at a hydropower plant.” Journal
of Hydraulic Research, 42:1, 61-69
LAROCQUE L.A., IMRAN J. and CHAUDHRY M. (2013). “3D numerical simulation of partial breach
dam-break flow using the LES and k–ϵ turbulence models.” Jl of Hydraulic Research, 51:2, 145-157
LI S., LAI Y., WEBER L., MATOS SILVA J. and PATEL V.C. (2004). “Validation of a threedimensional numerical model for water-pump intakes.” Journal of Hydraulic Research, 42:3, 282-292
NOVAK P., GUINOT V., JEFFREY A. and REEVE D.E. (2010). “Hydraulic modelling – An
introduction.” Spon Press, London and New York, ISBN 978-0-419-25010-4, 616 pp.

PDF Download
Numerical Investigation of the Local Scour for Tripod Pile Foundation.

Numerical Investigation of the Local Scour for Tripod Pile Foundation.

Hassan, Waqed H.; Fadhe, Zahraa Mohammad; Thiab, Rifqa F.; Mahdi, Karrar

초록

This work investigates numerically a local scour moves in irregular waves around tripods. It is constructed and proven to use the numerical model of the seabed-tripodfluid with an RNG k turbulence model. The present numerical model then examines the flow velocity distribution and scour characteristics. After that, the suggested computational model Flow-3D is a useful tool for analyzing and forecasting the maximum scour development and the flow field in random waves around tripods. The scour values affecting the foundations of the tripod must be studied and calculated, as this phenomenon directly and negatively affects the structure of the structure and its design life. The lower diagonal braces and the main column act as blockages, increasing the flow accelerations underneath them. This increases the number of particles that are moved, which in turn creates strong scouring in the area. The numerical model has a good agreement with the experimental model, with a maximum percentage of error of 10% between the experimental and numerical models. In addition, Based on dimensional analysis parameters, an empirical equation has been devised to forecast scour depth with flow depth, median size ratio, Keulegan-Carpenter (Kc), Froud number flow, and wave velocity that the results obtained in this research at various flow velocities and flow depths demonstrated that the maximum scour depth rate depended on wave height with rising velocities and decreasing particle sizes (d50) and the scour depth attains its steady-current value for Vw < 0.75. As the Froude number rises, the maximum scour depth will be large.

주제어

BUILDING foundations; SURFACE waves (Seismic waves); FLOW velocity; RANDOM fields; DIMENSIONAL analysis; FROUDE number; OCEAN waves

키워드

출판물

Mathematical Modelling of Engineering Problems, 2024, Vol 11, Issue 4, p903

ISSN 2369-0739

저자 소속기관

  • 1 Civil Engineering Department, Faculty of Engineering, University of Warith Al-Anbiyaa, Kerbala 56001, Iraq
  • 2 Civil Engineering Department, Faculty of Engineering, University of Kerbala, Kerbala 56001, Iraq
  • 3 Department of Radiological Techniques, College of Health and Medical Techniques, Al-Zahraa University for Women, Karbala 56100, Iraq
  • 4 Soil Physics and Land Management Group, Wageningen University & Research, Wageningen 6708 PB, Netherlands
Figure (17): Stream Lines Indicating Average Flow Speed in the Model with Various Nose shapes, Measured at Mid-Depth and at the Flow Surface Level, at a Flow Rate of 78 Liters per Second.

Conducting experimental and numerical studies to analyze theimpact of the base nose shape on flow hydraulics in PKW weirusing FLOW-3D

FLOW-3D를 사용하여 PKW 둑의 흐름 수력학에 대한 베이스 노즈 모양의 영향을 분석하기 위한 실험 및 수치 연구 수행

Behshad Mardasi 1
Rasoul Ilkhanipour Zeynali 2
Majid Heydari 3

Abstract

Weirs are essential structures used to manage excess water flow from behind dams to downstream areas. Enhancing discharge efficiency often involves extending the effective length of Piano Key Weirs (PKW) in dams or regulating flow within irrigation and drainage networks. This study employed both numerical and laboratory investigations to assess the impact of different base nose shapes installed beneath the outlet keys and varying Input to output key width ratios (Wi/Wo) on discharges ranging from 5 to 80 liters per second. Furthermore, the study aimed to achieve research objectives and compare the performance of Piano Key Weirs with Ogee Weir. For numerical simulation, the optimal number of cells for meshing was determined, and an appropriate turbulence model was selected. The results indicated that the numerical model accurately simulated the laboratory sample with a high degree of precision. Moreover, the numerical model closely approximated PKW for all parameters Q, H, and Cd compared to the laboratory sample. The findings revealed that in laboratory models with a maximum discharge area of 80 liters per second, the weir with Wi/Wo=1.2 and a flow head value of 285 mm exhibited the lowest value, whereas the weir with Wi/Wo=0.71 and a flow head value of 305 mm showed the highest, attributed to the higher discharge in the input-output ratio. Additionally, as the ratio of flow head to weir height H/P increased, the discharge coefficient Cd decreased. Comparing the flow conditions in weirs with different base nose shapes, it was observed that the weir with a spindle nose shape (PKW1.2S) outperformed the PKW with a flat (PKW1.2), semi-cylindrical (PKW1.2CL) and triangular base nose (PKW1.2TR). The results emphasized that models featuring semi-cylindrical and flat noses exhibited notable flow deviation and abrupt disruption upon impact with the nose. However, this effect was significantly reduced in models equipped with triangular and spindle-shaped noses. Also, the coefficient of discharge in PKW1.2S and PKW1.2TR weirs, compared to the PKW1.20 weir, increased by 27% and 20%, respectively.

웨어는 댐 뒤에서 하류 지역으로의 과도한 물 흐름을 관리하는 데 사용되는 필수 구조물입니다. 배출 효율을 높이는 데에는 댐의 피아노 키 위어(PKW) 유효 길이를 연장하거나 관개 및 배수 네트워크 내 흐름을 조절하는 것이 포함됩니다.

이 연구에서는 콘센트 키 아래에 설치된 다양한 베이스 노즈 모양과 초당 5~80리터 범위의 배출에 대한 다양한 입력 대 출력 키 너비 비율(Wi/Wo)의 영향을 평가하기 위해 수치 및 실험실 조사를 모두 사용했습니다. 또한 본 연구에서는 연구 목적을 달성하고 Piano Key Weir와 Ogee Weir의 성능을 비교하는 것을 목표로 했습니다.

수치 시뮬레이션을 위해 메시 생성을 위한 최적의 셀 수를 결정하고 적절한 난류 모델을 선택했습니다. 결과는 수치 모델이 높은 정밀도로 실험실 샘플을 정확하게 시뮬레이션했음을 나타냅니다. 더욱이, 수치 모델은 실험실 샘플과 비교하여 모든 매개변수 Q, H 및 Cd에 대해 PKW에 매우 근접했습니다.

연구 결과, 최대 배출 면적이 초당 80리터인 실험실 모델에서는 Wi/Wo=1.2, 플로우 헤드 값이 285mm인 웨어가 가장 낮은 값을 나타냈고, Wi/Wo=0.71 및 a인 웨어는 가장 낮은 값을 나타냈습니다. 플로우 헤드 값은 305mm로 가장 높은 것으로 나타났는데, 이는 입출력 비율의 높은 토출량에 기인합니다. 또한, 웨어 높이에 대한 유수두 비율 H/P가 증가함에 따라 유출계수 Cd는 감소하였다.

베이스 노즈 모양이 다른 웨어의 흐름 조건을 비교해 보면, 스핀들 노즈 모양(PKW1.2S)의 웨어가 평면(PKW1.2), 반원통형(PKW1.2CL) 및 삼각형 모양의 PKW보다 성능이 우수한 것으로 관찰되었습니다. 베이스 노즈(PKW1.2TR) 결과는 반원통형 및 편평한 노즈를 특징으로 하는 모델이 노즈에 충격을 가할 때 눈에 띄는 흐름 편차와 급격한 중단을 나타냄을 강조했습니다.

그러나 삼각형 및 방추형 노즈를 장착한 모델에서는 이러한 효과가 크게 감소했습니다. 또한 PKW1.20보에 비해 PKW1.2S보와 PKW1.2TR보의 유출계수는 각각 27%, 20% 증가하였다.

Keywords

Piano Key Weir, Base Nose Shape, Flow Hydraulics, Numerical Model, Triangular
Nose Shape, Flat Nose Shape, Semi-Cylindrical Nose Shape, Spindle Nose Shape

Figure (17): Stream Lines Indicating Average Flow Speed in the Model with Various Nose shapes, Measured at Mid-Depth and at the Flow Surface Level, at a Flow Rate of 78 Liters per Second.
Figure (17): Stream Lines Indicating Average Flow Speed in the Model with Various Nose shapes, Measured at Mid-Depth and at the Flow Surface Level, at a Flow Rate of 78 Liters per Second.

Reference

  1. Chow, V.T. (1959). “Open channel hydraulics.” McGraw-Hill Book Company, New York,
    NY.
  2. Ouamane, A., and Lempérière, F. (2006). “Design of a new economic shape of weir.” Proc.,
    Intl. Symp. on Dams in the Societies of the 21st Century, 463-470, Barcelona, Spain.
  3. Crookston, B. M., Anderson, A., Shearin-Feimster, L., and Tullis, B. P. (2014). “Mitigation
    investigation of flow-induced vibrations at a rehabilitated spillway.” Proc., 5th IAHR Intl.
    Symp. on Hydraulic Structures, Univ. of Queensland Brisbane, Brisbane, Australia.
  4. Machiels, O. (2012). “Experimental study of the hydraulic behaviour of Piano Key Weirs.”
    Ph.D. Dissertation, Faculty of Applied Science, University of Liège, Liège, Belgium.
  5. Blanc, P., and Lempérière, F. (2001). “Labyrinth spillways have a promising future.” Intl. J.
    of Hydropower and Dams, 8(4), 129-131.
  6. Muslu, Y. (2001). “Numerical analysis for lateral weir flow.” J. of Irrigation and Drainage
    Eng., ASCE, 127, 246.
  7. Erpicum, S., Machiels, O., Dewals, B., Pirotton, M., and Archambeau, P. (2012).
    “Numerical and physical hydraulic modeling of Piano Key Weirs.” Proc., ASIA 2012 – 4th
    Intl. Conf. on Water Resources and Renewable Energy Development in Asia, Chiang Mai,
    Thailand.
  8. Tullis, J.P., Amanian, N., and Waldron, D. (1995). “Design of Labyrinth Spillways.” J. of
    Hydraulic Eng., ASCE, 121.
  9. Lux, F.L., and Hinchcliff, D. (1985). “Design and construction of labyrinth spillways.”
    Proc., 15th Intl. Congress on Large Dams, ICOLD, Vol. 4, 249-274, Paris, France.
  10. Erpicum, S., Laugier, F., Ho to Khanh, M., & Pfister, M. (2017). Labyrinth and Piano Key
    Weirs III–PKW 2017. CRC Press, Boca Raton, FL.
  11. Kabiri-Samani, A., and Javaheri, A. (2012). “Discharge coefficient for free and submerged flow over Piano Key weirs.” Hydraulic Research J., 50(1), 114-120.
  12. Hien, T.C., Son, H.T., and Khanh, M.H.T. (2006). “Results of some piano Key weirs
    hydraulic model tests in Vietnam.” Proc., 22nd ICOLD Congress, CIGB/ICOLD,
    Barcelona, Spain.
  13. Laugier, F., Lochu, A., Gille, C., Leite Ribeiro, M., and Boillat, J-L. (2009). “Design and
    construction of a labyrinth PKW spillway at St-Marc Dam.” Hydropower and Dams J.,
    15(5), 100-107.
  14. Cicero, G.M., Menon, J.M., Luck, M., and Pinchard, T. (2011). “Experimental study of side
    and scale effects on hydraulic performances of a Piano Key Weir.” In: Erpicum, S., Laugier,
    F., Boillat, J-L, Pirotton, M., Reverchon, B., and Schleiss, A-J (Eds.), Labyrinth and Piano
    Key Weirs, 167-172, CRC Press, London.
  15. Pralong, J., Vermeulen, J., Blancher, B., Laugier, F., Erpicum, S., Machiels, O., Pirotton,
    M., Boillat, J.L, Leite Ribeiro, M., and Schleiss, A.J. (2011). “A naming convention for the
    piano key weirs geometrical parameters.” In: Erpicum, S., Laugier, F., Boillat, J-L, Pirotton,
    M., Reverchon, B., and Schleiss, A-J (Eds.), Labyrinth and Piano Key Weirs, 271-278,
    CRC Press, London.
  16. Denys, F. J. M., and Basson, G. R. (2018). “Transient hydrodynamics of Piano Key Weirs.”
    Proc., 7th IAHR Intl. Symp. on Hydraulic Structures, ISHS2018, 518-527,
    DigitalCommons@USU, Logan, UT.
  17. Anderson, A., and Tullis, B. P. (2018). “Finite crest length weir nappe oscillation.” J. of
    Hydraulic Eng., ASCE, 144(6), 04018020. https://doi.org/10.1061/(ASCE)HY.1943-
    7900.0001461
  18. Erpicum, S., Laugier, F., Boillat, J.-L., Pirotton, M., Reverchon, B., and Schleiss, A. J.
    (2011). “Labyrinth and Piano Key Weirs–PKW 2011.” CRC Press, Boca Raton, FL.
  19. Aydin, C.M., and Emiroglu, M.E. (2011). “Determination of capacity of labyrinth side weir
    by CFD.” Flow Measurement and Instrumentation, 29, 1-8.
  20. Cicero, G.M., Delisle, J.R., Lefebvre, V., and Vermeulen, J. (2013). “Experimental and
    numerical study of the hydraulic performance of a trapezoidal PKW.” Proc., Intl. Workshop
    on Labyrinths and Piano Key Weirs PKW II 2013, 265-272, CRC Press.
  21. Anderson, R. M. (2011). “Piano Key Weir Head Discharge Relationships.” Master’s Thesis,
    Utah State University, Logan, Utah.
  22. Crookston, B.M., Anderson, R.M., and Tullis, B.P. (2018). “Free-flow discharge estimation
    method for Piano Key weir geometries.” J. of Hydro-environment Research, 19, 160-167
PDF 원문 보기
Fig. 3. Free surface and substrate profiles in all Sp and Ls cases at t = 1 s, t = 3 s, and t = 5 s, arranged left to right (note: the colour contours correspond to the horizontal component of the flow velocity (u), expressed in m/s).

Numerical investigation of dam break flow over erodible beds with diverse substrate level variations

다양한 기질 수준 변화를 갖는 침식성 층 위의 댐 파손 흐름에 대한 수치 조사

Alireza Khoshkonesh1, Blaise Nsom2, Saeid Okhravi3*, Fariba Ahmadi Dehrashid4, Payam Heidarian5,
Silvia DiFrancesco6
1 Department of Geography, School of Social Sciences, History, and Philosophy, Birkbeck University of London, London, UK.
2 Université de Bretagne Occidentale. IRDL/UBO UMR CNRS 6027. Rue de Kergoat, 29285 Brest, France.
3 Institute of Hydrology, Slovak Academy of Sciences, Dúbravská cesta 9, 84104, Bratislava, Slovak Republic.
4Department of Water Science and Engineering, Faculty of Agriculture, Bu-Ali Sina University, 65178-38695, Hamedan, Iran.
5 Department of Civil, Environmental, Architectural Engineering and Mathematics, University of Brescia, 25123 Brescia, Italy.
6Niccol`o Cusano University, via Don C. Gnocchi 3, 00166 Rome, Italy. * Corresponding author. Tel.: +421-944624921. E-mail: saeid.okhravi@savba.sk

Abstract

This study aimed to comprehensively investigate the influence of substrate level difference and material composition on dam break wave evolution over two different erodible beds. Utilizing the Volume of Fluid (VOF) method, we tracked free surface advection and reproduced wave evolution using experimental data from the literature. For model validation, a comprehensive sensitivity analysis encompassed mesh resolution, turbulence simulation methods, and bed load transport equations. The implementation of Large Eddy Simulation (LES), non-equilibrium sediment flux, and van Rijn’s (1984) bed load formula yielded higher accuracy compared to alternative approaches. The findings emphasize the significant effect of substrate level difference and material composition on dam break morphodynamic characteristics. Decreasing substrate level disparity led to reduced flow velocity, wavefront progression, free surface height, substrate erosion, and other pertinent parameters. Initial air entrapment proved substantial at the wavefront, illustrating pronounced air-water interaction along the bottom interface. The Shields parameter experienced a one-third reduction as substrate level difference quadrupled, with the highest near-bed concentration observed at the wavefront. This research provides fresh insights into the complex interplay of factors governing dam break wave propagation and morphological changes, advancing our comprehension of this intricate phenomenon.

이 연구는 두 개의 서로 다른 침식층에 대한 댐 파괴파 진화에 대한 기질 수준 차이와 재료 구성의 영향을 종합적으로 조사하는 것을 목표로 했습니다. VOF(유체량) 방법을 활용하여 자유 표면 이류를 추적하고 문헌의 실험 데이터를 사용하여 파동 진화를 재현했습니다.

모델 검증을 위해 메쉬 해상도, 난류 시뮬레이션 방법 및 침대 하중 전달 방정식을 포함하는 포괄적인 민감도 분석을 수행했습니다. LES(Large Eddy Simulation), 비평형 퇴적물 플럭스 및 van Rijn(1984)의 하상 부하 공식의 구현은 대체 접근 방식에 비해 더 높은 정확도를 산출했습니다.

연구 결과는 댐 붕괴 형태역학적 특성에 대한 기질 수준 차이와 재료 구성의 중요한 영향을 강조합니다. 기판 수준 차이가 감소하면 유속, 파면 진행, 자유 표면 높이, 기판 침식 및 기타 관련 매개변수가 감소했습니다.

초기 공기 포집은 파면에서 상당한 것으로 입증되었으며, 이는 바닥 경계면을 따라 뚜렷한 공기-물 상호 작용을 보여줍니다. 기판 레벨 차이가 4배로 증가함에 따라 Shields 매개변수는 1/3로 감소했으며, 파면에서 가장 높은 베드 근처 농도가 관찰되었습니다.

이 연구는 댐 파괴파 전파와 형태학적 변화를 지배하는 요인들의 복잡한 상호 작용에 대한 새로운 통찰력을 제공하여 이 복잡한 현상에 대한 이해를 향상시킵니다.

Keywords

Dam break; Substrate level difference; Erodible bed; Sediment transport; Computational fluid dynamics CFD.

Fig. 3. Free surface and substrate profiles in all Sp and Ls cases at t = 1 s, t = 3 s, and t = 5 s, arranged left to right (note: the colour contours correspond to the horizontal component of the flow velocity (u), expressed in m/s).
Fig. 3. Free surface and substrate profiles in all Sp and Ls cases at t = 1 s, t = 3 s, and t = 5 s, arranged left to right (note: the colour contours correspond to the horizontal component of the flow velocity (u), expressed in m/s).
PDF Download

REFERENCES

Aleixo, R., Soares-Frazão, S., Zech, Y., 2010. Velocity profiles in
dam-break flows: water and sediment layers. In: Proc. Int. Conf.
on Fluvial Hydraulics “River Flow 2010”, pp. 533–540.
An, S., Ku, H., Julien, P.Y., 2015. Numerical modelling of local
scour caused by submerged jets. Maejo Int. J. Sci. Technol., 9, 3,
328–343.
Bahmanpouri, F., Daliri, M., Khoshkonesh, A., Namin, M.M.,
Buccino, M., 2021. Bed compaction effect on dam break flow over
erodible bed; experimental and numerical modeling. J. Hydrol.,
594, 125645. https://doi.org/10.1016/j.jhydrol.2020.125645
Baklanov, A., 2007. Environmental risk and assessment modelling
– scientific needs and expected advancements. In: Ebel, A.,
Davitashvili, T. (Eds.): Air, Water and Soil Quality Modelling
for Risk and Impact Assessment Springer, Dordrecht, pp. 29–44.
Biscarini, C., Di Francesco, S., Nardi, F., Manciola, P., 2013.
Detailed simulation of complex hydraulic problems with
macroscopic and mesoscopic mathematical methods. Math.
Probl. Eng., 928309. https://doi.org/10.1155/2013/928309
Cao, Z., Pender, G., Wallis, S., Carling, P., 2004. Computational
dam-break hydraulics over erodible sediment bed. J. Hydraul.
Eng., 130, 7, 689–703.
Catucci, D., Briganti, R., Heller, V., 2021. Numerical validation of novel
scaling laws for air entrainment in water. Proc. R. Soc. A, 477, 2255,20210339. https://doi.org/10.1098/rspa.2021.0339
Dehrashid, F.A., Heidari, M., Rahimi, H., Khoshkonesh, A., Yuan,
S., Tang, X., Lu, C., Wang, X., 2023. CFD modeling the flow
dynamics in an open channel with double-layered vegetation.
Model. Earth Syst. Environ., 9, 1, 543–555.
Desombre, J., Morichon, D., Mory, M., 2013. RANS v2-f simulation
of a swash event: Detailed flow structure. Coastal Eng., 71, 1–12.
Dodangeh, E., Afzalimehr, H., 2022. Incipient motion of sediment
particles in the presence of bed forms under decelerating and
accelerating flows. J. Hydrol. Hydromech., 70, 1, 89–102.
Dong, Z., Wang, J., Vetsch, D.F., Boes, R.M., Tan, G., 2019.
Numerical simulation of air entrainment on stepped
spillways. In: E-proceedings of the 38th IAHR World Congress
(pp. 1494). September 1–6, 2019, Panama City, Panama. DOI:
10.3850/38WC092019-0755
Flow3D [computer software]. 2023. Santa Fe, NM: Flow Science,
Inc.
Fraccarollo, L., Capart, H., 2002. Riemann wave description of
erosional dam-break flows. J. Fluid Mech., 461, 183–228.
Gu, Z., Wang, T., Meng, W., Yu, C.H., An, R., 2023. Numerical
investigation of silted-up dam-break flow with different silted-up
sediment heights. Water Supply, 23, 2, 599–614.
Gualtieri, P., De Felice, S., Pasquino, V., Doria, G.P., 2018. Use of
conventional flow resistance equations and a model for the
Nikuradse roughness in vegetated flows at high submergence. J.
Hydrol. Hydromech., 66, 1, 107–120.
Heller, V., 2011. Scale effects in physical hydraulic engineering
models. J. Hydraul. Res., 49, 3, 293–306.
Hirt, C.W., 2003. Modeling turbulent entrainment of air at a free
surface. Flow Science, Inc.
Hirt, C.W., Nichols, B.D., 1981. Volume of fluid (VOF) method for
the dynamics of free boundaries. J. Comput. Phys., 39, 1, 201–
225.
Issakhov, A., Zhandaulet, Y., Nogaeva, A., 2018. Numerical
simulation of dam break flow for various forms of the obstacle
by VOF method. Int. J. Multiphase Flow, 109, 191–206.
Khayyer, A., Gotoh, H., 2010. On particle-based simulation of a dam
break over a wet bed. J. Hydraul. Res., 48, 2, 238–249.
Khoshkonesh, A., Daliri, M., Riaz, K., Dehrashid, F.A.,
Bahmanpouri, F., Di Francesco, S., 2022. Dam-break flow
dynamics over a stepped channel with vegetation. J. Hydrol., 613,128395. https://doi.org/10.1016/j.jhydrol.2022.128395
Khoshkonesh, A., Nsom, B., Gohari, S., Banejad, H., 2019.
A comprehensive study on dam-break flow over dry and wet
beds. Ocean Eng., 188, 106279.
https://doi.org/10.1016/j.oceaneng.2019.106279
Khoshkonesh, A., Sadeghi, S.H., Gohari, S., Karimpour, S., Oodi,
S., Di Francesco, S., 2023. Study of dam-break flow over a
vegetated channel with and without a drop. Water Resour.
Manage., 37, 5, 2107–2123.
Khosravi, K., Chegini, A.H.N., Cooper, J., Mao, L., Habibnejad, M.,
Shahedi, K., Binns, A., 2021. A laboratory investigation of bedload transport of gravel sediments under dam break flow. Int. J.
Sediment Res., 36, 2, 229–234.
Kim, Y., Zhou, Z., Hsu, T.J., Puleo, J.A., 2017. Large eddy
simulation of dam‐break‐driven swash on a rough‐planar beach.
J. Geophys. Res.: Oceans, 122, 2, 1274–1296.
Kocaman, S., Ozmen-Cagatay, H., 2012. The effect of lateral
channel contraction on dam break flows: Laboratory experiment.
J. Hydrol., 432, 145–153.
Leal, J.G., Ferreira, R.M., Cardoso, A.H., 2006. Dam-break wavefront celerity. J. Hydraul. Eng., 132, 1, 69–76.
Leal, J.G.A.B., Ferreira, R.M., Cardoso, A.H., 2003. Dam-break
wave propagation over a cohesionless erodible bed. In: Proc.
30rd IAHR Congress, 100, 261–268.
Li, Y. L., Ma, Y., Deng, R., Jiang, D.P., Hu, Z., 2019. Research on
dam-break induced tsunami bore acting on the triangular
breakwater based on high order 3D CLSVOF-THINC/WLICIBM approaching. Ocean Eng., 182, 645–659.
Li, Y.L., Yu, C.H., 2019. Research on dam-break flow induced front
wave impacting a vertical wall based on the CLSVOF and level
set methods. Ocean Eng., 178, 442–462.
Mei, S., Chen, S., Zhong, Q., Shan, Y., 2022. Detailed numerical
modeling for breach hydrograph and morphology evolution
during landslide dam breaching. Landslides, 19, 12, 2925–2949.
Meng, W., Yu, C.H., Li, J., An, R., 2022. Three-dimensional simulation
of silted-up dam-break flow striking a rigid structure. Ocean Eng.,
261, 112042. https://doi.org/10.1016/j.oceaneng.2022.112042
Meyer-Peter, E., Müller, R., 1948. Formulas for bed-load transport.
In: IAHSR 2nd meeting, Stockholm, appendix 2. IAHR.
Nielsen, P., 1984. Field measurements of time-averaged suspended
sediment concentrations under waves. Coastal Eng., 8, 1, 51–72.
Nielsen, P., 2018. Bed shear stress, surface shape and velocity field
near the tips of dam-breaks, tsunami and wave runup. Coastal
Eng., 138, 126–131.
Nsom, B., Latrache, N., Ramifidisoa, L., Khoshkonesh, A., 2019.
Analytical solution to the stability of gravity-driven stratified
flow of two liquids over an inclined plane. In: 24th French
Mechanics Congress in Brest. Brest, p. 244178.
Nsom, B., Ravelo, B., Ndong, W., 2008. Flow regimes in horizontal
viscous dam-break flow of Cayous mud. Appl. Rheol., 18, 4,
43577-1. https://doi.org/10.1515/arh-2008-0012
Oguzhan, S., Aksoy, A.O., 2020. Experimental investigation of the
effect of vegetation on dam break flood waves. J. Hydrol.
Hydromech., 68, 3, 231–241.
Okhravi, S., Gohari, S., Alemi, M., Maia, R., 2022. Effects of bedmaterial gradation on clear water scour at single and group of
piles. J. Hydrol. Hydromech., 70, 1, 114–127.
Okhravi, S., Gohari, S., Alemi, M., Maia, R., 2023. Numerical
modeling of local scour of non-uniform graded sediment for two
arrangements of pile groups. Int. J. Sediment Res., 38, 4, 597–614.
Parambath, A., 2010. Impact of tsunamis on near shore wind power
units. Master’s Thesis. Texas A&M University. Available
electronically from https://hdl.handle.net/1969.1/ETD-TAMU2010-12-8919
Pintado-Patiño, J.C., Puleo, J.A., Krafft, D., Torres-Freyermuth, A.,

  • Hydrodynamics and sediment transport under a dambreak-driven swash: An experimental study. Coastal Eng., 170,
  • https://doi.org/10.1016/j.coastaleng.2021.103986
    Riaz, K., Aslam, H.M.S., Yaseen, M.W., Ahmad, H.H.,
    Khoshkonesh, A., Noshin, S., 2022. Flood frequency analysis
    and hydraulic design of bridge at Mashan on river Kunhar. Arch.
    Hydroengineering Environ. Mech., 69, 1, 1–12.
    Ritter, A., 1892. Die Fortpflanzung der Wasserwellen. Zeitschrift
    des Vereines Deutscher Ingenieure, 36, 33, 947–954. (In
    German.)
    Smagorinsky, J., 1963. General circulation experiments with the
    primitive equations: I. The basic experiment. Mon. Weather
    Rev., 91, 3, 99–164.
    Soulsby, R.L., 1997. Dynamics of marine sands: a manual for
    practical applications. Oceanogr. Lit. Rev., 9, 44, 947.
    Spinewine, B., Capart, H., 2013. Intense bed-load due to a sudden
    dam-break. J. Fluid Mech., 731, 579–614.
    Van Rijn, L.C., 1984. Sediment transport, part I: bed load transport.
    J. Hydraul. Eng., 110, 10, 1431–1456.
    Vosoughi, F., Rakhshandehroo, G., Nikoo, M.R., Sadegh, M.,
  • Experimental study and numerical verification of
    silted-up dam break. J. Hydrol., 590, 125267.
    https://doi.org/10.1016/j.jhydrol.2020.125267
    Wu, W., Wang, S.S., 2008. One-dimensional explicit finite-volume
    model for sediment transport. J. Hydraul. Res., 46, 1, 87–98.
    Xu, T., Huai, W., Liu, H., 2023. MPS-based simulation of
    dam-break wave propagation over wet beds with a
    sediment layer. Ocean Eng., 281, 115035.
    https://doi.org/10.1016/j.oceaneng.2023.115035
    Yang, S., Yang, W., Qin, S., Li, Q., Yang, B., 2018. Numerical study
    on characteristics of dam-break wave. Ocean Eng., 159, 358–371.
    Yao, G.F., 2004. Development of new pressure-velocity solvers in
    FLOW-3D. Flow Science, Inc., USA.
PDF Download
Figure 1 | Schematic of the present research model with dimensions and macro-roughnesses installed.

On the hydraulic performance of the inclined drops: the effect of downstreammacro-roughness elements

경사 낙하의 수력학적 성능: 하류 거시 거칠기 요소의 영향

Farhoud Kalateh a,*, Ehsan Aminvash a and Rasoul Daneshfaraz b
a Faculty of Civil Engineering, University of Tabriz, Tabriz, Iran
b Faculty of Engineering, University of Maragheh, Maragheh, Iran
*Corresponding author. E-mail: f.kalateh@gmail.com

ABSTRACT

The main goal of the present study is to investigate the effects of macro-roughnesses downstream of the inclined drop through numerical models. Due to the vital importance of geometrical properties of the macro-roughnesses in the hydraulic performance and efficient energy dissipation downstream of inclined drops, two different geometries of macro-roughnesses, i.e., semi-circular and triangular geometries, have been investigated using the Flow-3D model. Numerical simulation showed that with the flow rate increase and relative critical depth, the flow energy consumption has decreased. Also, relative energy dissipation increases with the increase in height and slope angle, so that this amount of increase in energy loss compared to the smooth bed in semi-circular and triangular elements is 86.39 and 76.80%, respectively, in the inclined drop with a height of 15 cm and 86.99 and 65.78% in the drop with a height of 20 cm. The Froude number downstream on the uneven bed has been dramatically reduced, so this amount of reduction has been approximately 47 and 54% compared to the control condition. The relative depth of the downstream has also increased due to the turbulence of the flow on the uneven bed with the increase in the flow rate.

본 연구의 주요 목표는 수치 모델을 통해 경사 낙하 하류의 거시 거칠기 효과를 조사하는 것입니다. 수력학적 성능과 경사 낙하 하류의 효율적인 에너지 소산에서 거시 거칠기의 기하학적 특성이 매우 중요하기 때문에 두 가지 서로 다른 거시 거칠기 형상, 즉 반원형 및 삼각형 형상이 Flow를 사용하여 조사되었습니다.

3D 모델 수치 시뮬레이션을 통해 유량이 증가하고 상대 임계 깊이가 증가함에 따라 유동 에너지 소비가 감소하는 것으로 나타났습니다. 또한, 높이와 경사각이 증가함에 따라 상대적인 에너지 소산도 증가하는데, 반원형 요소와 삼각형 요소에서 평활층에 비해 에너지 손실의 증가량은 경사낙하에서 각각 86.39%와 76.80%입니다.

높이 15cm, 높이 20cm의 드롭에서 86.99%, 65.78%입니다. 고르지 못한 베드 하류의 프루드 수가 극적으로 감소하여 이 감소량은 대조 조건에 비해 약 47%와 54%였습니다. 유속이 증가함에 따라 고르지 못한 층에서의 흐름의 난류로 인해 하류의 상대적 깊이도 증가했습니다.

Key words

flow energy dissipation, Froude number, inclined drop, numerical simulation

Figure 1 | Schematic of the present research model with dimensions and macro-roughnesses installed.
Figure 1 | Schematic of the present research model with dimensions and macro-roughnesses installed.
Figure 2 | Meshing, boundary condition, and solution field network
Figure 2 | Meshing, boundary condition, and solution field network

REFERENCES

Abbaspour, A., Taghavianpour, T. & Arvanaghi, H. 2019 Experimental study of the hydraulic jump on the reverse bed with porous screens.
Applied Water Science 9, 155.
Abbaspour, A., Shiravani, P. & Hosseinzadeh Dalir, A. 2021 Experimental study of the energy dissipation on rough ramps. ISH Journal of
Hydraulic Engineering 27, 334–342.
Akib, S., Ahmed, A. A., Imran, H. M., Mahidin, M. F., Ahmed, H. S. & Rahman, S. 2015 Properties of a hydraulic jump over apparent
corrugated beds. Dam Engineering 25, 65–77.
AlTalib, A. N., Mohammed, A. Y. & Hayawi, H. A. 2015 Hydraulic jump and energy dissipation downstream stepped weir. Flow
Measurement and Instrumentation 69, 101616.
Bayon-Barrachina, A. & Lopez-Jimenez, P. A. 2015 Numerical analysis of hydraulic jumps using OpenFOAM. Journal of Hydroinformatics
17, 662–678.
Canovaro, F. & Solari, L. 2007 Dissipative analogies between a schematic macro-roughness arrangement and step–pool morphology. Earth
Surface Processes and Landforms: The Journal of the British Geomorphological Research Group 32, 1628–1640.
Daneshfaraz, R., Ghaderi, A., Akhtari, A. & Di Francesco, S. 2020 On the effect of block roughness in ogee spill-ways with flip buckets. Fluids
5, 182.
Daneshfaraz, R., Aminvash, E., Di Francesco, S., Najibi, A. & Abraham, J. 2021a Three-dimensional study of the effect of block roughness
geometry on inclined drop. Numerical Methods in Civil Engineering 6, 1–9.
Daneshfaraz, R., Aminvash, E., Ghaderi, A., Abraham, J. & Bagherzadeh, M. 2021b SVM performance for predicting the effect of horizontal
screen diameters on the hydraulic parameters of a vertical drop. Applied Science 11, 4238.
Daneshfaraz, R., Aminvash, E., Ghaderi, A., Kuriqi, A. & Abraham, J. 2021c Three-dimensional investigation of hydraulic properties of
vertical drop in the presence of step and grid dissipators. Symmetry 13, 895.
Dey, S. & Sarkar, A. 2008 Characteristics of turbulent flow in submerged jumps on rough beds. Journal of Engineering Mechanics 134, 49–59.
Ead, S. A. & Rajaratnam, N. 2002 Hydraulic jumps on corrugated beds. Journal of Hydraulic Engineering 128, 656–663.
Fang, H., Han, X., He, G. & Dey, S. 2018 Influence of permeable beds on hydraulically macro-rough flow. Journal of Fluid Mechanics 847,
552–590.
Federico, I., Marrone, S., Colagrossi, A., Aristodemo, F. & Antuono, M. 2019 Simulating 2D open-channel flows through an SPH model.
European Journal of Mechanics-B/Fluids 34, 35–46.
Ghaderi, A., Dasineh, M., Aristodemo, F. & Aricò, C. 2021 Numerical simulations of the flow field of a submerged hydraulic jump over
triangular macroroughnesses. Water 13, 674.
Ghare, A. D., Ingl, R. N., Porey, P. D. & Gokhale, S. S. 2010 Block ramp design for efficient energy dissipation. Journal of Energy Dissipation
136, 1–5.
Habibzadeh, A., Rajaratnam, N. & Loewen, M. 2019 Characteristics of the flow field downstream of free and submerged hydraulic jumps.
Proceedings of the Institution of Civil Engineers-Water Management 172, 180–194.
Hajiahmadi, A., Ghaeini-Hessaroeyeh, M. & Khanjani, M. J. 2021 Experimental evaluation of vertical shaft efficiency in vortex flow energy
dissipation. International Journal of Civil Engineering 19, 1445–1455.

Katourani, S. & Kashefipour, S. M. 2012 Effect of the geometric characteristics of baffle on hydraulic flow condition in baffled apron drop.
Irrigation Sciences and Engineering 37, 51–59.
Kurdistani, S. M., Varaki, M. E. & Moayedi Moshkaposhti, M. 2024 Apron and macro roughness as scour countermeasures downstream of
block ramps. ISH Journal of Hydraulic Engineering 1–9.
Lopardo, R. A. 2013 Extreme velocity fluctuations below free hydraulic jumps. Journal of Engineering 1–5.
Mahmoudi-Rad, M. & Najafzadeh, M. 2023 Experimental evaluation of the energy dissipation efficiency of the vortex flow section of drop
shafts. Scientific Reports 13, 1679.
Matin, M. A., Hasan, M. & Islam, M. R. 2018 Experiment on hydraulic jump in sudden expansion in a sloping rectangular channel. Journal of
Civil Engineering 36, 65–77.
Moghadam, K. F., Banihashemi, M. A., Badiei, P. & Shirkavand, A. 2019 A numerical approach to solve fluid-solid two-phase flows using time
splitting projection method with a pressure correction technique. Progress in Computational Fluid Dynamics, an International Journal
19, 357–367.
Moghadam, K. F., Banihashemi, M. A., Badiei, P. & Shirkavand, A. 2020 A time-splitting pressure-correction projection method for complete
two-fluid modeling of a local scour hole. International Journal of Sediment Research 35, 395–407.
Moradi-SabzKoohi, A., Kashefipour, S. M. & Bina, M. 2011 Experimental comparison of energy dissipation on drop structures. JWSS –
Isfahan University of Technology 15, 209–223. (in Persian).
Mouaze, D., Murzyn, F. & Chaplin, J. R. 2005 Free surface length scale estimation in hydraulic jumps. Journal of Fluids Engineering 127,
1191–1193.
Nicosia, A., Carollo, F. G. & Ferro, V. 2023 Effects of boulder arrangement on flow resistance due to macro-scale bed roughness. Water 15,
349.
Ohtsu, I. & Yasuda, Y. 1991 Hydraulic jump in sloping channel. Journal of Hydraulic Engineering 117, 905–921.
Pagliara, S. & Palermo, M. 2012 Effect of stilling basin geometry on the dissipative process in the presence of block ramps. Journal of
Irrigation and Drainage Engineering 138, 1027–1031.
Pagliara, S., Das, R. & Palermo, M. 2008 Energy dissipation on submerged block ramps. Journal of Irrigation and Drainage Engineering 134,
527–532.
Pagliara, S., Roshni, T. & Palermo, M. 2015 Energy dissipation over large-scale roughness for both transition and uniform flow conditions.
International Journal of Civil Engineering 13, 341–346.
Parsaie, A., Dehdar-Behbahani, S. & Haghiabi, A. H. 2016 Numerical modeling of cavitation on spillway’s flip bucket. Frontiers of Structural
and Civil Engineering 10, 438–444.
Pourabdollah, N., Heidarpour, M. & Abedi Koupai, J. 2018 Characteristics of free and submerged hydraulic jumps in different stilling basins.
In: Proceedings of the Institution of Civil Engineers-Water Management. Thomas Telford Ltd, pp. 1–11.
Roushangar, K. & Ghasempour, R. 2019 Evaluation of the impact of channel geometry and rough elements arrangement in hydraulic jump
energy dissipation via SVM. Journal of Hydroinformatics 21, 92–103.
Samadi-Boroujeni, H., Ghazali, M., Gorbani, B. & Nafchi, R. F. 2013 Effect of triangular corrugated beds on the hydraulic jump
characteristics. Canadian Journal of Civil Engineering 40, 841–847.
Shekari, Y., Javan, M. & Eghbalzadeh, A. 2014 Three-dimensional numerical study of submerged hydraulic jumps. Arabian Journal for
Science and Engineering 39, 6969–6981.
Tokyay, N. D., Evcimen, T. U. & Şimsek, Ç. 2011 Forced hydraulic jump on non-protruding rough beds. Canadian Journal of Civil
Engineering 38, 1136–1144.
Wagner, W. E. 1956 Hydraulic model studies of the check intake structure-potholes East canal. Bureau of Reclamation Hydraulic Laboratory
Report Hyd, 411.
Witt, A., Gulliver, J. S. & Shen, L. 2018 Numerical investigation of vorticity and bubble clustering in an air-entraining hydraulic jump.
Computers & Fluids 172, 162–180.

PDF Download
Fig. 9 From: An Investigation on Hydraulic Aspects of Rectangular Labyrinth Pool and Weir Fishway Using FLOW-3D

An Investigation on Hydraulic Aspects of Rectangular Labyrinth Pool and Weir Fishway Using FLOW-3D

Abstract

웨어의 두 가지 서로 다른 배열(즉, 직선형 웨어와 직사각형 미로 웨어)을 사용하여 웨어 모양, 웨어 간격, 웨어의 오리피스 존재, 흐름 영역에 대한 바닥 경사와 같은 기하학적 매개변수의 영향을 평가했습니다.

유량과 수심의 관계, 수심 평균 속도의 변화와 분포, 난류 특성, 어도에서의 에너지 소산. 흐름 조건에 미치는 영향을 조사하기 위해 FLOW-3D® 소프트웨어를 사용하여 전산 유체 역학 시뮬레이션을 수행했습니다.

수치 모델은 계산된 표면 프로파일과 속도를 문헌의 실험적으로 측정된 값과 비교하여 검증되었습니다. 수치 모델과 실험 데이터의 결과, 급락유동의 표면 프로파일과 표준화된 속도 프로파일에 대한 평균 제곱근 오차와 평균 절대 백분율 오차가 각각 0.014m와 3.11%로 나타나 수치 모델의 능력을 확인했습니다.

수영장과 둑의 흐름 특성을 예측합니다. 각 모델에 대해 L/B = 1.83(L: 웨어 거리, B: 수로 폭) 값에서 급락 흐름이 발생할 수 있고 L/B = 0.61에서 스트리밍 흐름이 발생할 수 있습니다. 직사각형 미로보 모델은 기존 모델보다 무차원 방류량(Q+)이 더 큽니다.

수중 흐름의 기존 보와 직사각형 미로 보의 경우 Q는 각각 1.56과 1.47h에 비례합니다(h: 보 위 수심). 기존 웨어의 풀 내 평균 깊이 속도는 직사각형 미로 웨어의 평균 깊이 속도보다 높습니다.

그러나 주어진 방류량, 바닥 경사 및 웨어 간격에 대해 난류 운동 에너지(TKE) 및 난류 강도(TI) 값은 기존 웨어에 비해 직사각형 미로 웨어에서 더 높습니다. 기존의 웨어는 직사각형 미로 웨어보다 에너지 소산이 더 낮습니다.

더 낮은 TKE 및 TI 값은 미로 웨어 상단, 웨어 하류 벽 모서리, 웨어 측벽과 채널 벽 사이에서 관찰되었습니다. 보와 바닥 경사면 사이의 거리가 증가함에 따라 평균 깊이 속도, 난류 운동 에너지의 평균값 및 난류 강도가 증가하고 수영장의 체적 에너지 소산이 감소했습니다.

둑에 개구부가 있으면 평균 깊이 속도와 TI 값이 증가하고 풀 내에서 가장 높은 TKE 범위가 감소하여 두 모델 모두에서 물고기를 위한 휴식 공간이 더 넓어지고(TKE가 낮아짐) 에너지 소산율이 감소했습니다.

원문보기

Two different arrangements of the weir (i.e., straight weir and rectangular labyrinth weir) were used to evaluate the effects of geometric parameters such as weir shape, weir spacing, presence of an orifice at the weir, and bed slope on the flow regime and the relationship between discharge and depth, variation and distribution of depth-averaged velocity, turbulence characteristics, and energy dissipation at the fishway. Computational fluid dynamics simulations were performed using FLOW-3D® software to examine the effects on flow conditions. The numerical model was validated by comparing the calculated surface profiles and velocities with experimentally measured values from the literature. The results of the numerical model and experimental data showed that the root-mean-square error and mean absolute percentage error for the surface profiles and normalized velocity profiles of plunging flows were 0.014 m and 3.11%, respectively, confirming the ability of the numerical model to predict the flow characteristics of the pool and weir. A plunging flow can occur at values of L/B = 1.83 (L: distance of the weir, B: width of the channel) and streaming flow at L/B = 0.61 for each model. The rectangular labyrinth weir model has larger dimensionless discharge values (Q+) than the conventional model. For the conventional weir and the rectangular labyrinth weir at submerged flow, Q is proportional to 1.56 and 1.47h, respectively (h: the water depth above the weir). The average depth velocity in the pool of a conventional weir is higher than that of a rectangular labyrinth weir. However, for a given discharge, bed slope, and weir spacing, the turbulent kinetic energy (TKE) and turbulence intensity (TI) values are higher for a rectangular labyrinth weir compared to conventional weir. The conventional weir has lower energy dissipation than the rectangular labyrinth weir. Lower TKE and TI values were observed at the top of the labyrinth weir, at the corner of the wall downstream of the weir, and between the side walls of the weir and the channel wall. As the distance between the weirs and the bottom slope increased, the average depth velocity, the average value of turbulent kinetic energy and the turbulence intensity increased, and the volumetric energy dissipation in the pool decreased. The presence of an opening in the weir increased the average depth velocity and TI values and decreased the range of highest TKE within the pool, resulted in larger resting areas for fish (lower TKE), and decreased the energy dissipation rates in both models.

1 Introduction

Artificial barriers such as detour dams, weirs, and culverts in lakes and rivers prevent fish from migrating and completing the upstream and downstream movement cycle. This chain is related to the life stage of the fish, its location, and the type of migration. Several riverine fish species instinctively migrate upstream for spawning and other needs. Conversely, downstream migration is a characteristic of early life stages [1]. A fish ladder is a waterway that allows one or more fish species to cross a specific obstacle. These structures are constructed near detour dams and other transverse structures that have prevented such migration by allowing fish to overcome obstacles [2]. The flow pattern in fish ladders influences safe and comfortable passage for ascending fish. The flow’s strong turbulence can reduce the fish’s speed, injure them, and delay or prevent them from exiting the fish ladder. In adult fish, spawning migrations are usually complex, and delays are critical to reproductive success [3].

Various fish ladders/fishways include vertical slots, denil, rock ramps, and pool weirs [1]. The choice of fish ladder usually depends on many factors, including water elevation, space available for construction, and fish species. Pool and weir structures are among the most important fish ladders that help fish overcome obstacles in streams or rivers and swim upstream [1]. Because they are easy to construct and maintain, this type of fish ladder has received considerable attention from researchers and practitioners. Such a fish ladder consists of a sloping-floor channel with series of pools directly separated by a series of weirs [4]. These fish ladders, with or without underwater openings, are generally well-suited for slopes of 10% or less [1, 2]. Within these pools, flow velocities are low and provide resting areas for fish after they enter the fish ladder. After resting in the pools, fish overcome these weirs by blasting or jumping over them [2]. There may also be an opening in the flooded portion of the weir through which the fish can swim instead of jumping over the weir. Design parameters such as the length of the pool, the height of the weir, the slope of the bottom, and the water discharge are the most important factors in determining the hydraulic structure of this type of fish ladder [3]. The flow over the weir depends on the flow depth at a given slope S0 and the pool length, either “plunging” or “streaming.” In plunging flow, the water column h over each weir creates a water jet that releases energy through turbulent mixing and diffusion mechanisms [5]. The dimensionless discharges for plunging (Q+) and streaming (Q*) flows are shown in Fig. 1, where Q is the total discharge, B is the width of the channel, w is the weir height, S0 is the slope of the bottom, h is the water depth above the weir, d is the flow depth, and g is the acceleration due to gravity. The maximum velocity occurs near the top of the weir for plunging flow. At the water’s surface, it drops to about half [6].

figure 1
Fig. 1

Extensive experimental studies have been conducted to investigate flow patterns for various physical geometries (i.e., bed slope, pool length, and weir height) [2]. Guiny et al. [7] modified the standard design by adding vertical slots, orifices, and weirs in fishways. The efficiency of the orifices and vertical slots was related to the velocities at their entrances. In the laboratory experiments of Yagci [8], the three-dimensional (3D) mean flow and turbulence structure of a pool weir fishway combined with an orifice and a slot is investigated. It is shown that the energy dissipation per unit volume and the discharge have a linear relationship.

Considering the beneficial characteristics reported in the limited studies of researchers on the labyrinth weir in the pool-weir-type fishway, and knowing that the characteristics of flow in pool-weir-type fishways are highly dependent on the geometry of the weir, an alternative design of the rectangular labyrinth weir instead of the straight weirs in the pool-weir-type fishway is investigated in this study [7, 9]. Kim [10] conducted experiments to compare the hydraulic characteristics of three different weir types in a pool-weir-type fishway. The results show that a straight, rectangular weir with a notch is preferable to a zigzag or trapezoidal weir. Studies on natural fish passes show that pass ability can be improved by lengthening the weir’s crest [7]. Zhong et al. [11] investigated the semi-rigid weir’s hydraulic performance in the fishway’s flow field with a pool weir. The results showed that this type of fishway performed better with a lower invert slope and a smaller radius ratio but with a larger pool spacing.

Considering that an alternative method to study the flow characteristics in a fishway with a pool weir is based on numerical methods and modeling from computational fluid dynamics (CFD), which can easily change the geometry of the fishway for different flow fields, this study uses the powerful package CFD and the software FLOW-3D to evaluate the proposed weir design and compare it with the conventional one to extend the application of the fishway. The main objective of this study was to evaluate the hydraulic performance of the rectangular labyrinth pool and the weir with submerged openings in different hydraulic configurations. The primary objective of creating a new weir configuration for suitable flow patterns is evaluated based on the swimming capabilities of different fish species. Specifically, the following questions will be answered: (a) How do the various hydraulic and geometric parameters relate to the effects of water velocity and turbulence, expressed as turbulent kinetic energy (TKE) and turbulence intensity (TI) within the fishway, i.e., are conventional weirs more affected by hydraulics than rectangular labyrinth weirs? (b) Which weir configurations have the greatest effect on fish performance in the fishway? (c) In the presence of an orifice plate, does the performance of each weir configuration differ with different weir spacing, bed gradients, and flow regimes from that without an orifice plate?

2 Materials and Methods

2.1 Physical Model Configuration

This paper focuses on Ead et al. [6]’s laboratory experiments as a reference, testing ten pool weirs (Fig. 2). The experimental flume was 6 m long, 0.56 m wide, and 0.6 m high, with a bottom slope of 10%. Field measurements were made at steady flow with a maximum flow rate of 0.165 m3/s. Discharge was measured with magnetic flow meters in the inlets and water level with point meters (see Ead et al. [6]. for more details). Table 1 summarizes the experimental conditions considered for model calibration in this study.

figure 2
Fig. 2

Table 1 Experimental conditions considered for calibration

Full size table

2.2 Numerical Models

Computational fluid dynamics (CFD) simulations were performed using FLOW-3D® v11.2 to validate a series of experimental liner pool weirs by Ead et al. [6] and to investigate the effects of the rectangular labyrinth pool weir with an orifice. The dimensions of the channel and data collection areas in the numerical models are the same as those of the laboratory model. Two types of pool weirs were considered: conventional and labyrinth. The proposed rectangular labyrinth pool weirs have a symmetrical cross section and are sized to fit within the experimental channel. The conventional pool weir model had a pool length of l = 0.685 and 0.342 m, a weir height of w = 0.141 m, a weir width of B = 0.56 m, and a channel slope of S0 = 5 and 10%. The rectangular labyrinth weirs have the same front width as the offset, i.e., a = b = c = 0.186 m. A square underwater opening with a width of 0.05 m and a depth of 0.05 m was created in the middle of the weir. The weir configuration considered in the present study is shown in Fig. 3.

figure 3
Fig. 3

2.3 Governing Equations

FLOW-3D® software solves the Navier–Stokes–Reynolds equations for three-dimensional analysis of incompressible flows using the fluid-volume method on a gridded domain. FLOW -3D® uses an advanced free surface flow tracking algorithm (TruVOF) developed by Hirt and Nichols [12], where fluid configurations are defined in terms of a VOF function F (x, y, z, t). In this case, F (fluid fraction) represents the volume fraction occupied by the fluid: F = 1 in cells filled with fluid and F = 0 in cells without fluid (empty areas) [4, 13]. The free surface area is at an intermediate value of F. (Typically, F = 0.5, but the user can specify a different intermediate value.) The equations in Cartesian coordinates (x, y, z) applicable to the model are as follows:

�f∂�∂�+∂(���x)∂�+∂(���y)∂�+∂(���z)∂�=�SOR

(1)

∂�∂�+1�f(��x∂�∂�+��y∂�∂�+��z∂�∂�)=−1�∂�∂�+�x+�x

(2)

∂�∂�+1�f(��x∂�∂�+��y∂�∂�+��z∂�∂�)=−1�∂�∂�+�y+�y

(3)

∂�∂�+1�f(��x∂�∂�+��y∂�∂�+��z∂�∂�)=−1�∂�∂�+�z+�z

(4)

where (u, v, w) are the velocity components, (Ax, Ay, Az) are the flow area components, (Gx, Gy, Gz) are the mass accelerations, and (fx, fy, fz) are the viscous accelerations in the directions (x, y, z), ρ is the fluid density, RSOR is the spring term, Vf is the volume fraction associated with the flow, and P is the pressure. The k–ε turbulence model (RNG) was used in this study to solve the turbulence of the flow field. This model is a modified version of the standard k–ε model that improves performance. The model is a two-equation model; the first equation (Eq. 5) expresses the turbulence’s energy, called turbulent kinetic energy (k) [14]. The second equation (Eq. 6) is the turbulent dissipation rate (ε), which determines the rate of dissipation of kinetic energy [15]. These equations are expressed as follows Dasineh et al. [4]:

∂(��)∂�+∂(����)∂��=∂∂��[������∂�∂��]+��−�ε

(5)

∂(�ε)∂�+∂(�ε��)∂��=∂∂��[�ε�eff∂ε∂��]+�1εε��k−�2ε�ε2�

(6)

In these equations, k is the turbulent kinetic energy, ε is the turbulent energy consumption rate, Gk is the generation of turbulent kinetic energy by the average velocity gradient, with empirical constants αε = αk = 1.39, C1ε = 1.42, and C2ε = 1.68, eff is the effective viscosity, μeff = μ + μt [15]. Here, μ is the hydrodynamic density coefficient, and μt is the turbulent density of the fluid.

2.4 Meshing and the Boundary Conditions in the Model Setup

The numerical area is divided into three mesh blocks in the X-direction. The meshes are divided into different sizes, a containing mesh block for the entire spatial domain and a nested block with refined cells for the domain of interest. Three different sizes were selected for each of the grid blocks. By comparing the accuracy of their results based on the experimental data, the reasonable mesh for the solution domain was finally selected. The convergence index method (GCI) evaluated the mesh sensitivity analysis. Based on this method, many researchers, such as Ahmadi et al. [16] and Ahmadi et al. [15], have studied the independence of numerical results from mesh size. Three different mesh sizes with a refinement ratio (r) of 1.33 were used to perform the convergence index method. The refinement ratio is the ratio between the larger and smaller mesh sizes (r = Gcoarse/Gfine). According to the recommendation of Celik et al. [17], the recommended number for the refinement ratio is 1.3, which gives acceptable results. Table 2 shows the characteristics of the three mesh sizes selected for mesh sensitivity analysis.Table 2 Characteristics of the meshes tested in the convergence analysis

Full size table

The results of u1 = umax (u1 = velocity component along the x1 axis and umax = maximum velocity of u1 in a section perpendicular to the invert of the fishway) at Q = 0.035 m3/s, × 1/l = 0.66, and Y1/b = 0 in the pool of conventional weir No. 4, obtained from the output results of the software, were used to evaluate the accuracy of the calculation range. As shown in Fig. 4, x1 = the distance from a given weir in the x-direction, Y1 = the water depth measured in the y-direction, Y0 = the vertical distance in the Cartesian coordinate system, h = the water column at the crest, b = the distance between the two points of maximum velocity umax and zero velocity, and l = the pool length.

figure 4
Fig. 4

The apparent index of convergence (p) in the GCI method is calculated as follows:

�=ln⁡(�3−�2)(�2−�1)/ln⁡(�)

(7)

f1, f2, and f3 are the hydraulic parameters obtained from the numerical simulation (f1 corresponds to the small mesh), and r is the refinement ratio. The following equation defines the convergence index of the fine mesh:

GCIfine=1.25|ε|��−1

(8)

Here, ε = (f2 − f1)/f1 is the relative error, and f2 and f3 are the values of hydraulic parameters considered for medium and small grids, respectively. GCI12 and GCI23 dimensionless indices can be calculated as:

GCI12=1.25|�2−�1�1|��−1

(9)

Then, the independence of the network is preserved. The convergence index of the network parameters obtained by Eqs. (7)–(9) for all three network variables is shown in Table 3. Since the GCI values for the smaller grid (GCI12) are lower compared to coarse grid (GCI23), it can be concluded that the independence of the grid is almost achieved. No further change in the grid size of the solution domain is required. The calculated values (GCI23/rpGCI12) are close to 1, which shows that the numerical results obtained are within the convergence range. As a result, the meshing of the solution domain consisting of a block mesh with a mesh size of 0.012 m and a block mesh within a larger block mesh with a mesh size of 0.009 m was selected as the optimal mesh (Fig. 5).Table 3 GCI calculation

Full size table

figure 5
Fig. 5

The boundary conditions applied to the area are shown in Fig. 6. The boundary condition of specific flow rate (volume flow rate-Q) was used for the inlet of the flow. For the downstream boundary, the flow output (outflow-O) condition did not affect the flow in the solution area. For the Zmax boundary, the specified pressure boundary condition was used along with the fluid fraction = 0 (P). This type of boundary condition considers free surface or atmospheric pressure conditions (Ghaderi et al. [19]). The wall boundary condition is defined for the bottom of the channel, which acts like a virtual wall without friction (W). The boundary between mesh blocks and walls were considered a symmetrical condition (S).

figure 6
Fig. 6

The convergence of the steady-state solutions was controlled during the simulations by monitoring the changes in discharge at the inlet boundary conditions. Figure 7 shows the time series plots of the discharge obtained from the Model A for the three main discharges from the numerical results. The 8 s to reach the flow equilibrium is suitable for the case of the fish ladder with pool and weir. Almost all discharge fluctuations in the models are insignificant in time, and the flow has reached relative stability. The computation time for the simulations was between 6 and 8 h using a personal computer with eight cores of a CPU (Intel Core i7-7700K @ 4.20 GHz and 16 GB RAM).

figure 7
Fig. 7

3 Results

3.1 Verification of Numerical Results

Quantitative outcomes, including free surface and normalized velocity profiles obtained using FLOW-3D software, were reviewed and compared with the results of Ead et al. [6]. The fourth pool was selected to present the results and compare the experiment and simulation. For each quantity, the percentage of mean absolute error (MAPE (%)) and root-mean-square error (RMSE) are calculated. Equations (10) and (11) show the method used to calculate the errors.

MAPE(%)100×1�∑1�|�exp−�num�exp|

(10)

RMSE(−)1�∑1�(�exp−�num)2

(11)

Here, Xexp is the value of the laboratory data, Xnum is the numerical data value, and n is the amount of data. As shown in Fig. 8, let x1 = distance from a given weir in the x-direction and Y1 = water depth in the y-direction from the bottom. The trend of the surface profiles for each of the numerical results is the same as that of the laboratory results. The surface profiles of the plunging flows drop after the flow enters and then rises to approach the next weir. The RMSE and MAPE error values for Model A are 0.014 m and 3.11%, respectively, indicating acceptable agreement between numerical and laboratory results. Figure 9 shows the velocity vectors and plunging flow from the numerical results, where x and y are horizontal and vertical to the flow direction, respectively. It can be seen that the jet in the fish ladder pool has a relatively high velocity. The two vortices, i.e., the enclosed vortex rotating clockwise behind the weir and the surface vortex rotating counterclockwise above the jet, are observed for the regime of incident flow. The point where the jet meets the fish passage bed is shown in the figure. The normalized velocity profiles upstream and downstream of the impact points are shown in Fig. 10. The figure shows that the numerical results agree well with the experimental data of Ead et al. [6].

figure 8
Fig. 8
figure 9
Fig. 9
figure 10
Fig. 10

3.2 Flow Regime and Discharge-Depth Relationship

Depending on the geometric shape of the fishway, including the distance of the weir, the slope of the bottom, the height of the weir, and the flow conditions, the flow regime in the fishway is divided into three categories: dipping, transitional, and flow regimes [4]. In the plunging flow regime, the flow enters the pool through the weir, impacts the bottom of the fishway, and forms a hydraulic jump causing two eddies [2, 20]. In the streamwise flow regime, the surface of the flow passing over the weir is almost parallel to the bottom of the channel. The transitional regime has intermediate flow characteristics between the submerged and flow regimes. To predict the flow regime created in the fishway, Ead et al. [6] proposed two dimensionless parameters, Qt* and L/w, where Qt* is the dimensionless discharge, L is the distance between weirs, and w is the height of the weir:

��∗=���0���

(12)

Q is the total discharge, B is the width of the channel, S0 is the slope of the bed, and g is the gravity acceleration. Figure 11 shows different ranges for each flow regime based on the slope of the bed and the distance between the pools in this study. The results of Baki et al. [21], Ead et al. [6] and Dizabadi et al. [22] were used for this comparison. The distance between the pools affects the changes in the regime of the fish ladder. So, if you decrease the distance between weirs, the flow regime more likely becomes. This study determined all three flow regimes in a fish ladder. When the corresponding range of Qt* is less than 0.6, the flow regime can dip at values of L/B = 1.83. If the corresponding range of Qt* is greater than 0.5, transitional flow may occur at L/B = 1.22. On the other hand, when Qt* is greater than 1, streamwise flow can occur at values of L/B = 0.61. These observations agree well with the results of Baki et al. [21], Ead et al. [6] and Dizabadi et al. [22].

figure 11
Fig. 11

For plunging flows, another dimensionless discharge (Q+) versus h/w given by Ead et al. [6] was used for further evaluation:

�+=��ℎ�ℎ=23�d�

(13)

where h is the water depth above the weir, and Cd is the discharge coefficient. Figure 12a compares the numerical and experimental results of Ead et al. [6]. In this figure, Rehbock’s empirical equation is used to estimate the discharge coefficient of Ead et al. [6].

�d=0.57+0.075ℎ�

(14)

figure 12
Fig. 12

The numerical results for the conventional weir (Model A) and the rectangular labyrinth weir (Model B) of this study agree well with the laboratory results of Ead et al. [6]. When comparing models A and B, it is also found that a rectangular labyrinth weir has larger Q + values than the conventional weir as the length of the weir crest increases for a given channel width and fixed headwater elevation. In Fig. 12b, Models A and B’s flow depth plot shows the plunging flow regime. The power trend lines drawn through the data are the best-fit lines. The data shown in Fig. 12b are for different bed slopes and weir geometries. For the conventional weir and the rectangular labyrinth weir at submerged flow, Q can be assumed to be proportional to 1.56 and 1.47h, respectively. In the results of Ead et al. [6], Q is proportional to 1.5h. If we assume that the flow through the orifice is Qo and the total outflow is Q, the change in the ratio of Qo/Q to total outflow for models A and B can be shown in Fig. 13. For both models, the flow through the orifice decreases as the total flow increases. A logarithmic trend line was also found between the total outflow and the dimensionless ratio Qo/Q.

figure 13
Fig. 13

3.3 Depth-Averaged Velocity Distributions

To ensure that the target fish species can pass the fish ladder with maximum efficiency, the average velocity in the fish ladder should be low enough [4]. Therefore, the average velocity in depth should be as much as possible below the critical swimming velocities of the target fishes at a constant flow depth in the pool [20]. The contour plot of depth-averaged velocity was used instead of another direction, such as longitudinal velocity because fish are more sensitive to depth-averaged flow velocity than to its direction under different hydraulic conditions. Figure 14 shows the distribution of depth-averaged velocity in the pool for Models A and B in two cases with and without orifice plates. Model A’s velocity within the pool differs slightly in the spanwise direction. However, no significant variation in velocity was observed. The flow is gradually directed to the sides as it passes through the rectangular labyrinth weir. This increases the velocity at the sides of the channel. Therefore, the high-velocity zone is located at the sides. The low velocity is in the downstream apex of the weir. This area may be suitable for swimming target fish. The presence of an opening in the weir increases the flow velocity at the opening and in the pool’s center, especially in Model A. The flow velocity increase caused by the models’ opening varied from 7.7 to 12.48%. Figure 15 illustrates the effect of the inverted slope on the averaged depth velocity distribution in the pool at low and high discharge. At constant discharge, flow velocity increases with increasing bed slope. In general, high flow velocity was found in the weir toe sidewall and the weir and channel sidewalls.

figure 14
Fig. 14
figure 15
Fig. 15

On the other hand, for a constant bed slope, the high-velocity area of the pool increases due to the increase in runoff. For both bed slopes and different discharges, the most appropriate path for fish to travel from upstream to downstream is through the middle of the cross section and along the top of the rectangular labyrinth weirs. The maximum dominant velocities for Model B at S0 = 5% were 0.83 and 1.01 m/s; at S0 = 10%, they were 1.12 and 1.61 m/s at low and high flows, respectively. The low mean velocities for the same distance and S0 = 5 and 10% were 0.17 and 0.26 m/s, respectively.

Figure 16 shows the contour of the averaged depth velocity for various distances from the weir at low and high discharge. The contour plot shows a large variation in velocity within short distances from the weir. At L/B = 0.61, velocities are low upstream and downstream of the top of the weir. The high velocities occur in the side walls of the weir and the channel. At L/B = 1.22, the low-velocity zone displaces the higher velocity in most of the pool. Higher velocities were found only on the sides of the channel. As the discharge increases, the velocity zone in the pool becomes wider. At L/B = 1.83, there is an area of higher velocities only upstream of the crest and on the sides of the weir. At high discharge, the prevailing maximum velocities for L/B = 0.61, 1.22, and 1.83 were 1.46, 1.65, and 1.84 m/s, respectively. As the distance between weirs increases, the range of maximum velocity increases.

figure 16
Fig. 16

On the other hand, the low mean velocity for these distances was 0.27, 0.44, and 0.72 m/s, respectively. Thus, the low-velocity zone decreases with increasing distance between weirs. Figure 17 shows the pattern distribution of streamlines along with the velocity contour at various distances from the weir for Q = 0.05 m3/s. A stream-like flow is generally formed in the pool at a small distance between weirs (L/B = 0.61). The rotation cell under the jet forms clockwise between the two weirs. At the distances between the spillways (L/B = 1.22), the transition regime of the flow is formed. The transition regime occurs when or shortly after the weir is flooded. The rotation cell under the jet is clockwise smaller than the flow regime and larger than the submergence regime. At a distance L/B = 1.83, a plunging flow is formed so that the plunging jet dips into the pool and extends downstream to the center of the pool. The clockwise rotation of the cell is bounded by the dipping jet of the weir and is located between the bottom and the side walls of the weir and the channel.

figure 17
Fig. 17

Figure 18 shows the average depth velocity bar graph for each weir at different bed slopes and with and without orifice plates. As the distance between weirs increases, all models’ average depth velocity increases. As the slope of the bottom increases and an orifice plate is present, the average depth velocity in the pool increases. In addition, the average pool depth velocity increases as the discharge increases. Among the models, Model A’s average depth velocity is higher than Model B’s. The variation in velocity ranged from 8.11 to 12.24% for the models without an orifice plate and from 10.26 to 16.87% for the models with an orifice plate.

figure 18
Fig. 18

3.4 Turbulence Characteristics

The turbulent kinetic energy is one of the important parameters reflecting the turbulent properties of the flow field [23]. When the k value is high, more energy and a longer transit time are required to migrate the target species. The turbulent kinetic energy is defined as follows:

�=12(�x′2+�y′2+�z′2)

(15)

where ux, uy, and uz are fluctuating velocities in the x, y, and z directions, respectively. An illustration of the TKE and the effects of the geometric arrangement of the weir and the presence of an opening in the weir is shown in Fig. 19. For a given bed slope, in Model A, the highest TKE values are uniformly distributed in the weir’s upstream portion in the channel’s cross section. In contrast, for the rectangular labyrinth weir (Model B), the highest TKE values are concentrated on the sides of the pool between the crest of the weir and the channel wall. The highest TKE value in Models A and B is 0.224 and 0.278 J/kg, respectively, at the highest bottom slope (S0 = 10%). In the downstream portion of the conventional weir and within the crest of the weir and the walls of the rectangular labyrinth, there was a much lower TKE value that provided the best conditions for fish to recover in the pool between the weirs. The average of the lowest TKE for bottom slopes of 5 and 10% in Model A is 0.041 and 0.056 J/kg, and for Model B, is 0.047 and 0.064 J/kg. The presence of an opening in the weirs reduces the area of the highest TKE within the pool. It also increases the resting areas for fish (lower TKE). The highest TKE at the highest bottom slope in Models A and B with an orifice is 0.208 and 0.191 J/kg, respectively.

figure 19
Fig. 19

Figure 20 shows the effect of slope on the longitudinal distribution of TKE in the pools. TKE values significantly increase for a given discharge with an increasing bottom slope. Thus, for a low bed slope (S0 = 5%), a large pool area has expanded with average values of 0.131 and 0.168 J/kg for low and high discharge, respectively. For a bed slope of S0 = 10%, the average TKE values are 0.176 and 0.234 J/kg. Furthermore, as the discharge increases, the area with high TKE values within the pool increases. Lower TKE values are observed at the apex of the labyrinth weir, at the corner of the wall downstream of the weir, and between the side walls of the weir and the channel wall for both bottom slopes. The effect of distance between weirs on TKE is shown in Fig. 21. Low TKE values were observed at low discharge and short distances between weirs. Low TKE values are located at the top of the rectangular labyrinth weir and the downstream corner of the weir wall. There is a maximum value of TKE at the large distances between weirs, L/B = 1.83, along the center line of the pool, where the dip jet meets the bottom of the bed. At high discharge, the maximum TKE value for the distance L/B = 0.61, 1.22, and 1.83 was 0.246, 0.322, and 0.417 J/kg, respectively. In addition, the maximum TKE range increases with the distance between weirs.

figure 20
Fig. 20
figure 21
Fig. 21

For TKE size, the average value (TKEave) is plotted against q in Fig. 22. For all models, the TKE values increase with increasing q. For example, in models A and B with L/B = 0.61 and a slope of 10%, the TKE value increases by 41.66 and 86.95%, respectively, as q increases from 0.1 to 0.27 m2/s. The TKE values in Model B are higher than Model A for a given discharge, bed slope, and weir distance. The TKEave in Model B is higher compared to Model A, ranging from 31.46 to 57.94%. The presence of an orifice in the weir reduces the TKE values in both weirs. The intensity of the reduction is greater in Model B. For example, in Models A and B with L/B = 0.61 and q = 0.1 m2/s, an orifice reduces TKEave values by 60.35 and 19.04%, respectively. For each model, increasing the bed slope increases the TKEave values in the pool. For example, for Model B with q = 0.18 m2/s, increasing the bed slope from 5 to 10% increases the TKEave value by 14.34%. Increasing the distance between weirs increases the TKEave values in the pool. For example, in Model B with S0 = 10% and q = 0.3 m2/s, the TKEave in the pool increases by 34.22% if you increase the distance between weirs from L/B = 0.61 to L/B = 0.183.

figure 22
Fig. 22

Cotel et al. [24] suggested that turbulence intensity (TI) is a suitable parameter for studying fish swimming performance. Figure 23 shows the plot of TI and the effects of the geometric arrangement of the weir and the presence of an orifice. In Model A, the highest TI values are found upstream of the weirs and are evenly distributed across the cross section of the channel. The TI values increase as you move upstream to downstream in the pool. For the rectangular labyrinth weir, the highest TI values were concentrated on the sides of the pool, between the top of the weir and the side wall of the channel, and along the top of the weir. Downstream of the conventional weir, within the apex of the weir, and at the corners of the walls of the rectangular labyrinth weir, the percentage of TI was low. At the highest discharge, the average range of TI in Models A and B was 24–45% and 15–62%, respectively. The diversity of TI is greater in the rectangular labyrinth weir than the conventional weir. Fish swimming performance is reduced due to higher turbulence intensity. However, fish species may prefer different disturbance intensities depending on their swimming abilities; for example, Salmo trutta prefers a disturbance intensity of 18–53% [25]. Kupferschmidt and Zhu [26] found a higher range of TI for fishways, such as natural rock weirs, of 40–60%. The presence of an orifice in the weir increases TI values within the pool, especially along the middle portion of the cross section of the fishway. With an orifice in the weir, the average range of TI in Models A and B was 28–59% and 22–73%, respectively.

figure 23
Fig. 23

The effect of bed slope on TI variation is shown in Fig. 24. TI increases in different pool areas as the bed slope increases for a given discharge. For a low bed slope (S0 = 5%), a large pool area has increased from 38 to 63% and from 56 to 71% for low and high discharge, respectively. For a bed slope of S0 = 10%, the average values of TI are 45–67% and 61–73% for low and high discharge, respectively. Therefore, as runoff increases, the area with high TI values within the pool increases. A lower TI is observed for both bottom slopes in the corner of the wall, downstream of the crest walls, and between the side walls in the weir and channel. Figure 25 compares weir spacing with the distribution of TI values within the pool. The TI values are low at low flows and short distances between weirs. A maximum value of TI occurs at long spacing and where the plunging stream impinges on the bed and the area around the bed. TI ranges from 36 to 57%, 58–72%, and 47–76% for the highest flow in a wide pool area for L/B = 0.61, 1.22, and 1.83, respectively.

figure 24
Fig. 24
figure 25
Fig. 25

The average value of turbulence intensity (TIave) is plotted against q in Fig. 26. The increase in TI values with the increase in q values is seen in all models. For example, the average values of TI for Models A and B at L/B = 0.61 and slope of 10% increased from 23.9 to 33.5% and from 42 to 51.8%, respectively, with the increase in q from 0.1 to 0.27 m2/s. For a given discharge, a given gradient, and a given spacing of weirs, the TIave is higher in Model B than Model A. The presence of an orifice in the weirs increases the TI values in both types. For example, in Models A and B with L/B = 0.61 and q = 0.1 m2/s, the presence of an orifice increases TIave from 23.9 to 37.1% and from 42 to 48.8%, respectively. For each model, TIave in the pool increases with increasing bed slope. For Model B with q = 0.18 m2/s, TIave increases from 37.5 to 45.8% when you increase the invert slope from 5 to 10%. Increasing the distance between weirs increases the TIave in the pool. In Model B with S0 = 10% and q = 0.3 m2/s, the TIave in the pool increases from 51.8 to 63.7% as the distance between weirs increases from L/B = 0.61 to L/B = 0.183.

figure 26
Fig. 26

3.5 Energy Dissipation

To facilitate the passage of various target species through the pool of fishways, it is necessary to pay attention to the energy dissipation of the flow and to keep the flow velocity in the pool slow. The average volumetric energy dissipation (k) in the pool is calculated using the following basic formula:

�=����0��

(16)

where ρ is the water density, and H is the average water depth of the pool. The change in k versus Q for all models at two bottom slopes, S0 = 5%, and S0 = 10%, is shown in Fig. 27. Like the results of Yagci [8] and Kupferschmidt and Zhu [26], at a constant bottom slope, the energy dissipation in the pool increases with increasing discharge. The trend of change in k as a function of Q from the present study at a bottom gradient of S0 = 5% is also consistent with the results of Kupferschmidt and Zhu [26] for the fishway with rock weir. The only difference between the results is the geometry of the fishway and the combination of boulders instead of a solid wall. Comparison of the models shows that the conventional model has lower energy dissipation than the rectangular labyrinth for a given discharge. Also, increasing the distance between weirs decreases the volumetric energy dissipation for each model with the same bed slope. Increasing the slope of the bottom leads to an increase in volumetric energy dissipation, and an opening in the weir leads to a decrease in volumetric energy dissipation for both models. Therefore, as a guideline for volumetric energy dissipation, if the value within the pool is too high, the increased distance of the weir, the decreased slope of the bed, or the creation of an opening in the weir would decrease the volumetric dissipation rate.

figure 27
Fig. 27

To evaluate the energy dissipation inside the pool, the general method of energy difference in two sections can use:

ε=�1−�2�1

(17)

where ε is the energy dissipation rate, and E1 and E2 are the specific energies in Sects. 1 and 2, respectively. The distance between Sects. 1 and 2 is the same. (L is the distance between two upstream and downstream weirs.) Figure 28 shows the changes in ε relative to q (flow per unit width). The rectangular labyrinth weir (Model B) has a higher energy dissipation rate than the conventional weir (Model A) at a constant bottom gradient. For example, at S0 = 5%, L/B = 0.61, and q = 0.08 m3/s.m, the energy dissipation rate in Model A (conventional weir) was 0.261. In Model B (rectangular labyrinth weir), however, it was 0.338 (22.75% increase). For each model, the energy dissipation rate within the pool increases as the slope of the bottom increases. For Model B with L/B = 1.83 and q = 0.178 m3/s.m, the energy dissipation rate at S0 = 5% and 10% is 0.305 and 0.358, respectively (14.8% increase). Figure 29 shows an orifice’s effect on the pools’ energy dissipation rate. With an orifice in the weir, both models’ energy dissipation rates decreased. Thus, the reduction in energy dissipation rate varied from 7.32 to 9.48% for Model A and from 8.46 to 10.57 for Model B.

figure 28
Fig. 28
figure 29
Fig. 29

4 Discussion

This study consisted of entirely of numerical analysis. Although this study was limited to two weirs, the hydraulic performance and flow characteristics in a pooled fishway are highlighted by the rectangular labyrinth weir and its comparison with the conventional straight weir. The study compared the numerical simulations with laboratory experiments in terms of surface profiles, velocity vectors, and flow characteristics in a fish ladder pool. The results indicate agreement between the numerical and laboratory data, supporting the reliability of the numerical model in capturing the observed phenomena.

When the configuration of the weir changes to a rectangular labyrinth weir, the flow characteristics, the maximum and minimum area, and even the location of each hydraulic parameter change compared to a conventional weir. In the rectangular labyrinth weir, the flow is gradually directed to the sides as it passes the weir. This increases the velocity at the sides of the channel [21]. Therefore, the high-velocity area is located on the sides. In the downstream apex of the weir, the flow velocity is low, and this area may be suitable for swimming target fish. However, no significant change in velocity was observed at the conventional weir within the fish ladder. This resulted in an average increase in TKE of 32% and an average increase in TI of about 17% compared to conventional weirs.

In addition, there is a slight difference in the flow regime for both weir configurations. In addition, the rectangular labyrinth weir has a higher energy dissipation rate for a given discharge and constant bottom slope than the conventional weir. By reducing the distance between the weirs, this becomes even more intense. Finally, the presence of an orifice in both configurations of the weir increased the flow velocity at the orifice and in the middle of the pool, reducing the highest TKE value and increasing the values of TI within the pool of the fish ladder. This resulted in a reduction in volumetric energy dissipation for both weir configurations.

The results of this study will help the reader understand the direct effects of the governing geometric parameters on the hydraulic characteristics of a fishway with a pool and weir. However, due to the limited configurations of the study, further investigation is needed to evaluate the position of the weir’s crest on the flow direction and the difference in flow characteristics when combining boulders instead of a solid wall for this type of labyrinth weir [26]. In addition, hydraulic engineers and biologists must work together to design an effective fishway with rectangular labyrinth configurations. The migration habits of the target species should be considered when designing the most appropriate design [27]. Parametric studies and field observations are recommended to determine the perfect design criteria.

The current study focused on comparing a rectangular labyrinth weir with a conventional straight weir. Further research can explore other weir configurations, such as variations in crest position, different shapes of labyrinth weirs, or the use of boulders instead of solid walls. This would help understand the influence of different geometric parameters on hydraulic characteristics.

5 Conclusions

A new layout of the weir was evaluated, namely a rectangular labyrinth weir compared to a straight weir in a pool and weir system. The differences between the weirs were highlighted, particularly how variations in the geometry of the structures, such as the shape of the weir, the spacing of the weir, the presence of an opening at the weir, and the slope of the bottom, affect the hydraulics within the structures. The main findings of this study are as follows:

  • The calculated dimensionless discharge (Qt*) confirmed three different flow regimes: when the corresponding range of Qt* is smaller than 0.6, the regime of plunging flow occurs for values of L/B = 1.83. (L: distance of the weir; B: channel width). When the corresponding range of Qt* is greater than 0.5, transitional flow occurs at L/B = 1.22. On the other hand, if Qt* is greater than 1, the streaming flow is at values of L/B = 0.61.
  • For the conventional weir and the rectangular labyrinth weir with the plunging flow, it can be assumed that the discharge (Q) is proportional to 1.56 and 1.47h, respectively (h: water depth above the weir). This information is useful for estimating the discharge based on water depth in practical applications.
  • In the rectangular labyrinth weir, the high-velocity zone is located on the side walls between the top of the weir and the channel wall. A high-velocity variation within short distances of the weir. Low velocity occurs within the downstream apex of the weir. This area may be suitable for swimming target fish.
  • As the distance between weirs increased, the zone of maximum velocity increased. However, the zone of low speed decreased. The prevailing maximum velocity for a rectangular labyrinth weir at L/B = 0.61, 1.22, and 1.83 was 1.46, 1.65, and 1.84 m/s, respectively. The low mean velocities for these distances were 0.27, 0.44, and 0.72 m/s, respectively. This finding highlights the importance of weir spacing in determining the flow characteristics within the fishway.
  • The presence of an orifice in the weir increased the flow velocity at the orifice and in the middle of the pool, especially in a conventional weir. The increase ranged from 7.7 to 12.48%.
  • For a given bottom slope, in a conventional weir, the highest values of turbulent kinetic energy (TKE) are uniformly distributed in the upstream part of the weir in the cross section of the channel. In contrast, for the rectangular labyrinth weir, the highest TKE values were concentrated on the sides of the pool between the crest of the weir and the channel wall. The highest TKE value for the conventional and the rectangular labyrinth weir was 0.224 and 0.278 J/kg, respectively, at the highest bottom slope (S0 = 10%).
  • For a given discharge, bottom slope, and weir spacing, the average values of TI are higher for the rectangular labyrinth weir than for the conventional weir. At the highest discharge, the average range of turbulence intensity (TI) for the conventional and rectangular labyrinth weirs was between 24 and 45% and 15% and 62%, respectively. This reveals that the rectangular labyrinth weir may generate more turbulent flow conditions within the fishway.
  • For a given discharge and constant bottom slope, the rectangular labyrinth weir has a higher energy dissipation rate than the conventional weir (22.75 and 34.86%).
  • Increasing the distance between weirs decreased volumetric energy dissipation. However, increasing the gradient increased volumetric energy dissipation. The presence of an opening in the weir resulted in a decrease in volumetric energy dissipation for both model types.

Availability of data and materials

Data is contained within the article.

References

  1. Katopodis C (1992) Introduction to fishway design, working document. Freshwater Institute, Central Arctic Region
  2. Marriner, B.A.; Baki, A.B.M.; Zhu, D.Z.; Thiem, J.D.; Cooke, S.J.; Katopodis, C.: Field and numerical assessment of turning pool hydraulics in a vertical slot fishway. Ecol. Eng. 63, 88–101 (2014). https://doi.org/10.1016/j.ecoleng.2013.12.010Article Google Scholar 
  3. Dasineh, M.; Ghaderi, A.; Bagherzadeh, M.; Ahmadi, M.; Kuriqi, A.: Prediction of hydraulic jumps on a triangular bed roughness using numerical modeling and soft computing methods. Mathematics 9, 3135 (2021)Article Google Scholar 
  4. Silva, A.T.; Bermúdez, M.; Santos, J.M.; Rabuñal, J.R.; Puertas, J.: Pool-type fishway design for a potamodromous cyprinid in the Iberian Peninsula: the Iberian barbel—synthesis and future directions. Sustainability 12, 3387 (2020). https://doi.org/10.3390/su12083387Article Google Scholar 
  5. Santos, J.M.; Branco, P.; Katopodis, C.; Ferreira, T.; Pinheiro, A.: Retrofitting pool-and-weir fishways to improve passage performance of benthic fishes: effect of boulder density and fishway discharge. Ecol. Eng. 73, 335–344 (2014). https://doi.org/10.1016/j.ecoleng.2014.09.065Article Google Scholar 
  6. Ead, S.; Katopodis, C.; Sikora, G.; Rajaratnam, N.J.J.: Flow regimes and structure in pool and weir fishways. J. Environ. Eng. Sci. 3, 379–390 (2004)Article Google Scholar 
  7. Guiny, E.; Ervine, D.A.; Armstrong, J.D.: Hydraulic and biological aspects of fish passes for Atlantic salmon. J. Hydraul. Eng. 131, 542–553 (2005)Article Google Scholar 
  8. Yagci, O.: Hydraulic aspects of pool-weir fishways as ecologically friendly water structure. Ecol. Eng. 36, 36–46 (2010). https://doi.org/10.1016/j.ecoleng.2009.09.007Article Google Scholar 
  9. Dizabadi, S.; Hakim, S.S.; Azimi, A.H.: Discharge characteristics and structure of flow in labyrinth weirs with a downstream pool. Flow Meas. Instrum. 71, 101683 (2020). https://doi.org/10.1016/j.flowmeasinst.2019.101683Article Google Scholar 
  10. Kim, J.H.: Hydraulic characteristics by weir type in a pool-weir fishway. Ecol. Eng. 16, 425–433 (2001). https://doi.org/10.1016/S0925-8574(00)00125-7Article Google Scholar 
  11. Zhong, Z.; Ruan, T.; Hu, Y.; Liu, J.; Liu, B.; Xu, W.: Experimental and numerical assessment of hydraulic characteristic of a new semi-frustum weir in the pool-weir fishway. Ecol. Eng. 170, 106362 (2021). https://doi.org/10.1016/j.ecoleng.2021.106362Article Google Scholar 
  12. Hirt, C.W.; Nichols, B.D.: Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 39, 201–225 (1981). https://doi.org/10.1016/0021-9991(81)90145-5Article Google Scholar 
  13. Roache, P.J.: Perspective: a method for uniform reporting of grid refinement studies. J. Fluids Eng. 1994(116), 405–413 (1994)Article Google Scholar 
  14. Guo, S.; Chen, S.; Huang, X.; Zhang, Y.; Jin, S.: CFD and experimental investigations of drag force on spherical leak detector in pipe flows at high Reynolds number. Comput. Model. Eng. Sci. 101(1), 59–80 (2014)Google Scholar 
  15. Ahmadi, M.; Kuriqi, A.; Nezhad, H.M.; Ghaderi, A.; Mohammadi, M.: Innovative configuration of vertical slot fishway to enhance fish swimming conditions. J. Hydrodyn. 34, 917–933 (2022). https://doi.org/10.1007/s42241-022-0071-yArticle Google Scholar 
  16. Ahmadi, M.; Ghaderi, A.; MohammadNezhad, H.; Kuriqi, A.; Di Francesco, S.J.W.: Numerical investigation of hydraulics in a vertical slot fishway with upgraded configurations. Water 13, 2711 (2021)Article Google Scholar 
  17. Celik, I.B.; Ghia, U.; Roache, P.J.; Freitas, C.J.J.: Procedure for estimation and reporting of uncertainty due to discretization in CFD applications. J. Fluids Eng. Trans. ASME (2008). https://doi.org/10.1115/1.2960953Article Google Scholar 
  18. Li, S.; Yang, J.; Ansell, A.: Evaluation of pool-type fish passage with labyrinth weirs. Sustainability (2022). https://doi.org/10.3390/su14031098Article Google Scholar 
  19. Ghaderi, A.; Dasineh, M.; Aristodemo, F.; Aricò, C.: Numerical simulations of the flow field of a submerged hydraulic jump over triangular macroroughnesses. Water 13(5), 674 (2021)Article Google Scholar 
  20. Branco, P.; Santos, J.M.; Katopodis, C.; Pinheiro, A.; Ferreira, M.T.: Pool-type fishways: two different morpho-ecological cyprinid species facing plunging and streaming flows. PLoS ONE 8, e65089 (2013). https://doi.org/10.1371/journal.pone.0065089Article Google Scholar 
  21. Baki, A.B.M.; Zhu, D.Z.; Harwood, A.; Lewis, A.; Healey, K.: Rock-weir fishway I: flow regimes and hydraulic characteristics. J. Ecohydraulics 2, 122–141 (2017). https://doi.org/10.1080/24705357.2017.1369182Article Google Scholar 
  22. Dizabadi, S.; Azimi, A.H.: Hydraulic and turbulence structure of triangular labyrinth weir-pool fishways. River Res. Appl. 36, 280–295 (2020). https://doi.org/10.1002/rra.3581Article Google Scholar 
  23. Faizal, W.M.; Ghazali, N.N.N.; Khor, C.Y.; Zainon, M.Z.; Ibrahim, N.B.; Razif, R.M.: Turbulent kinetic energy of flow during inhale and exhale to characterize the severity of obstructive sleep apnea patient. Comput. Model. Eng. Sci. 136(1), 43–61 (2023)Google Scholar 
  24. Cotel, A.J.; Webb, P.W.; Tritico, H.: Do brown trout choose locations with reduced turbulence? Trans. Am. Fish. Soc. 135, 610–619 (2006). https://doi.org/10.1577/T04-196.1Article Google Scholar 
  25. Hargreaves, D.M.; Wright, N.G.: On the use of the k–ε model in commercial CFD software to model the neutral atmospheric boundary layer. J. Wind Eng. Ind. Aerodyn. 95, 355–369 (2007). https://doi.org/10.1016/j.jweia.2006.08.002Article Google Scholar 
  26. Kupferschmidt, C.; Zhu, D.Z.: Physical modelling of pool and weir fishways with rock weirs. River Res. Appl. 33, 1130–1142 (2017). https://doi.org/10.1002/rra.3157Article Google Scholar 
  27. Romão, F.; Quaresma, A.L.; Santos, J.M.; Amaral, S.D.; Branco, P.; Pinheiro, A.N.: Multislot fishway improves entrance performance and fish transit time over vertical slots. Water (2021). https://doi.org/10.3390/w13030275Article Google Scholar 

Download references

원문보기
비선형 파력의 영향에 따른 잔해 언덕 방파제 형상의 효과에 대한 수치 분석

비선형 파력의 영향에 따른 잔해 언덕 방파제 형상의 효과에 대한 수치 분석

Numerical Analysis of the Effects of Rubble Mound Breakwater Geometry Under the Effect of Nonlinear Wave Force

Arabian Journal for Science and EngineeringAims and scopeSubmit manuscript

Cite this article

Abstract

Assessing the interaction of waves and porous offshore structures such as rubble mound breakwaters plays a critical role in designing such structures optimally. This study focused on the effect of the geometric parameters of a sloped rubble mound breakwater, including the shape of the armour, method of its arrangement, and the breakwater slope. Thus, three main design criteria, including the wave reflection coefficient (Kr), transmission coefficient (Kt), and depreciation wave energy coefficient (Kd), are discussed. Based on the results, a decrease in wavelength reduced the Kr and increased the Kt and Kd. The rubble mound breakwater with the Coreloc armour layer could exhibit the lowest Kr compared to other armour geometries. In addition, a decrease in the breakwater slope reduced the Kr and Kd by 3.4 and 1.25%, respectively. In addition, a decrease in the breakwater slope from 33 to 25° increased the wave breaking height by 6.1% on average. Further, a decrease in the breakwater slope reduced the intensity of turbulence depreciation. Finally, the armour geometry and arrangement of armour layers on the breakwater with its different slopes affect the wave behaviour and interaction between the wave and breakwater. Thus, layering on the breakwater and the correct use of the geometric shapes of the armour should be considered when designing such structures.

파도와 잔해 더미 방파제와 같은 다공성 해양 구조물의 상호 작용을 평가하는 것은 이러한 구조물을 최적으로 설계하는 데 중요한 역할을 합니다. 본 연구는 경사진 잔해 둔덕 방파제의 기하학적 매개변수의 효과에 초점을 맞추었는데, 여기에는 갑옷의 형태, 배치 방법, 방파제 경사 등이 포함된다. 따라서 파동 반사 계수(Kr), 투과 계수(Kt) 및 감가상각파 에너지 계수(Kd)에 대해 논의합니다. 결과에 따르면 파장이 감소하면 K가 감소합니다.r그리고 K를 증가시켰습니다t 및 Kd. Coreloc 장갑 층이 있는 잔해 언덕 방파제는 가장 낮은 K를 나타낼 수 있습니다.r 다른 갑옷 형상과 비교했습니다. 또한 방파제 경사가 감소하여 K가 감소했습니다.r 및 Kd 각각 3.4%, 1.25% 증가했다. 또한 방파제 경사가 33°에서 25°로 감소하여 파도 파쇄 높이가 평균 6.1% 증가했습니다. 또한, 방파제 경사의 감소는 난류 감가상각의 강도를 감소시켰다. 마지막으로, 경사가 다른 방파제의 장갑 형상과 장갑 층의 배열은 파도 거동과 파도와 방파제 사이의 상호 작용에 영향을 미칩니다. 따라서 이러한 구조를 설계 할 때 방파제에 층을 쌓고 갑옷의 기하학적 모양을 올바르게 사용하는 것을 고려해야합니다.

Keywords

  • Rubble mound breakwater
  • Computational fluid dynamics
  • Armour layer
  • Wave reflection coefficient
  • Wave transmission coefficient
  • Wave energy dissipation coefficient

References

  1. Sollitt, C.K.; Cross, R.H.: Wave transmission through permeable breakwaters. In Coastal Engineering. pp. 1827–1846. (1973)
  2. Sulisz, W.: Wave reflection and transmission at permeable breakwaters of arbitrary cross-section. Coast. Eng. 9(4), 371–386 (1985)Article  Google Scholar 
  3. Kobayashi, N.; Wurjanto, A.: Numerical model for waves on rough permeable slopes. J. Coast. Res.149–166. (1990)
  4. Wurjanto, A.; Kobayashi, N.: Irregular wave reflection and runup on permeable slopes. J. Waterw. Port Coast. Ocean Eng. 119(5), 537–557 (1993)Article  Google Scholar 
  5. van Gent, M.R.: Numerical modelling of wave interaction with dynamically stable structures. In Coastal Engineering 1996. pp. 1930–1943. (1997)
  6. Liu, P.L.F.; Wen, J.: Nonlinear diffusive surface waves in porous media. J. Fluid Mech. 347, 119–139 (1997)Article  MathSciNet  MATH  Google Scholar 
  7. Troch, P.; De Rouck, J.: Development of two-dimensional numerical wave flume for wave interaction with rubble mound breakwaters. In Coastal Engineering. pp. 1638–1649. (1999)
  8. Liu, P.L.F.; Lin, P.; Chang, K.A.; Sakakiyama, T.: Numerical modeling of wave interaction with porous structures. J. Waterw. Port Coast. Ocean Eng. 125(6), 322–330 (1999)Article  Google Scholar 
  9. Abdolmaleki, K.; Thiagarajan, K.P.; Morris-Thomas, M.T.: Simulation of the dam break problem and impact flows using a Navier-Stokes solver. Simulation 13, 17 (2004)Google Scholar 
  10. Higuera, P.; Lara, J.L.; Losada, I.J.: Realistic wave generation and active wave absorption for Navier-Stokes models: application to OpenFOAM®. Coast. Eng. 71, 102–118 (2013)Article  Google Scholar 
  11. Higuera, P.; Lara, J.L.; Losada, I.J.: Three-dimensional interaction of waves and porous coastal structures using OpenFOAM®. Part II: application. Coast. Eng. 83, 259–270 (2014)Article  Google Scholar 
  12. Gui, Q.; Dong, P.; Shao, S.; Chen, Y.: Incompressible SPH simulation of wave interaction with porous structure. Ocean Eng. 110, 126–139 (2015)Article  Google Scholar 
  13. Dentale, F.; Donnarumma, G.; Carratelli, E.P.; Reale, F.: A numerical method to analyze the interaction between sea waves and rubble mound emerged breakwaters. WSEAS Trans. Fluid Mech 10, 106–116 (2015)Google Scholar 
  14. Dentale, F.; Reale, F.; Di Leo, A.; Carratelli, E.P.: A CFD approach to rubble mound breakwater design. Int. J. Naval Archit. Ocean Eng. 10(5), 644–650 (2018)Article  Google Scholar 
  15. Koley, S.: Wave transmission through multilayered porous breakwater under regular and irregular incident waves. Eng. Anal. Bound. Elem. 108, 393–401 (2019)Article  MathSciNet  MATH  Google Scholar 
  16. Koley, S.; Panduranga, K.; Almashan, N.; Neelamani, S.; Al-Ragum, A.: Numerical and experimental modeling of water wave interaction with rubble mound offshore porous breakwaters. Ocean Eng. 218, 108218 (2020)Article  Google Scholar 
  17. Pourteimouri, P.; Hejazi, K.: Development of an integrated numerical model for simulating wave interaction with permeable submerged breakwaters using extended Navier-Stokes equations. J. Mar. Sci. Eng. 8(2), 87 (2020)Article  Google Scholar 
  18. Cao, D.; Yuan, J.; Chen, H.: Towards modelling wave-induced forces on an armour layer unit of rubble mound coastal revetments. Ocean Eng. 239, 109811 (2021)Article  Google Scholar 
  19. Díaz-Carrasco, P.; Eldrup, M.R.; Andersen, T.L.: Advance in wave reflection estimation for rubble mound breakwaters: the importance of the relative water depth. Coast. Eng. 168, 103921 (2021)Article  Google Scholar 
  20. Vieira, F.; Taveira-Pinto, F.; Rosa-Santos, P.: Damage evolution in single-layer cube armoured breakwaters with a regular placement pattern. Coast. Eng. 169, 103943 (2021)Article  Google Scholar 
  21. Booshi, S.; Ketabdari, M.J.: Modeling of solitary wave interaction with emerged porous breakwater using PLIC-VOF method. Ocean Eng. 241, 110041 (2021)Article  Google Scholar 
  22. Aristodemo, F.; Filianoti, P.; Tripepi, G.; Gurnari, L.; Ghaderi, A.: On the energy transmission by a submerged barrier interacting with a solitary wave. Appl. Ocean Res. 122, 103123 (2022)Article  Google Scholar 
  23. Teixeira, P.R.; Didier, E.: Numerical analysis of performance of an oscillating water column wave energy converter inserted into a composite breakwater with rubble mound foundation. Ocean Eng. 278, 114421 (2023)Article  Google Scholar 
  24. Burgan, H.I.: Numerical modeling of structural irregularities on unsymmetrical buildings. Tehnički vjesnik 28(3), 856–861 (2021)Google Scholar 
  25. Jones, I.P.: CFDS-Flow3D user guide. (1994)
  26. Al Shaikhli, H.I.; Khassaf, S.I.: Stepped mound breakwater simulation by using flow 3D. Eurasian J. Eng. Technol. 6, 60–68 (2022)Google Scholar 
  27. Hirt, C.W.; Nichols, B.D.: Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 39(1), 201–225 (1981)Article  MATH  Google Scholar 
  28. Ghaderi, A.; Dasineh, M.; Aristodemo, F.; Aricò, C.: Numerical simulations of the flow field of a submerged hydraulic jump over triangular macroroughnesses. Water 13(5), 674 (2021)Article  Google Scholar 
  29. Yakhot, V.; Orszag, S.A.; Thangam, S.; Gatski, T.B.; Speziale, C.G.: Development of turbulence models for shear flows by a double expansion technique. Phys. Fluids A 4(7), 1510–1520 (1992)Article  MathSciNet  MATH  Google Scholar 
  30. Van der Meer, J.W.; Stam, C.J.M.: Wave runup on smooth and rock slopes of coastal structures. J. Waterw. Port Coast. Ocean Eng. 118(5), 534–550 (1992)Article  Google Scholar 
  31. Goda, Y.; Suzuki, Y. Estimation of incident and reflected waves in random wave experiments. In: ASCE, Proceedings of 15th International Conference on Coastal Engineering, (Honolulu, Hawaii). vol. 1, pp. 828–845. (1976)
  32. Zanuttigh, B.; Van der Meer, J.W.: Wave reflection from coastal structures. In: AA.VV., Proceedings of the XXX International Conference on Coastal Engineering, World Scientific, (San Diego, CA, USA, September 2006). pp. 4337–4349. (2006)
  33. Seelig W.N.; Ahrens J.P.: Estimation of wave reflection and energy dissipation coefficients for beaches, revetments, and breakwaters. CERC, Technical Paper, Fort Belvoir. vol. 81, p. 41 (1981)
  34. Mase, H.: Random wave runup height on gentle slope. J. Waterw. Port Coast. Ocean Eng. 115(5), 649–661 (1989)Article  Google Scholar 
원본 바로가기
Open Channels Flow에서의 콘크리트 캔버스 거동 연구

Study of Concrete Canvas Behavior in Open Channels Flow

Document Type : Research Paper

Authors

1 Imam Hosein Uni

2 Researcher of Imam Hossein University, Faculty of Engineering and Passive Defense

3 Ivanki University, Iran.

 10.22124/JCR.2023.24324.1618

Abstract

개방 수로의 심한 수력 구배는 침전으로 인한 심각한 침식과 문제를 일으킵니다. 패브릭 콘크리트는 기존의 콘크리트 표면을 대체할 수 있는 높은 실행 속도를 가진 독특한 제품입니다. 이 제품의 기계적 저항 매개변수에 따르면 부식 요인에 대한 우수한 내구성 외에도 직물 콘크리트의 응용 분야 중 하나는 운하 및 수로 암거 표면에 사용하는 것입니다. 이 연구에서는 먼저 사다리꼴 단면의 개방 채널 흐름을 직선 경로 상태의 3가지 공통 채널 형상, 굴곡 및 편차가 있는 경로, 마지막으로 채널 하단의 높이가 변경된 채널 경로를 포함하는 9가지 시나리오에서 시뮬레이션합니다. 각 주에서 flow-3d 소프트웨어를 사용한 흐름 난류 모델링과 함께 3개의 서로 다른 흐름 체제가 조사되었습니다.

FLOW-3D 소프트웨어를 사용한 유동 난류 모델링과 함께 다양한 유동 체제가 조사되었습니다. ABAQUS 소프트웨어를 사용하여 패브릭 콘크리트 구성요소와 연결 영역을 모델링하고, 콘크리트 표면과 취약한 연결 영역에 동일한 힘을 가하여 생성된 응력의 양을 확인했습니다. 결과는 생성된 응력이 직물 콘크리트의 인장 및 압축 응력 용량에 비해 매우 낮다는 것을 보여줍니다. 흐름과 콘크리트의 수력 연구를 검증하기 위해 관련 실험실 결과가 사용되었습니다.

Severe hydraulic gradients in open channels cause severe bed erosion and problems caused by sedimentation. Fabric concrete is a unique product with high execution speed that can replace traditional concrete surfaces. According to the mechanical resistance parameters of this product, in addition to its good durability against corrosive factors, one of the applications of fabric concrete is its use on the surface of canals and water course culverts. In this research, first, the flow of open channels in trapezoidal section is simulated under 9 scenarios, which include 3 common channel geometries in the state of a straight path, a path with bends and deviations, and finally, a channel path with a change in height at the bottom of the channel. In each of the states, 3 different flow regimes have been investigated along with flow turbulence modeling using flow-3d software.

Different flow regimes have been investigated along with flow turbulence modeling using flow-3d software. Using ABAQUS software, fabric concrete components and their connection areas have been modeled, and by applying forces equated to the concrete surface and vulnerable connection areas, the amount of created stresses has been checked. The results show that the created stresses are very low compared to the tensile and compressive stress capacity of fabric concrete. In order to validate the hydraulic studies of flow and concrete, the relevant laboratory results have been used.

Keywords

Main Subjects

원본 보기 바로가기
Study on the critical sediment concentration determining the optimal transport capability of submarine sediment flows with different particle size composition

Study on the critical sediment concentration determining the optimal transport capability of submarine sediment flows with different particle size composition

Yupeng Ren abc, Huiguang Zhou cd, Houjie Wang ab, Xiao Wu ab, Guohui Xu cd, Qingsheng Meng cd

Abstract

해저 퇴적물 흐름은 퇴적물을 심해로 운반하는 주요 수단 중 하나이며, 종종 장거리를 이동하고 수십 또는 수백 킬로미터에 걸쳐 상당한 양의 퇴적물을 운반합니다. 그것의 강력한 파괴력은 종종 이동 과정에서 잠수함 유틸리티에 심각한 손상을 초래합니다.

퇴적물 흐름의 퇴적물 농도는 주변 해수와의 밀도차를 결정하며, 이 밀도 차이는 퇴적물 흐름의 흐름 능력을 결정하여 이송된 퇴적물의 최종 퇴적 위치에 영향을 미칩니다. 본 논문에서는 다양한 미사 및 점토 중량비(미사/점토 비율이라고 함)를 갖는 다양한 퇴적물 농도의 퇴적물 흐름을 수로 테스트를 통해 연구합니다.

우리의 테스트 결과는 특정 퇴적물 구성에 대해 퇴적물 흐름이 가장 빠르게 이동하는 임계 퇴적물 농도가 있음을 나타냅니다. 4가지 미사/점토 비율 각각에 대한 임계 퇴적물 농도와 이에 상응하는 최대 속도가 구해집니다. 결과는 점토 함량이 임계 퇴적물 농도와 선형적으로 음의 상관 관계가 있음을 나타냅니다.

퇴적물 농도가 증가함에 따라 퇴적물의 흐름 거동은 흐름 상태에서 붕괴된 상태로 변환되고 흐름 거동이 변화하는 두 탁한 현탁액의 유체 특성은 모두 Bingham 유체입니다.

또한 본 논문에서는 퇴적물 흐름 내 입자 배열을 분석하여 위에서 언급한 결과에 대한 미시적 설명도 제공합니다.

Submarine sediment flows is one of the main means for transporting sediment to the deep sea, often traveling long-distance and transporting significant volumes of sediment for tens or even hundreds of kilometers. Its strong destructive force often causes serious damage to submarine utilities on its course of movement. The sediment concentration of the sediment flow determines its density difference with the ambient seawater, and this density difference determines the flow ability of the sediment flow, and thus affects the final deposition locations of the transported sediment. In this paper, sediment flows of different sediment concentration with various silt and clay weight ratios (referred to as silt/clay ratio) are studied using flume tests. Our test results indicate that there is a critical sediment concentration at which sediment flows travel the fastest for a specific sediment composition. The critical sediment concentrations and their corresponding maximum velocities for each of the four silt/clay ratios are obtained. The results further indicate that the clay content is linearly negatively correlated with the critical sediment concentration. As the sediment concentration increases, the flow behaviors of sediment flows transform from the flow state to the collapsed state, and the fluid properties of the two turbid suspensions with changing flow behaviors are both Bingham fluids. Additionally, this paper also provides a microscopic explanation of the above-mentioned results by analyzing the arrangement of particles within the sediment flow.

Introduction

Submarine sediment flows are important carriers for sea floor sediment movement and may carry and transport significant volumes of sediment for tens or even hundreds of kilometers (Prior et al., 1987; Pirmez and Imran, 2003; Zhang et al., 2018). Earthquakes, storms, and floods may all trigger submarine sediment flow events (Hsu et al., 2008; Piper and Normark, 2009; Pope et al., 2017b; Gavey et al., 2017). Sediment flows have strong forces during the movement, which will cause great harm to submarine structures such as cables and pipelines (Pope et al., 2017a). It was first confirmed that the cable breaking event caused by the sediment flow occurred in 1929. The sediment flow triggered by the Grand Banks earthquake damaged 12 cables. According to the time sequence of the cable breaking, the maximum velocity of the sediment flow is as high as 28 m/s (Heezen and Ewing, 1952; Kuenen, 1952; Heezen et al., 1954). Subsequent research shows that the lowest turbidity velocity that can break the cable also needs to reach 19 m/s (Piper et al., 1988). Since then, there have been many damage events of submarine cables and oil and gas pipelines caused by sediment flows in the world (Hsu et al., 2008; Carter et al., 2012; Cattaneo et al., 2012; Carter et al., 2014). During its movement, the sediment flow will gradually deposit a large amount of sediment carried by it along the way, that is, the deposition process of the sediment flow. On the one hand, this process brings a large amount of terrestrial nutrients and other materials to the ocean, while on the other hand, it causes damage and burial to benthic organisms, thus forming the largest sedimentary accumulation on Earth – submarine fans, which are highly likely to become good reservoirs for oil and gas resources (Daly, 1936; Yuan et al., 2010; Wu et al., 2022). The study on sediment flows (such as, the study of flow velocity and the forces acting on seabed structures) can provide important references for the safe design of seabed structures, the protection of submarine ecosystems, and exploration of turbidity sediments related oil and gas deposits. Therefore, it is of great significance to study the movement of sediment flows.

The sediment flow, as a highly sediment-concentrated fluid flowing on the sea floor, has a dense bottom layer and a dilute turbulent cloud. Observations at the Monterey Canyon indicated that the sediment flow can maintain its movement over long distances if its bottom has a relatively high sediment concentration. This dense bottom layer can be very destructive along its movement path to any facilities on the sea floor (Paull et al., 2018; Heerema et al., 2020; Wang et al., 2020). The sediment flow mentioned in this research paper is the general term of sediment density flow.

The sediment flow, which occurs on the seafloor, has the potential to cause erosion along its path. In this process, the suspended sediment is replenished, allowing the sediment flow to maintain its continuous flow capacity (Zhao et al., 2018). The dynamic force of sediment flow movement stem from its own gravity and density difference with surrounding water. In cases that the gravity drive of the slope is absent (on a flat sea floor), the flow velocity and distance of sediment flows are essentially determined by the sediment composition and concentration of the sediment flows as previous studies have demonstrated. Ilstad et al. (2004) conducted underwater flow tests in a sloped tank and employed high speed video camera to perform particle tracking. The results indicated that the premixed sand-rich and clay-rich slurries demonstrated different flow velocity and flow behavior. Using mixed kaolinite(d50 = 6 μm) and silica flour(d50 = 9 μm) in three compositions with total volumetric concentration ranged 22% or 28%, Felix and Peakall (2006) carried out underwater flow tests in a 5° slope Perspex channel and found that the flow ability of sediment flows is different depending on sediment compositions and concentrations. Sumner et al. (2009) used annular flume experiments to investigate the depositional dynamics and deposits of waning sediment-laden flows, finding that decelerating fast flows with fixed sand content and variable mud content resulted in four different deposit types. Chowdhury and Testik (2011) used lock-exchange tank, and experimented the kaolin clay sediment flows in the concentration range of 25–350 g/L, and predicted the fluid mud sediment flows propagation characteristics, but this study focused on giving sediment flows propagate phase transition time parameters, and is limited to clay. Lv et al. (2017) found through experiments that the rheological properties and flow behavior of kaolin clay (d50 = 3.7 μm) sediment flows were correlated to clay concentrations. In the field monitoring conducted by Liu et al. (2023) at the Manila Trench in the South China Sea in 2021, significant differences in the velocity, movement distance, and flow morphology of turbidity currents were observed. These differences may be attributed to variations in the particle composition of the turbidity currents.

On low and gentle slopes, although sediment flow with sand as the main sediment composition moves faster, it is difficult to propagate over long distances because sand has greater settling velocity and subaqueous angle of repose. Whereas the sediment flows with silt and clay as main composition may maintain relatively stable currents. Although its movement speed is slow, it has the ability to propagate over long distances because of the low settling rate of the fine particles (Ilstad et al., 2004; Liu et al., 2023). In a field observation at the Gaoping submarine canyon, the sediments collected from the sediment flows exhibited grain size gradation and the sediment was mostly composed of silt and clay (Liu et al., 2012). At the largest deltas in the world, for instance, the Mississippi River Delta, the sediments are mainly composed of silt and clay, which generally distributed along the coast in a wide range and provided the sediment sources for further distribution. The sediment flows originated and transported sediment from the coast to the deep sea are therefore share the same sediment compositions as delta sediments. To study the sediment flows composed of silt and clay is of great importance.

The sediment concentration of the sediment flows determines the density difference between the sediment flows and the ambient water and plays a key role in its flow ability. For the sediment flow with sediment composed of silt and clay, low sediment concentration means low density and therefore leads to low flow ability; however, although high sediment concentration results in high density, since there is cohesion between fine particles, it changes fluid properties and leads to low flow ability as well. Therefore, there should be a critical sediment concentration with mixed composition of silt and clay, at which the sediment flow maintains its strongest flow capacity and have the highest movement speed. In other words, the two characteristics of particle diameter and concentration of the sediment flow determine its own motion ability, which, if occurs, may become the most destructive force to submarine structures.

The objectives of this work was to study how the sediment composition (measured in relative weight of silt and clay, and referred as silt/clay ratio) and sediment concentration affect flow ability and behavior of the sediment flows, and to quantify the critical sediment concentration at which the sediment flows reached the greatest flow velocity under the experiment setting. We used straight flume without slope and conducted a series of flume tests with varying sediment compositions (silt-rich or clay-rich) and concentrations (96 to 1212 g/L). Each sediment flow sample was tested and analyzed for rheological properties using a rheometer, in order to characterize the relationship between flow behavior and rheological properties. Combined with the particle diameter, density and viscosity characteristics of the sediment flows measured in the experiment, a numerical modeling study is conducted, which are mutually validated with the experimental results.

The sediment concentration determines the arrangements of the sediment particles in the turbid suspension, and the arrangement impacts the fluid properties of the turbid suspension. The microscopic mode of particle arrangement in the turbid suspension can be constructed to further analyze the relationship between the fluid properties of turbid suspension and the flow behaviors of the sediment flow, and then characterize the critical sediment concentration at which the sediment flow runs the fastest. A simplified microscopic model of particle arrangement in turbid suspension was constructed to analyze the microscopic arrangement characteristics of sediment particles in turbid suspension with the fastest velocity.

Section snippets

Equipment and materials

The sediment flows flow experiments were performed in a Perspex channel with smooth transparent walls. The layout and dimensions of the experimental set-up were shown in Fig. 1. The bottom of the channel was flat and straight, and a gate was arranged to separate the two tanks. In order to study the flow capacity of turbidity currents from the perspective of their own composition (particle size distribution and concentration), we used a straight channel instead of an inclined one, to avoid any

Relationship between sediment flow flow velocity and sediment concentration

After the sediment flow is generated, its movement in the first half (50 cm) of the channel is relatively stable, and there is obvious shock diffusion in the second half. The reason is that the excitation wave (similar to the surge) will be formed during the sediment flow movement, and its speed is much faster than the speed of the sediment flow head. When the excitation wave reaches the tail of the channel, it will be reflected, thus affecting the subsequent flow of the sediment flow.

Sediment flows motion simulation based on FLOW-3D

As a relatively mature 3D fluid simulation software, FLOW-3D can accurately predict the free surface flow, and has been used to simulate the movement process of sediment flows for many times (Heimsund, 2007). The model adopted in this paper is RNG turbulence model, which can better deal with the flow with high strain rate and is suitable for the simulation of sediment flows with variable shape during movement. The governing equations of the numerical model involved include continuity equation,

Conclusions

In this study, we conducted a series of sediment flow flume tests with mixed silt and clay sediment samples in four silt/clay ratios on a flat slope. Rheological measurements were carried out on turbid suspension samples and microstructure analysis of the sediment particle arrangements was conducted, we concluded that:

  • (1)The flow velocity of the sediment flow is controlled by the sediment concentration and its own particle diameter composition, the flow velocity increased with the increase of the

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Natural Science Foundation of China [Grant no. 42206055]; the National Natural Science Foundation of China [Grant no. 41976049]; and the National Natural Science Foundation of China [Grant no. 42272327].

References (39)

There are more references available in the full text version of this article.

원본 보기
Figure 1. Three-dimensional finite element model of local scouring of semi-exposed submarine cable.

반노출 해저케이블의 국부 정련과정 및 영향인자에 대한 수치적 연구

Numerical Study of the Local Scouring Process and Influencing Factors of Semi-Exposed Submarine Cables

by Qishun Li,Yanpeng Hao *,Peng Zhang,Haotian Tan,Wanxing Tian,Linhao Chen andLin Yang

School of Electric Power Engineering, South China University of Technology, Guangzhou 510640, China

*Author to whom correspondence should be addressed.J. Mar. Sci. Eng.2023, 11(7), 1349; https://doi.org/10.3390/jmse11071349

Received: 10 June 2023 / Revised: 19 June 2023 / Accepted: 27 June 2023 / Published: 1 July 2023(This article belongs to the Section Ocean Engineering)

일부 수식이 손상되어 표시될 수 있습니다. 이 경우 원문을 참조하시기 바랍니다.

Abstract

Local scouring might result in the spanning of submarine cables, endangering their mechanical and electrical properties. In this contribution, a three-dimensional computational fluid dynamics simulation model is developed using FLOW-3D, and the scouring process of semi-exposed submarine cables is investigated. The effects of the sediment critical Shields number, sediment density, and ocean current velocity on local scouring are discussed, and variation rules for the submarine cables’ spanning time are provided. The results indicate that three scouring holes are formed around the submarine cables. The location of the bottom of the holes corresponds to that of the maximum shear velocity. The continuous development of scouring holes at the wake position leads to the spanning of the submarine cables. The increase in the sediment’s critical Shields number and sediment density, as well as the decrease in the ocean current velocity, will extend the time for maintaining the stability of the upstream scouring hole and retard the development velocity of the wake position and downstream scouring holes. The spanning time has a cubic relationship with the sediment’s critical Shields number, a linear relationship with the sediment density, and an exponential relationship with the ocean current velocity. In this paper, the local scouring process of semi-exposed submarine cables is studied, which provides a theoretical basis for the operation and maintenance of submarine cables.

Keywords: 

submarine cable; local scouring; numerical simulation; computational fluid dynamics

1. Introduction

As a key piece of equipment in cross-sea power grids, submarine cables are widely used to connect autonomous power grids, supply power to islands or offshore platforms, and transmit electric power generated by marine renewable energy installations to onshore substations [1]. Once submarine cables break down due to natural disasters or human-made damage, the normal operation of other marine electric power equipment connected to them may be affected. These chain reactions will cause great economic losses and serious social impacts [2].

To protect submarine cables, they are usually buried 1 to 3 m below the seabed [3]. However, submarine cables are still confronted with potential threats from the complex subsea environment. Under the influence of fishing, anchor damage, ocean current scouring, and other factors, the sediment above submarine cables will always inevitably migrate. When a submarine cable is partially exposed, the scouring at this position will be exacerbated; eventually, it will cause the submarine cable to span. According to a field investigation of the 500 kV oil-filled submarine cable that is part of the Hainan networking system, the total length of the span is 49 m [4]. Under strong ocean currents, spanning submarine cables may experience vortex-induced vibrations. Fatigue stress caused by vortex-induced vibrations may lead to metal sheath rupture [5], which endangers the mechanical and electrical properties of submarine cables. Therefore, understanding the local scouring processes of partially exposed submarine cables is crucial for predicting scouring patterns. This is the basis for developing effective operation and maintenance strategies for submarine cables.

The mechanism and influencing factors of sediment erosion have been examined by researchers around the world. In 1988, Sumer [6] conducted experiments to show that the shedding vortex in the wake of a pipeline would increase the Shields parameter by 3–4 times, which would result in severe scouring. In 1991, Chiew [7] performed experiments to prove that the maximum scouring depth could be obtained when the pipeline was located on a flat bed and was scoured by a unidirectional water flow. Based on the test results, they provided a prediction formula for the maximum scouring depth. In 2003, Mastbergen [8] proposed a one-dimensional, steady-state numerical model of turbidity currents, which considered the negative pore pressures in the seabed. The calculated results of this model were basically consistent with the actual scouring of a submarine canyon. In 2007, Dey [9] presented a semitheoretical model for the computation of the maximum clear-water scour depth below underwater pipelines in uniform sediments under a steady flow, and the predicted scour depth in clear water satisfactorily agreed with the observed values. In 2008, Dey [10] conducted experiments on clear-water scour below underwater pipelines under a steady flow and obtained a variation pattern of the depth of the scouring hole. In 2008, Liang [11] used a two-dimensional numerical simulation to study the scouring process of a tube bundle under the action of currents and waves. They discovered that, compared with the scouring of a single tube, the scouring depth of the tube bundle was deeper, and the scouring time was longer. In 2012, Yang [12] found that placing rubber sheets under pipes can greatly accelerate their self-burial. The rubber sheets had the best performance when their length was about 1.5 times the size of the pipe. In 2020, Li [13] investigated the two-dimensional local scour beneath two submarine pipelines in tandem under wave-plus-current conditions via numerical simulation. They found that for conditions involving waves plus a low-strength current, the scour pattern beneath the two pipelines behaved like that in the pure-wave condition. Conversely, when the current had equal strength to the wave-induced flow, the scour pattern beneath the two pipelines resembled that in the pure-current condition. In 2020, Guan [14] studied and discussed the interactive coupling effects among a vibrating pipeline, flow field, and scour process through experiments, and the experimental data showed that the evolution of the scour hole had significant influences on the pipeline vibrations. In 2021, Liu [15] developed a two-dimensional finite element numerical model and researched the local scour around a vibrating pipeline. The numerical results showed that the maximum vibration amplitude of the pipeline could reach about 1.2 times diameter, and the maximum scour depth occurred on the wake side of the vibrating pipeline. In 2021, Huang [16] carried out two-dimensional numerical simulations to investigate the scour beneath a single pipeline and piggyback pipelines subjected to an oscillatory flow condition at a KC number of 11 and captured typical steady-streaming structures around the pipelines due to the oscillatory flow condition. In 2021, Cui [17] investigated the characteristics of the riverbed scour profile for a pipeline buried at different depths under the condition of riverbed sediments with different particle sizes. The results indicated that, in general, the equilibrium scour depth changed in a spoon shape with the gradual increase in the embedment ratio. In 2022, Li [18] used numerical simulation to study the influence of the burial depth of partially buried pipelines on the surrounding flow field, but they did not investigate the scour depth. In 2022, Zhu [19] performed experiments to prove that the scour hole propagation rate under a pipeline decreases with an increasing pipeline embedment ratio and rises with the KC number. In 2022, Najafzadeh [20] proposed equations for the prediction of the scouring propagation rate around pipelines due to currents based on a machine learning model, and the prediction results were consistent with the experimental data. In 2023, Ma [21] used the computational fluid dynamics coarse-grained discrete element method to simulate the scour process around a pipeline. The results showed that this method can effectively reduce the considerable need for computing resources and excessive computation time. In 2023, through numerical simulations, Hu [22] discovered that the water velocity and the pipeline diameter had a significant effect on the depth of scouring.

In the preceding works, the researchers investigated the mechanism of sediment scouring and the effect of various factors on the local scouring of submarine pipelines. However, submarine cables are buried beneath the seabed, while submarine pipelines are erected above the seabed. The difference in laying methods leads to a large discrepancy between their local scouring processes. Therefore, the conclusions of the above investigations are not applicable to the local scouring of submarine cables. Currently, there is no report on the research of the local scouring of partially exposed submarine cables.

In this paper, a three-dimensional computational fluid dynamics (CFD) finite element model, based on two-phase flow, is established using FLOW-3D. The local scouring process of semi-exposed submarine cables under steady-state ocean currents is studied, and the variation rules of the depth and the shape of the scouring holes, as well as the shear velocity with time, are obtained. By setting different critical Shields numbers of the sediment, different sediment densities, and different ocean current velocities, the change rule of the scouring holes’ development rate and the time required for the spanning of submarine cables are explored.

2. Sediment Scouring Model

In the sediment scouring model, the sediment is set as the dispersed particle, which is regarded as a kind of quasifluid. In this context, sediment scouring is considered as a two-phase flow process between the liquid phase and solid particle phase. The sediment in this process is further divided into two categories: one is suspended in the fluid, and the other is deposited on the bottom.When the local Shields number of sediment is greater than the critical Shields number, the deposited sediment will be transformed into the suspended sediment under the action of ocean currents. The calculation formulae of the local Shields numbers θ and the critical Shields numbers 

θcr of sediment is given as [23,24

]

𝜃=𝑈2𝑓(𝜌𝑠/𝜌𝑓−1)𝑔𝑑50,�=��2(��/��−1)��50,(1)

𝜃𝑐𝑟=0.31+1.2𝐷∗+0.055(1−𝑒−0.02𝐷∗),���=0.31+1.2�*+0.055(1−�−0.02�*),(2)

𝐷∗=𝑑50𝜌𝑓(𝜌𝑠−𝜌𝑓)𝑔/𝜇2−−−−−−−−−−−−−−√3,�*=�50��(��−��)�/�23,(3)where 

Uf is the shearing velocity of bed surface, 

ρs is the density of the sediment particle, 

ρf is the fluid density, g is the acceleration of gravity, d

50 is the median size of sediment, and μ is the dynamic viscosity of sediment.And each sediment particle suspended in the fluid obeys the equations for mass conservation and energy conservation

∂𝑐𝑠∂𝑡+∇⋅(𝑢𝑐𝑠)=0,∂��∂�+∇⋅(�¯��)=0,(4)

∂𝑢𝑠∂𝑡+𝑢⋅∇𝑢𝑠=−1𝜌𝑠∇𝑃+𝐹−𝐾𝑓𝑠𝜌𝑠𝑢𝑟,∂��∂�+�¯⋅∇��=−1��∇�+�−�������,(5)where 

cs is the concentration of the sediment particle, 

𝑢�¯ is the mean velocity vector of the fluid and the sediment particle, 

us is the velocity of the sediment particle, 

fs is the volume fraction of the sediment particle, P is the pressure, F is the volumetric and viscous force, K is the drag force, and 

ur is the relative velocity.

3. Numerical Setup and Modeling

In this paper, a three-dimensional submarine cable local scouring simulation model is established by FLOW-3D. Based on the numerical simulation, the process of the submarine cable, which gradually changes from semi-exposed to the spanning state under the steady-state ocean current, is studied. The geometric modeling, the mesh division, the physical field setup, and the grid independent test of CFD numerical model are as follows.

3.1. Geometric Modeling and Mesh Division

A three-dimensional (3D) numerical model of the local scouring of a semi-exposed submarine cable is established, which is shown in Figure 1. The dimensions of the model are marked in Figure 1. The inlet direction of the ocean current is defined as the upstream of the submarine cable (referred to as upstream), and the outlet direction of the ocean current is defined as the downstream of the submarine cable (referred to as downstream).

Jmse 11 01349 g001 550

Figure 1. Three-dimensional finite element model of local scouring of semi-exposed submarine cable.

The submarine cable with a diameter of 0.2 m is positioned on sediment that is initially in a semi-exposed state. When the length of the span is short, the submarine cable will not show obvious deformation due to gravity or scouring from the ocean current. Therefore, the submarine cable surface is set as the fixed boundary. The model’s left boundary is set as the inlet, the right boundary is set as the outlet, the front and rear boundaries are set as symmetry, and the bottom boundary is set as the non-slip wall. Since the water depth above the submarine cable is more than 0.6 m in practice, the top boundary of the model is also set as symmetry. The sediment near the inlet and the outlet will be carried by ocean currents, which leads to the abnormal scouring terrain. At each end of the sediment, a baffle (thickness of 3 cm) is installed to ensure that the simulation results can reflect the real situation.

Due to the fact that the flow field around the semi-exposed submarine cable is not a simple two-dimensional symmetrical distribution, it should be solved by three-dimensional numerical simulation. Considering the accuracy and efficiency of the calculation, the size of mesh is set to 0.02 m. The total number of meshes after the dissection is 133,254.

3.2. Physical Field Setup

The CFD finite element model contains four physical field modules: sediment scouring module, gravity and non-inertial reference frame module, density evaluation module, and viscosity and turbulence module. In this paper, the renormalization group (RNG) k–ε turbulence model is used, which has high computational accuracy for turbulent vortices. Therefore, this turbulence model is suitable for calculating the sediment scouring process around the semi-exposed submarine cable [25]. The key parameters of the numerical simulation are referring to the survey results of submarine sediments in the Korean Peninsula [26], as listed in Table 1.Table 1. Key parameters of numerical simulation.

Table

3.3. Mesh Independent Test

In order to eliminate errors caused by the quantity of grids in the calculation process, two sizes of mesh are set on the validation model, and the scour profiles under different mesh sizes are compared. The validation model is shown in Figure 2, and the scouring terrain under different mesh size is given in Figure 3.

Jmse 11 01349 g002 550

Figure 2. Validation model.

Jmse 11 01349 g003 550

Figure 3. Scouring terrain under different mesh sizes.

It can be seen from Figure 3 that with the increase in the number of meshes, the scouring terrain of the verification model changes slightly, and the scouring depth is basically unchanged. Considering the accuracy of the numerical simulation and the calculation’s time cost, it is reasonable to consider setting the mesh size to 0.02 m.

4. Results and Analysis

4.1. Analysis of Local Scouring Process

Based on the CFD finite element numerical simulation, the local scouring process of the submarine cable under the steady-state ocean current is analyzed. The end time of the simulation is 9 h, the initial time step is 0.01 s, and the fluid velocity is 0.40 m/s. Simulation results are saved every minute. Figure 4 illustrates the scouring terrain around the semi-exposed submarine cable, which has been scoured by the steady-state current for 5 h.

Jmse 11 01349 g004 550

Figure 4. Scouring terrain around semi-exposed submarine cable (scour for 5 h).

As can be seen from Figure 4, three scouring holes were separately formed in the upstream wake position and downstream of the semi-exposed submarine cable. The scouring holes are labeled according to their locations. The variation of the scouring terrain around the semi-exposed submarine cable over time is given in Figure 5. The red circle in the picture corresponds to the position of the submarine cable, and the red box in the legend marks the time when the submarine cable is spanning.

Jmse 11 01349 g005 550

Figure 5. Variation of scouring terrain around semi-exposed submarine cable adapted to time.

From Figure 5, in the first hour of scouring, the upstream (−0.5 m to −0.1 m) and downstream (0.43 m to 1.5 m) scouring holes appeared. The upstream scouring hole was relatively flat with depth of 0.04 m. The depth of the downstream scouring hole increased with the increase in distance, and the maximum depth was 0.13 m. The scouring hole that developed at the wake position was very shallow, and its depth was only 0.007 m.

In the second hour of scouring, the upstream scouring hole’s depth remained nearly constant. The depth of the downstream scouring hole only increased by 0.002 m. The scouring hole at the wake position developed steadily, and its depth increased from 0.007 m to 0.014 m.

The upstream and downstream scouring holes did not continue to develop during the third to the sixth hour. Compared to the first two hours, the development of scouring holes at the wake position accelerated significantly, with an average growth rate of 0.028 m/h. The growth rate in the fifth hour of the scouring hole at the wake position was slightly faster than the other times. After 6 h of scouring, the sediment on the right side of the submarine cable had been hollowed out.

In the seventh and the eighth hour of scouring, the upstream scouring hole’s depth increased slightly, the downstream scouring hole still remained stable, and the depth of the scouring hole at wake position increased by 0.019 m. The sediment under the submarine cable was gradually eroded as well. By the end of the eighth hour, the lower right part of the submarine cable had been exposed to water as well.

At 8 h 21 min of the scouring, the submarine cable was completely spanned, and the scouring holes were connected to each other. Within the next 10 min, the development of the scouring holes sped up significantly, and the maximum depth of scouring holes increased greatly to 0.27 m.

In reference [17], researchers have studied the local scouring process of semi-buried pipelines in sandy riverbeds through experiments. The test results show that the scouring process can be divided into a start-up stage, micropore formation stage, extension stage, and equilibrium stage. In this paper, the first three stages are simulated, and the results are in good agreement with the experiment, which proves the accuracy of the present numerical model.

In this research, the velocity of ocean currents at the sediment surface is defined as the shear velocity, which plays an important role in the process of local scouring. Figure 6 provides visual data on how the shear velocity varies over time.

Jmse 11 01349 g006 550

Figure 6. Shear velocity changes in the scouring process.

The semi-exposed submarine cable protrudes from the seabed, which makes the shear velocity of its surface much higher than other locations. After the submarine cable is spanned, the shear velocity of the scouring hole surface below it is taken. This is the reason for the sudden change of shear velocity at the submarine cable’s location in Figure 6.The shear velocity in the initial state of the upstream scouring hole is obviously greater than in subsequent times. After 1 h of scouring, the shear velocity in the upstream scouring hole rapidly decreased from 1.1 × 10

−2 m/s to 3.98 × 10

−3 m/s and remained stable until the end of the sixth hour. This phenomenon explains why the upstream scouring hole developed rapidly in the first hour but remained stable for the following 5 h.The shear velocity in the downstream scouring hole reduced at first and then increased; its initial value was 1.41 × 10

−2 m/s. It took approximately 5 h for the shear velocity to stabilize, and the stable shear velocity was 2.26 × 10

−3 m/s. Therefore, compared with the upstream scouring hole, the downstream scouring hole was deeper and required more time to reach stability.The initial shear velocity in the scouring hole at the wake position was only 7.1 × 10

−3 m/s, which almost does not change in the first hour. This leads to a very slow development of the scouring hole at the wake position in the early stages. The maximum shear velocity in this scouring hole gradually increased to 1.05 × 10

−2 m/s from the second to the fifth hour, and then decreased to 6.61 × 10

−3 m/s by the end of the eighth hour. This is why the scouring hole at the wake position grows fastest around the fifth hour. Consistent with the pattern of change in the scouring hole’s terrain, the location of the maximal shear velocity also shifted to the right with time.

The shear velocity of all three scouring holes rose dramatically in the last hour. Combined with the terrain in Figure 5, this can be attributed to the complete spanning of the submarine cable.

From Equations (3)–(5), one can see the movement of the sediment is related directly with the sediment’s critical Shields number, sediment density, and ocean current velocity. Based on the parameters in Table 1, the influence of the above parameters on the local scouring process of semi-exposed submarine cables will be discussed.

4.2. Influence Factors

4.2.1. Sediment’s Critical Shields Number

The sediment’s critical Shields number 

θcr is set as 0.02, 0.03, 0.04, 0.05, 0.06, and 0.07, and the variations of scouring terrain over time under each 

θcr are displayed in Figure 7.

Jmse 11 01349 g007 550

Figure 7. Influence of sediment’s critical Shields number 

θcr on local scouring around semi-exposed submarine cable: (a

) θcr = 0.02; (b

) θcr = 0.03; (c

) θcr = 0.04; (d

) θcr = 0.05; (e

) θcr = 0.06; and (f

) θcr = 0.07.From Figure 7, one can see that a change in 

θcr will affect the depth of the upstream scouring hole and the development speed of the scouring hole at the wake position, but it will have no significant impact on the expansion of the downstream scouring hole.Under conditions of different 

θcr, the upstream scouring hole will reach a temporary plateau within 1 h, at which time the stable depth will be about 0.04 m. When 

θcr ≤ 0.05, the upstream scouring hole will continue to expand after a few hours. The stable time is obviously affected by 

θcr, which will gradually increase from 1 h to 11 h with the increase in 

θcr. The terrain of the upstream scouring hole will gradually convert to deep on the left and to shallow on the right. Since the scouring hole at the wake position has not been stable, its state at the time of submarine cable spanning is studied emphatically. In the whole process of scouring, the scouring hole at the wake position continues to develop and does not reach a stable state. With the increase in 

θcr, the development velocity of the scouring hole at the wake position will decrease considerably. Its average evolution velocity decreases from 3.88 cm/h to 1.62 cm/h, and its depth decreases from 21.9 cm to 18.8 cm. Under the condition of each 

θcr, the downstream scouring hole will stabilize within 1 h, and the stable depth will be basically unchanged (all about 13.5 cm).As 

θcr increases, so does the sediment’s ability to withstand shearing forces, which will cause it to become increasingly difficult to be eroded or carried away by ocean currents. This effect has been directly reflected in the depth of scouring holes (upstream and wake position). Due to the blocking effect of semi-exposed submarine cables, the wake is elongated, which is why the downstream scouring hole develops before the scouring hole at the wake position and quickly reaches a stable state. However, due to the high wake intensity, this process is not significantly affected by the change of 

θcr.

4.2.2. Sediment Density

The density of sediment 

ρs is set as 1550 kg/m

3, 1600 kg/m

3, 1650 kg/m

3, 1700 kg/m

3, 1750 kg/m

3, and 1800 kg/m

3, and the variation of scouring terrain over time under each 

ρs are displayed in Figure 8.

Jmse 11 01349 g008 550

Figure 8. Influence of sediment density 

ρs on local scouring around semi-exposed submarine cable: (a

) ρs = 1550 kg/m

3; (b) ρs = 1600 kg/m

3; (c) ρs = 1650 kg/m

3; (d) ρs = 1700 kg/m

3; (e) ρs = 1750 kg/m

3; and (f

) ρs = 1800 kg/m

3.From Figure 8, one can see that a change in 

ρs will also affect the depth of the upstream scouring hole and the development speed of the scouring hole at the wake position. In addition, it can even have an impact on the downstream scouring hole depth.Under different 

ρs conditions, the upstream scouring hole will always reach a temporary stable state in 1 h, at which time the stable depth will be 0.04 m. When 

ρs ≤ 1750 kg/m

3, the upstream scouring hole will continue to expand after a few hours. The stabilization time of upstream scouring hole is more clearly affected by 

ρs, which will gradually increase from 3 h to 13 h with the increase in 

ρs. The terrain of the upstream scouring hole will gradually change to deep on the left and to shallow on the right. Since the scouring hole at the wake position has not been stable, its state at the time of the submarine cable spanning is studied emphatically, too. In the whole process of scouring, the scouring hole at the wake position continues to develop and does not reach a stable state. When 

ρs is large, the development rate of scouring hole obviously decreased with time. With the increase in 

ρs, the development velocity of the scouring hole at the wake position reduces from 3.38 cm/h to 1.14 cm/h, and the depth of this scouring hole declines from 20 cm to 15 cm. As 

ρs increases, the stabilization time of the downstream scouring hole increases from less than 1 h to about 2 h, but the stabilization depth of the downstream scouring hole remains essentially the same (all around 13.5 cm).As can be seen from Equation (1), the increase in 

ρs will reduce the Shields number, thus weakening the shear action of the sediment by the ocean current, which explains the extension of the stability time of the upstream scouring hole. At the same time, with the increase in the depth of scouring hole at the wake position, its shear velocity will decreases. Therefore, under a larger 

ρs value, the development speed of scouring hole at the wake position will decrease significantly with time. Possibly for the same reason, 

ρs can affect the development rate of downstream scouring hole.

4.2.3. Ocean Current Velocity

The ocean current velocity v is set as 0.35 m/s, 0.40 m/s, 0.45 m/s, 0.50 m/s, 0.55 m/s, and 0.60 m/s. Figure 9 presents the variation in scouring terrain with time for each v.

Jmse 11 01349 g009 550

Figure 9. Influence of ocean current velocity v on local scouring around semi-exposed submarine cable: (a) v = 0.35 m/s; (b) v = 0.40 m/s; (c) v = 0.45 m/s; (d) v = 0.50 m/s; (e) v = 0.55 m/s; and (f) v = 0.60 m/s.

Changes in v affect the depth of the upstream and downstream scouring holes, as well as the development velocity of the wake position and downstream scouring holes.

When v ≤ 0.45 m/s, the upstream scouring hole will reach a temporary stable state within 1 h, at which point the stable depth will be 0.04 m. The stabilization time of the upstream scouring hole is affected by v, which will gradually decrease from 15 h to 3 h with the increase in v. When v > 0.45 m/s, the upstream scouring hole is going to expand continuously. With the increase in v, its average development velocity increases from 6.68 cm/h to 8.66 cm/h, and its terrain changes to deep on the left and to shallow on the right. When the submarine cable is spanning, special attention should be paid to the depth of the scouring hole at the wake position. Throughout whole scouring process, the scouring hole at the wake position continues to develop and does not reach a stable state. With the increase in v, the depth of scouring hole at the wake position will increase from 14 cm to 20 cm, and the average development velocity will increase from 0.91 cm/h to 10.43 cm/h. As v increases, the time required to stabilize the downstream scouring hole is shortened from 1to 2 h to less than 1 h, but the stable depth is remains nearly constant at 13.5 cm.

An increase in v will increase the shear velocity. Therefore, when the depth of the scouring hole increases, the shear velocity in the hole will also increase, which can deepen both the upstream and downstream scouring hole. According to Equation (1), the Shields number is proportional to the square of the shear velocity. The increase in shear velocity significantly intensifies local scouring, which increases the development rate of scouring holes at the wake position and downstream.

4.3. Variation Rule of Spanning Time

In this paper, the spanning time is defined as the time taken for a semi-exposed submarine cable (initial state) to become a spanning submarine cable. Figure 10 illustrates the effect of the above parameters on the spanning time of the semi-exposed submarine cable.

Jmse 11 01349 g010 550

Figure 10. Influence of different parameters on spanning time of the semi-exposed submarine cable: (a) Sediment critical Shields number; (b) Sediment density; and (c) Ocean current velocity.From Figure 10a, the spanning time monotonically increases with the increase in the critical Shields number of sediment. However, the slope of the curve decreases first and then increases, and the inflection point is at 

θcr = 4.59 × 10

−2. The relationship between spanning time t and sediment’s critical Shields number 

θcr can be formulated by a cubic function as shown in Equation (6):

𝑡=−2.98+6.76𝜃𝑐𝑟−1.45𝜃2𝑐𝑟+0.11𝜃3𝑐𝑟.�=−2.98+6.76���−1.45���2+0.11���3.(6)It can be seen from Figure 10b that with the increase in the sediment density, the spanning time increases monotonically and linearly. The relationship between the spanning time t and the sediment’s density 

ρs can be formulated by the first order function as shown in Equation (7):

𝑡=−41.59+30.54𝜌𝑠.�=−41.59+30.54��.(7)Figure 10c shows that with the increase in the ocean current velocity, the spanning time decreases monotonically. The slope of the curve increases with the increase in the ocean current velocity, so it can be considered that there is saturation of the ocean current velocity effect. The relationship between the spanning time t and the ocean current velocity v can be formulated by the exponential function

𝑡=0.15𝑣−4.38.�=0.15�−4.38.(8)

5. Conclusions

In this paper, a three-dimensional CFD finite element numerical simulation model is established, which is used to research the local scouring process of the semi-exposed submarine cable under the steady-state ocean current. The relationship between shear velocity and scouring terrain is discussed, the influence of sediment critical Shields number, sediment density and ocean current velocity on the local scouring process is analyzed, and the variation rules of the spanning time of the semi-exposed submarine cable is given. The conclusions are as follows:

  • Under the steady-state ocean currents, scouring holes will be formed at the upstream, wake position and downstream of the semi-exposed submarine cable. The upstream and downstream scouring holes develop faster, which will reach a temporary stable state at about 1 h after the start of the scouring. The scouring hole at the wake position will continue to expand at a slower rate and eventually lead to the spanning of the submarine cable.
  • There is a close relationship between the distribution of shear velocity and the scouring terrain. As the local scouring process occurs, the location of the maximum shear velocity within the scouring hole shifts and causes the bottom of the hole to move as well.
  • When the sediment’s critical Shields number and density are significantly large and ocean current velocity is sufficiently low, the duration of the stable state of the upstream scouring hole will be prolonged, and the average development velocity of the scouring holes at the wake position and downstream will be reduced.
  • The relationship between the spanning time and the critical Shields number θcr can be formulated as a cubic function, in which the curve’s inflection point is θcr = 4.59 × 10−2. The relationship between spanning time and sediment density can be formulated as a linear function. The relationship between spanning time and ocean current velocity can be formulated by exponential function.

Based on the conclusions of this paper, even when it is too late to take measures or when the exposed position of the submarine cable cannot be located, the degree of burial depth development still can be predicted. This prediction is important for the operation and maintenance of the submarine cable. However, the study still leaves something to be desired. Only the local scouring process under the steady-state ocean current was studied, which is an extreme condition. In practice, exposed submarine cables are more likely to be scoured by reciprocating ocean currents. In the future, we will investigate the local scouring of submarine cables under the reciprocating ocean current.

Author Contributions

Conceptualization, Y.H. and Q.L.; methodology, Q.L., P.Z. and H.T.; software, Q.L.; validation, Q.L., L.C. and W.T.; writing—original draft preparation, Q.L.; writing—review and editing, Y.H. and Q.L.; supervision, Y.H. and L.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the [Smart Grid Joint Fund Key Project between National Natural Science Foundation of China and State Grid Corporation] grant number [U1766220].

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data supporting the reported results cannot be shared at this time, as they have been used in producing more publications on this research.

Acknowledgments

This work is supported by the Smart Grid Joint Fund Key Project of the National Natural Science Foundation of China and State Grid Corporation (Grant No. U1766220).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Taormina, B.; Bald, J.; Want, A.; Thouzeau, G.; Lejart, M.; Desroy, N.; Carlier, A. A review of potential impacts of submarine power cables on the marine environment: Knowledge gaps, recommendations and future directions. Renew. Sust. Energ. Rev. 2018, 96, 380–391. [Google Scholar] [CrossRef]
  2. Gulski, E.; Anders, G.J.; Jogen, R.A.; Parciak, J.; Siemiński, J.; Piesowicz, E.; Paszkiewicz, S.; Irska, I. Discussion of electrical and thermal aspects of offshore wind farms’ power cables reliability. Renew. Sust. Energ. Rev. 2021, 151, 111580. [Google Scholar] [CrossRef]
  3. Wang, W.; Yan, X.; Li, S.; Zhang, L.; Ouyang, J.; Ni, X. Failure of submarine cables used in high-voltage power transmission: Characteristics, mechanisms, key issues and prospects. IET Gener. Transm. Distrib. 2021, 15, 1387–1402. [Google Scholar] [CrossRef]
  4. Chen, H.; Chen, Z.; Lu, H.; Wu, C.; Liang, J. Protection method for submarine cable detection and exposed suspension problem in Qiongzhou straits. Telecom Pow. Technol. 2019, 36, 60–61+63. [Google Scholar]
  5. Zhu, J.; Ren, B.; Dong, P.; Chen, W. Vortex-induced vibrations of a free spanning submarine power cable. Ocean Eng. 2023, 272, 113792. [Google Scholar] [CrossRef]
  6. Sumer, B.M.; Jensen, H.R.; Mao, Y.; Fredsøe, J. Effect of lee-wake on scour below pipelines in current. J. Waterw. Port Coast. Ocean. Eng. 1988, 114, 599–614. [Google Scholar] [CrossRef]
  7. Chiew, Y.M. Prediction of maximum scour depth at submarine pipelines. J. Hydraul. Eng. 1991, 117, 452–466. [Google Scholar] [CrossRef]
  8. Mastbergen, D.R.; Vandenberg, J.H. Breaching in fine sands and the generation of sustained turbidity currents in submarine canyons. Sedimentology 2003, 50, 625–637. [Google Scholar] [CrossRef]
  9. Dey, S.; Singh, N.P. Clear-water scour depth below underwater pipelines. J. Hydro-Env. Res. 2007, 1, 157–162. [Google Scholar] [CrossRef]
  10. Dey, S.; Singh, N.P. Clear-water scour below underwater pipelines under steady flow. J. Hydraul. Eng. 2008, 134, 588–600. [Google Scholar] [CrossRef]
  11. Liang, D.; Cheng, L. Numerical study of scour around a pipeline bundle. Proc. Inst. Civil Eng. Mar. Eng. 2008, 161, 89–95. [Google Scholar] [CrossRef]
  12. Yang, L.; Guo, Y.; Shi, B.; Kuang, C.; Xu, W.; Cao, S. Study of scour around submarine pipeline with a rubber plate or rigid spoiler in wave conditions. J. Waterw. Port Coast. Ocean Eng. 2012, 138, 484–490. [Google Scholar] [CrossRef]
  13. Li, Y.; Ong, M.C.; Fuhrman, D.R.; Larsen, B.E. Numerical investigation of wave-plus-current induced scour beneath two submarine pipelines in tandem. Coast. Eng. 2020, 156, 103619. [Google Scholar] [CrossRef]
  14. Guan, D.; Hsieh, S.C.; Chiew, Y.M.; Low, Y.M.; Wei, M. Local scour and flow characteristics around pipeline subjected to vortex-induced vibrations. J. Hydraul. Eng. 2020, 146, 04019048. [Google Scholar] [CrossRef]
  15. Liu, M.M.; Jin, X.; Wang, L.; Yang, F.; Tang, J. Numerical investigation of local scour around a vibrating pipeline under steady currents. Ocean Eng. 2021, 221, 108546. [Google Scholar] [CrossRef]
  16. Huang, J.; Yin, G.; Ong, M.C.; Myrhaug, D.; Jia, X. Numerical investigation of scour beneath pipelines subjected to an oscillatory flow condition. J. Mar. Sci. Eng. 2021, 9, 1102. [Google Scholar] [CrossRef]
  17. Cui, F.; Du, Y.; Hao, X.; Peng, S.; Bao, Z.; Peng, S. Experimental study on local scour and related mechanical effects at river-crossing underwater oil and gas pipelines. Adv. Civ. Eng. 2021, 2021, 6689212. [Google Scholar] [CrossRef]
  18. Li, B.; Ma, H. Scouring mechanism of suspended and partially-buried pipelines under steady flow. Coast. Eng. 2022, 177, 104201. [Google Scholar] [CrossRef]
  19. Najafzadeh, M.; Oliveto, G. Scour propagation rates around offshore pipelines exposed to currents by applying data-driven models. Water 2022, 14, 493. [Google Scholar] [CrossRef]
  20. Zhu, Y.; Xie, L.; Wong, T.; Su, T. Development of three-dimensional scour below pipelines in regular waves. J. Mar. Sci. Eng. 2022, 10, 124. [Google Scholar] [CrossRef]
  21. Ma, H.; Li, B. CFD-CGDEM coupling model for scour process simulation of submarine pipelines. Ocean Eng. 2023, 271, 113789. [Google Scholar] [CrossRef]
  22. Hu, K.; Bai, X.; Vaz, M.A. Numerical simulation on the local scour processing and influencing factors of submarine pipeline. J. Mar. Sci. Eng. 2023, 11, 234. [Google Scholar] [CrossRef]
  23. Yang, B.; Gao, F.; Wu, Y. Experimental study on local scour of sandy seabed under submarine pipeline in unidirectional currents. Eng. Mech. 2008, 25, 206–210. [Google Scholar]
  24. Cheng, Y.; Wang, X.; Luo, W.; Huang, X.; Lyu, X. Experimental study of local scour around a downstream inclined pile under combined waves and current. Adv. Eng. Sci. 2021, 53, 64–71. [Google Scholar]
  25. Lu, Y.; Zhou, L.; Shen, X. Different turbulence models for simulating a liquid-liquid hydro cyclone. J. Tsinghua Univ. 2001, 41, 105–109. [Google Scholar]
  26. Yun, D.H.; Kim, Y.T. Experimental study on settlement and scour characteristics of artificial reef with different reinforcement type and soil type. Geotext. Geomembr. 2018, 46, 448–454. [Google Scholar] [CrossRef]
원문 바로가기
Validity evaluation of popular liquid-vapor phase change models for cryogenic self-pressurization process

극저온 자체 가압 공정을 위한 인기 있는 액체-증기 상 변화 모델의 타당성 평가

액체-증기 상 변화 모델은 밀폐된 용기의 자체 가압 프로세스 시뮬레이션에 매우 큰 영향을 미칩니다. Hertz-Knudsen 관계, 에너지 점프 모델 및 그 파생물과 같은 널리 사용되는 액체-증기 상 변화 모델은 실온 유체를 기반으로 개발되었습니다. 액체-증기 전이를 통한 극저온 시뮬레이션에 널리 적용되었지만 각 모델의 성능은 극저온 조건에서 명시적으로 조사 및 비교되지 않았습니다. 본 연구에서는 171가지 일반적인 액체-증기 상 변화 모델을 통합한 통합 다상 솔버가 제안되었으며, 이를 통해 이러한 모델을 실험 데이터와 직접 비교할 수 있습니다. 증발 및 응축 모델의 예측 정확도와 계산 속도를 평가하기 위해 총 <>개의 자체 가압 시뮬레이션이 수행되었습니다. 압력 예측은 최적화 전략이 서로 다른 모델 계수에 크게 의존하는 것으로 나타났습니다. 에너지 점프 모델은 극저온 자체 가압 시뮬레이션에 적합하지 않은 것으로 나타났습니다. 평균 편차와 CPU 소비량에 따르면 Lee 모델과 Tanasawa 모델은 다른 모델보다 안정적이고 효율적인 것으로 입증되었습니다.

원문 바로가기
Elsevier

International Journal of Heat and Mass Transfer

Volume 181, December 2021, 121879

International Journal of Heat and Mass Transfer

Validity evaluation of popular liquid-vapor phase change models for cryogenic self-pressurization process

Author links open overlay panelZhongqi Zuo, Jingyi Wu, Yonghua HuangShow moreAdd to MendeleyShareCite

https://doi.org/10.1016/j.ijheatmasstransfer.2021.121879Get rights and content

Abstract

Liquid-vapor phase change models vitally influence the simulation of self-pressurization processes in closed containers. Popular liquid-vapor phase change models, such as the Hertz-Knudsen relation, energy jump model, and their derivations were developed based on room-temperature fluids. Although they had widely been applied in cryogenic simulations with liquid-vapor transitions, the performance of each model was not explicitly investigated and compared yet under cryogenic conditions. A unified multi-phase solver incorporating four typical liquid-vapor phase change models has been proposed in the present study, which enables direct comparison among those models against experimental data. A total number of 171 self-pressurization simulations were conducted to evaluate the evaporation and condensation models’ prediction accuracy and calculation speed. It was found that the pressure prediction highly depended on the model coefficients, whose optimization strategies differed from each other. The energy jump model was found inadequate for cryogenic self-pressurization simulations. According to the average deviation and CPU consumption, the Lee model and the Tanasawa model were proven to be more stable and more efficient than the others.

Introduction

The liquid-vapor phase change of cryogenic fluids is widely involved in industrial applications, such as the hydrogen transport vehicles [1], shipborne liquid natural gas (LNG) containers [2] and on-orbit cryogenic propellant tanks [3]. These applications require cryogenic fluids to be stored for weeks to months. Although high-performance insulation measures are adopted, heat inevitably enters the tank via radiation and conduction. The self-pressurization in the tank induced by the heat leakage eventually causes the venting loss of the cryogenic fluids and threatens the safety of the craft in long-term missions. To reduce the boil-off loss and extend the cryogenic storage duration, a more comprehensive understanding of the self-pressurization mechanism is needed.

Due to the difficulties and limitations in implementing cryogenic experiments, numerical modeling is a convenient and powerful way to study the self-pressurization process of cryogenic fluids. However, how the phase change models influence the mass and heat transfer under cryogenic conditions is still unsettled [4]. As concluded by Persad and Ward [5], a seemingly slight variation in the liquid-vapor phase change models can lead to erroneous predictions.

Among the liquid-vapor phase change models, the kinetic theory gas (KTG) based models and the energy jump model are the most popular ones used in recent self-pressurization simulations [6]. The KTG based models, also known as the Hertz-Knudsen relation models, were developed on the concept of the Maxwell-Boltzmann distribution of the gas molecular [7]. The Hertz-Knudsen relation has evolved to several models, including the Schrage model [8], the Tanasawa model [9], the Lee model [10] and the statistical rate theory (SRT) [11], which will be described in Section 2.2. Since the Schrage model and the Lee model are embedded and configured as the default ones in the commercial CFD solvers Flow-3D® and Ansys Fluent® respectively, they have been widely used in self-pressurization simulations for liquid nitrogen [12], [13] and liquid hydrogen [14], [15]. The major drawback of the KTG models lies in the difficulty of selecting model coefficients, which were reported in a considerably wide range spanning three magnitudes even for the same working fluid [16], [17], [18], [19], [20], [21]. Studies showed that the liquid level, pressure and mass transfer rate are directly influenced by the model coefficients [16], [22], [23], [24], [25]. Wrong coefficients will lead to deviation or even divergence of the results. The energy jump model is also known as the thermal limitation model. It assumes that the evaporation and condensation at the liquid-vapor interface are induced only by heat conduction. The model is widely adopted in lumped node simulations due to its simplicity [6], [26], [27]. To improve the accuracy of mass flux prediction, the energy jump model was modified by including the convection heat transfer [28], [29]. However, the convection correlations are empirical and developed mainly for room-temperature fluids. Whether the correlation itself can be precisely applied in cryogenic simulations still needs further investigation.

Fig. 1 summarizes the cryogenic simulations involving the modeling of evaporation and condensation processes in recent years. The publication has been increasing rapidly. However, the characteristics of each evaporation and condensation model are not explicitly revealed when simulating self-pressurization. A comparative study of the phase change models is highly needed for cryogenic fluids for a better simulation of the self-pressurization processes.

In the present paper, a unified multi-phase solver incorporating four typical liquid-vapor phase change models, namely the Tanasawa model, the Lee model, the energy jump model, and the modified energy jump model has been proposed, which enables direct comparison among different models. The models are used to simulate the pressure and temperature evolutions in an experimental liquid nitrogen tank in normal gravity, which helps to evaluate themselves in the aspects of accuracy, calculation speed and robustness.

Section snippets

Governing equations for the self-pressurization tank

In the present study, both the fluid domain and the solid wall of the tank are modeled and discretized. The heat transportation at the solid boundaries is considered to be irrelevant with the nearby fluid velocity. Consequently, two sets of the solid and the fluid governing equations can be decoupled and solved separately. The pressures in the cryogenic container are usually from 100 kPa to 300 kPa. Under these conditions, the Knudsen number is far smaller than 0.01, and the fluids are

Self-pressurization results and phase change model comparison

This section compares the simulation results by different phase change models. Section 3.1 compares the pressure and temperature outputs from two KTG based models, namely the Lee model and the Tanasawa model. Section 3.2 presents the pressure predictions from the energy transport models, namely the energy jump model and the modified energy jump model, and compares pressure prediction performances between the KTG based models and the energy transport models. Section 3.3 evaluates the four models 

Conclusion

A unified vapor-liquid-solid multi-phase numerical solver has been accomplished for the self pressurization simulation in cryogenic containers. Compared to the early fluid-only solver, the temperature prediction in the vicinity of the tank wall improves significantly. Four liquid-vapor phase change models were integrated into the solver, which enables fair and effective comparison for performances between each other. The pressure and temperature prediction accuracies, and the calculation speed

CRediT authorship contribution statement

Zhongqi Zuo: Data curation, Formal analysis, Writing – original draft, Validation. Jingyi Wu: Conceptualization, Writing – review & editing, Validation. Yonghua Huang: Conceptualization, Formal analysis, Writing – review & editing, Validation.

Declaration of Competing Interest

Authors declare that they have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled, “Validity evaluation of popular liquid-vapor phase change models for cryogenic self-pressurization process”.

Acknowledgement

This project is supported by the National Natural Science Foundation of China (No. 51936006).

References (40)

There are more references available in the full text version of this article.

Cited by (7)

Intrusion of fine sediments into river bed and its effect on river environment – a research review

미세한 퇴적물이 강바닥에 침투하고 하천 환경에 미치는 영향 – 연구 검토

Intrusion of fine sediments into river bed and its effect on river environment – a research review

Nilav Karna,K.S. Hari Prasad, Sanjay Giri & A.S. Lodhi

Abstract

Fine sediments enter into the river through various sources such as channel bed, bank, and catchment. It has been regarded as a type of pollution in river. Fine sediments present in a river have a significant effect on river health. Benthic micro-organism, plants, and large fishes, all are part of food chain of river biota. Any detrimental effect on any of these components of food chain misbalances the entire riverine ecosystem. Numerous studies have been carried out on the various environmental aspects of rivers considering the presence of fine sediment in river flow. The present paper critically reviews many of these aspects to understand the various environmental impacts of suspended sediment on river health, flora and fauna.

Keywords: 

  1. Introduction
    The existence of fine sediment in a river system is a natural phenomenon. But in many cases it is exacerbated by the manmade activities. The natural cause of fines being in flow generally keeps the whole system in equilibrium except during some calamites whereas anthropogenic activities leading to fines entering into the flow puts several adverse impacts on the entire river system and its ecology. Presence of fines in flow is considered as a type of pollution in water. In United States,
    the fine sediment in water along with other non point source pollution is considered as a major obstacle in providing quality water for fishes and recreation activities (Diplas and Parker 1985).
    Sediments in a river are broadly of two types, organic and inorganic, and they both move in two ways either along the bed of the channel called bed load or in suspension called suspended load and their movements depend upon fluid flow and sediment characteristics. Further many investigators have divided the materials in suspension into two different types.
    One which originates from channel bed and bank is called bed material suspended load and another that migrates from feeding catchment area is called wash load. A general perception is that wash loads are very fine materials like clay, silt but it may not always be true (Woo et al. 1986). In general, suspended materials are of size less than 2 mm. The impact of sand on the various aspects of river is comparatively less than that of silt and clay. The latter are chemically active and good carrier of many contaminants and nutrients such as dioxins, phosphorous, heavy and trace metals, polychlorinated biphenyl (PCBs), radionuclide, etc. (Foster and Charlesworth 1996; Horowitz et al. 1995; Owens et al. 2001; Salomons and Förstner 1984; Stone and Droppo 1994; Thoms 1987). Foy and Bailey-Watt (1998) reported that out of 129 lakes in England and Wales, 69% have phosphorous contamination. Ten percent lakes, rivers, and bays of United States have sediment contaminants with chemicals as reported by USEPA. Several field and experimental studies have been conducted
    considering, sand, silt, and clay as suspended material. Hence, the subject reported herein is based on considering the fine sediment size smaller than 2 mm.
    Fine sediments have the ability to alter the hydraulics of the flow. Presence of fines in flow can change the magnitude of turbulence, it can change the friction resistance to flow. Fines can change the mobility and permeability of the bed material. In some extreme cases, fines in flow may even change the morphology of the river (Doeg and Koehn 1994; Nuttall 1972; Wright and Berrie 1987). Fines in the flow adversely affect the producer by increasing the turbidity, hindering the
    photosynthesis process by limiting the light penetration. This is ultimately reflected in the entire food ecosystem of river (Davis-Colley et al. 1992; Van Niewenhuyre and Laparrieve 1986). In addition, abrasion due to flowing sediment kills the aquatic flora (Edwards 1969; Brookes 1986). Intrusion of fines into the pores of river bed reduces space for several invertebrates, affects the spawning process (Petts 1984; Richards and Bacon 1994; Schalchli 1992). There are several other direct
    or indirect, short-term or long-term impacts of fines in river.
    The present paper reports the physical/environmental significance of fines in river. The hydraulic significance of presence of fines in the river has been reviewed in another paper (Effect of fine sediments on river hydraulics – a research review – http://dx.doi.org/10.1080/09715010.2014.982001).

References

  • Adams, J.N., and Beschta, R.L. (1980). “Gravel bed composition in oregon coastal streams.” Can. J. Fish. Aquat.Sci., 37, 1514–1521.10.1139/f80-196  [Crossref], [Web of Science ®], [Google Scholar]
  • Alabaster, J.S., and Llyod, R.L. (1980). Water quality criteria for fresh water, Butterworth, London, 297. [Google Scholar]
  • Aldridge, D.W., Payne, B.S., and Miller, A.C. (1987). “The effects of intermittent exposure to suspended solids and turbulence on three species of freshwater mussels.” Environ. Pollution, 45, 17–28.10.1016/0269-7491(87)90013-3  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Barton, B.A. (1977). “Short-term effects of highway construction on the limnology of a small stream in southern Ontario.” Freshwater Biol., 7, 99–108.10.1111/fwb.1977.7.issue-2  [Crossref], [Web of Science ®], [Google Scholar]
  • Bash, J., Berman, C., and Bolton, S. (2001). Effects of turbidity and suspended solids on salmonids, Center for Streamside Studies, University of Washington, Seattle, WA. [Google Scholar]
  • Baxter, C.V., and Hauer, F.R. (2000). “Geomorphology, hyporheic exchange, and selection of spawning habitat by bull trout (Salvelinus confuentus).” Can. J. Fish. Aquat.Sci., 57, 1470–1481.10.1139/f00-056  [Crossref], [Web of Science ®], [Google Scholar]
  • Berkman, H.E., and Rabeni, C.F. (1987). “Effect of siltation on stream fish communities.” Environ. Biol. Fish., 18, 285–294.10.1007/BF00004881  [Crossref], [Web of Science ®], [Google Scholar]
  • Beschta, R.L., and Jackson, W.L. (1979). “The intrusion of fine sediments into a stable gravel bed.” J. Fish. Res. Board Can., 36, 204–210.10.1139/f79-030  [Crossref], [Google Scholar]
  • Boon, P.J. (1988). “The impact of river regulation on invertebrate communities in the UK.” Reg. River Res. Manage., 2, 389–409.10.1002/(ISSN)1099-1646  [Crossref], [Google Scholar]
  • Brookes, A. (1986). “Response of aquatic vegetation to sedimentation downstream from river channelization works in England and Wales.” Biol. Conserv., 38, 352–367. [Crossref], [Web of Science ®], [Google Scholar]
  • Bruton, M.N. (1985). “The effects of suspensoids on fish.” Hydrobiologia, 125, 221–241.10.1007/BF00045937  [Crossref], [Web of Science ®], [Google Scholar]
  • Carling, P.A. (1984). “Deposition of fine and coarse sand in an open-work gravel bed.” Can. J. Fish. Aquat. Sci., 41, 263–270.10.1139/f84-030  [Crossref], [Web of Science ®], [Google Scholar]
  • Carling, P.A., and McCahon, C.P. (1987). “Natural siltation of brown trout (Salmo trutta L.) spawning gravels during low-flow conditions.” Regulated streams, J.F. Craig and J.B. Kemper, eds., Plenum Press, New York, NY, 229–244.10.1007/978-1-4684-5392-8  [Crossref], [Google Scholar]
  • Carter, J., Owens, P.N., Walling, D.E., and Leeks, G.J.L. (2003). “Fingerprinting suspended sediment sources in a large urban river system.” Sci. Total Environ., 314–316, 513–534.10.1016/S0048-9697(03)00071-8  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Chang, H.H. (1988). Fluvial processes in river engineering, Krieger, Malabar Florida, 432. [Google Scholar]
  • Chapman, D.W. (1988). “Critical review of variables used to define effects of fines in redds of large salmonids.” Trans. Am. Fish. Soc., 117, 1–21.10.1577/1548-8659(1988)117<0001:CROVUT>2.3.CO;2  [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Church, M.A., Mclean, D.G., and Wolcott, J.F. (1987). “River bed gravel sampling and analysis.” Sediment transport in gravel-bed rivers, C.R. Thorne, J.C. Bathrust, and R.D. Hey, eds., John Willey, Chichester, 43–79. [Google Scholar]
  • Cline, L.D., Short, R.A., and Ward, J.V. (1982). “The influence of highway construction on the macroinvertebrates and epilithic algae of a high mountain stream.” Hydrobiologia, 96, 149–159.10.1007/BF02185430  [Crossref], [Web of Science ®], [Google Scholar]
  • Collins, A.L., Walling, D.E., and Leeks, G.J.L. (1997). “Fingerprinting the origin of fluvial suspended sediment in larger river basins: combining assessment of spatial provenance and source type.” Geografiska Annaler, 79A, 239–254.10.1111/1468-0459.00020  [Crossref], [Google Scholar]
  • Cordone, A.J., and Kelly, D.W. (1961). “The influence of inorganic sediment on the aquatic life of stream.” Calif. Fish Game, 47, 189–228. [Google Scholar]
  • Culp, J.M., Wrona, F.J., and Davies, R.W. (1985). “Response of stream benthos and drift to fine sediment depositionversus transport.” Can. J. Zool., 64, 1345–1351. [Crossref], [Web of Science ®], [Google Scholar]
  • Davies-Colley, R.J., Hickey, C.W., Quinn, J.M., and Ryan, P.A. (1992). “Effects of clay discharges on streams.” Hydrobiologia, 248, 215–234.10.1007/BF00006149  [Crossref], [Web of Science ®], [Google Scholar]
  • Dhamotharan, S., Wood, A., Parker, G., and Stefan, H. (1980). Bed load transport in a model gravel stream. Project Report No. 190. St. Anthony Falls Hydraulic Laboratory, University of Minnesota. [Google Scholar]
  • Diplas, P., and Parker, G. (1985). Pollution of gravel spawning grounds due to fine sediment. Project Report, No. 240. St. Anthony Falls Laboratory, University of Minnesota, Minneapolis, MN. [Google Scholar]
  • Doeg, T.J., and Koehn, J.D. (1994). “Effects of draining and desilting a small weir on downstream fish and macroinvertebrates.” Reg. River Res. Manage., 9, 263–277.10.1002/(ISSN)1099-1646  [Crossref], [Web of Science ®], [Google Scholar]
  • Droppo, I.G. (2001). “Rethinking what constitutes suspended sediment.” Hydrol. Process., 15, 1551–1564.10.1002/(ISSN)1099-1085  [Crossref], [Web of Science ®], [Google Scholar]
  • Droppo, I.G., and Ongley, E.D. (1994). “Flocculation of suspended sediment in rivers of southeastern Canada.” Water Res., 28, 1799–1809.10.1016/0043-1354(94)90253-4  [Crossref], [Web of Science ®], [Google Scholar]
  • Einstein, H.A. (1968). “Deposition of suspended particles in a gravel bed.” J. Hydraul. Eng., 94, 1197–1205. [Google Scholar]
  • Erman, D.C., and Ligon, F.K. (1988). “Effects of discharge fluctuation and the addition of fine sediment on stream fish and macroinvertebrates below a water-filtration facility.” Environ. Manage., 12, 85–97.10.1007/BF01867380  [Crossref], [Web of Science ®], [Google Scholar]
  • Farnsworth, K.L., and Milliman, J.D. (2003). “Effects of climatic and anthropogenic change on small mountainous rivers: the Salinas River example.” Global Planet. Change, 39, 53–64.10.1016/S0921-8181(03)00017-1  [Crossref], [Web of Science ®], [Google Scholar]
  • Foster, I.D.L., and Charlesworth, S.M. (1996). “Heavy metals in the hydrological cycle: trends and explanation.” Hydrol. Process., 10, 227–261.10.1002/(ISSN)1099-1085  [Crossref], [Web of Science ®], [Google Scholar]
  • Foy, R.H., and Bailey-Watts, A.E. (1998). “Observations on the spatial and temporal variation in the phosphorus status of lakes in the British Isles.” Soil Use Manage., 14, 131–138.10.1111/sum.1998.14.issue-s4  [Crossref], [Web of Science ®], [Google Scholar]
  • Frostick, L.E., Lucas, P.M., and Reid, I. (1984). “The infiltration of fine matrices into coarse-grained alluvial sediments and its implications for stratigraphical interpretation.” J. Geol. Soc. London, 141, 955–965.10.1144/gsjgs.141.6.0955  [Crossref], [Web of Science ®], [Google Scholar]
  • Gagnier, D.L., and Bailey, R.C. (1994). “Balancing loss of information and gains in efficiency in characterizing stream sediment samples.” J. North Am. Benthol. Soc., 13, 170–180.10.2307/1467236  [Crossref], [Web of Science ®], [Google Scholar]
  • Gammon, J.R. (1970). The effect of inorganic sediment on stream biota. Environmental Protection Agency, Water Pollution Control Research, Series, 18050 DWC 12/70. USGPO, Washington, DC. [Google Scholar]
  • Graham, A.A. (1990). “Siltation of stone-surface periphyton in rivers by clay-sized particles from low concentrations in suspention.” Hydrobiologia, 199, 107–115.10.1007/BF00005603  [Crossref], [Web of Science ®], [Google Scholar]
  • Greig, S.M., Sear, D.A., and Carling, P.A. (2005). “The impact of fine sediment accumulation on the survival of incubating salmon progeny: Implications for sediment management.” Sci. Total Environ., 344, 241–258.10.1016/j.scitotenv.2005.02.010  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Harrod, T.R., and Theurer, F.D. (2002). “Sediment.” Agriculture, hydrology and water quality, P.M. Haygarth and S.C. Jarvis, eds., CABI, Wallingford, 502. [Crossref], [Google Scholar]
  • Horowitz, A.J., Elrick, K.A., Robbins, J.A., and Cook, R.B. (1995). “Effect of mining and related activities on the sediment trace element geochemistry of Lake Coeur D’Alene, Idaho, USA part II: Subsurface sediments.” Hydrol. Process., 9, 35–54.10.1002/(ISSN)1099-1085  [Crossref], [Web of Science ®], [Google Scholar]
  • Hynes, H.B.N. (1970). The ecology of running waters, Liverpool University Press, Liverpool. [Google Scholar]
  • Khullar, N.K. (2002). “Effect of wash load on transport of uniform and nonuniform sediments.” Ph.D. thesis, Indian Institute of Technology Roorkee. [Google Scholar]
  • Kondolf, G.M. (1995). “Managing bedload sediment in regulated rivers: Examples from California, USA.” Geophys. Monograph, 89, 165–176. [Google Scholar]
  • Kondolf, G.M. (1997). “Hungry water: effects of dams and gravel mining on river channels.” Environ. Manage., 21, 533–551.10.1007/s002679900048  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Langer, O.E. (1980). “Effects of sedimentation on salmonid stream life.” Report on the Technical Workshop on Suspended Solids and the Aquatic Environment, K. Weagle, ed., Whitehorse. [Google Scholar]
  • Lemly, A.D. (1982). “Modification of benthic insect communities in polluted streams: combined effects of sedimentation and nutrient enrichment.” Hydrobiologia, 87, 229–245.10.1007/BF00007232  [Crossref], [Web of Science ®], [Google Scholar]
  • Levasseur, M., Bergeron, N.E., Lapointe, M.F., and Bérubé, F. (2006). “Effects of silt and very fine sand dynamics in Atlantic salmon (Salmo salar) redds on embryo hatching success.” Can. J. Fish. Aquat. Sci., 63, 1450–1459.10.1139/f06-050  [Crossref], [Web of Science ®], [Google Scholar]
  • Lewis, K. (1973a). “The effect of suspended coal particles on the life forms of the aquatic moss Eurhynchium riparioides (Hedw.).” Fresh Water Biol., 3, 251–257.10.1111/fwb.1973.3.issue-3  [Crossref], [Google Scholar]
  • Lewis, K. (1973b). “The effect of suspended coal particles on the life forms of the aquatic moss Eurhynchium riparioides (Hedw.).” Fresh Water Biol., 3, 391–395.10.1111/fwb.1973.3.issue-4  [Crossref], [Google Scholar]
  • Lisle, T. (1980). “Sedimentation of Spawning Areas during Storm Flows, Jacoby Creek, North Coastal California.” Presented at the fall meeting of the American Geophysical Union, San Francisco, CA. [Google Scholar]
  • Marchant, R. (1989). “Changes in the benthic invertebrate communities of the thomson river, southeastern Australia, after dam construction.” Reg. River Res. Manage., 4, 71–89.10.1002/(ISSN)1099-1646  [Crossref], [Google Scholar]
  • McNeil, W.J., and Ahnell, W.H. (1964). Success of pink salmon spawning relative to size of spawning bed material. US Fish and Wildlife Service. Special Scientific Report, Fisheries 469. Washington, DC. [Google Scholar]
  • Milhous, R.T. (1973). “Sediment transport in a gravel bottomed stream.” Ph.D. thesis, Oregon State University, Corvallis, OR. [Google Scholar]
  • Milliman, J.D., and Syvitski, J.P.M. (1992). “Geomorphic/tectonic control of sediment discharge to the oceans: the importance of small mountainous rivers.” J. Geol., 100, 525–544.10.1086/jg.1992.100.issue-5  [Crossref], [Web of Science ®], [Google Scholar]
  • Mohnakrishnan, A. (2001). Reservoir sedimentation, Seminar on Reservoir Sedimentation, Ooty. [Google Scholar]
  • Mohta, J.A., Wallbrink, P.J., Hairsine, P.B., and Grayson, R.B. (2003). “Determining the sources of suspended sediment in a forested catchment in southeastern Australia.” Water Resour. Res., 39, 1056. [Web of Science ®], [Google Scholar]
  • Morris, G.L. (1993). “A global perspective of sediment control measures in reservoirs.” Notes on sediment management in reservoirs, S. Fan and G. Morris, eds., Water Resources Publications, Colorado, 13–44. [Google Scholar]
  • Morris, L.G., and Fan, J. (2010). Reservoir Sedimentation hand book – design and management of dams, reservoirs and watershed for sustainable use. McGraw-Hill, 440 and 499. [Google Scholar]
  • Newcombe, C.P., and Macdonald, D.D. (1991). “Effects of suspended sediments on aquatic ecosystems.” North Am. J. Fish. Manage., 11, 72–82.10.1577/1548-8675(1991)011<0072:EOSSOA>2.3.CO;2  [Taylor & Francis Online], [Google Scholar]
  • Nuttal, P.M. (1972). “The effects of sand deposition upon the macroinvertebrate fauna of the River Camel, Cornwall.” Freshwater Biol., 2, 181–186.10.1111/fwb.1972.2.issue-3  [Crossref], [Google Scholar]
  • Olsson, T.I., and Petersen, B. (1986). “Effects of gravel size and peat material on embryo survival and alevin emergence of brown trout, Salmo trutta L.” Hydrobiologia, 135, 9–14.10.1007/BF00006453  [Crossref], [Web of Science ®], [Google Scholar]
  • Owens, P.N., Walling, D.E., and Leeks, G.J.L. (2000). “Tracing fluvial suspended sediment sources in the catchment of the River Tweed, Scotland, using composite fingerprints and a numerical mixing model.” Tracers in eomorphology, I.D.L. Foster, ed., Wiley, Chichester, 291–308. [Google Scholar]
  • Owens, P.N., Walling, D.E., Carton, J., Meharg, A.A., Wright, J., and Leeks, G.J.L. (2001). “Downstream changes in the transport and storage of sediment-associated contaminants (P, Cr and PCBs) in agricultural and industrialized drainage basins.” Sci. Total Environ., 266, 177–186.10.1016/S0048-9697(00)00729-4  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Petts, G.E. (1984). Impounded rivers: Perspectives for ecological management, Wiley, Chichester, 326. [Google Scholar]
  • Phillips, J.M., and Walling, D.E. (1995). “An assessment of the effects of sample collection, storage and resuspension on the representativeness of measurements of the effective particle size distribution of fluvial suspended sediment.” Water Res., 29, 2498–2508.10.1016/0043-1354(95)00087-2  [Crossref], [Web of Science ®], [Google Scholar]
  • Quinn, J.M., Davies-Coley, R.J., Hickey, C.W., Vickers, M.L., and Ryan, P.A. (1992). “Effects of clay discharges on streams.” Hydrobiologia, 248, 235–247.10.1007/BF00006150  [Crossref], [Web of Science ®], [Google Scholar]
  • Reiser, D.W., and White, R.G. (1990). “Effects of stream flow reduction on Chinook salmon egg incubation and fry quality.” Rivers, 1, 110–118. [Google Scholar]
  • Richards, C., and Bacon, K.L. (1994). “Influence of fine sediment on macroibvertebrates colonization of surface and hyporheic stream substrate.” Great Basin Nat., 54, 106–113. [Google Scholar]
  • Richards, C., Host, G.H., and Arthur, J.W. (1993). “Identification of predominant environmental factors structuring stream macroinvertebrate communities within a large agricultural catchment.” Freshwater Biol., 29, 285–294.10.1111/fwb.1993.29.issue-2  [Crossref], [Web of Science ®], [Google Scholar]
  • Rosenberg, D.M., and Wiens, A.P. (1978). “Effects of sediment addition on macrobenthic invertebrates in a Northern Canadian River.” Water Res., 12, 753–763.10.1016/0043-1354(78)90024-6  [Crossref], [Web of Science ®], [Google Scholar]
  • Ryan, P.A. (1991). “Environmental effects of sediment on New Zealand streams: A review.” New Zeal. J. Mar. Freshwater Res., 25, 207–221.10.1080/00288330.1991.9516472  [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Salomons, W., and Förstner, U. (1984). Metals in the hydrocycle, Sringer Verglag, New York, NY.10.1007/978-3-642-69325-0  [Crossref], [Google Scholar]
  • Schalchli, U. (1992). “The clogging of coarse gravel river beds by fine sediment.” Hydrobiologia, 235–236, 189–197.10.1007/BF00026211  [Crossref], [Web of Science ®], [Google Scholar]
  • Scrivener, J.C., and Brownlee, M.J. (1989). “Effects of forest harvesting on spawning gravel and incubation survival of chum (Oncorhynchus keta) andcoho salmon (O. kisutch) in Carnation Creek, British Columbia.” Can. J. Fish. Aquat. Sci., 46, 681–696.10.1139/f89-087  [Crossref], [Web of Science ®], [Google Scholar]
  • Sear, D.A. (1993). “Fine sediment infiltration into gravel spawning beds within a regulated river experiencing floods: Ecological implications for salmonids.” Reg Rivers Res. Manage., 8, 373–390.10.1002/(ISSN)1099-1646  [Crossref], [Google Scholar]
  • Soutar, R.G. (1989). “Afforestation and sediment yields in British fresh waters.” Soil Use Manage., 5, 82–86.10.1111/sum.1989.5.issue-2  [Crossref], [Web of Science ®], [Google Scholar]
  • Stone, M., and Droppo, I.G. (1994). “In-channel surficial fine-grained sediment laminae: Part II: Chemical characteristics and implications for contaminant transport in fluvial systems.” Hydrol. Process., 8, 113–124.10.1002/(ISSN)1099-1085  [Crossref], [Web of Science ®], [Google Scholar]
  • Thoms, M.C. (1987). “Channel sedimentation within the urbanized River Tame, UK.” Reg. Rivers Res. Manage., 1, 229–246.10.1002/(ISSN)1099-1646  [Crossref], [Google Scholar]
  • Trimble, S.W. (1983). “A sediment budget for Coon Creek, Driftless area, Wisconsin, 1853–1977.” Am. J. Sci., 283, 454–474.10.2475/ajs.283.5.454  [Crossref], [Web of Science ®], [Google Scholar]
  • U.S. Department of Health, Education and Welfare. (1965). Environmental Health Practices in recreational Areas, Public Health Service, Publication No. 1195. [Google Scholar]
  • Van Nieuwenhuyse, E.E., and LaPerriere, J.D. (1986). “Effects of placer gold mining on primary production in subarctic streams of Alaska.” J. Am. Water Res. Assoc., 22, 91–99. [Crossref], [Google Scholar]
  • Vörösmarty, C.J., Meybeck, M., Fekete, B., Sharma, K., Green, P., and Syvitski, J.P.M. (2003). “Anthropogenic sediment retention: major global impact from registered river impoundments.” Global Planet. Change, 39, 169–190.10.1016/S0921-8181(03)00023-7  [Crossref], [Web of Science ®], [Google Scholar]
  • Walling, D.E. (1995). “Suspended sediment yields in a changing environment.” Changing river channels, A. Gurnell and G. Petts, eds., Wiley, Chichester, 149–176. [Google Scholar]
  • Walling, D.E., and Moorehead, D.W. (1989). “The particle size characteristics of fluvial suspended sediment: an overview.” Hydrobiologia, 176–177, 125–149.10.1007/BF00026549  [Crossref], [Web of Science ®], [Google Scholar]
  • Walling, D.E., Owens, P.N., and Leeks, G.J.L. (1999). “Fingerprinting suspended sediment sources in the catchment of the River Ouse, Yorkshire, UK.” Hydrol. Process., 13, 955–975.10.1002/(ISSN)1099-1085  [Crossref], [Web of Science ®], [Google Scholar]
  • Walling, D.E., Owens, P.N., Waterfall, B.D., Leeks, G.J.L., and Wass, P.D. (2000). “The particle size characteristics of fluvial suspended sediment in the Humber and Tweed catchments, UK.” Sci. Total Environ., 251–252, 205–222.10.1016/S0048-9697(00)00384-3  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Wilbur, C.G. (1983). Turbidity in the aquatic environment: an environmental factor in fresh and oceanic waters, Charles C. Thomas, Springfield, IL, 133. [Google Scholar]
  • Woo, H.S., Julien, P.Y., and Richardson, E.V. (1986). “Washload and fine sediment load.” J. Hydraul. Eng., 112, 541–545.10.1061/(ASCE)0733-9429(1986)112:6(541)  [Crossref], [Google Scholar]
  • Wood, P.J., and Armitage, P.D. (1997). “Biological effects of fine sediment in the lotic environment.” Environ. Manage., 21, 203–217.10.1007/s002679900019  [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Wooster, J.K., Dusterhoff, S.R., Cui, Y., Sklar, L.S., Dietrich, W.E., and Malko, M. (2008). “Sediment supply and relative size distribution effects on fine sediment infiltration into immobile gravels.” Water Res. Res., 44, 1–18. [Crossref], [Web of Science ®], [Google Scholar]
  • Wren, G.Daniel, Bennett, J.Sean, Barkdoll, D.Brian, and Khunle, A.Roger. (2000). Studies in suspended sediment and turbulence in open channel flows, USDA, Agriculture Research Service, Research Report No. 18. [Google Scholar]
  • Wright, J.F., and Berrie, A.D. (1987). “Ecological effects of groundwater pumping and a natural drought on the upper reaches of a chalk stream.” Reg. River Res. Manage., 1, 145–160.10.1002/(ISSN)1099-1646  [Crossref], [Google Scholar]
  • Zhang, H., Xia, M., Chen, S.J., Li, Z., and Xia, H.B. (1976). “Regulation of sediments in some medium and small-sized reservoirs on heavily silt-laden streams in China.” 12th International Commission on Large Dams (ICOLD) Congress, Q. 47, R. 32, Mexico City, 1123–1243. [Google Scholar]
Effects of pile-cap elevation on scour and turbulence around a complex bridge pier

복잡한 교각 주변의 세굴 및 난기류에 대한 말뚝 뚜껑 높이의 영향

ABSTRACT

이 연구에서는 세 가지 다른 말뚝 뚜껑 높이에서 직사각형 말뚝 캡이 있는 복잡한 부두 주변의 지역 세굴 및 관련 흐름 유체 역학을 조사합니다. 말뚝 캡 높이가 초기 모래층에 대해 선택되었으며, 말뚝 캡이 흐름에 노출되지 않고(사례 I), 부분적으로 노출되고(사례 II) 완전히 노출(사례 III)되도록 했습니다. 실험은 맑은 물 세굴 조건 하에서 재순환 수로에서 수행되었으며, 입자 이미지 유속계 (PIV) 기술을 사용하여 다른 수직면에서 순간 유속을 얻었습니다. 부분적으로 노출된 파일 캡 케이스는 최대 수세미 깊이(MSD)를 보여주었습니다. 사례 II에서 MSD가 발생한 이유는 난류 유동장 분석을 통해 밝혀졌는데, 이는 말뚝 캡이 흐름에 노출됨에 따라 더 높은 세굴 깊이를 담당하는 말뚝 가장자리에서 와류 생성에 지배적으로 영향을 미친다는 것을 보여주었습니다. 유동장에 대한 파일 캡의 영향은 평균 속도, 소용돌이, 레이놀즈 전단 응력 및 난류 운동 에너지 윤곽을 통해 사례 III에서 두드러지게 나타났지만 파일 캡이 베드에서 떨어져 있었기 때문에 파일 캡 모서리는 수세미에 직접적인 영향을 미치지 않았습니다.

In this study, the local scour and the associated flow hydrodynamics around a complex pier with rectangular pile-cap at three different pile-cap elevations are investigated. The pile-cap elevations were selected with respect to the initial sand bed, such that the pile-cap was unexposed (case I), partially exposed (case II), and fully exposed (case III) to the flow. The experiments were performed in a recirculating flume under clear-water scour conditions, and the instantaneous flow velocity was obtained at different vertical planes using the particle image velocimetry (PIV) technique. The partially exposed pile-cap case showed the maximum obtained scour-depth (MSD). The reason behind the MSD occurrence in case II was enunciated through the analysis of turbulent flow field which showed that as the pile-cap got exposed to the flow, it dominantly affected the generation of vortices from the pile-cap corners responsible for the higher scour depth. The effect of the pile-cap on the flow field was prominently seen in case III through the mean velocities, vorticity, Reynolds shear stresses and turbulent kinetic energy contours, but since the pile-cap was away from the bed, the pile-cap corners did not show any direct effect on the scour.

KEYWORDS: 

References

  • Adrian, R. J. (2013). Structure of turbulent boundary layers. In Jeremy G. Venditti, James L. Best, Michael Church, & Richard J. Hardy (Eds.), Coherent flow structures at earth’s surface (pp. 17–24). John Wiley and Sons. [Crossref], [Google Scholar]
  • Adrian, R. J., & Westerweel, J. (2011). Particle image velocimetry, No. 30. Cambridge University Press. [Google Scholar]
  • Alemi, M., & Maia, R. (2018). Numerical simulation of the flow and local scour process around single and complex bridge piers. International Journal of Civil Engineering, 16(5), 475–487. https://doi.org/10.1007/s40999-016-0137-8 [Crossref], [Google Scholar]
  • Alemi, M., Pêgo, J. P., & Maia, R. (2019). Numerical simulation of the turbulent flow around a complex bridge pier on the scoured bed. European Journal of Mechanics – B/Fluids, 76, 316–331. https://doi.org/10.1016/j.euromechflu.2019.03.011 [Crossref], [Web of Science ®], [Google Scholar]
  • Amini, A., Hamidi, S., Shirzadi, A., Behmanesh, J., & Akib, S. (2021). Efficiency of artificial neural networks in determining scour depth at composite bridge piers. International Journal of River Basin Management, 19(3), 327–333. https://doi.org/10.1080/15715124.2020.1742138 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Arneson, L. A., Zevenbergen, L. W., Lagasse, P. F., & Clopper, P. E. (2015). Evaluating scour at bridges, 5th ed. hydraulic engineering circular No. 18 (HEC-18). Federal Highway Administration. [Google Scholar]
  • Ataie-Ashtiani, B., & Aslani-Kordkandi, A. (2012). Flow field around side-by-side piers with and without a scour hole. European Journal of Mechanics – B/Fluids, 36, 152–166. https://doi.org/10.1016/j.euromechflu.2012.03.007 [Crossref], [Web of Science ®], [Google Scholar]
  • Ataie-Ashtiani, B., Baratian-Ghorghi, Z., & Beheshti, A. A. (2010). Experimental investigation of clear-water local scour of compound piers. Journal of Hydraulic Engineering, 136(6), 343–351. https://doi.org/10.1061/(ASCE)0733-9429(2010)136:6(343) [Crossref], [Web of Science ®], [Google Scholar]
  • Avallone, F., Discetti, S., Astarita, T., & Cardone, G. (2015). Convergence enhancement of single-pixel PIV with symmetric double correlation. Experiments in Fluids, 56(4), 71. https://doi.org/10.1007/s00348-015-1938-2 [Crossref], [Web of Science ®], [Google Scholar]
  • Beheshti, A. A., & Ataie-Ashtiani, B. (2010). Experimental study of three-dimensional flow field around a complex bridge pier. Journal of Engineering Mechanics, 136(2), 143–154. https://doi.org/10.1061/(ASCE)EM.1943-7889.0000073 [Crossref], [Web of Science ®], [Google Scholar]
  • Beheshti, A. A., & Ataie-Ashtiani, B. (2016). Scour hole influence on turbulent flow field around complex bridge piers. Flow, Turbulence and Combustion, 97(2), 451–474. https://doi.org/10.1007/s10494-016-9707-8 [Crossref], [Web of Science ®], [Google Scholar]
  • Cameron, S. M., Nikora, V. I., & Marusic, I. (2019). Drag forces on a bed particle in open-channel flow: Effects of pressure spatial fluctuations and very-large-scale motions. Journal of Fluid Mechanics, 863, 494–512. https://doi.org/10.1017/jfm.2018.1003 [Crossref], [Web of Science ®], [Google Scholar]
  • Cheng, N., & Emadzadeh, A. (2017). Laboratory measurements of vortex-induced sediment pickup rates. International Journal of Sediment Research, 32(1), 98–104. https://doi.org/10.1016/j.ijsrc.2016.04.005 [Crossref], [Web of Science ®], [Google Scholar]
  • Coleman, S. E. (2005). Clearwater local scour at complex piers. Journal of Hydraulic Engineering, 131(4), 330–334. https://doi.org/10.1061/(ASCE)0733-9429(2005)131:4(330) [Crossref], [Web of Science ®], [Google Scholar]
  • Das, S., & Mazumdar, A. (2015). Turbulence flow field around two eccentric circular piers in scour hole. International Journal of River Basin Management, 13(3), 343–361. https://doi.org/10.1080/15715124.2015.1012515 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Esmaeili Varaki, M., Radice, A., Samira Hossini, S., & Fazl Ola, R. (2019). Local scour at a complex pier with inclined columns footed on capped piles: Effect of the pile arrangement and of the cap thickness and elevation. ISH Journal of Hydraulic Engineering, 1–10. https://doi.org/10.1080/09715010.2019.1702109 [Taylor & Francis Online], [Google Scholar]
  • Ferraro, D., Tafarojnoruz, A., Gaudio, R., & Cardoso, A. H. (2013). Effects of pile cap thickness on the maximum scour depth at a complex pier. Journal of Hydraulic Engineering, 139(5), 482–491. https://doi.org/10.1061/(ASCE)HY.1943-7900.0000704 [Crossref], [Web of Science ®], [Google Scholar]
  • Gaudio, R., Tafarojnoruz, A., & Calomino, F. (2012). Combined flow-altering countermeasures against bridge pier scour. Journal of Hydraulic Research, 50(1), 35–43. https://doi.org/10.1080/00221686.2011.649548 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Gautam, P., Eldho, T., & Behera, M. (2016). Experimental study of scour around a complex pier with elliptical pile-cap. In J. Harris, R. Whitehouse, & S. Moxon (Eds.), Scour and Erosion: Proceedings of the 8th International Conference on Scour and Erosion (Oxford, UK, 12-15 September 2016) (pp. 759–765). CRC Press. [Crossref], [Google Scholar]
  • Gautam, P., Eldho, T. I., Mazumder, B. S., & Behera, M. R. (2019). Experimental study of flow and turbulence characteristics around simple and complex piers using PIV. Experimental Thermal and Fluid Science, 100, 193–206. https://doi.org/10.1016/j.expthermflusci.2018.09.010 [Crossref], [Web of Science ®], [Google Scholar]
  • Graf, W. H., & Istiarto, I. (2002). Flow pattern in the scour hole around a cylinder. Journal of Hydraulic Research, 40(1), 13–20. https://doi.org/10.1080/00221680209499869 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Hjulstrom, F. (1935). Study of the morphological activity of Rivers as illustrated by the River fyris bulletin, vol. 25. Geological Institute of Upsala. [Google Scholar]
  • Kumar, A., & Kothyari, U. C. (2012). Three-dimensional flow characteristics within the scour hole around circular uniform and compound piers. Journal of Hydraulic Engineering, 138(5), 420–429. https://doi.org/10.1061/(ASCE)HY.1943-7900.0000527 [Crossref], [Web of Science ®], [Google Scholar]
  • Mashahir, M. B., Zarrati, A. R., & Rezayi, M. J. (2004). Time development of scouring around a bridge pier protected by collar. In Proceedings 2nd International Conference on Scour and Erosion (ICSE-2). November 14–17, 2004, Singapore. [Google Scholar]
  • Melville, B. W. (2008). The physics of local scour at bridge piers. In Proceedings of the 4th International Conference on Scour and Erosion (ICSE-4). November 5-7, 2008, Tokyo, Japan (pp. 28–40). [Google Scholar]
  • Melville, B. W., & Chiew, Y. M. (1999). Time scale for local scour at bridge piers. Journal of Hydraulic Engineering, 125(1), 59–65. https://doi.org/10.1061/(ASCE)0733-9429(1999)125:1(59) [Crossref], [Web of Science ®], [Google Scholar]
  • Melville, B. W., & Raudkivi, A. J. (1977). Flow characteristics in local scour at bridge piers. Journal of Hydraulic Research, 15(4), 373–380. https://doi.org/10.1080/00221687709499641 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Moreno, M., Maia, R., & Couto, L. (2016a). Effects of relative column width and pile-cap elevation on local scour depth around complex piers. Journal of Hydraulic Engineering, 142(2), 04015051. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001080 [Crossref], [Web of Science ®], [Google Scholar]
  • Moreno, M., Maia, R., & Couto, L. (2016b). Prediction of equilibrium local scour depth at complex bridge piers. Journal of Hydraulic Engineering, 142(11), 04016045. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001153 [Crossref], [Web of Science ®], [Google Scholar]
  • Nezu, I., & Rodi, W. (1986). Open-channel flow measurements with a laser Doppler anemometer. Journal of Hydraulic Engineering, 112(5), 335–355. https://doi.org/10.1061/(ASCE)0733-9429(1986)112:5(335) [Crossref], [Web of Science ®], [Google Scholar]
  • Radice, A., & Tran, C. K. (2012). Study of sediment motion in scour hole of a circular pier. Journal of Hydraulic Research, 50(1), 44–51. https://doi.org/10.1080/00221686.2011.641764 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Richardson, J. R., & York, K. (1999). Hydrodynamic countermeasures for local pier scour. Transportation Research Record: Journal of the Transportation Research Board, 1690(1), 186–192. https://doi.org/10.3141/1690-21 [Crossref], [Google Scholar]
  • Saw, E., Debue, P., Kuzzay, D., Daviaud, F., & Dubrulle, B. (2018). On the universality of anomalous scaling exponents of structure functions in turbulent flows. Journal of Fluid Mechanics, 837, 657–669. https://doi.org/10.1017/jfm.2017.848 [Crossref], [Web of Science ®], [Google Scholar]
  • Schlichting, H. (1968). Boundary layer theory (Vol. 960). McGraw-Hill. [Google Scholar]
  • Sheppard, D. M., Demir, H., & Melville, B. W. (2011). Scour at wide piers and long skewed piers (Vol. 682). Transportation Research Board. [Google Scholar]
  • Tafarojnoruz, A., Gaudio, R., & Calomino, F. (2012). Bridge pier scour mitigation under steady and unsteady flow conditions. Acta Geophysica, 60(4), 1076–1097. https://doi.org/10.2478/s11600-012-0040-x [Crossref], [Web of Science ®], [Google Scholar]
  • Tafarojnoruz, A., Gaudio, R., & Dey, S. (2010). Flow-altering countermeasures against scour at bridge piers: A review. Journal of Hydraulic Research, 48(4), 441–452. https://doi.org/10.1080/00221686.2010.491645 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Tennekes, H., & Lumley, J. L. (1972). A first course in turbulence. MIT press. [Crossref], [Google Scholar]
  • Veerappadevaru, G., Gangadharaiah, T., & Jagadeesh, T. R. (2011). Vortex scouring process around bridge pier with a caisson. Journal of Hydraulic Research, 49(3), 378–383. https://doi.org/10.1080/00221686.2011.568195 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Veerappadevaru, G., Gangadharaiah, T., & Jagadeesh, T. R. (2012). Temporal variation of vortex scour process around caisson piers. Journal of Hydraulic Research, 50(2), 200–207. https://doi.org/10.1080/00221686.2012.666832 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Vijayasree, B. A., Eldho, T. I., Mazumder, B. S., & Ahmad, N. (2019). Influence of bridge pier shape on flow field and scour geometry. International Journal of River Basin Management, 17(1), 109–129. https://doi.org/10.1080/15715124.2017.1394315 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Yang, Y., Melville, B. W., Sheppard, D. M., & Shamseldin, A. Y. (2018). Clear-water local scour at skewed complex bridge piers. Journal of Hydraulic Engineering, 144(6), 04018019. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001458 [Crossref], [Web of Science ®], [Google Scholar]
  • Yang, Y., Melville, B. W., Macky, G. H., & Shamseldin, A. Y. (2020). Temporal evolution of clear-water local scour at aligned and skewed complex bridge piers. Journal of Hydraulic Engineering, 146(4), 04020026. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001732 [Crossref], [Web of Science ®], [Google Scholar]
원문 보기
Figure 11. Sketch of scour mechanism around USAF under random waves.

Scour Characteristics and Equilibrium Scour Depth Prediction around Umbrella Suction Anchor Foundation under Random Waves

by Ruigeng Hu 1,Hongjun Liu 2,Hao Leng 1,Peng Yu 3 andXiuhai Wang 1,2,*

1College of Environmental Science and Engineering, Ocean University of China, Qingdao 266000, China

2Key Lab of Marine Environment and Ecology (Ocean University of China), Ministry of Education, Qingdao 266000, China

3Qingdao Geo-Engineering Survering Institute, Qingdao 266100, China

*Author to whom correspondence should be addressed.

J. Mar. Sci. Eng. 2021, 9(8), 886; https://doi.org/10.3390/jmse9080886

Received: 6 July 2021 / Revised: 8 August 2021 / Accepted: 13 August 2021 / Published: 17 August 2021

(This article belongs to the Section Ocean Engineering)

Download 

Abstract

A series of numerical simulation were conducted to study the local scour around umbrella suction anchor foundation (USAF) under random waves. In this study, the validation was carried out firstly to verify the accuracy of the present model. Furthermore, the scour evolution and scour mechanism were analyzed respectively. In addition, two revised models were proposed to predict the equilibrium scour depth Seq around USAF. At last, a parametric study was carried out to study the effects of the Froude number Fr and Euler number Eu for the Seq. The results indicate that the present numerical model is accurate and reasonable for depicting the scour morphology under random waves. The revised Raaijmakers’s model shows good agreement with the simulating results of the present study when KCs,p < 8. The predicting results of the revised stochastic model are the most favorable for n = 10 when KCrms,a < 4. The higher Fr and Eu both lead to the more intensive horseshoe vortex and larger Seq.

Keywords: 

scour; numerical investigation; random waves; equilibrium scour depth; KC number

1. Introduction

The rapid expansion of cities tends to cause social and economic problems, such as environmental pollution and traffic jam. As a kind of clean energy, offshore wind power has developed rapidly in recent years. The foundation of offshore wind turbine (OWT) supports the upper tower, and suffers the cyclic loading induced by waves, tides and winds, which exerts a vital influence on the OWT system. The types of OWT foundation include the fixed and floating foundation, and the fixed foundation was used usually for nearshore wind turbine. After the construction of fixed foundation, the hydrodynamic field changes in the vicinity of the foundation, leading to the horseshoe vortex formation and streamline compression at the upside and sides of foundation respectively [1,2,3,4]. As a result, the neighboring soil would be carried away by the shear stress induced by vortex, and the scour hole would emerge in the vicinity of foundation. The scour holes increase the cantilever length, and weaken the lateral bearing capacity of foundation [5,6,7,8,9]. Moreover, the natural frequency of OWT system increases with the increase of cantilever length, causing the resonance occurs when the system natural frequency equals the wave or wind frequency [10,11,12]. Given that, an innovative foundation called umbrella suction anchor foundation (USAF) has been designed for nearshore wind power. The previous studies indicated the USAF was characterized by the favorable lateral bearing capacity with the low cost [6,13,14]. The close-up of USAF is shown in Figure 1, and it includes six parts: 1-interal buckets, 2-external skirt, 3-anchor ring, 4-anchor branch, 5-supporting rod, 6-telescopic hook. The detailed description and application method of USAF can be found in reference [13].

Jmse 09 00886 g001 550

Figure 1. The close-up of umbrella suction anchor foundation (USAF).

Numerical and experimental investigations of scour around OWT foundation under steady currents and waves have been extensively studied by many researchers [1,2,15,16,17,18,19,20,21,22,23,24]. The seabed scour can be classified as two types according to Shields parameter θ, i.e., clear bed scour (θ < θcr) or live bed scour (θ > θcr). Due to the set of foundation, the adverse hydraulic pressure gradient exists at upstream foundation edges, resulting in the streamline separation between boundary layer flow and seabed. The separating boundary layer ascended at upstream anchor edges and developed into the horseshoe vortex. Then, the horseshoe vortex moved downstream gradually along the periphery of the anchor, and the vortex shed off continually at the lee-side of the anchor, i.e., wake vortex. The core of wake vortex is a negative pressure center, liking a vacuum cleaner. Hence, the soil particles were swirled into the negative pressure core and carried away by wake vortexes. At the same time, the onset of scour at rear side occurred. Finally, the wake vortex became downflow when the turbulence energy could not support the survival of wake vortex. According to Tavouktsoglou et al. [25], the scale of pile wall boundary layer is proportional to 1/ln(Rd) (Rd is pile Reynolds), which means the turbulence intensity induced by the flow-structure interaction would decrease with Rd increases, but the effects of Rd can be neglected only if the flow around the foundation is fully turbulent [26]. According to previous studies [1,15,27,28,29,30,31,32], the scour development around pile foundation under waves was significantly influenced by Shields parameter θ and KC number simultaneously (calculated by Equation (1)). Sand ripples widely existed around pile under waves in the case of live bed scour, and the scour morphology is related with θ and KC. Compared with θ, KC has a greater influence on the scour morphology [21,27,28]. The influence mechanism of KC on the scour around the pile is reflected in two aspects: the horseshoe vortex at upstream and wake vortex shedding at downstream.

KC=UwmTD��=�wm��(1)

where, Uwm is the maximum velocity of the undisturbed wave-induced oscillatory flow at the sea bottom above the wave boundary layer, T is wave period, and D is pile diameter.

There are two prerequisites to satisfy the formation of horseshoe vortex at upstream pile edges: (1) the incoming flow boundary layer with sufficient thickness and (2) the magnitude of upstream adverse pressure gradient making the boundary layer separating [1,15,16,18,20]. The smaller KC results the lower adverse pressure gradient, and the boundary layer cannot separate, herein, there is almost no horseshoe vortex emerging at upside of pile. Sumer et al. [1,15] carried out several sets of wave flume experiments under regular and irregular waves respectively, and the experiment results show that there is no horseshoe vortex when KC is less than 6. While the scale and lifespan of horseshoe vortex increase evidently with the increase of KC when KC is larger than 6. Moreover, the wake vortex contributes to the scour at lee-side of pile. Similar with the case of horseshoe vortex, there is no wake vortex when KC is less than 6. The wake vortex is mainly responsible for scour around pile when KC is greater than 6 and less than O(100), while horseshoe vortex controls scour nearly when KC is greater than O(100).

Sumer et al. [1] found that the equilibrium scour depth was nil around pile when KC was less than 6 under regular waves for live bed scour, while the equilibrium scour depth increased with the increase of KC. Based on that, Sumer proposed an equilibrium scour depth predicting equation (Equation (2)). Carreiras et al. [33] revised Sumer’s equation with m = 0.06 for nonlinear waves. Different with the findings of Sumer et al. [1] and Carreiras et al. [33], Corvaro et al. [21] found the scour still occurred for KC ≈ 4, and proposed the revised equilibrium scour depth predicting equation (Equation (3)) for KC > 4.

Rudolph and Bos [2] conducted a series of wave flume experiments to investigate the scour depth around monopile under waves only, waves and currents combined respectively, indicting KC was one of key parameters in influencing equilibrium scour depth, and proposed the equilibrium scour depth predicting equation (Equation (4)) for low KC (1 < KC < 10). Through analyzing the extensive data from published literatures, Raaijmakers and Rudolph [34] developed the equilibrium scour depth predicting equation (Equation (5)) for low KC, which was suitable for waves only, waves and currents combined. Khalfin [35] carried out several sets of wave flume experiments to study scour development around monopile, and proposed the equilibrium scour depth predicting equation (Equation (6)) for low KC (0.1 < KC < 3.5). Different with above equations, the Khalfin’s equation considers the Shields parameter θ and KC number simultaneously in predicting equilibrium scour depth. The flow reversal occurred under through in one wave period, so sand particles would be carried away from lee-side of pile to upside, resulting in sand particles backfilled into the upstream scour hole [20,29]. Considering the backfilling effects, Zanke et al. [36] proposed the equilibrium scour depth predicting equation (Equation (7)) around pile by theoretical analysis, and the equation is suitable for the whole range of KC number under regular waves and currents combined.

S/D=1.3(1−exp([−m(KC−6)])�/�=1.3(1−exp(−�(��−6))(2)

where, m = 0.03 for linear waves.

S/D=1.3(1−exp([−0.02(KC−4)])�/�=1.3(1−exp(−0.02(��−4))(3)

S/D=1.3γKwaveKhw�/�=1.3��wave�ℎw(4)

where, γ is safety factor, depending on design process, typically γ = 1.5, Kwave is correction factor considering wave action, Khw is correction factor considering water depth.

S/D=1.5[tanh(hwD)]KwaveKhw�/�=1.5tanh(ℎw�)�wave�ℎw(5)

where, hw is water depth.

S/D=0.0753(θθcr−−−√−0.5)0.69KC0.68�/�=0.0753(��cr−0.5)0.69��0.68(6)

where, θ is shields parameter, θcr is critical shields parameter.

S/D=2.5(1−0.5u/uc)xrelxrel=xeff/(1+xeff)xeff=0.03(1−0.35ucr/u)(KC−6)⎫⎭⎬⎪⎪�/�=2.5(1−0.5�/��)��������=����/(1+����)����=0.03(1−0.35�cr/�)(��−6)(7)

where, u is near-bed orbital velocity amplitude, uc is critical velocity corresponding the onset of sediment motion.

S/D=1.3{1−exp[−0.03(KC2lnn+36)1/2−6]}�/�=1.31−exp−0.03(��2ln�+36)1/2−6(8)

where, n is the 1/n’th highest wave for random waves

For predicting equilibrium scour depth under irregular waves, i.e., random waves, Sumer and Fredsøe [16] found it’s suitable to take Equation (2) to predict equilibrium scour depth around pile under random waves with the root-mean-square (RMS) value of near-bed orbital velocity amplitude Um and peak wave period TP to calculate KC. Khalfin [35] recommended the RMS wave height Hrms and peak wave period TP were used to calculate KC for Equation (6). References [37,38,39,40] developed a series of stochastic theoretical models to predict equilibrium scour depth around pile under random waves, nonlinear random waves plus currents respectively. The stochastic approach thought the 1/n’th highest wave were responsible for scour in vicinity of pile under random waves, and the KC was calculated in Equation (8) with Um and mean zero-crossing wave period Tz. The results calculated by Equation (8) agree well with experimental values of Sumer and Fredsøe [16] if the 1/10′th highest wave was used. To author’s knowledge, the stochastic approach proposed by Myrhaug and Rue [37] is the only theoretical model to predict equilibrium scour depth around pile under random waves for the whole range of KC number in published documents. Other methods of predicting scour depth under random waves are mainly originated from the equation for regular waves-only, waves and currents combined, which are limited to the large KC number, such as KC > 6 for Equation (2) and KC > 4 for Equation (3) respectively. However, situations with relatively low KC number (KC < 4) often occur in reality, for example, monopile or suction anchor for OWT foundations in ocean environment. Moreover, local scour around OWT foundations under random waves has not yet been investigated fully. Therefore, further study are still needed in the aspect of scour around OWT foundations with low KC number under random waves. Given that, this study presents the scour sediment model around umbrella suction anchor foundation (USAF) under random waves. In this study, a comparison of equilibrium scour depth around USAF between this present numerical models and the previous theoretical models and experimental results was presented firstly. Then, this study gave a comprehensive analysis for the scour mechanisms around USAF. After that, two revised models were proposed according to the model of Raaijmakers and Rudolph [34] and the stochastic model developed by Myrhaug and Rue [37] respectively to predict the equilibrium scour depth. Finally, a parametric study was conducted to study the effects of the Froude number (Fr) and Euler number (Eu) to equilibrium scour depth respectively.

2. Numerical Method

2.1. Governing Equations of Flow

The following equations adopted in present model are already available in Flow 3D software. The authors used these theoretical equations to simulate scour in random waves without modification. The incompressible viscous fluid motion satisfies the Reynolds-averaged Navier-Stokes (RANS) equation, so the present numerical model solves RANS equations:

∂u∂t+1VF(uAx∂u∂x+vAy∂u∂y+wAz∂u∂z)=−1ρf∂p∂x+Gx+fx∂�∂�+1��(���∂�∂�+���∂�∂�+���∂�∂�)=−1�f∂�∂�+��+��(9)

∂v∂t+1VF(uAx∂v∂x+vAy∂v∂y+wAz∂v∂z)=−1ρf∂p∂y+Gy+fy∂�∂�+1��(���∂�∂�+���∂�∂�+���∂�∂�)=−1�f∂�∂�+��+��(10)

∂w∂t+1VF(uAx∂w∂x+vAy∂w∂y+wAz∂w∂z)=−1ρf∂p∂z+Gz+fz∂�∂�+1��(���∂�∂�+���∂�∂�+���∂�∂�)=−1�f∂�∂�+��+��(11)

where, VF is the volume fraction; u, v, and w are the velocity components in x, y, z direction respectively with Cartesian coordinates; Ai is the area fraction; ρf is the fluid density, fi is the viscous fluid acceleration, Gi is the fluid body acceleration (i = x, y, z).

2.2. Turbulent Model

The turbulence closure is available by the turbulent model, such as one-equation, the one-equation k-ε model, the standard k-ε model, RNG k-ε turbulent model and large eddy simulation (LES) model. The LES model requires very fine mesh grid, so the computational time is large, which hinders the LES model application in engineering. The RNG k-ε model can reduce computational time greatly with high accuracy in the near-wall region. Furthermore, the RNG k-ε model computes the maximum turbulent mixing length dynamically in simulating sediment scour model. Therefore, the RNG k-ε model was adopted to study the scour around anchor under random waves [41,42].

∂kT∂T+1VF(uAx∂kT∂x+vAy∂kT∂y+wAz∂kT∂z)=PT+GT+DiffkT−εkT∂��∂�+1��(���∂��∂�+���∂��∂�+���∂��∂�)=��+��+������−���(12)

∂εT∂T+1VF(uAx∂εT∂x+vAy∂εT∂y+wAz∂εT∂z)=CDIS1εTkT(PT+CDIS3GT)+Diffε−CDIS2ε2TkT∂��∂�+1��(���∂��∂�+���∂��∂�+���∂��∂�)=����1����(��+����3��)+�����−����2��2��(13)

where, kT is specific kinetic energy involved with turbulent velocity, GT is the turbulent energy generated by buoyancy; εT is the turbulent energy dissipating rate, PT is the turbulent energy, Diffε and DiffkT are diffusion terms associated with VF, Ai; CDIS1, CDIS2 and CDIS3 are dimensionless parameters, and CDIS1, CDIS3 have default values of 1.42, 0.2 respectively. CDIS2 can be obtained from PT and kT.

2.3. Sediment Scour Model

The sand particles may suffer four processes under waves, i.e., entrainment, bed load transport, suspended load transport, and deposition, so the sediment scour model should depict the above processes efficiently. In present numerical simulation, the sediment scour model includes the following aspects:

2.3.1. Entrainment and Deposition

The combination of entrainment and deposition determines the net scour rate of seabed in present sediment scour model. The entrainment lift velocity of sand particles was calculated as [43]:

ulift,i=αinsd0.3∗(θ−θcr)1.5∥g∥di(ρi−ρf)ρf−−−−−−−−−−−−√�lift,i=�����*0.3(�−�cr)1.5���(��−�f)�f(14)

where, αi is the entrainment parameter, ns is the outward point perpendicular to the seabed, d* is the dimensionless diameter of sand particles, which was calculated by Equation (15), θcr is the critical Shields parameter, g is the gravity acceleration, di is the diameter of sand particles, ρi is the density of seabed species.

d∗=di(∥g∥ρf(ρi−ρf)μ2f)1/3�*=��(��f(��−�f)�f2)1/3(15)

where μf is the fluid dynamic viscosity.

In Equation (14), the entrainment parameter αi confirms the rate at which sediment erodes when the given shear stress is larger than the critical shear stress, and the recommended value 0.018 was adopted according to the experimental data of Mastbergen and Von den Berg [43]. ns is the outward pointing normal to the seabed interface, and ns = (0,0,1) according to the Cartesian coordinates used in present numerical model.

The shields parameter was obtained from the following equation:

θ=U2f,m(ρi/ρf−1)gd50�=�f,m2(��/�f−1)��50(16)

where, Uf,m is the maximum value of the near-bed friction velocity; d50 is the median diameter of sand particles. The detailed calculation procedure of θ was available in Soulsby [44].

The critical shields parameter θcr was obtained from the Equation (17) [44]

θcr=0.31+1.2d∗+0.055[1−exp(−0.02d∗)]�cr=0.31+1.2�*+0.0551−exp(−0.02�*)(17)

The sand particles begin to deposit on seabed when the turbulence energy weaken and cann’t support the particles suspending. The setting velocity of the particles was calculated from the following equation [44]:

usettling,i=νfdi[(10.362+1.049d3∗)0.5−10.36]�settling,�=�f��(10.362+1.049�*3)0.5−10.36(18)

where νf is the fluid kinematic viscosity.

2.3.2. Bed Load Transport

This is called bed load transport when the sand particles roll or bounce over the seabed and always have contact with seabed. The bed load transport velocity was computed by [45]:

ubedload,i=qb,iδicb,ifb�bedload,�=�b,����b,��b(19)

where, qb,i is the bed load transport rate, which was obtained from Equation (20), δi is the bed load thickness, which was calculated by Equation (21), cb,i is the volume fraction of sand i in the multiple species, fb is the critical packing fraction of the seabed.

qb,i=8[∥g∥(ρi−ρfρf)d3i]1/2�b,�=8�(��−�f�f)��31/2(20)

δi=0.3d0.7∗(θθcr−1)0.5di��=0.3�*0.7(��cr−1)0.5��(21)

2.3.3. Suspended Load Transport

Through the following transport equation, the suspended sediment concentration could be acquired.

∂Cs,i∂t+∇(us,iCs,i)=∇∇(DfCs,i)∂�s,�∂�+∇(�s,��s,�)=∇∇(�f�s,�)(22)

where, Cs,i is the suspended sand particles mass concentration of sand i in the multiple species, us,i is the sand particles velocity of sand i, Df is the diffusivity.

The velocity of sand i in the multiple species could be obtained from the following equation:

us,i=u¯¯+usettling,ics,i�s,�=�¯+�settling,��s,�(23)

where, u¯�¯ is the velocity of mixed fluid-particles, which can be calculated by the RANS equation with turbulence model, cs,i is the suspended sand particles volume concentration, which was computed from Equation (24).

cs,i=Cs,iρi�s,�=�s,���(24)

3. Model Setup

The seabed-USAF-wave three-dimensional scour numerical model was built using Flow-3D software. As shown in Figure 2, the model includes sandy seabed, USAF model, sea water, two baffles and porous media. The dimensions of USAF are shown in Table 1. The sandy bed (210 m in length, 30 m in width and 11 m in height) is made up of uniform fine sand with median diameter d50 = 0.041 cm. The USAF model includes upper steel tube with the length of 20 m, which was installed in the middle of seabed. The location of USAF is positioned at 140 m from the upstream inflow boundary and 70 m from the downstream outflow boundary. Two baffles were installed at two ends of seabed. In order to eliminate the wave reflection basically, the porous media was set at the outflow side on the seabed.

Jmse 09 00886 g002 550

Figure 2. (a) The sketch of seabed-USAF-wave three-dimensional model; (b) boundary condation:Wv-wave boundary, S-symmetric boundary, O-outflow boundary; (c) USAF model.

Table 1. Numerical simulating cases.

Table

3.1. Mesh Geometric Dimensions

In the simulation of the scour under the random waves, the model includes the umbrella suction anchor foundation, seabed and fluid. As shown in Figure 3, the model mesh includes global mesh grid and nested mesh grid, and the total number of grids is 1,812,000. The basic procedure for building mesh grid consists of two steps. Step 1: Divide the global mesh using regular hexahedron with size of 0.6 × 0.6. The global mesh area is cubic box, embracing the seabed and whole fluid volume, and the dimensions are 210 m in length, 30 m in width and 32 m in height. The details of determining the grid size can see the following mesh sensitivity section. Step 2: Set nested fine mesh grid in vicinity of the USAF with size of 0.3 × 0.3 so as to shorten the computation cost and improve the calculation accuracy. The encryption range is −15 m to 15 m in x direction, −15 m to 15 m in y direction and 0 m to 32 m in z direction, respectively. In order to accurately capture the free-surface dynamics, such as the fluid-air interface, the volume of fluid (VOF) method was adopted for tracking the free water surface. One specific algorithm called FAVORTM (Fractional Area/Volume Obstacle Representation) was used to define the fractional face areas and fractional volumes of the cells which are open to fluid flow.

Jmse 09 00886 g003 550

Figure 3. The sketch of mesh grid.

3.2. Boundary Conditions

As shown in Figure 2, the initial fluid length is 210 m as long as seabed. A wave boundary was specified at the upstream offshore end. The details of determining the random wave spectrum can see the following wave parameters section. The outflow boundary was set at the downstream onshore end. The symmetry boundary was used at the top and two sides of the model. The symmetric boundaries were the better strategy to improve the computation efficiency and save the calculation cost [46]. At the seabed bottom, the wall boundary was adopted, which means the u = v = w= 0. Besides, the upper steel tube of USAF was set as no-slip condition.

3.3. Wave Parameters

The random waves with JONSWAP wave spectrum were used for all simulations as realistic representation of offshore conditions. The unidirectional JONSWAP frequency spectrum was described as [47]:

S(ω)=αg2ω5exp[−54(ωpω)4]γexp[−(ω−ωp)22σ2ω2p]�(�)=��2�5exp−54(�p�)4�exp−(�−�p)22�2�p2(25)

where, α is wave energy scale parameter, which is calculated by Equation (26), ω is frequency, ωp is wave spectrum peak frequency, which can be obtained from Equation (27). γ is wave spectrum peak enhancement factor, in this study γ = 3.3. σ is spectral width factor, σ equals 0.07 for ω ≤ ωp and 0.09 for ω > ωp respectively.

α=0.0076(gXU2)−0.22�=0.0076(���2)−0.22(26)

ωp=22(gU)(gXU2)−0.33�p=22(��)(���2)−0.33(27)

where, X is fetch length, U is average wind velocity at 10 m height from mean sea level.

In present numerical model, the input key parameters include X and U for wave boundary with JONSWAP wave spectrum. The objective wave height and period are available by different combinations of X and U. In this study, we designed 9 cases with different wave heights, periods and water depths for simulating scour around USAF under random waves (see Table 2). For random waves, the wave steepness ε and Ursell number Ur were acquired form Equations (28) and (29) respectively

ε=2πgHsT2a�=2���s�a2(28)

Ur=Hsk2h3w�r=�s�2ℎw3(29)

where, Hs is significant wave height, Ta is average wave period, k is wave number, hw is water depth. The Shield parameter θ satisfies θ > θcr for all simulations in current study, indicating the live bed scour prevails.

Table 2. Numerical simulating cases.

Table

3.4. Mesh Sensitivity

In this section, a mesh sensitivity analysis was conducted to investigate the influence of mesh grid size to results and make sure the calculation is mesh size independent and converged. Three mesh grid size were chosen: Mesh 1—global mesh grid size of 0.75 × 0.75, nested fine mesh grid size of 0.4 × 0.4, and total number of grids 1,724,000, Mesh 2—global mesh grid size of 0.6 × 0.6, nested fine mesh grid size of 0.3 × 0.3, and total number of grids 1,812,000, Mesh 3—global mesh grid size of 0.4 × 0.4, nested fine mesh grid size of 0.2 × 0.2, and total number of grids 1,932,000. The near-bed shear velocity U* is an important factor for influencing scour process [1,15], so U* at the position of (4,0,11.12) was evaluated under three mesh sizes. As the Figure 4 shown, the maximum error of shear velocity ∆U*1,2 is about 39.8% between the mesh 1 and mesh 2, and 4.8% between the mesh 2 and mesh 3. According to the mesh sensitivity criterion adopted by Pang et al. [48], it’s reasonable to think the results are mesh size independent and converged with mesh 2. Additionally, the present model was built according to prototype size, and the mesh size used in present model is larger than the mesh size adopted by Higueira et al. [49] and Corvaro et al. [50]. If we choose the smallest cell size, it will take too much time. For example, the simulation with Mesh3 required about 260 h by using a computer with Intel Xeon Scalable Gold 4214 CPU @24 Cores, 2.2 GHz and 64.00 GB RAM. Therefore, in this case, considering calculation accuracy and computation efficiency, the mesh 2 was chosen for all the simulation in this study.

Jmse 09 00886 g004 550

Figure 4. Comparison of near-bed shear velocity U* with different mesh grid size.

The nested mesh block was adopted for seabed in vicinity of the USAF, which was overlapped with the global mesh block. When two mesh blocks overlap each other, the governing equations are by default solved on the mesh block with smaller average cell size (i.e., higher grid resolution). It is should be noted that the Flow 3D software used the moving mesh captures the scour evolution and automatically adjusts the time step size to be as large as possible without exceeding any of the stability limits, affecting accuracy, or unduly increasing the effort required to enforce the continuity condition [51].

3.5. Model Validation

In order to verify the reliability of the present model, the results of present study were compared with the experimental data of Khosronejad et al. [52]. The experiment was conducted in an open channel with a slender vertical pile under unidirectional currents. The comparison of scour development between the present results and the experimental results is shown in Figure 5. The Figure 5 reveals that the present results agree well with the experimental data of Khosronejad et al. [52]. In the first stage, the scour depth increases rapidly. After that, the scour depth achieves a maximum value gradually. The equilibrium scour depth calculated by the present model is basically corresponding with the experimental results of Khosronejad et al. [52], although scour depth in the present model is slightly larger than the experimental results at initial stage.

Jmse 09 00886 g005 550

Figure 5. Comparison of time evolution of scour between the present study and Khosronejad et al. [52], Petersen et al. [17].

Secondly, another comparison was further conducted between the results of present study and the experimental data of Petersen et al. [17]. The experiment was carried out in a flume with a circular vertical pile in combined waves and current. Figure 4 shows a comparison of time evolution of scour depth between the simulating and the experimental results. As Figure 5 indicates, the scour depth in this study has good overall agreement with the experimental results proposed in Petersen et al. [17]. The equilibrium scour depth calculated by the present model is 0.399 m, which equals to the experimental value basically. Overall, the above verifications prove the present model is accurate and capable in dealing with sediment scour under waves.

In addition, in order to calibrate and validate the present model for hydrodynamic parameters, the comparison of water surface elevation was carried out with laboratory experiments conducted by Stahlmann [53] for wave gauge No. 3. The Figure 6 depicts the surface wave profiles between experiments and numerical model results. The comparison indicates that there is a good agreement between the model results and experimental values, especially the locations of wave crest and trough. Comparison of the surface elevation instructs the present model has an acceptable relative error, and the model is a calibrated in terms of the hydrodynamic parameters.

Jmse 09 00886 g006 550

Figure 6. Comparison of surface elevation between the present study and Stahlmann [53].

Finally, another comparison was conducted for equilibrium scour depth or maximum scour depth under random waves with the experimental data of Sumer and Fredsøe [16] and Schendel et al. [22]. The Figure 7 shows the comparison between the numerical results and experimental data of Run01, Run05, Run21 and Run22 in Sumer and Fredsøe [16] and test A05 and A09 in Schendel et al. [22]. As shown in Figure 7, the equilibrium scour depth or maximum scour depth distributed within the ±30 error lines basically, meaning the reliability and accuracy of present model for predicting equilibrium scour depth around foundation in random waves. However, compared with the experimental values, the present model overestimated the equilibrium scour depth generally. Given that, a calibration for scour depth was carried out by multiplying the mean reduced coefficient 0.85 in following section.

Jmse 09 00886 g007 550

Figure 7. Comparison of equilibrium (or maximum) scour depth between the present study and Sumer and Fredsøe [16], Schendel et al. [22].

Through the various examination for hydrodynamic and morphology parameters, it can be concluded that the present model is a validated and calibrated model for scour under random waves. Thus, the present numerical model would be utilized for scour simulation around foundation under random waves.

4. Numerical Results and Discussions

4.1. Scour Evolution

Figure 8 displays the scour evolution for case 1–9. As shown in Figure 8a, the scour depth increased rapidly at the initial stage, and then slowed down at the transition stage, which attributes to the backfilling occurred in scour holes under live bed scour condition, resulting in the net scour decreasing. Finally, the scour reached the equilibrium state when the amount of sediment backfilling equaled to that of scouring in the scour holes, i.e., the net scour transport rate was nil. Sumer and Fredsøe [16] proposed the following formula for the scour development under waves

St=Seq(1−exp(−t/Tc))�t=�eq(1−exp(−�/�c))(30)

where Tc is time scale of scour process.

Jmse 09 00886 g008 550

Figure 8. Time evolution of scour for case 1–9: (a) Case 1–5; (b) Case 6–9.

The computing time is 3600 s and the scour development curves in Figure 8 kept fluctuating, meaning it’s still not in equilibrium scour stage in these cases. According to Sumer and Fredsøe [16], the equilibrium scour depth can be acquired by fitting the data with Equation (30). From Figure 8, it can be seen that the scour evolution obtained from Equation (30) is consistent with the present study basically at initial stage, but the scour depth predicted by Equation (30) developed slightly faster than the simulating results and the Equation (30) overestimated the scour depth to some extent. Overall, the whole tendency of the results calculated by Equation (30) agrees well with the simulating results of the present study, which means the Equation (30) is applicable to depict the scour evolution around USAF under random waves.

4.2. Scour Mechanism under Random Waves

The scour morphology and scour evolution around USAF are similar under random waves in case 1~9. Taking case 7 as an example, the scour morphology is shown in Figure 9.

Jmse 09 00886 g009 550

Figure 9. Scour morphology under different times for case 7.

From Figure 9, at the initial stage (t < 1200 s), the scour occurred at upstream foundation edges between neighboring anchor branches. The maximum scour depth appeared at the lee-side of the USAF. Correspondingly, the sediments deposited at the periphery of the USAF, and the location of the maximum accretion depth was positioned at an angle of about 45° symmetrically with respect to the wave propagating direction in the lee-side of the USAF. After that, when t > 2400 s, the location of the maximum scour depth shifted to the upside of the USAF at an angle of about 45° with respect to the wave propagating direction.

According to previous studies [1,15,16,19,30,31], the horseshoe vortex, streamline compression and wake vortex shedding were responsible for scour around foundation. The Figure 10 displays the distribution of flow velocity in vicinity of foundation, which reflects the evolving processes of horseshoe vertex.

Jmse 09 00886 g010a 550
Jmse 09 00886 g010b 550

Figure 10. Velocity profile around USAF: (a) Flow runup and down stream at upstream anchor edges; (b) Horseshoe vortex at upstream anchor edges; (c) Flow reversal during wave through stage at lee side.

As shown in Figure 10, the inflow tripped to the upstream edges of the USAF and it was blocked by the upper tube of USAF. Then, the downflow formed the horizontal axis clockwise vortex and rolled on the seabed bypassing the tube, that is, the horseshoe vortex (Figure 11). The Figure 12 displays the turbulence intensity around the tube on the seabed. From Figure 12, it can be seen that the turbulence intensity was high-intensity with respect to the region of horseshoe vortex. This phenomenon occurred because of drastic water flow momentum exchanging in the horseshoe vortex. As a result, it created the prominent shear stress on the seabed, causing the local scour at the upstream edges of USAF. Besides, the horseshoe vortex moved downstream gradually along the periphery of the tube and the wake vortex shed off continually at the lee-side of the USAF, i.e., wake vortex.

Jmse 09 00886 g011 550

Figure 11. Sketch of scour mechanism around USAF under random waves.

Jmse 09 00886 g012 550

Figure 12. Turbulence intensity: (a) Turbulence intensity of horseshoe vortex; (b) Turbulence intensity of wake vortex; (c) Turbulence intensity of accretion area.

The core of wake vortex is a negative pressure center, liking a vacuum cleaner [11,42]. Hence, the soil particles were swirled into the negative pressure core and carried away by wake vortex. At the same time, the onset of scour at rear side occurred. Finally, the wake vortex became downflow at the downside of USAF. As is shown in Figure 12, the turbulence intensity was low where the downflow occurred at lee-side, which means the turbulence energy may not be able to support the survival of wake vortex, leading to accretion happening. As mentioned in previous section, the formation of horseshoe vortex was dependent with adverse pressure gradient at upside of foundation. As shown in Figure 13, the evaluated range of pressure distribution is −15 m to 15 m in x direction. The t = 450 s and t = 1800 s indicate that the wave crest and trough arrived at the upside and lee-side of the foundation respectively, and the t = 350 s was neither the wave crest nor trough. The adverse gradient pressure reached the maximum value at t = 450 s corresponding to the wave crest phase. In this case, it’s helpful for the wave boundary separating fully from seabed, which leads to the formation of horseshoe vortex with high turbulence intensity. Therefore, the horseshoe vortex is responsible for the local scour between neighboring anchor branches at upside of USAF. What’s more, due to the combination of the horseshoe vortex and streamline compression, the maximum scour depth occurred at the upside of the USAF with an angle of about 45° corresponding to the wave propagating direction. This is consistent with the findings of Pang et al. [48] and Sumer et al. [1,15] in case of regular waves. At the wave trough phase (t = 1800 s), the pressure gradient became positive at upstream USAF edges, which hindered the separating of wave boundary from seabed. In the meantime, the flow reversal occurred (Figure 10) and the adverse gradient pressure appeared at downstream USAF edges, but the magnitude of adverse gradient pressure at lee-side was lower than the upstream gradient pressure under wave crest. In this way, the intensity of horseshoe vortex behind the USAF under wave trough was low, which explains the difference of scour depth at upstream and downstream, i.e., the scour asymmetry. In other words, the scour asymmetry at upside and downside of USAF was attributed to wave asymmetry for random waves, and the phenomenon became more evident for nonlinear waves [21]. Briefly speaking, the vortex system at wave crest phase was mainly related to the scour process around USAF under random waves.

Jmse 09 00886 g013 550

Figure 13. Pressure distribution around USAF.

4.3. Equilibrium Scour Depth

The KC number is a key parameter for horseshoe vortex emerging and evolving under waves. According to Equation (1), when pile diameter D is fixed, the KC depends on the maximum near-bed velocity Uwm and wave period T. For random waves, the Uwm can be denoted by the root-mean-square (RMS) value of near-bed velocity amplitude Uwm,rms or the significant value of near-bed velocity amplitude Uwm,s. The Uwm,rms and Uwm,s for all simulating cases of the present study are listed in Table 3 and Table 4. The T can be denoted by the mean up zero-crossing wave period Ta, peak wave period Tp, significant wave period Ts, the maximum wave period Tm, 1/10′th highest wave period Tn = 1/10 and 1/5′th highest wave period Tn = 1/5 for random waves, so the different combinations of Uwm and T will acquire different KC. The Table 3 and Table 4 list 12 types of KC, for example, the KCrms,s was calculated by Uwm,rms and Ts. Sumer and Fredsøe [16] conducted a series of wave flume experiments to investigate the scour depth around monopile under random waves, and found the equilibrium scour depth predicting equation (Equation (2)) for regular waves was applicable for random waves with KCrms,p. It should be noted that the Equation (2) is only suitable for KC > 6 under regular waves or KCrms,p > 6 under random waves.

Table 3. Uwm,rms and KC for case 1~9.

Table

Table 4. Uwm,s and KC for case 1~9.

Table

Raaijmakers and Rudolph [34] proposed the equilibrium scour depth predicting model (Equation (5)) around pile under waves, which is suitable for low KC. The format of Equation (5) is similar with the formula proposed by Breusers [54], which can predict the equilibrium scour depth around pile at different scour stages. In order to verify the applicability of Raaijmakers’s model for predicting the equilibrium scour depth around USAF under random waves, a validation of the equilibrium scour depth Seq between the present study and Raaijmakers’s equation was conducted. The position where the scour depth Seq was evaluated is the location of the maximum scour depth, and it was depicted in Figure 14. The Figure 15 displays the comparison of Seq with different KC between the present study and Raaijmakers’s model.

Jmse 09 00886 g014 550

Figure 14. Sketch of the position where the Seq was evaluated.

Jmse 09 00886 g015a 550
Jmse 09 00886 g015b 550

Figure 15. Comparison of the equilibrium scour depth between the present model and the model of Raaijmakers and Rudolph [34]: (a) KCrms,s, KCrms,a; (b) KCrms,p, KCrms,m; (c) KCrms,n = 1/10, KCrms,n = 1/5; (d) KCs,s, KCs,a; (e) KCs,p, KCs,m; (f) KCs,n = 1/10, KCs,n = 1/5.

As shown in Figure 15, there is an error in predicting Seq between the present study and Raaijmakers’s model, and Raaijmakers’s model underestimates the results generally. Although the error exists, the varying trend of Seq with KC obtained from Raaijmakers’s model is consistent with the present study basically. What’s more, the error is minimum and the Raaijmakers’s model is of relatively high accuracy for predicting scour around USAF under random waves by using KCs,p. Based on this, a further revision was made to eliminate the error as much as possible, i.e., add the deviation value ∆S/D in the Raaijmakers’s model. The revised equilibrium scour depth predicting equation based on Raaijmakers’s model can be written as

S′eq/D=1.95[tanh(hD)](1−exp(−0.012KCs,p))+ΔS/D�eq′/�=1.95tanh(ℎ�)(1−exp(−0.012��s,p))+∆�/�(31)

As the Figure 16 shown, through trial-calculation, when ∆S/D = 0.05, the results calculated by Equation (31) show good agreement with the simulating results of the present study. The maximum error is about 18.2% and the engineering requirements have been met basically. In order to further verify the accuracy of the revised model for large KC (KCs,p > 4) under random waves, a validation between the revised model and the previous experimental results [21]. The experiment was conducted in a flume (50 m in length, 1.0 m in width and 1.3 m in height) with a slender vertical pile (D = 0.1 m) under random waves. The seabed is composed of 0.13 m deep layer of sand with d50 = 0.6 mm and the water depth is 0.5 m for all tests. The significant wave height is 0.12~0.21 m and the KCs,p is 5.52~11.38. The comparison between the predicting results by Equation (31) and the experimental results of Corvaro et al. [21] is shown in Figure 17. From Figure 17, the experimental data evenly distributes around the predicted results and the prediction accuracy is favorable when KCs,p < 8. However, the gap between the predicting results and experimental data becomes large and the Equation (31) overestimates the equilibrium scour depth to some extent when KCs,p > 8.

Jmse 09 00886 g016 550

Figure 16. Comparison of Seq between the simulating results and the predicting values by Equation (31).

Jmse 09 00886 g017 550

Figure 17. Comparison of Seq/D between the Experimental results of Corvaro et al. [21] and the predicting values by Equation (31).

In ocean environment, the waves are composed of a train of sinusoidal waves with different frequencies and amplitudes. The energy of constituent waves with very large and very small frequencies is relatively low, and the energy of waves is mainly concentrated in a certain range of moderate frequencies. Myrhaug and Rue [37] thought the 1/n’th highest wave was responsible for scour and proposed the stochastic model to predict the equilibrium scour depth around pile under random waves for full range of KC. Noteworthy is that the KC was denoted by KCrms,a in the stochastic model. To verify the application of the stochastic model for predicting scour depth around USAF, a validation between the simulating results of present study and predicting results by the stochastic model with n = 2,3,5,10,20,500 was carried out respectively.

As shown in Figure 18, compared with the simulating results, the stochastic model underestimates the equilibrium scour depth around USAF generally. Although the error exists, the varying trend of Seq with KCrms,a obtained from the stochastic model is consistent with the present study basically. What’s more, the gap between the predicting values by stochastic model and the simulating results decreases with the increase of n, but for large n, for example n = 500, the varying trend diverges between the predicting values and simulating results, meaning it’s not feasible only by increasing n in stochastic model to predict the equilibrium scour depth around USAF.

Jmse 09 00886 g018 550

Figure 18. Comparison of Seq between the simulating results and the predicting values by Equation (8).

The Figure 19 lists the deviation value ∆Seq/D′ between the predicting values and simulating results with different KCrms,a and n. Then, fitted the relationship between the ∆S′and n under different KCrms,a, and the fitting curve can be written by Equation (32). The revised stochastic model (Equation (33)) can be acquired by adding ∆Seq/D′ to Equation (8).

ΔSeq/D=0.052*exp(−n/6.566)+0.068∆�eq/�=0.052*exp(−�/6.566)+0.068(32)

S′eq¯/D=S′eq/D+0.052*exp(−n/6.566)+0.068�eq′¯/�=�eq′/�+0.052*exp(−�/6.566)+0.068(33)

Jmse 09 00886 g019 550

Figure 19. The fitting line between ∆S′and n.

The comparison between the predicting results by Equation (33) and the simulating results of present study is shown in Figure 20. According to the Figure 20, the varying trend of Seq with KCrms,a obtained from the stochastic model is consistent with the present study basically. Compared with predicting results by the stochastic model, the results calculated by Equation (33) is favorable. Moreover, comparison with simulating results indicates that the predicting results are the most favorable for n = 10, which is consistent with the findings of Myrhaug and Rue [37] for equilibrium scour depth predicting around slender pile in case of random waves.

Jmse 09 00886 g020 550

Figure 20. Comparison of Seq between the simulating results and the predicting values by Equation (33).

In order to further verify the accuracy of the Equation (33) for large KC (KCrms,a > 4) under random waves, a validation was conducted between the Equation (33) and the previous experimental results of Sumer and Fredsøe [16] and Corvaro et al. [21]. The details of experiments conducted by Corvaro et al. [21] were described in above section. Sumer and Fredsøe [16] investigated the local scour around pile under random waves. The experiments were conducted in a wave basin with a slender vertical pile (D = 0.032, 0.055 m). The seabed is composed of 0.14 m deep layer of sand with d50 = 0.2 mm and the water depth was maintained at 0.5 m. The JONSWAP wave spectrum was used and the KCrms,a was 5.29~16.95. The comparison between the predicting results by Equation (33) and the experimental results of Sumer and Fredsøe [16] and Corvaro et al. [21] are shown in Figure 21. From Figure 21, contrary to the case of low KCrms,a (KCrms,a < 4), the error between the predicting values and experimental results increases with decreasing of n for KCrms,a > 4. Therefore, the predicting results are the most favorable for n = 2 when KCrms,a > 4.

Jmse 09 00886 g021 550

Figure 21. Comparison of Seq between the experimental results of Sumer and Fredsøe [16] and Corvaro et al. [21] and the predicting values by Equation (33).

Noteworthy is that the present model was built according to prototype size, so the errors between the numerical results and experimental data of References [16,21] may be attribute to the scale effects. In laboratory experiments on scouring process, it is typically impossible to ensure a rigorous similarity of all physical parameters between the model and prototype structure, leading to the scale effects in the laboratory experiments. To avoid a cohesive behaviour, the bed material was not scaled geometrically according to model scale. As a consequence, the relatively large-scaled sediments sizes may result in the overestimation of bed load transport and underestimation of suspended load transport compared with field conditions. What’s more, the disproportional scaled sediment presumably lead to the difference of bed roughness between the model and prototype, and thus large influences for wave boundary layer on the seabed and scour process. Besides, according to Corvaro et al. [21] and Schendel et al. [55], the pile Reynolds numbers and Froude numbers both affect the scour depth for the condition of non fully developed turbulent flow in laboratory experiments.

4.4. Parametric Study

4.4.1. Influence of Froude Number

As described above, the set of foundation leads to the adverse pressure gradient appearing at upstream, leading to the wave boundary layer separating from seabed, then horseshoe vortex formatting and the horseshoe vortex are mainly responsible for scour around foundation (see Figure 22). The Froude number Fr is the key parameter to influence the scale and intensity of horseshoe vortex. The Fr under waves can be calculated by the following formula [42]

Fr=UwgD−−−√�r=�w��(34)

where Uw is the mean water particle velocity during 1/4 cycle of wave oscillation, obtained from the following formula. Noteworthy is that the root-mean-square (RMS) value of near-bed velocity amplitude Uwm,rms is used for calculating Uwm.

Uw=1T/4∫0T/4Uwmsin(t/T)dt=2πUwm�w=1�/4∫0�/4�wmsin(�/�)��=2��wm(35)

Jmse 09 00886 g022 550

Figure 22. Sketch of flow field at upstream USAF edges.

Tavouktsoglou et al. [25] proposed the following formula between Fr and the vertical location of the stagnation y

yh∝Fer�ℎ∝�r�(36)

where e is constant.

The Figure 23 displays the relationship between Seq/D and Fr of the present study. In order to compare with the simulating results, the experimental data of Corvaro et al. [21] was also depicted in Figure 23. As shown in Figure 23, the equilibrium scour depth appears a logarithmic increase as Fr increases and approaches the mathematical asymptotic value, which is also consistent with the experimental results of Corvaro et al. [21]. According to Figure 24, the adverse pressure gradient pressure at upstream USAF edges increases with the increase of Fr, which is benefit for the wave boundary layer separating from seabed, resulting in the high-intensity horseshoe vortex, hence, causing intensive scour around USAF. Based on the previous study of Tavouktsoglou et al. [25] for scour around pile under currents, the high Fr leads to the stagnation point is closer to the mean sea level for shallow water, causing the stronger downflow kinetic energy. As mentioned in previous section, the energy of downflow at upstream makes up the energy of the subsequent horseshoe vortex, so the stronger downflow kinetic energy results in the more intensive horseshoe vortex. Therefore, the higher Fr leads to the more intensive horseshoe vortex by influencing the position of stagnation point y presumably. Qi and Gao [19] carried out a series of flume tests to investigate the scour around pile under regular waves, and proposed the fitting formula between Seq/D and Fr as following

lg(Seq/D)=Aexp(B/Fr)+Clg(�eq/�)=�exp(�/�r)+�(37)

where A, B and C are constant.

Jmse 09 00886 g023 550

Figure 23. The fitting curve between Seq/D and Fr.

Jmse 09 00886 g024 550

Figure 24. Sketch of adverse pressure gradient at upstream USAF edges.

Took the Equation (37) to fit the simulating results with A = −0.002, B = 0.686 and C = −0.808, and the results are shown in Figure 23. From Figure 23, the simulating results evenly distribute around the Equation (37) and the varying trend of Seq/D and Fr in present study is consistent with Equation (37) basically, meaning the Equation (37) is applicable to express the relationship of Seq/D with Fr around USAF under random waves.

4.4.2. Influence of Euler Number

The Euler number Eu is the influencing factor for the hydrodynamic field around foundation. The Eu under waves can be calculated by the following formula. The Eu can be represented by the Equation (38) for uniform cylinders [25]. The root-mean-square (RMS) value of near-bed velocity amplitude Um,rms is used for calculating Um.

Eu=U2mgD�u=�m2��(38)

where Um is depth-averaged flow velocity.

The Figure 25 displays the relationship between Seq/D and Eu of the present study. In order to compare with the simulating results, the experimental data of Sumer and Fredsøe [16] and Corvaro et al. [21] were also plotted in Figure 25. As shown in Figure 25, similar with the varying trend of Seq/D and Fr, the equilibrium scour depth appears a logarithmic increase as Eu increases and approaches the mathematical asymptotic value, which is also consistent with the experimental results of Sumer and Fredsøe [16] and Corvaro et al. [21]. According to Figure 24, the adverse pressure gradient pressure at upstream USAF edges increases with the increasing of Eu, which is benefit for the wave boundary layer separating from seabed, inducing the high-intensity horseshoe vortex, hence, causing intensive scour around USAF.

Jmse 09 00886 g025 550

Figure 25. The fitting curve between Seq/D and Eu.

Therefore, the variation of Fr and Eu reflect the magnitude of adverse pressure gradient pressure at upstream. Given that, the Equation (37) also was used to fit the simulating results with A = 8.875, B = 0.078 and C = −9.601, and the results are shown in Figure 25. From Figure 25, the simulating results evenly distribute around the Equation (37) and the varying trend of Seq/D and Eu in present study is consistent with Equation (37) basically, meaning the Equation (37) is also applicable to express the relationship of Seq/D with Eu around USAF under random waves. Additionally, according to the above description of Fr, it can be inferred that the higher Fr and Eu both lead to the more intensive horseshoe vortex by influencing the position of stagnation point y presumably.

5. Conclusions

A series of numerical models were established to investigate the local scour around umbrella suction anchor foundation (USAF) under random waves. The numerical model was validated for hydrodynamic and morphology parameters by comparing with the experimental data of Khosronejad et al. [52], Petersen et al. [17], Sumer and Fredsøe [16] and Schendel et al. [22]. Based on the simulating results, the scour evolution and scour mechanisms around USAF under random waves were analyzed respectively. Two revised models were proposed according to the model of Raaijmakers and Rudolph [34] and the stochastic model developed by Myrhaug and Rue [37] to predict the equilibrium scour depth around USAF under random waves. Finally, a parametric study was carried out with the present model to study the effects of the Froude number Fr and Euler number Eu to the equilibrium scour depth around USAF under random waves. The main conclusions can be described as follows.(1)

The packed sediment scour model and the RNG k−ε turbulence model were used to simulate the sand particles transport processes and the flow field around UASF respectively. The scour evolution obtained by the present model agrees well with the experimental results of Khosronejad et al. [52], Petersen et al. [17], Sumer and Fredsøe [16] and Schendel et al. [22], which indicates that the present model is accurate and reasonable for depicting the scour morphology around UASF under random waves.(2)

The vortex system at wave crest phase is mainly related to the scour process around USAF under random waves. The maximum scour depth appeared at the lee-side of the USAF at the initial stage (t < 1200 s). Subsequently, when t > 2400 s, the location of the maximum scour depth shifted to the upside of the USAF at an angle of about 45° with respect to the wave propagating direction.(3)

The error is negligible and the Raaijmakers’s model is of relatively high accuracy for predicting scour around USAF under random waves when KC is calculated by KCs,p. Given that, a further revision model (Equation (31)) was proposed according to Raaijmakers’s model to predict the equilibrium scour depth around USAF under random waves and it shows good agreement with the simulating results of the present study when KCs,p < 8.(4)

Another further revision model (Equation (33)) was proposed according to the stochastic model established by Myrhaug and Rue [37] to predict the equilibrium scour depth around USAF under random waves, and the predicting results are the most favorable for n = 10 when KCrms,a < 4. However, contrary to the case of low KCrms,a, the predicting results are the most favorable for n = 2 when KCrms,a > 4 by the comparison with experimental results of Sumer and Fredsøe [16] and Corvaro et al. [21].(5)

The same formula (Equation (37)) is applicable to express the relationship of Seq/D with Eu or Fr, and it can be inferred that the higher Fr and Eu both lead to the more intensive horseshoe vortex and larger Seq.

Author Contributions

Conceptualization, H.L. (Hongjun Liu); Data curation, R.H. and P.Y.; Formal analysis, X.W. and H.L. (Hao Leng); Funding acquisition, X.W.; Writing—original draft, R.H. and P.Y.; Writing—review & editing, X.W. and H.L. (Hao Leng); The final manuscript has been approved by all the authors. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Fundamental Research Funds for the Central Universities (grant number 202061027) and the National Natural Science Foundation of China (grant number 41572247).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Sumer, B.M.; Fredsøe, J.; Christiansen, N. Scour Around Vertical Pile in Waves. J. Waterw. Port. Coast. Ocean Eng. 1992, 118, 15–31. [Google Scholar] [CrossRef]
  2. Rudolph, D.; Bos, K. Scour around a monopile under combined wave-current conditions and low KC-numbers. In Proceedings of the 6th International Conference on Scour and Erosion, Amsterdam, The Netherlands, 1–3 November 2006; pp. 582–588. [Google Scholar]
  3. Nielsen, A.W.; Liu, X.; Sumer, B.M.; Fredsøe, J. Flow and bed shear stresses in scour protections around a pile in a current. Coast. Eng. 2013, 72, 20–38. [Google Scholar] [CrossRef]
  4. Ahmad, N.; Bihs, H.; Myrhaug, D.; Kamath, A.; Arntsen, Ø.A. Three-dimensional numerical modelling of wave-induced scour around piles in a side-by-side arrangement. Coast. Eng. 2018, 138, 132–151. [Google Scholar] [CrossRef]
  5. Li, H.; Ong, M.C.; Leira, B.J.; Myrhaug, D. Effects of Soil Profile Variation and Scour on Structural Response of an Offshore Monopile Wind Turbine. J. Offshore Mech. Arct. Eng. 2018, 140, 042001. [Google Scholar] [CrossRef]
  6. Li, H.; Liu, H.; Liu, S. Dynamic analysis of umbrella suction anchor foundation embedded in seabed for offshore wind turbines. Géoméch. Energy Environ. 2017, 10, 12–20. [Google Scholar] [CrossRef]
  7. Fazeres-Ferradosa, T.; Rosa-Santos, P.; Taveira-Pinto, F.; Vanem, E.; Carvalho, H.; Correia, J.A.F.D.O. Editorial: Advanced research on offshore structures and foundation design: Part 1. Proc. Inst. Civ. Eng. Marit. Eng. 2019, 172, 118–123. [Google Scholar] [CrossRef]
  8. Chavez, C.E.A.; Stratigaki, V.; Wu, M.; Troch, P.; Schendel, A.; Welzel, M.; Villanueva, R.; Schlurmann, T.; De Vos, L.; Kisacik, D.; et al. Large-Scale Experiments to Improve Monopile Scour Protection Design Adapted to Climate Change—The PROTEUS Project. Energies 2019, 12, 1709. [Google Scholar] [CrossRef][Green Version]
  9. Wu, M.; De Vos, L.; Chavez, C.E.A.; Stratigaki, V.; Fazeres-Ferradosa, T.; Rosa-Santos, P.; Taveira-Pinto, F.; Troch, P. Large Scale Experimental Study of the Scour Protection Damage Around a Monopile Foundation Under Combined Wave and Current Conditions. J. Mar. Sci. Eng. 2020, 8, 417. [Google Scholar] [CrossRef]
  10. Sørensen, S.P.H.; Ibsen, L.B. Assessment of foundation design for offshore monopiles unprotected against scour. Ocean Eng. 2013, 63, 17–25. [Google Scholar] [CrossRef]
  11. Prendergast, L.; Gavin, K.; Doherty, P. An investigation into the effect of scour on the natural frequency of an offshore wind turbine. Ocean Eng. 2015, 101, 1–11. [Google Scholar] [CrossRef][Green Version]
  12. Fazeres-Ferradosa, T.; Chambel, J.; Taveira-Pinto, F.; Rosa-Santos, P.; Taveira-Pinto, F.; Giannini, G.; Haerens, P. Scour Protections for Offshore Foundations of Marine Energy Harvesting Technologies: A Review. J. Mar. Sci. Eng. 2021, 9, 297. [Google Scholar] [CrossRef]
  13. Yang, Q.; Yu, P.; Liu, Y.; Liu, H.; Zhang, P.; Wang, Q. Scour characteristics of an offshore umbrella suction anchor foundation under the combined actions of waves and currents. Ocean Eng. 2020, 202, 106701. [Google Scholar] [CrossRef]
  14. Yu, P.; Hu, R.; Yang, J.; Liu, H. Numerical investigation of local scour around USAF with different hydraulic conditions under currents and waves. Ocean Eng. 2020, 213, 107696. [Google Scholar] [CrossRef]
  15. Sumer, B.M.; Christiansen, N.; Fredsøe, J. The horseshoe vortex and vortex shedding around a vertical wall-mounted cylinder exposed to waves. J. Fluid Mech. 1997, 332, 41–70. [Google Scholar] [CrossRef]
  16. Sumer, B.M.; Fredsøe, J. Scour around Pile in Combined Waves and Current. J. Hydraul. Eng. 2001, 127, 403–411. [Google Scholar] [CrossRef]
  17. Petersen, T.U.; Sumer, B.M.; Fredsøe, J. Time scale of scour around a pile in combined waves and current. In Proceedings of the 6th International Conference on Scour and Erosion, Paris, France, 27–31 August 2012. [Google Scholar]
  18. Petersen, T.U.; Sumer, B.M.; Fredsøe, J.; Raaijmakers, T.C.; Schouten, J.-J. Edge scour at scour protections around piles in the marine environment—Laboratory and field investigation. Coast. Eng. 2015, 106, 42–72. [Google Scholar] [CrossRef]
  19. Qi, W.; Gao, F. Equilibrium scour depth at offshore monopile foundation in combined waves and current. Sci. China Ser. E Technol. Sci. 2014, 57, 1030–1039. [Google Scholar] [CrossRef][Green Version]
  20. Larsen, B.E.; Fuhrman, D.R.; Baykal, C.; Sumer, B.M. Tsunami-induced scour around monopile foundations. Coast. Eng. 2017, 129, 36–49. [Google Scholar] [CrossRef][Green Version]
  21. Corvaro, S.; Marini, F.; Mancinelli, A.; Lorenzoni, C.; Brocchini, M. Hydro- and Morpho-dynamics Induced by a Vertical Slender Pile under Regular and Random Waves. J. Waterw. Port. Coast. Ocean Eng. 2018, 144, 04018018. [Google Scholar] [CrossRef]
  22. Schendel, A.; Welzel, M.; Schlurmann, T.; Hsu, T.-W. Scour around a monopile induced by directionally spread irregular waves in combination with oblique currents. Coast. Eng. 2020, 161, 103751. [Google Scholar] [CrossRef]
  23. Fazeres-Ferradosa, T.; Taveira-Pinto, F.; Romão, X.; Reis, M.; das Neves, L. Reliability assessment of offshore dynamic scour protections using copulas. Wind. Eng. 2018, 43, 506–538. [Google Scholar] [CrossRef]
  24. Fazeres-Ferradosa, T.; Welzel, M.; Schendel, A.; Baelus, L.; Santos, P.R.; Pinto, F.T. Extended characterization of damage in rubble mound scour protections. Coast. Eng. 2020, 158, 103671. [Google Scholar] [CrossRef]
  25. Tavouktsoglou, N.S.; Harris, J.M.; Simons, R.R.; Whitehouse, R.J.S. Equilibrium Scour-Depth Prediction around Cylindrical Structures. J. Waterw. Port. Coast. Ocean Eng. 2017, 143, 04017017. [Google Scholar] [CrossRef][Green Version]
  26. Ettema, R.; Melville, B.; Barkdoll, B. Scale Effect in Pier-Scour Experiments. J. Hydraul. Eng. 1998, 124, 639–642. [Google Scholar] [CrossRef]
  27. Umeda, S. Scour Regime and Scour Depth around a Pile in Waves. J. Coast. Res. Spec. Issue 2011, 64, 845–849. [Google Scholar]
  28. Umeda, S. Scour process around monopiles during various phases of sea storms. J. Coast. Res. 2013, 165, 1599–1604. [Google Scholar] [CrossRef]
  29. Baykal, C.; Sumer, B.; Fuhrman, D.R.; Jacobsen, N.; Fredsøe, J. Numerical simulation of scour and backfilling processes around a circular pile in waves. Coast. Eng. 2017, 122, 87–107. [Google Scholar] [CrossRef][Green Version]
  30. Miles, J.; Martin, T.; Goddard, L. Current and wave effects around windfarm monopile foundations. Coast. Eng. 2017, 121, 167–178. [Google Scholar] [CrossRef][Green Version]
  31. Miozzi, M.; Corvaro, S.; Pereira, F.A.; Brocchini, M. Wave-induced morphodynamics and sediment transport around a slender vertical cylinder. Adv. Water Resour. 2019, 129, 263–280. [Google Scholar] [CrossRef]
  32. Yu, T.; Zhang, Y.; Zhang, S.; Shi, Z.; Chen, X.; Xu, Y.; Tang, Y. Experimental study on scour around a composite bucket foundation due to waves and current. Ocean Eng. 2019, 189, 106302. [Google Scholar] [CrossRef]
  33. Carreiras, J.; Larroudé, P.; Seabra-Santos, F.; Mory, M. Wave Scour Around Piles. In Proceedings of the Coastal Engineering 2000, American Society of Civil Engineers (ASCE), Sydney, Australia, 16–21 July 2000; pp. 1860–1870. [Google Scholar]
  34. Raaijmakers, T.; Rudolph, D. Time-dependent scour development under combined current and waves conditions—Laboratory experiments with online monitoring technique. In Proceedings of the 4th International Conference on Scour and Erosion, Tokyo, Japan, 5–7 November 2008; pp. 152–161. [Google Scholar]
  35. Khalfin, I.S. Modeling and calculation of bed score around large-diameter vertical cylinder under wave action. Water Resour. 2007, 34, 357. [Google Scholar] [CrossRef][Green Version]
  36. Zanke, U.C.; Hsu, T.-W.; Roland, A.; Link, O.; Diab, R. Equilibrium scour depths around piles in noncohesive sediments under currents and waves. Coast. Eng. 2011, 58, 986–991. [Google Scholar] [CrossRef]
  37. Myrhaug, D.; Rue, H. Scour below pipelines and around vertical piles in random waves. Coast. Eng. 2003, 48, 227–242. [Google Scholar] [CrossRef]
  38. Myrhaug, D.; Ong, M.C.; Føien, H.; Gjengedal, C.; Leira, B.J. Scour below pipelines and around vertical piles due to second-order random waves plus a current. Ocean Eng. 2009, 36, 605–616. [Google Scholar] [CrossRef]
  39. Myrhaug, D.; Ong, M.C. Random wave-induced onshore scour characteristics around submerged breakwaters using a stochastic method. Ocean Eng. 2010, 37, 1233–1238. [Google Scholar] [CrossRef]
  40. Ong, M.C.; Myrhaug, D.; Hesten, P. Scour around vertical piles due to long-crested and short-crested nonlinear random waves plus a current. Coast. Eng. 2013, 73, 106–114. [Google Scholar] [CrossRef]
  41. Yakhot, V.; Orszag, S.A. Renormalization group analysis of turbulence. I. Basic theory. J. Sci. Comput. 1986, 1, 3–51. [Google Scholar] [CrossRef]
  42. Yakhot, V.; Smith, L.M. The renormalization group, the e-expansion and derivation of turbulence models. J. Sci. Comput. 1992, 7, 35–61. [Google Scholar] [CrossRef]
  43. Mastbergen, D.R.; Berg, J.V.D. Breaching in fine sands and the generation of sustained turbidity currents in submarine canyons. Sedimentology 2003, 50, 625–637. [Google Scholar] [CrossRef]
  44. Soulsby, R. Dynamics of Marine Sands; Thomas Telford Ltd.: London, UK, 1998. [Google Scholar] [CrossRef]
  45. Van Rijn, L.C. Sediment Transport, Part I: Bed Load Transport. J. Hydraul. Eng. 1984, 110, 1431–1456. [Google Scholar] [CrossRef][Green Version]
  46. Zhang, Q.; Zhou, X.-L.; Wang, J.-H. Numerical investigation of local scour around three adjacent piles with different arrangements under current. Ocean Eng. 2017, 142, 625–638. [Google Scholar] [CrossRef]
  47. Yu, Y.X.; Liu, S.X. Random Wave and Its Applications to Engineering, 4th ed.; Dalian University of Technology Press: Dalian, China, 2011. [Google Scholar]
  48. Pang, A.; Skote, M.; Lim, S.; Gullman-Strand, J.; Morgan, N. A numerical approach for determining equilibrium scour depth around a mono-pile due to steady currents. Appl. Ocean Res. 2016, 57, 114–124. [Google Scholar] [CrossRef]
  49. Higuera, P.; Lara, J.L.; Losada, I.J. Three-dimensional interaction of waves and porous coastal structures using Open-FOAM®. Part I: Formulation and validation. Coast. Eng. 2014, 83, 243–258. [Google Scholar] [CrossRef]
  50. Corvaro, S.; Crivellini, A.; Marini, F.; Cimarelli, A.; Capitanelli, L.; Mancinelli, A. Experimental and Numerical Analysis of the Hydrodynamics around a Vertical Cylinder in Waves. J. Mar. Sci. Eng. 2019, 7, 453. [Google Scholar] [CrossRef][Green Version]
  51. Flow3D User Manual, version 11.0.3; Flow Science, Inc.: Santa Fe, NM, USA, 2013.
  52. Khosronejad, A.; Kang, S.; Sotiropoulos, F. Experimental and computational investigation of local scour around bridge piers. Adv. Water Resour. 2012, 37, 73–85. [Google Scholar] [CrossRef]
  53. Stahlmann, A. Experimental and Numerical Modeling of Scour at Foundation Structures for Offshore Wind Turbines. Ph.D. Thesis, Franzius-Institute for Hydraulic, Estuarine and Coastal Engineering, Leibniz Universität Hannover, Hannover, Germany, 2013. [Google Scholar]
  54. Breusers, H.N.C.; Nicollet, G.; Shen, H. Local Scour Around Cylindrical Piers. J. Hydraul. Res. 1977, 15, 211–252. [Google Scholar] [CrossRef]
  55. Schendel, A.; Hildebrandt, A.; Goseberg, N.; Schlurmann, T. Processes and evolution of scour around a monopile induced by tidal currents. Coast. Eng. 2018, 139, 65–84. [Google Scholar] [CrossRef]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

      

MDPI and ACS Style

Hu, R.; Liu, H.; Leng, H.; Yu, P.; Wang, X. Scour Characteristics and Equilibrium Scour Depth Prediction around Umbrella Suction Anchor Foundation under Random Waves. J. Mar. Sci. Eng. 2021, 9, 886. https://doi.org/10.3390/jmse9080886

AMA Style

Hu R, Liu H, Leng H, Yu P, Wang X. Scour Characteristics and Equilibrium Scour Depth Prediction around Umbrella Suction Anchor Foundation under Random Waves. Journal of Marine Science and Engineering. 2021; 9(8):886. https://doi.org/10.3390/jmse9080886Chicago/Turabian Style

Hu, Ruigeng, Hongjun Liu, Hao Leng, Peng Yu, and Xiuhai Wang. 2021. “Scour Characteristics and Equilibrium Scour Depth Prediction around Umbrella Suction Anchor Foundation under Random Waves” Journal of Marine Science and Engineering 9, no. 8: 886. https://doi.org/10.3390/jmse9080886

Find Other Styles

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

For more information on the journal statistics, click here.

Multiple requests from the same IP address are counted as one view.

원본 다운로드
Study on Hydrodynamic Performance of Unsymmetrical Double Vertical Slotted Barriers

침수된 강성 식생을 갖는 개방 수로 흐름의 특성에 대한 3차원 수치 시뮬레이션

A 3-D numerical simulation of the characteristics of open channel flows with submerged rigid vegetation

Journal of Hydrodynamics volume 33, pages833–843 (2021)Cite this article

Abstract

이 백서는 Flow-3D를 적용하여 다양한 흐름 배출 및 식생 시나리오가 흐름 속도(세로, 가로 및 수직 속도 포함)에 미치는 영향을 조사합니다.

실험적 측정을 통한 검증 후 식생직경, 식생높이, 유량방류량에 대한 민감도 분석을 수행하였다. 종방향 속도의 경우 흐름 구조에 가장 큰 영향을 미치는 것은 배출보다는 식생 직경에서 비롯됩니다.

그러나 식생 높이는 수직 분포의 변곡점을 결정합니다. 식생지 내 두 지점, 즉 상류와 하류의 횡속도를 비교하면 수심에 따른 대칭적인 패턴을 확인할 수 있다. 식생 지역의 가로 및 세로 유체 순환 패턴을 포함하여 흐름 또는 식생 시나리오와 관계없이 수직 속도에 대해서도 동일한 패턴이 관찰됩니다.

또한 식생의 직경이 클수록 이러한 패턴이 더 분명해집니다. 상부 순환은 초목 캐노피 근처에서 발생합니다. 식생지역의 가로방향과 세로방향의 순환에 관한 이러한 발견은 침수식생을 통한 3차원 유동구조를 밝혀준다.

This paper applies the Flow-3D to investigate the impacts of different flow discharge and vegetation scenarios on the flow velocity (including the longitudinal, transverse and vertical velocities). After the verification by using experimental measurements, a sensitivity analysis is conducted for the vegetation diameter, the vegetation height and the flow discharge. For the longitudinal velocity, the greatest impact on the flow structure originates from the vegetation diameter, rather than the discharge. The vegetation height, however, determines the inflection point of the vertical distribution. Comparing the transverse velocities at two positions in the vegetated area, i.e., the upstream and the downstream, a symmetric pattern is identified along the water depth. The same pattern is also observed for the vertical velocity regardless of the flow or vegetation scenario, including both transverse and vertical fluid circulation patterns in the vegetated area. Moreover, the larger the vegetation diameter is, the more evident these patterns become. The upper circulation occurs near the vegetation canopy. These findings regarding the circulations along the transverse and vertical directions in the vegetated region shed light on the 3-D flow structure through the submerged vegetation.

Key words

  • Submerged rigid vegetation
  • longitudinal velocity
  • transverse velocity
  • vertical velocity
  • open channel

References

  1. Angelina A., Jordanova C. S. J. Experimental study of bed load transport through emergent vegetation [J]. Journal of Hydraulic Engineering, ASCE, 2003, 129(6): 474–478.Article Google Scholar 
  2. Li Y., Wang Y., Anim D. O. et al. Flow characteristics in different densities of submerged flexible vegetation from an open-channel flume study of artificial plants [J]. Geomorphology, 2014, 204: 314–324.Article Google Scholar 
  3. Bai F., Yang Z., Huai W. et al. A depth-averaged two dimensional shallow water model to simulate flow-rigid vegetation interactions [J]. Procedia Engineering, 2016, 154: 482–489.Article Google Scholar 
  4. Huai W. X., Song S., Han J. et al. Prediction of velocity distribution in straight open-channel flow with partial vegetation by singular perturbation method [J]. Applied Mathematics and Mechanics (Engilsh Editon), 2016, 37(10): 1315–1324.Article MathSciNet Google Scholar 
  5. Wang P. F., Wang C. Numerical model for flow through submerged vegetation regions in a shallow lake [J]. Journal of Hydrodynamics, 2011, 23(2): 170–178.Article Google Scholar 
  6. Wang W. J., Cui X. Y., Dong F. et al. Predictions of bulk velocity for open channel flow through submerged vegetation [J]. Journal of Hydrodynamics, 2020, 32(4): 795–799.Article Google Scholar 
  7. Zhang M., Li C. W., Shen Y. Depth-averaged modeling of free surface flows in open channels with emerged and submerged vegetation [J]. Applied Mathematical Modelling, 2013, 37(1–2): 540–553.Article MathSciNet Google Scholar 
  8. Huai W., Wang W., Hu Y. et al. Analytical model of the mean velocity distribution in an open channel with double-layered rigid vegetation [J]. Advances in Water Resources, 2014, 69: 106–113.Article Google Scholar 
  9. Panigrahi K., Khatua K. K. Prediction of velocity distribution in straight channel with rigid vegetation [J]. Aquatic Procedia, 2015, 4: 819–825.Article Google Scholar 
  10. Huai W. X., Zeng Y. H., Xu Z. G. et al. Three-layer model for vertical velocity distribution in open channel flow with submerged rigid vegetation [J]. Advances in Water Resources, 2009, 32(4): 487–492.Article Google Scholar 
  11. Chen S. C., Kuo Y. M., Li Y. H. Flow characteristics within different configurations of submerged flexible vegetation [J]. Journal of Hydrology, 2011, 398(1–2): 124–134.Article Google Scholar 
  12. Yagci O., Tschiesche U., Kabdasli M. S. The role of different forms of natural riparian vegetation on turbulence and kinetic energy characteristics [J]. Advances in Water Resources, 2010, 33(5): 601–614.Article Google Scholar 
  13. Wu F. S. Characteristics of flow resistance in open channels with non-submerged rigid vegetation [J]. Journal of Hydrodynamics, 2008, 20(2): 239–245.Article Google Scholar 
  14. Huai W., Hu Y., Zeng Y. et al. Velocity distribution for open channel flows with suspended vegetation [J]. Advances in Water Resources, 2012, 49: 56–61.Article Google Scholar 
  15. Pu J. H., Hussain A., Guo Y. K. et al. Submerged flexible vegetation impact on open channel flow velocity distribution: An analytical modelling study on drag and friction [J]. Water Science and Engineering, 2019, 12(2): 121–128.Article Google Scholar 
  16. Zhang M. L., Li C. W., Shen Y. M. A 3D non-linear k-ε turbulent model for prediction of flow and mass transport in channel with vegetation [J]. Applied Mathematical Modelling, 2010, 34(4): 1021–1031.Article MathSciNet Google Scholar 
  17. Anjum N., Tanaka N. Numerical investigation of velocity distribution of turbulent flow through vertically double-layered vegetation [J]. Water Science and Engineering, 2019, 12(4): 319–329.Article Google Scholar 
  18. Wang W., Huai W. X., Gao M. Numerical investigation of flow through vegetated multi-stage compound channel [J]. Journal of Hydrodynamics, 2014, 26(3): 467–473.Article Google Scholar 
  19. Ghani U., Anjum N., Pasha G. A. et al. Numerical investigation of the flow characteristics through discontinuous and layered vegetation patches of finite width in an open channel [J]. Environmental Fluid Mechanics, 2019, 19(6): 1469–1495.Article Google Scholar 
  20. Aydin M. C., Emiroglu M. E. Determination of capacity of labyrinth side weir by CFD [J]. Flow Measurement and Instrumentation, 2013, 29: 1–8.Article Google Scholar 
  21. Hao W. L., Wu W. Q., Zhu C. J. et al. Experimental study on vertical distribution of flow velocity in vegetated river channel [J]. Water Resources and Power, 2015, 33(2): 85–88(in Chinese).Google Scholar 
  22. Pietri L., Petroff A., Amielh M. et al. Turbulent flows interacting with varying density canopies [J]. Mécanique and Industries, 2009, 10(3–4): 181–185.Article Google Scholar 
  23. Li Y., Du W., Yu Z. et al. Impact of flexible emergent vegetation on the flow turbulence and kinetic energy characteristics in a flume experiment [J]. Journal of Hydro-environment Research, 2015, 9(3): 354–367.Article Google Scholar 
  24. Li W. Q., Wang D., Jiao J. L. et al. Effects of vegetation patch density on flow velocity characteristics in an open channel [J]. Journal of Hydrodynamics, 2018, 31(5): 1052–1059.Article Google Scholar 
  25. Langre E. D., Gutierrez A., Cossé J. On the scaling of drag reduction by reconfiguration in plants [J]. Comptes Rendus Mécanique, 2012, 340(1–2): 35–40.Article Google Scholar 
  26. Fathi-Maghadam M., Kouwen N. Nonrigid, nonsubmerged, vegetative roughness on floodplains [J]. Journal of Hydraulic Engineering, ASCE, 1997, 123(1): 51–57.Article Google Scholar 
  27. Liang D., Wu X. A random walk simulation of scalar mixing in flows through submerged vegetations [J]. Journal of Hydrodynamics, 2014, 26(3): 343–350.Article MathSciNet Google Scholar 
  28. Ghisalberti M., Nepf H. Mass transport in vegetated shear flows [J]. Environmental Fluid Mechanics, 2005, 5(6): 527–551.Article Google Scholar 
원문보기
Figure 2 Modeling the plant with cylindrical tubes at the bottom of the canal.

Optimized Vegetation Density to Dissipate Energy of Flood Flow in Open Canals

열린 운하에서 홍수 흐름의 에너지를 분산시키기 위해 최적화된 식생 밀도

Mahdi Feizbahr,1Navid Tonekaboni,2Guang-Jun Jiang,3,4and Hong-Xia Chen3,4
Academic Editor: Mohammad Yazdi

Abstract

강을 따라 식생은 조도를 증가시키고 평균 유속을 감소시키며, 유동 에너지를 감소시키고 강 횡단면의 유속 프로파일을 변경합니다. 자연의 많은 운하와 강은 홍수 동안 초목으로 덮여 있습니다. 운하의 조도는 식물의 영향을 많이 받기 때문에 홍수시 유동저항에 큰 영향을 미친다. 식물로 인한 흐름에 대한 거칠기 저항은 흐름 조건과 식물에 따라 달라지므로 모델은 유속, 유속 깊이 및 수로를 따라 식생 유형의 영향을 고려하여 유속을 시뮬레이션해야 합니다. 총 48개의 모델을 시뮬레이션하여 근관의 거칠기 효과를 조사했습니다. 결과는 속도를 높임으로써 베드 속도를 감소시키는 식생의 영향이 무시할만하다는 것을 나타냅니다.

Abstract

Vegetation along the river increases the roughness and reduces the average flow velocity, reduces flow energy, and changes the flow velocity profile in the cross section of the river. Many canals and rivers in nature are covered with vegetation during the floods. Canal’s roughness is strongly affected by plants and therefore it has a great effect on flow resistance during flood. Roughness resistance against the flow due to the plants depends on the flow conditions and plant, so the model should simulate the current velocity by considering the effects of velocity, depth of flow, and type of vegetation along the canal. Total of 48 models have been simulated to investigate the effect of roughness in the canal. The results indicated that, by enhancing the velocity, the effect of vegetation in decreasing the bed velocity is negligible, while when the current has lower speed, the effect of vegetation on decreasing the bed velocity is obviously considerable.

1. Introduction

Considering the impact of each variable is a very popular field within the analytical and statistical methods and intelligent systems [1–14]. This can help research for better modeling considering the relation of variables or interaction of them toward reaching a better condition for the objective function in control and engineering [15–27]. Consequently, it is necessary to study the effects of the passive factors on the active domain [28–36]. Because of the effect of vegetation on reducing the discharge capacity of rivers [37], pruning plants was necessary to improve the condition of rivers. One of the important effects of vegetation in river protection is the action of roots, which cause soil consolidation and soil structure improvement and, by enhancing the shear strength of soil, increase the resistance of canal walls against the erosive force of water. The outer limbs of the plant increase the roughness of the canal walls and reduce the flow velocity and deplete the flow energy in vicinity of the walls. Vegetation by reducing the shear stress of the canal bed reduces flood discharge and sedimentation in the intervals between vegetation and increases the stability of the walls [38–41].

One of the main factors influencing the speed, depth, and extent of flood in this method is Manning’s roughness coefficient. On the other hand, soil cover [42], especially vegetation, is one of the most determining factors in Manning’s roughness coefficient. Therefore, it is expected that those seasonal changes in the vegetation of the region will play an important role in the calculated value of Manning’s roughness coefficient and ultimately in predicting the flood wave behavior [43–45]. The roughness caused by plants’ resistance to flood current depends on the flow and plant conditions. Flow conditions include depth and velocity of the plant, and plant conditions include plant type, hardness or flexibility, dimensions, density, and shape of the plant [46]. In general, the issue discussed in this research is the optimization of flood-induced flow in canals by considering the effect of vegetation-induced roughness. Therefore, the effect of plants on the roughness coefficient and canal transmission coefficient and in consequence the flow depth should be evaluated [47, 48].

Current resistance is generally known by its roughness coefficient. The equation that is mainly used in this field is Manning equation. The ratio of shear velocity to average current velocity  is another form of current resistance. The reason for using the  ratio is that it is dimensionless and has a strong theoretical basis. The reason for using Manning roughness coefficient is its pervasiveness. According to Freeman et al. [49], the Manning roughness coefficient for plants was calculated according to the Kouwen and Unny [50] method for incremental resistance. This method involves increasing the roughness for various surface and plant irregularities. Manning’s roughness coefficient has all the factors affecting the resistance of the canal. Therefore, the appropriate way to more accurately estimate this coefficient is to know the factors affecting this coefficient [51].

To calculate the flow rate, velocity, and depth of flow in canals as well as flood and sediment estimation, it is important to evaluate the flow resistance. To determine the flow resistance in open ducts, Manning, Chézy, and Darcy–Weisbach relations are used [52]. In these relations, there are parameters such as Manning’s roughness coefficient (n), Chézy roughness coefficient (C), and Darcy–Weisbach coefficient (f). All three of these coefficients are a kind of flow resistance coefficient that is widely used in the equations governing flow in rivers [53].

The three relations that express the relationship between the average flow velocity (V) and the resistance and geometric and hydraulic coefficients of the canal are as follows:where n, f, and c are Manning, Darcy–Weisbach, and Chézy coefficients, respectively. V = average flow velocity, R = hydraulic radius, Sf = slope of energy line, which in uniform flow is equal to the slope of the canal bed,  = gravitational acceleration, and Kn is a coefficient whose value is equal to 1 in the SI system and 1.486 in the English system. The coefficients of resistance in equations (1) to (3) are related as follows:

Based on the boundary layer theory, the flow resistance for rough substrates is determined from the following general relation:where f = Darcy–Weisbach coefficient of friction, y = flow depth, Ks = bed roughness size, and A = constant coefficient.

On the other hand, the relationship between the Darcy–Weisbach coefficient of friction and the shear velocity of the flow is as follows:

By using equation (6), equation (5) is converted as follows:

Investigation on the effect of vegetation arrangement on shear velocity of flow in laboratory conditions showed that, with increasing the shear Reynolds number (), the numerical value of the  ratio also increases; in other words the amount of roughness coefficient increases with a slight difference in the cases without vegetation, checkered arrangement, and cross arrangement, respectively [54].

Roughness in river vegetation is simulated in mathematical models with a variable floor slope flume by different densities and discharges. The vegetation considered submerged in the bed of the flume. Results showed that, with increasing vegetation density, canal roughness and flow shear speed increase and with increasing flow rate and depth, Manning’s roughness coefficient decreases. Factors affecting the roughness caused by vegetation include the effect of plant density and arrangement on flow resistance, the effect of flow velocity on flow resistance, and the effect of depth [45, 55].

One of the works that has been done on the effect of vegetation on the roughness coefficient is Darby [56] study, which investigates a flood wave model that considers all the effects of vegetation on the roughness coefficient. There are currently two methods for estimating vegetation roughness. One method is to add the thrust force effect to Manning’s equation [47, 57, 58] and the other method is to increase the canal bed roughness (Manning-Strickler coefficient) [45, 59–61]. These two methods provide acceptable results in models designed to simulate floodplain flow. Wang et al. [62] simulate the floodplain with submerged vegetation using these two methods and to increase the accuracy of the results, they suggested using the effective height of the plant under running water instead of using the actual height of the plant. Freeman et al. [49] provided equations for determining the coefficient of vegetation roughness under different conditions. Lee et al. [63] proposed a method for calculating the Manning coefficient using the flow velocity ratio at different depths. Much research has been done on the Manning roughness coefficient in rivers, and researchers [49, 63–66] sought to obtain a specific number for n to use in river engineering. However, since the depth and geometric conditions of rivers are completely variable in different places, the values of Manning roughness coefficient have changed subsequently, and it has not been possible to choose a fixed number. In river engineering software, the Manning roughness coefficient is determined only for specific and constant conditions or normal flow. Lee et al. [63] stated that seasonal conditions, density, and type of vegetation should also be considered. Hydraulic roughness and Manning roughness coefficient n of the plant were obtained by estimating the total Manning roughness coefficient from the matching of the measured water surface curve and water surface height. The following equation is used for the flow surface curve:where  is the depth of water change, S0 is the slope of the canal floor, Sf is the slope of the energy line, and Fr is the Froude number which is obtained from the following equation:where D is the characteristic length of the canal. Flood flow velocity is one of the important parameters of flood waves, which is very important in calculating the water level profile and energy consumption. In the cases where there are many limitations for researchers due to the wide range of experimental dimensions and the variety of design parameters, the use of numerical methods that are able to estimate the rest of the unknown results with acceptable accuracy is economically justified.

FLOW-3D software uses Finite Difference Method (FDM) for numerical solution of two-dimensional and three-dimensional flow. This software is dedicated to computational fluid dynamics (CFD) and is provided by Flow Science [67]. The flow is divided into networks with tubular cells. For each cell there are values of dependent variables and all variables are calculated in the center of the cell, except for the velocity, which is calculated at the center of the cell. In this software, two numerical techniques have been used for geometric simulation, FAVOR™ (Fractional-Area-Volume-Obstacle-Representation) and the VOF (Volume-of-Fluid) method. The equations used at this model for this research include the principle of mass survival and the magnitude of motion as follows. The fluid motion equations in three dimensions, including the Navier–Stokes equations with some additional terms, are as follows:where  are mass accelerations in the directions x, y, z and  are viscosity accelerations in the directions x, y, z and are obtained from the following equations:

Shear stresses  in equation (11) are obtained from the following equations:

The standard model is used for high Reynolds currents, but in this model, RNG theory allows the analytical differential formula to be used for the effective viscosity that occurs at low Reynolds numbers. Therefore, the RNG model can be used for low and high Reynolds currents.

Weather changes are high and this affects many factors continuously. The presence of vegetation in any area reduces the velocity of surface flows and prevents soil erosion, so vegetation will have a significant impact on reducing destructive floods. One of the methods of erosion protection in floodplain watersheds is the use of biological methods. The presence of vegetation in watersheds reduces the flow rate during floods and prevents soil erosion. The external organs of plants increase the roughness and decrease the velocity of water flow and thus reduce its shear stress energy. One of the important factors with which the hydraulic resistance of plants is expressed is the roughness coefficient. Measuring the roughness coefficient of plants and investigating their effect on reducing velocity and shear stress of flow is of special importance.

Roughness coefficients in canals are affected by two main factors, namely, flow conditions and vegetation characteristics [68]. So far, much research has been done on the effect of the roughness factor created by vegetation, but the issue of plant density has received less attention. For this purpose, this study was conducted to investigate the effect of vegetation density on flow velocity changes.

In a study conducted using a software model on three density modes in the submerged state effect on flow velocity changes in 48 different modes was investigated (Table 1).

Table 1 

The studied models.

The number of cells used in this simulation is equal to 1955888 cells. The boundary conditions were introduced to the model as a constant speed and depth (Figure 1). At the output boundary, due to the presence of supercritical current, no parameter for the current is considered. Absolute roughness for floors and walls was introduced to the model (Figure 1). In this case, the flow was assumed to be nonviscous and air entry into the flow was not considered. After  seconds, this model reached a convergence accuracy of .

Figure 1 

The simulated model and its boundary conditions.

Due to the fact that it is not possible to model the vegetation in FLOW-3D software, in this research, the vegetation of small soft plants was studied so that Manning’s coefficients can be entered into the canal bed in the form of roughness coefficients obtained from the studies of Chow [69] in similar conditions. In practice, in such modeling, the effect of plant height is eliminated due to the small height of herbaceous plants, and modeling can provide relatively acceptable results in these conditions.

48 models with input velocities proportional to the height of the regular semihexagonal canal were considered to create supercritical conditions. Manning coefficients were applied based on Chow [69] studies in order to control the canal bed. Speed profiles were drawn and discussed.

Any control and simulation system has some inputs that we should determine to test any technology [70–77]. Determination and true implementation of such parameters is one of the key steps of any simulation [23, 78–81] and computing procedure [82–86]. The input current is created by applying the flow rate through the VFR (Volume Flow Rate) option and the output flow is considered Output and for other borders the Symmetry option is considered.

Simulation of the models and checking their action and responses and observing how a process behaves is one of the accepted methods in engineering and science [87, 88]. For verification of FLOW-3D software, the results of computer simulations are compared with laboratory measurements and according to the values of computational error, convergence error, and the time required for convergence, the most appropriate option for real-time simulation is selected (Figures 2 and 3 ).

Figure 2 

Modeling the plant with cylindrical tubes at the bottom of the canal.

Figure 3 

Velocity profiles in positions 2 and 5.

The canal is 7 meters long, 0.5 meters wide, and 0.8 meters deep. This test was used to validate the application of the software to predict the flow rate parameters. In this experiment, instead of using the plant, cylindrical pipes were used in the bottom of the canal.

The conditions of this modeling are similar to the laboratory conditions and the boundary conditions used in the laboratory were used for numerical modeling. The critical flow enters the simulation model from the upstream boundary, so in the upstream boundary conditions, critical velocity and depth are considered. The flow at the downstream boundary is supercritical, so no parameters are applied to the downstream boundary.

The software well predicts the process of changing the speed profile in the open canal along with the considered obstacles. The error in the calculated speed values can be due to the complexity of the flow and the interaction of the turbulence caused by the roughness of the floor with the turbulence caused by the three-dimensional cycles in the hydraulic jump. As a result, the software is able to predict the speed distribution in open canals.

2. Modeling Results

After analyzing the models, the results were shown in graphs (Figures 4–14 ). The total number of experiments in this study was 48 due to the limitations of modeling.


(d)


(a)


(b)


(c)


(d)


(a)


(b)


(c)


(d)

  • (a)
    (a)
  • (b)
    (b)
  • (c)
    (c)
  • (d)
    (d)

Figure 4 

Flow velocity profiles for canals with a depth of 1 m and flow velocities of 3–3.3 m/s. Canal with a depth of 1 meter and a flow velocity of (a) 3 meters per second, (b) 3.1 meters per second, (c) 3.2 meters per second, and (d) 3.3 meters per second.

Figure 5 

Canal diagram with a depth of 1 meter and a flow rate of 3 meters per second.

Figure 6 

Canal diagram with a depth of 1 meter and a flow rate of 3.1 meters per second.

Figure 7 

Canal diagram with a depth of 1 meter and a flow rate of 3.2 meters per second.

Figure 8 

Canal diagram with a depth of 1 meter and a flow rate of 3.3 meters per second.


(d)


(a)


(b)


(c)


(d)


(a)


(b)


(c)


(d)

  • (a)
    (a)
  • (b)
    (b)
  • (c)
    (c)
  • (d)
    (d)

Figure 9 

Flow velocity profiles for canals with a depth of 2 m and flow velocities of 4–4.3 m/s. Canal with a depth of 2 meters and a flow rate of (a) 4 meters per second, (b) 4.1 meters per second, (c) 4.2 meters per second, and (d) 4.3 meters per second.

Figure 10 

Canal diagram with a depth of 2 meters and a flow rate of 4 meters per second.

Figure 11 

Canal diagram with a depth of 2 meters and a flow rate of 4.1 meters per second.

Figure 12 

Canal diagram with a depth of 2 meters and a flow rate of 4.2 meters per second.

Figure 13 

Canal diagram with a depth of 2 meters and a flow rate of 4.3 meters per second.


(d)


(a)


(b)


(c)


(d)


(a)


(b)


(c)


(d)

  • (a)
    (a)
  • (b)
    (b)
  • (c)
    (c)
  • (d)
    (d)

Figure 14 

Flow velocity profiles for canals with a depth of 3 m and flow velocities of 5–5.3 m/s. Canal with a depth of 2 meters and a flow rate of (a) 4 meters per second, (b) 4.1 meters per second, (c) 4.2 meters per second, and (d) 4.3 meters per second.

To investigate the effects of roughness with flow velocity, the trend of flow velocity changes at different depths and with supercritical flow to a Froude number proportional to the depth of the section has been obtained.

According to the velocity profiles of Figure 5, it can be seen that, with the increasing of Manning’s coefficient, the canal bed speed decreases.

According to Figures 5 to 8, it can be found that, with increasing the Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of the models 1 to 12, which can be justified by increasing the speed and of course increasing the Froude number.

According to Figure 10, we see that, with increasing Manning’s coefficient, the canal bed speed decreases.

According to Figure 11, we see that, with increasing Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of Figures 5–10, which can be justified by increasing the speed and, of course, increasing the Froude number.

With increasing Manning’s coefficient, the canal bed speed decreases (Figure 12). But this deceleration is more noticeable than the deceleration of the higher models (Figures 5–8 and 10, 11), which can be justified by increasing the speed and, of course, increasing the Froude number.

According to Figure 13, with increasing Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of Figures 5 to 12, which can be justified by increasing the speed and, of course, increasing the Froude number.

According to Figure 15, with increasing Manning’s coefficient, the canal bed speed decreases.

Figure 15 

Canal diagram with a depth of 3 meters and a flow rate of 5 meters per second.

According to Figure 16, with increasing Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of the higher model, which can be justified by increasing the speed and, of course, increasing the Froude number.

Figure 16 

Canal diagram with a depth of 3 meters and a flow rate of 5.1 meters per second.

According to Figure 17, it is clear that, with increasing Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of the higher models, which can be justified by increasing the speed and, of course, increasing the Froude number.

Figure 17 

Canal diagram with a depth of 3 meters and a flow rate of 5.2 meters per second.

According to Figure 18, with increasing Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of the higher models, which can be justified by increasing the speed and, of course, increasing the Froude number.

Figure 18 

Canal diagram with a depth of 3 meters and a flow rate of 5.3 meters per second.

According to Figure 19, it can be seen that the vegetation placed in front of the flow input velocity has negligible effect on the reduction of velocity, which of course can be justified due to the flexibility of the vegetation. The only unusual thing is the unexpected decrease in floor speed of 3 m/s compared to higher speeds.


(c)


(a)


(b)


(c)


(a)


(b)


(c)

  • (a)
    (a)
  • (b)
    (b)
  • (c)
    (c)

Figure 19 

Comparison of velocity profiles with the same plant densities (depth 1 m). Comparison of velocity profiles with (a) plant densities of 25%, depth 1 m; (b) plant densities of 50%, depth 1 m; and (c) plant densities of 75%, depth 1 m.

According to Figure 20, by increasing the speed of vegetation, the effect of vegetation on reducing the flow rate becomes more noticeable. And the role of input current does not have much effect in reducing speed.


(c)


(a)


(b)


(c)


(a)


(b)


(c)

  • (a)
    (a)
  • (b)
    (b)
  • (c)
    (c)

Figure 20 

Comparison of velocity profiles with the same plant densities (depth 2 m). Comparison of velocity profiles with (a) plant densities of 25%, depth 2 m; (b) plant densities of 50%, depth 2 m; and (c) plant densities of 75%, depth 2 m.

According to Figure 21, it can be seen that, with increasing speed, the effect of vegetation on reducing the bed flow rate becomes more noticeable and the role of the input current does not have much effect. In general, it can be seen that, by increasing the speed of the input current, the slope of the profiles increases from the bed to the water surface and due to the fact that, in software, the roughness coefficient applies to the channel floor only in the boundary conditions, this can be perfectly justified. Of course, it can be noted that, due to the flexible conditions of the vegetation of the bed, this modeling can show acceptable results for such grasses in the canal floor. In the next directions, we may try application of swarm-based optimization methods for modeling and finding the most effective factors in this research [2, 7, 8, 15, 18, 89–94]. In future, we can also apply the simulation logic and software of this research for other domains such as power engineering [95–99].


(c)


(a)


(b)


(c)


(a)


(b)


(c)

  • (a)
    (a)
  • (b)
    (b)
  • (c)
    (c)

Figure 21 

Comparison of velocity profiles with the same plant densities (depth 3 m). Comparison of velocity profiles with (a) plant densities of 25%, depth 3 m; (b) plant densities of 50%, depth 3 m; and (c) plant densities of 75%, depth 3 m.

3. Conclusion

The effects of vegetation on the flood canal were investigated by numerical modeling with FLOW-3D software. After analyzing the results, the following conclusions were reached:(i)Increasing the density of vegetation reduces the velocity of the canal floor but has no effect on the velocity of the canal surface.(ii)Increasing the Froude number is directly related to increasing the speed of the canal floor.(iii)In the canal with a depth of one meter, a sudden increase in speed can be observed from the lowest speed and higher speed, which is justified by the sudden increase in Froude number.(iv)As the inlet flow rate increases, the slope of the profiles from the bed to the water surface increases.(v)By reducing the Froude number, the effect of vegetation on reducing the flow bed rate becomes more noticeable. And the input velocity in reducing the velocity of the canal floor does not have much effect.(vi)At a flow rate between 3 and 3.3 meters per second due to the shallow depth of the canal and the higher landing number a more critical area is observed in which the flow bed velocity in this area is between 2.86 and 3.1 m/s.(vii)Due to the critical flow velocity and the slight effect of the roughness of the horseshoe vortex floor, it is not visible and is only partially observed in models 1-2-3 and 21.(viii)As the flow rate increases, the effect of vegetation on the rate of bed reduction decreases.(ix)In conditions where less current intensity is passing, vegetation has a greater effect on reducing current intensity and energy consumption increases.(x)In the case of using the flow rate of 0.8 cubic meters per second, the velocity distribution and flow regime show about 20% more energy consumption than in the case of using the flow rate of 1.3 cubic meters per second.

Nomenclature

n:Manning’s roughness coefficient
C:Chézy roughness coefficient
f:Darcy–Weisbach coefficient
V:Flow velocity
R:Hydraulic radius
g:Gravitational acceleration
y:Flow depth
Ks:Bed roughness
A:Constant coefficient
:Reynolds number
∂y/∂x:Depth of water change
S0:Slope of the canal floor
Sf:Slope of energy line
Fr:Froude number
D:Characteristic length of the canal
G:Mass acceleration
:Shear stresses.

Data Availability

All data are included within the paper.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was partially supported by the National Natural Science Foundation of China under Contract no. 71761030 and Natural Science Foundation of Inner Mongolia under Contract no. 2019LH07003.

References

  1. H. Yu, L. Jie, W. Gui et al., “Dynamic Gaussian bare-bones fruit fly optimizers with abandonment mechanism: method and analysis,” Engineering with Computers, vol. 20, pp. 1–29, 2020.View at: Publisher Site | Google Scholar
  2. X. Zhao, D. Li, B. Yang, C. Ma, Y. Zhu, and H. Chen, “Feature selection based on improved ant colony optimization for online detection of foreign fiber in cotton,” Applied Soft Computing, vol. 24, pp. 585–596, 2014.View at: Publisher Site | Google Scholar
  3. J. Hu, H. Chen, A. A. Heidari et al., “Orthogonal learning covariance matrix for defects of grey wolf optimizer: insights, balance, diversity, and feature selection,” Knowledge-Based Systems, vol. 213, Article ID 106684, 2021.View at: Publisher Site | Google Scholar
  4. C. Yu, M. Chen, K. Chen et al., “SGOA: annealing-behaved grasshopper optimizer for global tasks,” Engineering with Computers, vol. 4, pp. 1–28, 2021.View at: Publisher Site | Google Scholar
  5. W. Shan, Z. Qiao, A. A. Heidari, H. Chen, H. Turabieh, and Y. Teng, “Double adaptive weights for stabilization of moth flame optimizer: balance analysis, engineering cases, and medical diagnosis,” Knowledge-Based Systems, vol. 8, Article ID 106728, 2020.View at: Google Scholar
  6. J. Tu, H. Chen, J. Liu et al., “Evolutionary biogeography-based whale optimization methods with communication structure: towards measuring the balance,” Knowledge-Based Systems, vol. 212, Article ID 106642, 2021.View at: Publisher Site | Google Scholar
  7. Y. Zhang, R. Liu, X. Wang et al., “Towards augmented kernel extreme learning models for bankruptcy prediction: algorithmic behavior and comprehensive analysis,” Neurocomputing, vol. 430, 2020.View at: Google Scholar
  8. H.-L. Chen, G. Wang, C. Ma, Z.-N. Cai, W.-B. Liu, and S.-J. Wang, “An efficient hybrid kernel extreme learning machine approach for early diagnosis of Parkinson׳s disease,” Neurocomputing, vol. 184, pp. 131–144, 2016.View at: Publisher Site | Google Scholar
  9. J. Xia, H. Chen, Q. Li et al., “Ultrasound-based differentiation of malignant and benign thyroid Nodules: an extreme learning machine approach,” Computer Methods and Programs in Biomedicine, vol. 147, pp. 37–49, 2017.View at: Publisher Site | Google Scholar
  10. C. Li, L. Hou, B. Y. Sharma et al., “Developing a new intelligent system for the diagnosis of tuberculous pleural effusion,” Computer Methods and Programs in Biomedicine, vol. 153, pp. 211–225, 2018.View at: Publisher Site | Google Scholar
  11. X. Xu and H.-L. Chen, “Adaptive computational chemotaxis based on field in bacterial foraging optimization,” Soft Computing, vol. 18, no. 4, pp. 797–807, 2014.View at: Publisher Site | Google Scholar
  12. M. Wang, H. Chen, B. Yang et al., “Toward an optimal kernel extreme learning machine using a chaotic moth-flame optimization strategy with applications in medical diagnoses,” Neurocomputing, vol. 267, pp. 69–84, 2017.View at: Publisher Site | Google Scholar
  13. L. Chao, K. Zhang, Z. Li, Y. Zhu, J. Wang, and Z. Yu, “Geographically weighted regression based methods for merging satellite and gauge precipitation,” Journal of Hydrology, vol. 558, pp. 275–289, 2018.View at: Publisher Site | Google Scholar
  14. F. J. Golrokh, G. Azeem, and A. Hasan, “Eco-efficiency evaluation in cement industries: DEA malmquist productivity index using optimization models,” ENG Transactions, vol. 1, 2020.View at: Google Scholar
  15. D. Zhao, L. Lei, F. Yu et al., “Chaotic random spare ant colony optimization for multi-threshold image segmentation of 2D Kapur entropy,” Knowledge-Based Systems, vol. 8, Article ID 106510, 2020.View at: Google Scholar
  16. Y. Zhang, R. Liu, X. Wang, H. Chen, and C. Li, “Boosted binary Harris hawks optimizer and feature selection,” Engineering with Computers, vol. 517, pp. 1–30, 2020.View at: Publisher Site | Google Scholar
  17. L. Hu, G. Hong, J. Ma, X. Wang, and H. Chen, “An efficient machine learning approach for diagnosis of paraquat-poisoned patients,” Computers in Biology and Medicine, vol. 59, pp. 116–124, 2015.View at: Publisher Site | Google Scholar
  18. L. Shen, H. Chen, Z. Yu et al., “Evolving support vector machines using fruit fly optimization for medical data classification,” Knowledge-Based Systems, vol. 96, pp. 61–75, 2016.View at: Publisher Site | Google Scholar
  19. X. Zhao, X. Zhang, Z. Cai et al., “Chaos enhanced grey wolf optimization wrapped ELM for diagnosis of paraquat-poisoned patients,” Computational Biology and Chemistry, vol. 78, pp. 481–490, 2019.View at: Publisher Site | Google Scholar
  20. Y. Xu, H. Chen, J. Luo, Q. Zhang, S. Jiao, and X. Zhang, “Enhanced Moth-flame optimizer with mutation strategy for global optimization,” Information Sciences, vol. 492, pp. 181–203, 2019.View at: Publisher Site | Google Scholar
  21. M. Wang and H. Chen, “Chaotic multi-swarm whale optimizer boosted support vector machine for medical diagnosis,” Applied Soft Computing Journal, vol. 88, Article ID 105946, 2020.View at: Publisher Site | Google Scholar
  22. Y. Chen, J. Li, H. Lu, and P. Yan, “Coupling system dynamics analysis and risk aversion programming for optimizing the mixed noise-driven shale gas-water supply chains,” Journal of Cleaner Production, vol. 278, Article ID 123209, 2020.View at: Google Scholar
  23. H. Tang, Y. Xu, A. Lin et al., “Predicting green consumption behaviors of students using efficient firefly grey wolf-assisted K-nearest neighbor classifiers,” IEEE Access, vol. 8, pp. 35546–35562, 2020.View at: Publisher Site | Google Scholar
  24. H.-J. Ma and G.-H. Yang, “Adaptive fault tolerant control of cooperative heterogeneous systems with actuator faults and unreliable interconnections,” IEEE Transactions on Automatic Control, vol. 61, no. 11, pp. 3240–3255, 2015.View at: Google Scholar
  25. H.-J. Ma and L.-X. Xu, “Decentralized adaptive fault-tolerant control for a class of strong interconnected nonlinear systems via graph theory,” IEEE Transactions on Automatic Control, vol. 66, 2020.View at: Google Scholar
  26. H. J. Ma, L. X. Xu, and G. H. Yang, “Multiple environment integral reinforcement learning-based fault-tolerant control for affine nonlinear systems,” IEEE Transactions on Cybernetics, vol. 51, pp. 1–16, 2019.View at: Publisher Site | Google Scholar
  27. J. Hu, M. Wang, C. Zhao, Q. Pan, and C. Du, “Formation control and collision avoidance for multi-UAV systems based on Voronoi partition,” Science China Technological Sciences, vol. 63, no. 1, pp. 65–72, 2020.View at: Publisher Site | Google Scholar
  28. C. Zhang, H. Li, Y. Qian, C. Chen, and X. Zhou, “Locality-constrained discriminative matrix regression for robust face identification,” IEEE Transactions on Neural Networks and Learning Systems, vol. 99, pp. 1–15, 2020.View at: Publisher Site | Google Scholar
  29. X. Zhang, D. Wang, Z. Zhou, and Y. Ma, “Robust low-rank tensor recovery with rectification and alignment,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 238–255, 2019.View at: Google Scholar
  30. X. Zhang, J. Wang, T. Wang, R. Jiang, J. Xu, and L. Zhao, “Robust feature learning for adversarial defense via hierarchical feature alignment,” Information Sciences, vol. 560, 2020.View at: Google Scholar
  31. X. Zhang, R. Jiang, T. Wang, and J. Wang, “Recursive neural network for video deblurring,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 03, p. 1, 2020.View at: Publisher Site | Google Scholar
  32. X. Zhang, T. Wang, J. Wang, G. Tang, and L. Zhao, “Pyramid channel-based feature attention network for image dehazing,” Computer Vision and Image Understanding, vol. 197-198, Article ID 103003, 2020.View at: Publisher Site | Google Scholar
  33. X. Zhang, T. Wang, W. Luo, and P. Huang, “Multi-level fusion and attention-guided CNN for image dehazing,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 3, p. 1, 2020.View at: Publisher Site | Google Scholar
  34. L. He, J. Shen, and Y. Zhang, “Ecological vulnerability assessment for ecological conservation and environmental management,” Journal of Environmental Management, vol. 206, pp. 1115–1125, 2018.View at: Publisher Site | Google Scholar
  35. Y. Chen, W. Zheng, W. Li, and Y. Huang, “Large group Activity security risk assessment and risk early warning based on random forest algorithm,” Pattern Recognition Letters, vol. 144, pp. 1–5, 2021.View at: Publisher Site | Google Scholar
  36. J. Hu, H. Zhang, Z. Li, C. Zhao, Z. Xu, and Q. Pan, “Object traversing by monocular UAV in outdoor environment,” Asian Journal of Control, vol. 25, 2020.View at: Google Scholar
  37. P. Tian, H. Lu, W. Feng, Y. Guan, and Y. Xue, “Large decrease in streamflow and sediment load of Qinghai-Tibetan Plateau driven by future climate change: a case study in Lhasa River Basin,” Catena, vol. 187, Article ID 104340, 2020.View at: Publisher Site | Google Scholar
  38. A. Stokes, C. Atger, A. G. Bengough, T. Fourcaud, and R. C. Sidle, “Desirable plant root traits for protecting natural and engineered slopes against landslides,” Plant and Soil, vol. 324, no. 1, pp. 1–30, 2009.View at: Publisher Site | Google Scholar
  39. T. B. Devi, A. Sharma, and B. Kumar, “Studies on emergent flow over vegetative channel bed with downward seepage,” Hydrological Sciences Journal, vol. 62, no. 3, pp. 408–420, 2017.View at: Google Scholar
  40. G. Ireland, M. Volpi, and G. Petropoulos, “Examining the capability of supervised machine learning classifiers in extracting flooded areas from Landsat TM imagery: a case study from a Mediterranean flood,” Remote Sensing, vol. 7, no. 3, pp. 3372–3399, 2015.View at: Publisher Site | Google Scholar
  41. L. Goodarzi and S. Javadi, “Assessment of aquifer vulnerability using the DRASTIC model; a case study of the Dezful-Andimeshk Aquifer,” Computational Research Progress in Applied Science & Engineering, vol. 2, no. 1, pp. 17–22, 2016.View at: Google Scholar
  42. K. Zhang, Q. Wang, L. Chao et al., “Ground observation-based analysis of soil moisture spatiotemporal variability across a humid to semi-humid transitional zone in China,” Journal of Hydrology, vol. 574, pp. 903–914, 2019.View at: Publisher Site | Google Scholar
  43. L. De Doncker, P. Troch, R. Verhoeven, K. Bal, P. Meire, and J. Quintelier, “Determination of the Manning roughness coefficient influenced by vegetation in the river Aa and Biebrza river,” Environmental Fluid Mechanics, vol. 9, no. 5, pp. 549–567, 2009.View at: Publisher Site | Google Scholar
  44. M. Fathi-Moghadam and K. Drikvandi, “Manning roughness coefficient for rivers and flood plains with non-submerged vegetation,” International Journal of Hydraulic Engineering, vol. 1, no. 1, pp. 1–4, 2012.View at: Google Scholar
  45. F.-C. Wu, H. W. Shen, and Y.-J. Chou, “Variation of roughness coefficients for unsubmerged and submerged vegetation,” Journal of Hydraulic Engineering, vol. 125, no. 9, pp. 934–942, 1999.View at: Publisher Site | Google Scholar
  46. M. K. Wood, “Rangeland vegetation-hydrologic interactions,” in Vegetation Science Applications for Rangeland Analysis and Management, vol. 3, pp. 469–491, Springer, 1988.View at: Publisher Site | Google Scholar
  47. C. Wilson, O. Yagci, H.-P. Rauch, and N. Olsen, “3D numerical modelling of a willow vegetated river/floodplain system,” Journal of Hydrology, vol. 327, no. 1-2, pp. 13–21, 2006.View at: Publisher Site | Google Scholar
  48. R. Yazarloo, M. Khamehchian, and M. R. Nikoodel, “Observational-computational 3d engineering geological model and geotechnical characteristics of young sediments of golestan province,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 03, 2017.View at: Google Scholar
  49. G. E. Freeman, W. H. Rahmeyer, and R. R. Copeland, “Determination of resistance due to shrubs and woody vegetation,” International Journal of River Basin Management, vol. 19, 2000.View at: Google Scholar
  50. N. Kouwen and T. E. Unny, “Flexible roughness in open channels,” Journal of the Hydraulics Division, vol. 99, no. 5, pp. 713–728, 1973.View at: Publisher Site | Google Scholar
  51. S. Hosseini and J. Abrishami, Open Channel Hydraulics, Elsevier, Amsterdam, Netherlands, 2007.
  52. C. S. James, A. L. Birkhead, A. A. Jordanova, and J. J. O’Sullivan, “Flow resistance of emergent vegetation,” Journal of Hydraulic Research, vol. 42, no. 4, pp. 390–398, 2004.View at: Publisher Site | Google Scholar
  53. F. Huthoff and D. Augustijn, “Channel roughness in 1D steady uniform flow: Manning or Chézy?,,” NCR-days, vol. 102, 2004.View at: Google Scholar
  54. M. S. Sabegh, M. Saneie, M. Habibi, A. A. Abbasi, and M. Ghadimkhani, “Experimental investigation on the effect of river bank tree planting array, on shear velocity,” Journal of Watershed Engineering and Management, vol. 2, no. 4, 2011.View at: Google Scholar
  55. A. Errico, V. Pasquino, M. Maxwald, G. B. Chirico, L. Solari, and F. Preti, “The effect of flexible vegetation on flow in drainage channels: estimation of roughness coefficients at the real scale,” Ecological Engineering, vol. 120, pp. 411–421, 2018.View at: Publisher Site | Google Scholar
  56. S. E. Darby, “Effect of riparian vegetation on flow resistance and flood potential,” Journal of Hydraulic Engineering, vol. 125, no. 5, pp. 443–454, 1999.View at: Publisher Site | Google Scholar
  57. V. Kutija and H. Thi Minh Hong, “A numerical model for assessing the additional resistance to flow introduced by flexible vegetation,” Journal of Hydraulic Research, vol. 34, no. 1, pp. 99–114, 1996.View at: Publisher Site | Google Scholar
  58. T. Fischer-Antze, T. Stoesser, P. Bates, and N. R. B. Olsen, “3D numerical modelling of open-channel flow with submerged vegetation,” Journal of Hydraulic Research, vol. 39, no. 3, pp. 303–310, 2001.View at: Publisher Site | Google Scholar
  59. U. Stephan and D. Gutknecht, “Hydraulic resistance of submerged flexible vegetation,” Journal of Hydrology, vol. 269, no. 1-2, pp. 27–43, 2002.View at: Publisher Site | Google Scholar
  60. F. G. Carollo, V. Ferro, and D. Termini, “Flow resistance law in channels with flexible submerged vegetation,” Journal of Hydraulic Engineering, vol. 131, no. 7, pp. 554–564, 2005.View at: Publisher Site | Google Scholar
  61. W. Fu-sheng, “Flow resistance of flexible vegetation in open channel,” Journal of Hydraulic Engineering, vol. S1, 2007.View at: Google Scholar
  62. P.-f. Wang, C. Wang, and D. Z. Zhu, “Hydraulic resistance of submerged vegetation related to effective height,” Journal of Hydrodynamics, vol. 22, no. 2, pp. 265–273, 2010.View at: Publisher Site | Google Scholar
  63. J. K. Lee, L. C. Roig, H. L. Jenter, and H. M. Visser, “Drag coefficients for modeling flow through emergent vegetation in the Florida Everglades,” Ecological Engineering, vol. 22, no. 4-5, pp. 237–248, 2004.View at: Publisher Site | Google Scholar
  64. G. J. Arcement and V. R. Schneider, Guide for Selecting Manning’s Roughness Coefficients for Natural Channels and Flood Plains, US Government Printing Office, Washington, DC, USA, 1989.
  65. Y. Ding and S. S. Y. Wang, “Identification of Manning’s roughness coefficients in channel network using adjoint analysis,” International Journal of Computational Fluid Dynamics, vol. 19, no. 1, pp. 3–13, 2005.View at: Publisher Site | Google Scholar
  66. E. T. Engman, “Roughness coefficients for routing surface runoff,” Journal of Irrigation and Drainage Engineering, vol. 112, no. 1, pp. 39–53, 1986.View at: Publisher Site | Google Scholar
  67. M. Feizbahr, C. Kok Keong, F. Rostami, and M. Shahrokhi, “Wave energy dissipation using perforated and non perforated piles,” International Journal of Engineering, vol. 31, no. 2, pp. 212–219, 2018.View at: Publisher Site | Google Scholar
  68. M. Farzadkhoo, A. Keshavarzi, H. Hamidifar, and M. Javan, “Sudden pollutant discharge in vegetated compound meandering rivers,” Catena, vol. 182, Article ID 104155, 2019.View at: Publisher Site | Google Scholar
  69. V. T. Chow, Open-channel Hydraulics, Mcgraw-Hill Civil Engineering Series, Chennai, TN, India, 1959.
  70. X. Zhang, R. Jing, Z. Li, Z. Li, X. Chen, and C.-Y. Su, “Adaptive pseudo inverse control for a class of nonlinear asymmetric and saturated nonlinear hysteretic systems,” IEEE/CAA Journal of Automatica Sinica, vol. 8, no. 4, pp. 916–928, 2020.View at: Google Scholar
  71. C. Zuo, Q. Chen, L. Tian, L. Waller, and A. Asundi, “Transport of intensity phase retrieval and computational imaging for partially coherent fields: the phase space perspective,” Optics and Lasers in Engineering, vol. 71, pp. 20–32, 2015.View at: Publisher Site | Google Scholar
  72. C. Zuo, J. Sun, J. Li, J. Zhang, A. Asundi, and Q. Chen, “High-resolution transport-of-intensity quantitative phase microscopy with annular illumination,” Scientific Reports, vol. 7, no. 1, pp. 7654–7722, 2017.View at: Publisher Site | Google Scholar
  73. B.-H. Li, Y. Liu, A.-M. Zhang, W.-H. Wang, and S. Wan, “A survey on blocking technology of entity resolution,” Journal of Computer Science and Technology, vol. 35, no. 4, pp. 769–793, 2020.View at: Publisher Site | Google Scholar
  74. Y. Liu, B. Zhang, Y. Feng et al., “Development of 340-GHz transceiver front end based on GaAs monolithic integration technology for THz active imaging array,” Applied Sciences, vol. 10, no. 21, p. 7924, 2020.View at: Publisher Site | Google Scholar
  75. J. Hu, H. Zhang, L. Liu, X. Zhu, C. Zhao, and Q. Pan, “Convergent multiagent formation control with collision avoidance,” IEEE Transactions on Robotics, vol. 36, no. 6, pp. 1805–1818, 2020.View at: Publisher Site | Google Scholar
  76. M. B. Movahhed, J. Ayoubinejad, F. N. Asl, and M. Feizbahr, “The effect of rain on pedestrians crossing speed,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 6, no. 3, 2020.View at: Google Scholar
  77. A. Li, D. Spano, J. Krivochiza et al., “A tutorial on interference exploitation via symbol-level precoding: overview, state-of-the-art and future directions,” IEEE Communications Surveys & Tutorials, vol. 22, no. 2, pp. 796–839, 2020.View at: Publisher Site | Google Scholar
  78. W. Zhu, C. Ma, X. Zhao et al., “Evaluation of sino foreign cooperative education project using orthogonal sine cosine optimized kernel extreme learning machine,” IEEE Access, vol. 8, pp. 61107–61123, 2020.View at: Publisher Site | Google Scholar
  79. G. Liu, W. Jia, M. Wang et al., “Predicting cervical hyperextension injury: a covariance guided sine cosine support vector machine,” IEEE Access, vol. 8, pp. 46895–46908, 2020.View at: Publisher Site | Google Scholar
  80. Y. Wei, H. Lv, M. Chen et al., “Predicting entrepreneurial intention of students: an extreme learning machine with Gaussian barebone harris hawks optimizer,” IEEE Access, vol. 8, pp. 76841–76855, 2020.View at: Publisher Site | Google Scholar
  81. A. Lin, Q. Wu, A. A. Heidari et al., “Predicting intentions of students for master programs using a chaos-induced sine cosine-based fuzzy K-Nearest neighbor classifier,” Ieee Access, vol. 7, pp. 67235–67248, 2019.View at: Publisher Site | Google Scholar
  82. Y. Fan, P. Wang, A. A. Heidari et al., “Rationalized fruit fly optimization with sine cosine algorithm: a comprehensive analysis,” Expert Systems with Applications, vol. 157, Article ID 113486, 2020.View at: Publisher Site | Google Scholar
  83. E. Rodríguez-Esparza, L. A. Zanella-Calzada, D. Oliva et al., “An efficient Harris hawks-inspired image segmentation method,” Expert Systems with Applications, vol. 155, Article ID 113428, 2020.View at: Publisher Site | Google Scholar
  84. S. Jiao, G. Chong, C. Huang et al., “Orthogonally adapted Harris hawks optimization for parameter estimation of photovoltaic models,” Energy, vol. 203, Article ID 117804, 2020.View at: Publisher Site | Google Scholar
  85. Z. Xu, Z. Hu, A. A. Heidari et al., “Orthogonally-designed adapted grasshopper optimization: a comprehensive analysis,” Expert Systems with Applications, vol. 150, Article ID 113282, 2020.View at: Publisher Site | Google Scholar
  86. A. Abbassi, R. Abbassi, A. A. Heidari et al., “Parameters identification of photovoltaic cell models using enhanced exploratory salp chains-based approach,” Energy, vol. 198, Article ID 117333, 2020.View at: Publisher Site | Google Scholar
  87. M. Mahmoodi and K. K. Aminjan, “Numerical simulation of flow through sukhoi 24 air inlet,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 03, 2017.View at: Google Scholar
  88. F. J. Golrokh and A. Hasan, “A comparison of machine learning clustering algorithms based on the DEA optimization approach for pharmaceutical companies in developing countries,” ENG Transactions, vol. 1, 2020.View at: Google Scholar
  89. H. Chen, A. A. Heidari, H. Chen, M. Wang, Z. Pan, and A. H. Gandomi, “Multi-population differential evolution-assisted Harris hawks optimization: framework and case studies,” Future Generation Computer Systems, vol. 111, pp. 175–198, 2020.View at: Publisher Site | Google Scholar
  90. J. Guo, H. Zheng, B. Li, and G.-Z. Fu, “Bayesian hierarchical model-based information fusion for degradation analysis considering non-competing relationship,” IEEE Access, vol. 7, pp. 175222–175227, 2019.View at: Publisher Site | Google Scholar
  91. J. Guo, H. Zheng, B. Li, and G.-Z. Fu, “A Bayesian approach for degradation analysis with individual differences,” IEEE Access, vol. 7, pp. 175033–175040, 2019.View at: Publisher Site | Google Scholar
  92. M. M. A. Malakoutian, Y. Malakoutian, P. Mostafapour, and S. Z. D. Abed, “Prediction for monthly rainfall of six meteorological regions and TRNC (case study: north Cyprus),” ENG Transactions, vol. 2, no. 2, 2021.View at: Google Scholar
  93. H. Arslan, M. Ranjbar, and Z. Mutlum, “Maximum sound transmission loss in multi-chamber reactive silencers: are two chambers enough?,,” ENG Transactions, vol. 2, no. 1, 2021.View at: Google Scholar
  94. N. Tonekaboni, M. Feizbahr, N. Tonekaboni, G.-J. Jiang, and H.-X. Chen, “Optimization of solar CCHP systems with collector enhanced by porous media and nanofluid,” Mathematical Problems in Engineering, vol. 2021, Article ID 9984840, 12 pages, 2021.View at: Publisher Site | Google Scholar
  95. Z. Niu, B. Zhang, J. Wang et al., “The research on 220GHz multicarrier high-speed communication system,” China Communications, vol. 17, no. 3, pp. 131–139, 2020.View at: Publisher Site | Google Scholar
  96. B. Zhang, Z. Niu, J. Wang et al., “Four‐hundred gigahertz broadband multi‐branch waveguide coupler,” IET Microwaves, Antennas & Propagation, vol. 14, no. 11, pp. 1175–1179, 2020.View at: Publisher Site | Google Scholar
  97. Z.-Q. Niu, L. Yang, B. Zhang et al., “A mechanical reliability study of 3dB waveguide hybrid couplers in the submillimeter and terahertz band,” Journal of Zhejiang University Science, vol. 1, no. 1, 1998.View at: Google Scholar
  98. B. Zhang, D. Ji, D. Fang, S. Liang, Y. Fan, and X. Chen, “A novel 220-GHz GaN diode on-chip tripler with high driven power,” IEEE Electron Device Letters, vol. 40, no. 5, pp. 780–783, 2019.View at: Publisher Site | Google Scholar
  99. M. Taleghani and A. Taleghani, “Identification and ranking of factors affecting the implementation of knowledge management engineering based on TOPSIS technique,” ENG Transactions, vol. 1, no. 1, 2020.View at: Google Scholar
PDF 원문보기
Strain rate magnitude at the free surface, illustrating Kelvin-Helmoltz (KH) shear instabilities.

On the reef scale hydrodynamics at Sodwana Bay, South Africa

Environmental Fluid Mechanics (2022)Cite this article

Abstract

The hydrodynamics of coral reefs strongly influences their biological functioning, impacting processes such as nutrient availability and uptake, recruitment success and bleaching. For example, coral reefs located in oligotrophic regions depend on upwelling for nutrient supply. Coral reefs at Sodwana Bay, located on the east coast of South Africa, are an example of high latitude marginal reefs. These reefs are subjected to complex hydrodynamic forcings due to the interaction between the strong Agulhas current and the highly variable topography of the region. In this study, we explore the reef scale hydrodynamics resulting from the bathymetry for two steady current scenarios at Two-Mile Reef (TMR) using a combination of field data and numerical simulations. The influence of tides or waves was not considered for this study as well as reef-scale roughness. Tilt current meters with onboard temperature sensors were deployed at selected locations within TMR. We used field observations to identify the dominant flow conditions on the reef for numerical simulations that focused on the hydrodynamics driven by mean currents. During the field campaign, southerly currents were the predominant flow feature with occasional flow reversals to the north. Northerly currents were associated with greater variability towards the southern end of TMR. Numerical simulations showed that Jesser Point was central to the development of flow features for both the northerly and southerly current scenarios. High current variability in the south of TMR during reverse currents is related to the formation of Kelvin-Helmholtz type shear instabilities along the outer edge of an eddy formed north of Jesser Point. Furthermore, downward vertical velocities were computed along the offshore shelf at TMR during southerly currents. Current reversals caused a change in vertical velocities to an upward direction due to the orientation of the bathymetry relative to flow directions.

Highlights

  • A predominant southerly current was measured at Two-Mile Reef with occasional reversals towards the north.
  • Field observations indicated that northerly currents are spatially varied along Two-Mile Reef.
  • Simulation of reverse currents show the formation of a separated flow due to interaction with Jesser Point with Kelvin–Helmholtz type shear instabilities along the seaward edge.

지금까지 Sodwana Bay에서 자세한 암초 규모 유체 역학을 모델링하려는 시도는 없었습니다. 이러한 모델의 결과는 규모가 있는 산호초 사이의 흐름이 산호초 건강에 어떤 영향을 미치는지 탐색하는 데 사용할 수 있습니다. 이 연구에서는 Sodwana Bay의 유체역학을 탐색하는 데 사용할 수 있는 LES 모델을 개발하기 위한 단계별 접근 방식을 구현합니다. 여기서 우리는 이 초기 단계에서 파도와 조수의 영향을 배제하면서 Agulhas 해류의 유체역학에 초점을 맞춥니다. 이 접근법은 흐름의 첫 번째 LES를 제시하고 Sodwana Bay의 산호초에서 혼합함으로써 향후 연구의 기초를 제공합니다.

This is a preview of subscription content, access via your institution.

References

  1. Anarde K, Myres H, Figlus J (2016) Tilt current meter field validation in the surf zone. In: AGU fall meeting abstracts, vol 2016, pp EP23A—-0950
  2. Blocken B (2018) LES over RANS in building simulation for outdoor and indoor applications: A foregone conclusion? Build Simul 11(5):821–870. https://doi.org/10.1007/s12273-018-0459-3Article Google Scholar 
  3. Booij N, Ris RC, Holthuijsen LH (1999) A third-generation wave model for coastal regions: 1. Model description and validation. J Geophys Res Ocean 104(C4):7649–7666. https://doi.org/10.1029/98JC02622Article Google Scholar 
  4. Bouffanais R (2010) Advances and challenges of applied large-eddy simulation. Comput Fluids 39:735–738. https://doi.org/10.1016/j.compfluid.2009.12.003Article Google Scholar 
  5. Celliers L, Schleyer MH (2002) Coral bleaching on high-latitude marginal reefs at Sodwana Bay, South Africa. Mar Pollut Bull 44:1380–1387Article Google Scholar 
  6. Celliers L, Schleyer MH (2008) Coral community structure and risk assessment of high-latitude reefs at Sodwana Bay, South Africa. Biodivers Conserv 17(13):3097–3117. https://doi.org/10.1007/s10531-007-9271-6Article Google Scholar 
  7. Chen SC (2018) Performance assessment of FLOW-3D and XFlow in the numerical modelling of fish-bone type fishway hydraulics https://doi.org/10.15142/T3HH1J
  8. Corbella S, Pringle J, Stretch DD (2015) Assimilation of ocean wave spectra and atmospheric circulation patterns to improve wave modelling. Coast Eng 100:1–10. https://doi.org/10.1016/j.coastaleng.2015.03.003Article Google Scholar 
  9. Davis KA, Pawlak G, Monismith SG (2021) Turbulence and coral reefs. Ann Rev Mar Sci. https://doi.org/10.1146/annurev-marine-042120-071823Article Google Scholar 
  10. Flow Science Inc (2018) FLOW-3D, Version 12.0 Users Manual. Santa Fe, NM, https://www.flow3d.com/
  11. Flow Science Inc (2019) FLOW-3D, Version 12.0 [Computer Software]. Santa Fe, NM, https://www.flow3d.com/
  12. Franco A, Moernaut J, Schneider-Muntau B, Strasser M, Gems B (2020) The 1958 Lituya Bay tsunami – pre-event bathymetry reconstruction and 3D numerical modelling utilising the computational fluid dynamics software Flow-3D. Nat Hazards Earth Syst Sci 20(8):2255–2279Article Google Scholar 
  13. Fringer OB, Gerritsen M, Street RL (2006) An unstructured-grid, finite-volume, nonhydrostatic, parallel coastal ocean simulator. Ocean Model 14(3):139–173Article Google Scholar 
  14. Fringer OB, Dawson CN, He R, Ralston DK, Zhang YJ (2019) The future of coastal and estuarine modeling: findings from a workshop. Ocean Model 143(September):101458. https://doi.org/10.1016/j.ocemod.2019.101458Article Google Scholar 
  15. Glassom D, Celliers L, Schleyer MH (2006) Coral recruitment patterns at Sodwana Bay, South Africa. Coral Reefs 25(3):485–492. https://doi.org/10.1007/s00338-006-0117-6Article Google Scholar 
  16. Gomes A, Pinho JLS, Valente T, do Carmo JS, Hegde VA (2020) Performance assessment of a semi-circular breakwater through CFD modelling. J Mar Sci Eng. https://doi.org/10.3390/jmse8030226Article Google Scholar 
  17. Green RH, Lowe RJ, Buckley ML (2018) Hydrodynamics of a tidally forced coral reef atoll. J Geophys Res Oceans 123(10):7084–7101. https://doi.org/10.1029/2018JC013946Article Google Scholar 
  18. Hansen AB, Carstensen S, Christensen DF, Aagaard T (2017) Performance of a tilt current meter in the surf zone. Coastal dynamics
  19. Hench JL, Rosman JH (2013) Observations of spatial flow patterns at the coral colony scale on a shallow reef flat. J Geophys Res Ocean 118(3):1142–1156. https://doi.org/10.1002/jgrc.20105Article Google Scholar 
  20. Hirt CW (1993) Volume-fraction techniques: powerful tools for wind engineering. J Wind Eng Ind Aerodyn 46–47:327–338. https://doi.org/10.1016/0167-6105(93)90298-3Article Google Scholar 
  21. Hirt CW, Sicilian JM (1985) A porosity technique for the definition of obstacles in rectangular cell meshes. In: Proceedings of 4th International Conference on Ship Hydrodynamics https://ci.nii.ac.jp/naid/10009570543/en/
  22. Hocker LO, Hruska MA (2004) Interleaving synchronous data and asynchronous data in a single data storage file
  23. Hossain MM, Staples AE (2020) Effects of coral colony morphology on turbulent flow dynamics. PLoS ONE 15(10):e0225676. https://doi.org/10.1371/journal.pone.0225676Article Google Scholar 
  24. Jacob B, Stanev EV (2021) Understanding the impact of bathymetric changes in the german bight on coastal hydrodynamics: one step toward realistic morphodynamic modeling. Front Mar Sci. https://doi.org/10.3389/fmars.2021.640214Article Google Scholar 
  25. Koehl MAR, Hadfield MG (2010) Hydrodynamics of larval settlement from a larva’s point of view. Integr Comp Biol 50(4):539–551. https://doi.org/10.1093/icb/icq101Article Google Scholar 
  26. Lim A, Wheeler AJ, Price DM, O’Reilly L, Harris K, Conti L (2020) Influence of benthic currents on cold-water coral habitats: a combined benthic monitoring and 3D photogrammetric investigation. Sci Rep 10(1):19433. https://doi.org/10.1038/s41598-020-76446-yArticle Google Scholar 
  27. Limer BD, Bloomberg J, Holstein DM (2020) The influence of eddies on coral larval retention in the flower garden banks. Front Mar Sci 7:372. https://doi.org/10.3389/fmars.2020.00372Article Google Scholar 
  28. Monismith SG (2007) Hydrodynamics of coral reefs. Annu Rev Fluid Mech 39(1):37–55. https://doi.org/10.1146/annurev.fluid.38.050304.092125Article Google Scholar 
  29. Morris T (2009) Physical oceanography of Sodwana Bay and its effect on larval transport and coral bleaching. PhD thesis, Cape Peninsula University of Technology
  30. Morris T, Lamont T, Roberts MJ (2013) Effects of deep-sea eddies on the northern KwaZulu-Natal shelf, South Africa. Afr J Mar Sci 35(3):343–350. https://doi.org/10.2989/1814232X.2013.827991Article Google Scholar 
  31. Perry C, Larcombe P (2003) Marginal and non-reef-building coral environments. Coral Reefs 22:427–432. https://doi.org/10.1007/s00338-003-0330-5Article Google Scholar 
  32. Pope SB (2001) Turbulent flows. Cambridge University Press, CambridgeGoogle Scholar 
  33. Porter SN (2009) Biogeography and potential factors regulating shallow subtidal reef communities in the Western Indian Ocean. PhD thesis, University of Cape Town
  34. Porter SN, Schleyer MH (2017) Long-term dynamics of a high-latitude coral reef community at Sodwana Bay, South Africa. Coral Reefs 36(2):369–382. https://doi.org/10.1007/s00338-016-1531-zArticle Google Scholar 
  35. Porter SN, Schleyer MH (2019) Environmental variation and how its spatial structure influences the cross-shelf distribution of high-latitude coral communities in South Africa. Diversity. https://doi.org/10.3390/d11040057Article Google Scholar 
  36. Ramsay PJ (1994) Marine geology of the Sodwana Bay shelf, southeast Africa. Mar Geol 120(3–4):225–247. https://doi.org/10.1016/0025-3227(94)90060-4Article Google Scholar 
  37. Ramsay PJ, Mason TR (1990) Development of a type zoning model for Zululand coral reefs, Sodwana Bay, South Africa. J Coastal Res 6(4):829–852Google Scholar 
  38. Reguero BG, Beck MW, Agostini VN, Kramer P, Hancock B (2018) Coral reefs for coastal protection: a new methodological approach and engineering case study in Grenada. J Environ Manag 210:146–161. https://doi.org/10.1016/j.jenvman.2018.01.024Article Google Scholar 
  39. Reidenbach M, Stocking J, Szczyrba L, Wendelken C (2021) Hydrodynamic interactions with coral topography and its impact on larval settlement. Coral Reefs 40:1–15. https://doi.org/10.1007/s00338-021-02069-yArticle Google Scholar 
  40. Reidenbach MA, Koseff JR, Koehl MAR (2009) Hydrodynamic forces on larvae affect their settlement on coral reefs in turbulent, wave-driven flow. Limnol Oceanogr 54(1):318–330. https://doi.org/10.4319/lo.2009.54.1.0318Article Google Scholar 
  41. Roberts H, Richardson J, Lagumbay R, Meselhe E, Ma Y (2013) Hydrodynamic and sediment transport modeling using FLOW-3D for siting and optimization of the LCA medium diversion at white ditch hydrodynamic and sediment transport modeling using FLOW-3D for siting and optimization of the LCA medium diversion at white D (December)
  42. Roberts MJ, Ribbink AJ, Morris T, Berg MAVD, Engelbrecht DC, Harding RT (2006) Oceanographic environment of the Sodwana Bay coelacanths (Latimeria chalumnae), South Africa: coelacanth research. South Afr J Sci 102(9):435–443Google Scholar 
  43. Rogers JS, Monismith SG, Feddersen F, Storlazzi CD (2013) Hydrodynamics of spur and groove formations on a coral reef. J Geophys Res Ocean 118(6):3059–3073. https://doi.org/10.1002/jgrc.20225Article Google Scholar 
  44. Rogers JS, Monismith SG, Koweek DA, Torres WI, Dunbar RB (2016) Thermodynamics and hydrodynamics in an atoll reef system and their influence on coral cover. Limnol Oceanogr 61(6):2191–2206. https://doi.org/10.1002/lno.10365Article Google Scholar 
  45. Schleyer MH, Celliers L (2003) Coral dominance at the reef-sediment interface in marginal coral communities at Sodwana Bay, South Africa. Mar Freshw Res 54(8):967–972. https://doi.org/10.1071/MF02049Article Google Scholar 
  46. Schleyer MH, Porter SN (2018) Chapter One – drivers of soft and stony coral community distribution on the high-latitude coral reefs of South Africa. advances in marine biology, vol 80, Academic Press, pp 1–55, https://doi.org/10.1016/bs.amb.2018.09.001
  47. Scott F, Antolinez JAA, McCall R, Storlazzi C, Reniers A, Pearson S (2020) Hydro-morphological characterization of coral reefs for wave runup prediction. Front Mar Sci 7:361. https://doi.org/10.3389/fmars.2020.00361Article Google Scholar 
  48. Sebens KP, Grace SP, Helmuth B, Maney EJ Jr, Miles JS (1998) Water flow and prey capture by three scleractinian corals, Madracis mirabilis, Montastrea cavernosa and Porites porites, in a field enclosure. Mar Biol 131(2):347–360Article Google Scholar 
  49. Smagorinsky J (1963) General circulation experiments with the primitive equations. Mon Weather Rev 91(3):99–164Article Google Scholar 
  50. Stocking J, Laforsch C, Sigl R, Reidenbach M (2018) The role of turbulent hydrodynamics and surface morphology on heat and mass transfer in corals. J R Soc Interface 15:20180448. https://doi.org/10.1098/rsif.2018.0448Article Google Scholar 
  51. Van Leer B (1977) Towards the ultimate conservative difference scheme III. Upstream-centered finite-difference schemes for ideal compressible flow. J Comput Phys 23(3):263–275. https://doi.org/10.1016/0021-9991(77)90094-8Article Google Scholar 
  52. Wells C, Pringle J, Stretch D (2021) Cold water temperature anomalies on the Sodwana reefs and their driving mechanisms. South Afr J Sci. https://doi.org/10.17159/sajs.2021/9304Article Google Scholar 
  53. Wyatt ASJ, Lowe RJ, Humphries S, Waite AM (2010) Particulate nutrient fluxes over a fringing coral reef: relevant scales of phytoplankton production and mechanisms of supply. Mar Ecol Prog Ser 405:113–130Article Google Scholar 
  54. Yao Y, He T, Deng Z, Chen L, Guo H (2019) Large eddy simulation modeling of tsunami-like solitary wave processes over fringing reefs. Nat Hazards Earth Syst Sci 19(6):1281–1295. https://doi.org/10.5194/nhess-19-1281-2019Article Google Scholar 
  55. Zhao Q, Tanimoto K (1998) Numerical simulation of breaking waves by large eddy simulation and vof method. Coastal Engineering Proceedings 1(26), 10.9753/icce.v26.%p, https://journals.tdl.org/icce/index.php/icce/article/view/5656

Text and image taken from Deoraj, et al. (2022), On the reef scale hydrodynamics at Sodwana Bay, South Africa. Preprint courtesy the authors.

Figure 2: 3D (left) and 2D (right) views of wave elevation using case C

CFD 접근법을 사용하여 파도에서 하이드로포일의 SEAKEEPING 성능

SYAFIQ ZIKRYAND FITRIADHY*
Faculty of Ocean Engineering Technology and Informatics, Universiti Malaysia Terengganu, 21030 Kuala
Terengganu, Terengganu, Malaysia
*
Corresponding author: naoe.afit@gmail.com http://doi.org/10.46754/umtjur.2021.07.017

Abstract

수중익선은 일반적으로 열악한 환경 조건으로 인해 승객의 편안함에 영향을 미칠 수 있는 높은 저항과 과도한 수직 운동(히브 및 피치)을 경험합니다. 따라서 복잡한 유체역학적 현상이 존재하기 때문에 파랑에서 수중익선의 내항성능을 규명할 필요가 있다.

이를 위해 수중익선 운동에 대한 CFD(Computational Fluid Dynamic) 해석을 제안한다. Froude Number 및 포일 받음각과 같은 여러 매개변수가 고려되었습니다.

그 결과 Froude Number의 후속 증가는 히브 및 피치 운동에 반비례한다는 것이 밝혀졌습니다. 본질적으로 이것은 높은 응답 진폭 연산자(RAO)의 형태로 제공되는 수중익선 항해 성능의 업그레이드로 이어졌습니다.

또한 포일 선수의 증가하는 각도는 히브 운동에 비례하는 반면, 포일 선미는 7.5o에서 낮은 히브 운동을 보였고, 그 다음으로 5o, 10o 순으로 나타났다. 피치모션의 경우 포일 보우의 증가는 5o에서 더 낮았고, 그 다음이 10o, 7.5o 순이었다. 포일 선미의 증가는 수중익선에 의한 피치 모션 경험에 비례했습니다.

일반적으로 이 CFD 시뮬레이션은 앞서 언급한 설계 매개변수와 관련하여 공해 상태에서 수중익선 설계의 운영 효율성을 보장하는 데 매우 유용합니다.

Keywords

CFD, hydrofoil, foil angle of attack, heave, pitch.

Figure 1: Overall mesh block being used in simulation
Figure 1: Overall mesh block being used in simulation
Figure 2: 3D (left) and 2D (right) views of wave elevation using case C
Figure 2: 3D (left) and 2D (right) views of wave elevation using case C

References

Djavareshkian, M. H., & Esmaeili, A. (2014). Heuristic optimization of submerged hydrofoil
using ANFIS–PSO. Ocean Engineering, 92, 55-63.
Fitriadhy, A., & Adam, N. A. (2017). Heave and pitch motions performance of a monotricat ship in
head-seas. International Journal of Automotive and Mechanical Engineering, 14, 4243-4258.
Islam, M., Jahra, F., & Hiscock, S. (2016). Data analysis methodologies for hydrodynamic
experiments in waves. Journal of Naval Architecture and Marine Engineering, 13(1),
1-15.
Koutsourakis, N., Bartzis, J. G., & Markatos, N. C. (2012). Evaluation of Reynolds stress, k-ε and
RNG k-ε turbulence models in street canyon flows using various experimental datasets.
Environmental fluid mechanics, 1-25.
Manual, F. D. U. (2011). Flow3D User Manual, v9. 4.2, Flow Science. Inc., Santa Fe, NM. Matveev, K., & Duncan, R. (2005). Development
of the tool for predicting hydrofoil system performance and simulating motion of hydrofoil-assisted boats. Paper presented at the High Speed and High Performance Ship and Craft Symposium, Everett/WA: ASNE, USA.
Seif, M., Mehdigholi, H., & Najafi, A. (2014). Experimental and numerical modeling of the
high speed planing vessel motion. Journal of Marine Engineering & Technology, 13(2), 62-
72.
Sun, X., Yao, C., Xiong, Y., & Ye, Q. (2017). Numerical and experimental study on
seakeeping performance of a swath vehicle in head waves. Applied Ocean Research, 68, 262-
275.
Vakilabadi, K. A., Khedmati, M. R., & Seif, M.S. (2014). Experimental study on heave and
pitch motion characteristics of a wave-piercing trimaran. Transactions of FAMENA, 38(3), 13-
26.
Yakhot, A., Rakib, S., & Flannery, W. (1994). LowReynolds number approximation for turbulent
eddy viscosity. Journal of scientific computing, 9(3), 283-292.
Yakhot, V., & Orszag, S. A. (1986). Renormalization group analysis of turbulence. I. Basic theory.
Journal of scientific computing, 1(1), 3-51.

PDF 원문보기
Fig. 2. Design of the grate inlet types studied: (a) R1, (b) R2, (c) R3, (d) R4, (e) R5, (f) R6, (g) R7 (source: based on geometries of Chaparro Andrade and Abaunza Tabares, 2021)

Three-dimensional Numerical Evaluation of Hydraulic Efficiency and Discharge Coefficient in Grate Inlets

쇠창살 격자 유입구의 수리효율 및 배출계수에 대한 3차원 수치적 평가

Melquisedec Cortés Zambrano*, Helmer Edgardo Monroy González,
Wilson Enrique Amaya Tequia
Faculty of Civil Engineering, Santo Tomas Tunja University. Address Av. Universitaria No. 45-202.
Tunja – Boyacá – Colombia

Abstract

홍수는 지반이동 및 이동의 원인 중 하나이며, 급속한 도시화 및 도시화로 인해 이전보다 빈번하게 발생할 수 있다. 도시 배수 시스템의 특성은 집수 요소가 결정적인 역할을 하는 범람의 발생 및 범위를 정의할 수 있습니다. 이 문서는 7가지 유형의 화격자 유입구의 수력 유입 효율 및 배출 계수에 대한 수치 조사를 제시합니다. FLOW-3D® 시뮬레이터는 Q = 24, 34.1, 44, 100, 200 및 300 L/s의 유속에서 풀 스케일로 격자를 테스트하는 데 사용되며 종방향 기울기가 1.0인 실험 프로토타입의 구성을 유지합니다. %, 1.5% 및 2.0% 및 고정 횡단 경사, 총 126개 모델. 그 결과를 바탕으로 종류별 및 종단경사 조건에 따른 수력유입구 효율곡선과 토출계수를 구성하였다. 결과는 다른 조사에서 제안된 경험적 공식으로 조정되어 프로토타입의 물리적 테스트 결과를 검증하는 역할을 합니다.

Floods are one of the causes of ground movement and displacement, and due to rapid urbanization and urban growth may occur more frequently than before. The characteristics of an urban drainage system can define the occurrence and extent of flooding, where catchment elements have a determining role. This document presents the numerical investigation of the hydraulic inlet efficiency and the discharge coefficient of seven types of grate inlets. The FLOW-3D® simulator is used to test the gratings at a full scale, under flow rates of Q = 24, 34.1, 44, 100, 200 and 300 L/s, preserving the configuration of the experimental prototype with longitudinal slopes of 1.0%, 1.5% and 2.0% and a fixed cross slope, for a total of 126 models. Based on the results, hydraulic inlet efficiency curves and discharge coefficients are constructed for each type and a longitudinal slope condition. The results are adjusted with empirical formulations proposed in other investigations, serving to verify the results of physical testing of prototypes.

Keywords

grate inlet, inlet efficiency, discharge coefficient, computational fluid dynamic, 3D modelling.

Fig. 1. Physical model of the experimental campaign (source: Chaparro Andrade and Abaunza Tabares, 2021)
Fig. 1. Physical model of the experimental campaign (source: Chaparro Andrade and Abaunza Tabares, 2021)
Fig. 2. Design of the grate inlet types studied: (a) R1, (b) R2, (c) R3, (d) R4, (e) R5, (f) R6, (g) R7 (source: based on geometries of Chaparro Andrade and Abaunza Tabares, 2021)
Fig. 2. Design of the grate inlet types studied: (a) R1, (b) R2, (c) R3, (d) R4, (e) R5, (f) R6, (g) R7 (source: based on geometries of Chaparro Andrade and Abaunza Tabares, 2021)
Fig. 4. Comparison between the results obtained during physical experimentation in prototype 7 and simulation results with FLOW-3D® (source: made with FlowSight® and photographic record by Chaparro Andrade and Abaunza Tabares, 2021)
Fig. 4. Comparison between the results obtained during physical experimentation in prototype 7 and simulation results with FLOW-3D® (source: made with FlowSight® and photographic record by Chaparro Andrade and Abaunza Tabares, 2021)
Fig. 6. Example of the results of flow depth and velocity vectors in the xy plane, for a stable flow condition in a grate inlet type and free surface configuration and flow regime, of some grating types (source: produced with FlowSight®)
Fig. 6. Example of the results of flow depth and velocity vectors in the xy plane, for a stable flow condition in a grate inlet type and free surface configuration and flow regime, of some grating types (source: produced with FlowSight®)

References

Alia Md., S., and Sabtu, N. (2020). Comparison of Different Methodologies for Determining the Efficiency of Gully Inlets. In F. M.
Nazri (Ed.), Proceedings of AICCE‘19: Transforming the Nation
for a Sustainable Tomorrow (Vol. 53, pp. 1275-1284). Springer
Nature Switzerland AG. https://doi.org/10.1007/978-3-030-
32816-0_99
Antunes do Carmo, J. S. (2020). Physical Modelling vs. Numerical Modelling: Complementarity and Learning. July. https://doi.
org/10.20944/preprints202007.0753.v1
Aragón-Hernández, J. L. (2013). Modelación numérica integrada de los procesos hidráulicos en el drenaje urbano [Universidad Politécnica de Cataluña]. In Doctoral Tesis. https://
upcommons.upc.edu/handle/2117/95059?locale-attribute=es
Argue, J. R., and Pezzaniti, D. (1996). How reliable are inlet
(hydraulic) models at representing stormwater flow? Science
of the Total Environment, 189-190, 355-359. https://doi.org/10.1016/0048-9697(96)05231-X
Banco Mundial, O. (2019). Agua: Panorama general. https://
www.bancomundial.org/es/topic/water/overview
Cárdenas-Quintero, M., Carvajal-Serna, L. F., and Marbello-Pérez, R. (2018). Evaluación numérica tridimensional de un
sumidero de reja de fondo (Three-Dimensional Numerical Assessment of Grate Inlet). SSRN Electronic Journal, November.
https://doi.org/10.2139/ssrn.3112980
Carvalho, R. F., Lopes, P., Leandro, J., and David, L. M. (2019).
Numerical Research of Flows into Gullies with Different Outlet Locations. Water, 11(2), 794. https://doi.org/10.3390/
w11040794
Chaparro Andrade, F. G., and Abaunza Tabares, K. V. (2021). Importancia de los sumideros, su funcionamiento y diseño en redes de alcantarillado caso de estudio sector nororiental Tunja.
Universidad Santo Tomás.
Cortés Zambrano, M., Amaya Tequia, W. E., and Gamba Fernández, D. S. (2020). Implementation of the hydraulic modelling of
urban drainage in the northeast sector, Tunja, Boyacá. Revista
Facultad de Ingeniería Universidad de Antioquia. https://doi.
org/10.17533/udea.redin.20200578
Cosco, C., Gómez, M., Russo, B., Tellez-Alvarez, J., Macchione, F., Costabile, P., and Costanzo, C. (2020). Discharge coefficients for specific grated inlets. Influence of the Froude
number. Urban Water Journal, 17(7), 656-668. https://doi.org/10.1080/1573062X.2020.1811881
Despotovic, J., Plavsic, J., Stefanovic, N., and Pavlovic, D. (2005).
Inefficiency of storm water inlets as a source of urban floods.
Water Science and Technology, 51(2), 139-145. https://doi.
org/10.2166/wst.2005.0041
Ellis, J. B., and Marsalek, J. (1996). Overview of urban drainage:
Environmental impacts and concerns, means of mitigation and
implementation policies. Journal of Hydraulic Research, 34(6),
723-732. https://doi.org/10.1080/00221689609498446
Fang, X., Jiang, S., and Alam, S. R. (2010). Numerical simulations of efficiency of curb-opening inlets. Journal of Hydraulic
Engineering, 136(1), 62-66. https://doi.org/10.1061/(ASCE)
HY.1943-7900.0000131
Faram, M. G., and Harwood, R. (2000). CFD for the Water Industry; The Role of CFD as a Tool for the Development of Wastewater Treatment Systems. Hydro International, 21-22.
Faram, M. G., and Harwood, R. (2002). Assessment of the
effectiveness of stormwater treatment chambers using
computational fluid dynamics. Global Solutions for Urban Drainage, 40644(September 2002), 1-14. https://doi.
org/10.1061/40644(2002)7
Flow Science, I. (2018). FLOW-3D® Version 12.0 Users Manual.
In FLOW-3D [Computer software]. https://www.flow3d.com
Flow Science, I. (2019). FLOW-3D® Version 12.0 [Computer software] (No. 12). https://www.flow3d.com
Ghanbari, R., and Heidarnejad, M. (2020). Experimental and numerical analysis of flow hydraulics in triangular and rectangular
piano key weirs. Water Science, 00(00), 1-7. https://doi.org/10.
1080/11104929.2020.1724649

Gómez, M., and Russo, B. (2005a). Comparative study of methodologies to determine inlet efficiency from test data. HEC-12
methodology vs UPC method. Water Resources Management,
Algarve, Portugal., 80(October 2014), 623-632. https://doi.
org/10.2495/WRM050621
Gómez, M., and Russo, B. (2005b). Comparative study among
different methodologies to determine storm sewer inlet efficiency from test data. 10th International Conference on Urban
Drainage, August, 21-26. https://www.researchgate.net/publication/255602448_Comparative_study_among_different_methodologies_to_determine_storm_sewer_inlet_efficiency_
from_test_data
Gómez, M., Recasens, J., Russo, B., and Martínez-Gomariz, E.
(2016). Assessment of inlet efficiency through a 3D simulation: Numerical and experimental comparison. Water Science
and Technology, 74(8), 1926-1935. https://doi.org/10.2166/
wst.2016.326
Gómez, M., and Russo, B. (2011). Methodology to estimate hydraulic efficiency of drain inlets. Proceedings of the Institution of
Civil Engineers: Water Management, 164(2), 81-90. https://doi.
org/10.1680/wama.900070
Gómez Valentin, M. (2007). Hidrología urbana. In Hidrología Urbana (pp. 135-147). Instituto Flumen.
Jakeman, A. J., Letcher, R. A., and Norton, J. P. (2006). Ten iterative steps in development and evaluation of environmental
models. Environmental Modelling and Software, 21, 602-614.
https://doi.org/10.1016/j.envsoft.2006.01.004
Jang, J. H., Hsieh, C. T., and Chang, T. H. (2019). The importance of gully flow modelling to urban flood simulation. Urban Water Journal, 16(5), 377-388. https://doi.org/10.1080/1573062X.2019.1669198
Kaushal, D. R., Thinglas, T., Tomita, Y., Kuchii, S., and Tsukamoto, H. (2012). Experimental investigation on optimization of
invert trap configuration for sewer solid management. Powder Technology, 215-216, 1-14. https://doi.org/10.1016/j.powtec.2011.08.029
Khazaee, I., and Mohammadiun, M. (2010). Effects of flow field
on open channel flow properties using numerical investigation
and experimental comparison. International Journal of Energy
and Environment, 1(6), 1083-1096. https://doi.org/10.1016/
S0031-9384(10)00122-8
Kleidorfer, M., Tscheikner-Gratl, F., Vonach, T., and Rauch, W.
(2018). What can we learn from a 500-year event? Experiences
from urban drainage in Austria. Water Science and Technology,
77(8), 2146-2154. https://doi.org/10.2166/wst.2018.138
Leitão, J. P., Simões, N. E., Pina, R. D., Ochoa-Rodriguez, S.,
Onof, C., and Sá Marques, A. (2017). Stochastic evaluation of
the impact of sewer inlets‘ hydraulic capacity on urban pluvial
flooding. Stochastic Environmental Research and Risk Assessment, 31(8), 1907-1922. https://doi.org/10.1007/s00477-016-
1283-x
Lopes, P., Leandro, J., Carvalho, R. F., Russo, B., and Gómez, M.
(2016). Assessment of the ability of a volume of fluid model to
reproduce the efficiency of a continuous transverse gully with
grate. Journal of Irrigation and Drainage Engineering, 142(10),
1-9. https://doi.org/10.1061/(ASCE)IR.1943-4774.0001058
Mohsin, M., and Kaushal, D. R. (2016). 3D CFD validation of invert trap efficiency for sewer solid management using VOF model. Water Science and Engineering, 9(2), 106-114. https://doi.
org/10.1016/j.wse.2016.06.006
Palla, A., Colli, M., Candela, A., Aronica, G. T., and Lanza, L.
G. (2018). Pluvial flooding in urban areas: the role of surface
drainage efficiency. Journal of Flood Risk Management, 11,
S663-S676. https://doi.org/10.1111/jfr3.12246
Russo, B. (2010). Design of surface drainage systems according
to hazard criteria related to flooding of urban areas [Universitat
Politècnica de Catalunya]. https://dialnet.unirioja.es/servlet/
tesis?codigo=258828
Sedano-Cruz, K., Carvajal-Escoar, Y., and Ávila Díaz, A. J. (2013).
ANÁLISIS DE ASPECTOS QUE INCREMENTAN EL RIESGO
DE INUNDACIONES EN COLOMBIA. Luna Azul, 37, 219-218.
https://www.redalyc.org/articulo.oa?id=321729206014
Spaliviero, F., May, R. W. P., Escarameia, M. (2000). Spacing of road gullies. Hydraulic performance of BS EN 124 gully gratings. HR Walingford, 44(0). https://doi.org/10.13140/
RG.2.1.1344.0889
Téllez-Álvarez, J., Gómez, M., and Russo, B. (2020). Quantification of energy loss in two grated inlets under pressure. Water
(Switzerland), 12(6). https://doi.org/10.3390/w12061601
Téllez Álvarez, J., Gómez, V., Russo, B., and Redondo, J. M.
(2003). Performance assessment of numerical modelling
for hydraulic efficiency of a grated inlet. 1, 6-8. https://doi.org/10.16309/j.cnki.issn.1007-1776.2003.03.004
Téllez Álvarez, J., Gómez Valentin, M., Paindelli, A., and Russo,
B. (2017). ACTIVIDAD EXPERIMENTAL DE I+D+i EN INGENIERÍA
HIDRÁULICA EN ESPAÑA. In L. J. Balairón Pérez and D. López
Gómez (Eds.), Seminario 2017, Comunicaciones de las líneas prioritarias (pp. 41-43). Universitat Politècnica de València.
https://doi.org/10.1017/CBO9781107415324.004
Téllez Álvarez, J., Gómez Valentin, M., and Russo, B. (2019).
Modelling of Surcharge Flow Through Grated Inlet. In P. Gourbesville and G. Caignaert (Eds.), Advances in Hydroinformati-

cs. Springer, Singapore. https://doi.org/10.1007/978-981-
4451-42-0
UNDRR, I., and CRED, I. (2018). Pérdidas económicas, pobreza y
Desastres 1998 – 2017 (Vol. 6, Issue 1). https://doi.org/10.12962/
j23373520.v6i1.22451
Vyzikas, T., and Greaves, D. (2018). Numerial Modelling.
In D. Greaves and G. Iglesias (Eds.), Wave and Tidal Energy (pp. 289-363). John Wiley and Sons Ltd. https://doi.
org/10.1002/9781119014492
Yakhot, V., and Orszag, S. A. (1986). Renormalization Group Analysis of Turbulence. I . Basic Theory. Journal of Scientific Computing, 1(1), 3-51. https://doi.org/10.1007/BF01061452
Yakhot, V., and Smith, L. M. (1992). The renormalization group,
the ɛ-expansion and derivation of turbulence models. Journal
of Scientific Computing, 7(l), 35-61. https://doi.org/10.1007/
BF01060210
Yazdanfar, Z., and Sharma, A. (2015). Urban drainage system
planning and design – Challenges with climate change and urbanization: A review. Water Science and Technology, 72(2), 165-https://doi.org/10.2166/wst.2015.207

원문 PDF 보기
Numerical Simulation of Local Scour Around Square Artificial Reef

사각 인공어초 주변 국지세굴 수치모의

Abstract

인공어초(Artificial Reef, ARs)는 연안 어업 자원을 복원하고 생태 환경을 복원하기 위한 핵심 인공 구조물 중 하나입니다. 그러나 많은 AR이 세굴로 인해 안정성과 기능을 상실한 것으로 밝혀졌다. 

AR의 기능적 효과를 보장하기 위해서는 서로 다른 흐름 조건에서 세굴로 인한 매장과 같은 AR의 불안정성을 연구하는 것이 매우 중요합니다.

FLOW-3D에 의해 확립된 3차원 수치 모델은 정상류에서 AR 주변의 국부 세굴 특성을 연구하는 데 사용됩니다. RNG k-ε 난류 모델로 닫힌 RANS 방정식은 하나의 AR 주변의 안정적인 유동장을 시뮬레이션하기 위해 설정됩니다. 

시뮬레이션 결과는 이전 실험 결과와 비교되었으며 좋은 일치를 보여줍니다. 그 다음에, 세굴 특성, 평형 세굴 깊이 및 최대 세굴 체적에 대한 AR의 개구수 및 입사각의 영향을 조사하였다. 결과는 개구수가 증가함에 따라 세굴 깊이와 세굴 부피가 감소함을 나타냅니다. 

또한 수치적 결과를 바탕으로 AR의 개구수가 평형 세굴깊이와 최대 세굴량에 미치는 영향에 대한 실증식을 제시하였다. 입사각의 변화는 AR의 가장 상류 코너에서 베드 전단 응력의 변화에 ​​영향을 미칠 것입니다. 베드 전단 응력이 클수록 세굴이 더 강해집니다. 

본 연구는 증강현실의 최적화된 공학적 설계 및 구축을 위한 이론적 지원과 실질적인 지침을 제공할 것이다. 결과는 개구수가 증가함에 따라 세굴 깊이와 세굴 부피가 감소함을 나타냅니다. 또한 수치적 결과를 바탕으로 AR의 개구수가 평형 세굴깊이와 최대 세굴량에 미치는 영향에 대한 실증식을 제시하였다. 

입사각의 변화는 AR의 가장 상류 코너에서 베드 전단 응력의 변화에 ​​영향을 미칠 것입니다. 베드 전단 응력이 클수록 세굴이 더 강해집니다. 본 연구는 증강현실의 최적화된 공학적 설계 및 구축을 위한 이론적 지원과 실질적인 지침을 제공할 것이다. 

결과는 개구수가 증가함에 따라 세굴 깊이와 세굴 부피가 감소함을 나타냅니다. 또한 수치적 결과를 바탕으로 AR의 개구수가 평형 세굴깊이와 최대 세굴량에 미치는 영향에 대한 실증식을 제시하였다. 입사각의 변화는 AR의 가장 상류 코너에서 베드 전단 응력의 변화에 ​​영향을 미칠 것입니다. 

베드 전단 응력이 클수록 세굴이 더 강해집니다. 본 연구는 증강현실의 최적화된 공학적 설계 및 구축을 위한 이론적 지원과 실질적인 지침을 제공할 것이다. 입사각의 변화는 AR의 가장 상류 코너에서 베드 전단 응력의 변화에 ​​영향을 미칠 것입니다. 

베드 전단 응력이 클수록 세굴이 더 강해집니다. 본 연구는 증강현실의 최적화된 공학적 설계 및 구축을 위한 이론적 지원과 실질적인 지침을 제공할 것이다. 입사각의 변화는 AR의 가장 상류 코너에서 베드 전단 응력의 변화에 ​​영향을 미칠 것입니다. 베드 전단 응력이 클수록 세굴이 더 강해집니다. 

본 연구는 증강현실의 최적화된 공학적 설계 및 구축을 위한 이론적 지원과 실질적인 지침을 제공할 것이다.

Numerical Simulation of Local Scour Around Square Artificial Reef
Numerical Simulation of Local Scour Around Square Artificial Reef

Artificial reefs (ARs) are one of the key man-made constructs to restore the offshore fishery resources and recover the ecological environment. However, it is found that many ARs lost their stability and function due to scour. In order to ensure the functional effect of ARs, it is of great significance to study the instability of ARs, like burying caused by scour in different flow conditions. The three-dimensional numerical model established by FLOW-3D is used to study the local scour characteristics around the AR in steady currents. The RANS equations, closed with the RNG k-ε turbulence model, are established for simulating a stable flow field around one AR. The simulation results are compared with previous experimental results and shows good agreement. Then, the effect of the opening number and the incident angles of ARs on the scour characteristics, the equilibrium scour depth and maximum scour volume are investigated. The results indicate that the scour depth and scour volume decrease with the increasing opening number. Moreover, the empirical equations of the effect of the opening number of the AR on the equilibrium scour depth and maximum scour volume are proposed based on the numerical results. The change of the incident angles will affect the change of bed shear stress at the most upstream corner of the AR. The greater bed shear stress results in a more intense scour. This study will provide theoretical support, and practical guidance for the optimized engineering design and construction of ARs.

Mingda Yang,Yanli Tang,Fenfang Zhao,Shiji Xu,Guangjie Fang

키워드:

인공 어초 , 수치 시뮬레이션 , 로컬 세굴 , 세굴 부피 , 개방 수 , 공격 각도,컴퓨터 시뮬레이션

DOI 원문보기
Figure 5. Schematic view of flap and support structure [32]

Design Optimization of Ocean Renewable Energy Converter Using a Combined Bi-level Metaheuristic Approach

결합된 Bi-level 메타휴리스틱 접근법을 사용한 해양 재생 에너지 변환기의 설계 최적화

Erfan Amini a1, Mahdieh Nasiri b1, Navid Salami Pargoo a, Zahra Mozhgani c, Danial Golbaz d, Mehrdad Baniesmaeil e, Meysam Majidi Nezhad f, Mehdi Neshat gj, Davide Astiaso Garcia h, Georgios Sylaios i

Abstract

In recent years, there has been an increasing interest in renewable energies in view of the fact that fossil fuels are the leading cause of catastrophic environmental consequences. Ocean wave energy is a renewable energy source that is particularly prevalent in coastal areas. Since many countries have tremendous potential to extract this type of energy, a number of researchers have sought to determine certain effective factors on wave converters’ performance, with a primary emphasis on ambient factors. In this study, we used metaheuristic optimization methods to investigate the effects of geometric factors on the performance of an Oscillating Surge Wave Energy Converter (OSWEC), in addition to the effects of hydrodynamic parameters. To do so, we used CATIA software to model different geometries which were then inserted into a numerical model developed in Flow3D software. A Ribed-surface design of the converter’s flap is also introduced in this study to maximize wave-converter interaction. Besides, a Bi-level Hill Climbing Multi-Verse Optimization (HCMVO) method was also developed for this application. The results showed that the converter performs better with greater wave heights, flap freeboard heights, and shorter wave periods. Additionally, the added ribs led to more wave-converter interaction and better performance, while the distance between the flap and flume bed negatively impacted the performance. Finally, tracking the changes in the five-dimensional objective function revealed the optimum value for each parameter in all scenarios. This is achieved by the newly developed optimization algorithm, which is much faster than other existing cutting-edge metaheuristic approaches.

Keywords

Wave Energy Converter

OSWEC

Hydrodynamic Effects

Geometric Design

Metaheuristic Optimization

Multi-Verse Optimizer

1. Introduction

The increase in energy demand, the limitations of fossil fuels, as well as environmental crises, such as air pollution and global warming, are the leading causes of calling more attention to harvesting renewable energy recently [1], [2], [3]. While still in its infancy, ocean wave energy has neither reached commercial maturity nor technological convergence. In recent decades, remarkable progress has been made in the marine energy domain, which is still in the early stage of development, to improve the technology performance level (TPL) [4], [5]and technology readiness level (TRL) of wave energy converters (WECs). This has been achieved using novel modeling techniques [6], [7], [8], [9], [10], [11], [12], [13], [14] to gain the following advantages [15]: (i) As a source of sustainable energy, it contributes to the mix of energy resources that leads to greater diversity and attractiveness for coastal cities and suppliers. [16] (ii) Since wave energy can be exploited offshore and does not require any land, in-land site selection would be less expensive and undesirable visual effects would be reduced. [17] (iii) When the best layout and location of offshore site are taken into account, permanent generation of energy will be feasible (as opposed to using solar energy, for example, which is time-dependent) [18].

In general, the energy conversion process can be divided into three stages in a WEC device, including primary, secondary, and tertiary stages [19], [20]. In the first stage of energy conversion, which is the subject of this study, the wave power is converted to mechanical power by wave-structure interaction (WSI) between ocean waves and structures. Moreover, the mechanical power is transferred into electricity in the second stage, in which mechanical structures are coupled with power take-off systems (PTO). At this stage, optimal control strategies are useful to tune the system dynamics to maximize power output [10], [13], [12]. Furthermore, the tertiary energy conversion stage revolves around transferring the non-standard AC power into direct current (DC) power for energy storage or standard AC power for grid integration [21], [22]. We discuss only the first stage regardless of the secondary and tertiary stages. While Page 1 of 16 WECs include several categories and technologies such as terminators, point absorbers, and attenuators [15], [23], we focus on oscillating surge wave energy converters (OSWECs) in this paper due to its high capacity for industrialization [24].

Over the past two decades, a number of studies have been conducted to understand how OSWECs’ structures and interactions between ocean waves and flaps affect converters performance. Henry et al.’s experiment on oscillating surge wave energy converters is considered as one of the most influential pieces of research [25], which demonstrated how the performance of oscillating surge wave energy converters (OSWECs) is affected by seven different factors, including wave period, wave power, flap’s relative density, water depth, free-board of the flap, the gap between the tubes, gap underneath the flap, and flap width. These parameters were assessed in their two models in order to estimate the absorbed energy from incoming waves [26], [27]. In addition, Folly et al. investigated the impact of water depth on the OSWECs performance analytically, numerically, and experimentally. According to this and further similar studies, the average annual incident wave power is significantly reduced by water depth. Based on the experimental results, both the surge wave force and the power capture of OSWECs increase in shallow water [28], [29]. Following this, Sarkar et al. found that under such circumstances, the device that is located near the coast performs much better than those in the open ocean [30]. On the other hand, other studies are showing that the size of the converter, including height and width, is relatively independent of the location (within similar depth) [31]. Subsequently, Schmitt et al. studied OSWECs numerically and experimentally. In fact, for the simulation of OSWEC, OpenFOAM was used to test the applicability of Reynolds-averaged Navier-Stokes (RANS) solvers. Then, the experimental model reproduced the numerical results with satisfying accuracy [32]. In another influential study, Wang et al. numerically assessed the effect of OSWEC’s width on their performance. According to their findings, as converter width increases, its efficiency decreases in short wave periods while increases in long wave periods [33]. One of the main challenges in the analysis of the OSWEC is the coupled effect of hydrodynamic and geometric variables. As a result, numerous cutting-edge geometry studies have been performed in recent years in order to find the optimal structure that maximizes power output and minimizes costs. Garcia et al. reviewed hull geometry optimization studies in the literature in [19]. In addition, Guo and Ringwood surveyed geometric optimization methods to improve the hydrodynamic performance of OSWECs at the primary stage [14]. Besides, they classified the hull geometry of OSWECs based on Figure 1. Subsequently, Whittaker et al. proposed a different design of OSWEC called Oyster2. There have been three examples of different geometries of oysters with different water depths. Based on its water depth, they determined the width and height of the converter. They also found that in the constant wave period the less the converter’s width, the less power captures the converter has [34]. Afterward, O’Boyle et al. investigated a type of OSWEC called Oyster 800. They compared the experimental and numerical models with the prototype model. In order to precisely reproduce the shape, mass distribution, and buoyancy properties of the prototype, a 40th-scale experimental model has been designed. Overall, all the models were fairly accurate according to the results [35].

Inclusive analysis of recent research avenues in the area of flap geometry has revealed that the interaction-based designs of such converters are emerging as a novel approach. An initiative workflow is designed in the current study to maximizing the wave energy extrication by such systems. To begin with, a sensitivity analysis plays its role of determining the best hydrodynamic values for installing the converter’s flap. Then, all flap dimensions and characteristics come into play to finalize the primary model. Following, interactive designs is proposed to increase the influence of incident waves on the body by adding ribs on both sides of the flap as a novel design. Finally, a new bi-level metaheuristic method is proposed to consider the effects of simultaneous changes in ribs properties and other design parameters. We hope this novel approach will be utilized to make big-scale projects less costly and justifiable. The efficiency of the method is also compared with four well known metaheuristic algorithms and out weight them for this application.

This paper is organized as follows. First, the research methodology is introduced by providing details about the numerical model implementation. To that end, we first introduced the primary model’s geometry and software details. That primary model is later verified with a benchmark study with regard to the flap angle of rotation and water surface elevation. Then, governing equations and performance criteria are presented. In the third part of the paper, we discuss the model’s sensitivity to lower and upper parts width (we proposed a two cross-sectional design for the flap), bottom elevation, and freeboard. Finally, the novel optimization approach is introduced in the final part and compared with four recent metaheuristic algorithms.

2. Numerical Methods

In this section, after a brief introduction of the numerical software, Flow3D, boundary conditions are defined. Afterwards, the numerical model implementation, along with primary model properties are described. Finally, governing equations, as part of numerical process, are discussed.

2.1. Model Setup

FLOW-3D is a powerful and comprehensive CFD simulation platform for studying fluid dynamics. This software has several modules to solve many complex engineering problems. In addition, modeling complex flows is simple and effective using FLOW-3D’s robust meshing capabilities [36]. Interaction between fluid and moving objects might alter the computational range. Dynamic meshes are used in our modeling to take these changes into account. At each time step, the computational node positions change in order to adapt the meshing area to the moving object. In addition, to choose mesh dimensions, some factors are taken into account such as computational accuracy, computational time, and stability. The final grid size is selected based on the detailed procedure provided in [37]. To that end, we performed grid-independence testing on a CFD model using three different mesh grid sizes of 0.01, 0.015, and 0.02 meters. The problem geometry and boundary conditions were defined the same, and simulations were run on all three grids under the same conditions. The predicted values of the relevant variable, such as velocity, was compared between the grids. The convergence behavior of the numerical solution was analyzed by calculating the relative L2 norm error between two consecutive grids. Based on the results obtained, it was found that the grid size of 0.02 meters showed the least error, indicating that it provided the most accurate and reliable solution among the three grids. Therefore, the grid size of 0.02 meters was selected as the optimal spatial resolution for the mesh grid.

In this work, the flume dimensions are 10 meters long, 0.1 meters wide, and 2.2 meters high, which are shown in figure2. In addition, input waves with linear characteristics have a height of 0.1 meters and a period of 1.4 seconds. Among the linear wave methods included in this software, RNGk-ε and k- ε are appropriate for turbulence model. The research of Lopez et al. shows that RNGk- ε provides the most accurate simulation of turbulence in OSWECs [21]. We use CATIA software to create the flap primary model and other innovative designs for this project. The flap measures 0.1 m x 0.65 m x 0.360 m in x, y and z directions, respectively. In Figure 3, the primary model of flap and its dimensions are shown. In this simulation, five boundaries have been defined, including 1. Inlet, 2. Outlet, 3. Converter flap, 4. Bed flume, and 5. Water surface, which are shown in figure 2. Besides, to avoid wave reflection in inlet and outlet zones, Flow3D is capable of defining some areas as damping zones, the length of which has to be one to one and a half times the wavelength. Therefore, in the model, this length is considered equal to 2 meters. Furthermore, there is no slip in all the boundaries. In other words, at every single time step, the fluid velocity is zero on the bed flume, while it is equal to the flap velocity on the converter flap. According to the wave theory defined in the software, at the inlet boundary, the water velocity is called from the wave speed to be fed into the model.

2.2. Verification

In the current study, we utilize the Schmitt experimental model as a benchmark for verification, which was developed at the Queen’s University of Belfast. The experiments were conducted on the flap of the converter, its rotation, and its interaction with the water surface. Thus, the details of the experiments are presented below based up on the experimental setup’s description [38]. In the experiment, the laboratory flume has a length of 20m and a width of 4.58m. Besides, in order to avoid incident wave reflection, a wave absorption source is devised at the end of the left flume. The flume bed, also, includes two parts with different slops. The flap position and dimensions of the flume can be seen in Figure4. In addition, a wave-maker with 6 paddles is installed at one end. At the opposite end, there is a beach with wire meshes. Additionally, there are 6 indicators to extract the water level elevation. In the flap model, there are three components: the fixed support structure, the hinge, and the flap. The flap measures 0.1m x 0.65m x 0.341m in x, y and z directions, respectively. In Figure5, the details are given [32]. The support structure consists of a 15 mm thick stainless steel base plate measuring 1m by 1.4m, which is screwed onto the bottom of the tank. The hinge is supported by three bearing blocks. There is a foam centerpiece on the front and back of the flap which is sandwiched between two PVC plates. Enabling changes of the flap, three metal fittings link the flap to the hinge. Moreover, in this experiment, the selected wave is generated based on sea wave data at scale 1:40. The wave height and the wave period are equal to 0.038 (m) and 2.0625 (s), respectively, which are tantamount to a wave with a period of 13 (s) and a height of 1.5 (m).

Two distinct graphs illustrate the numerical and experi-mental study results. Figure6 and Figure7 are denoting the angle of rotation of flap and surface elevation in computational and experimental models, respectively. The two figures roughly represent that the numerical and experimental models are a good match. However, for the purpose of verifying the match, we calculated the correlation coefficient (C) and root mean square error (RMSE). According to Figure6, correlation coefficient and RMSE are 0.998 and 0.003, respectively, and in Figure7 correlation coefficient and RMSE are respectively 0.999 and 0.001. Accordingly, there is a good match between the numerical and empirical models. It is worth mentioning that the small differences between the numerical and experimental outputs may be due to the error of the measuring devices and the calibration of the data collection devices.

Including continuity equation and momentum conserva- tion for incompressible fluid are given as [32], [39]:(1)

where P represents the pressure, g denotes gravitational acceleration, u represents fluid velocity, and Di is damping coefficient. Likewise, the model uses the same equation. to calculate the fluid velocity in other directions as well. Considering the turbulence, we use the two-equation model of RNGK- ε. These equations are:

(3)��t(��)+����(����)=����[�eff�������]+��-��and(4)���(��)+����(����)=����[�eff�������]+�1�∗����-��2��2�Where �2� and �1� are constants. In addition, �� and �� represent the turbulent Prandtl number of � and k, respectively.

�� also denote the production of turbulent kinetic energy of k under the effect of velocity gradient, which is calculated as follows:(5)��=�eff[�����+�����]�����(6)�eff=�+��(7)�eff=�+��where � is molecular viscosity,�� represents turbulence viscosity, k denotes kinetic energy, and ∊∊ is energy dissipation rate. The values of constant coefficients in the two-equation RNGK ∊-∊ model is as shown in the Table 1 [40].Table 2.

Table 1. Constant coefficients in RNGK-∊ model

Factors��0�1�2������
Quantity0.0124.381.421.681.391.390.084

Table 2. Flap properties

Joint height (m)0.476
Height of the center of mass (m)0.53
Weight (Kg)10.77

It is worth mentioning that the volume of fluid method is used to separate water and air phases in this software [41]. Below is the equation of this method [40].(8)����+����(���)=0where α and 1 − α are portion of water phase and air phase, respectively. As a weighting factor, each fluid phase portion is used to determine the mixture properties. Finally, using the following equations, we calculate the efficiency of converters [42], [34], [43]:(9)�=14|�|2�+�2+(�+�a)2(�n2-�2)2where �� represents natural frequency, I denotes the inertia of OSWEC, Ia is the added inertia, F is the complex wave force, and B denotes the hydrodynamic damping coefficient. Afterward, the capture factor of the converter is calculated by [44]:(10)��=�1/2��2����gw where �� represents the capture factor, which is the total efficiency of device per unit length of the wave crest at each time step [15], �� represent the dimensional amplitude of the incident wave, w is the flap’s width, and Cg is the group velocity of the incident wave, as below:(11)��=��0·121+2�0ℎsinh2�0ℎwhere �0 denotes the wave number, h is water depth, and H is the height of incident waves.

According to previous sections ∊,����-∊ modeling is used for all models simulated in this section. For this purpose, the empty boundary condition is used for flume walls. In order to preventing wave reflection at the inlet and outlet of the flume, the length of wave absorption is set to be at least one incident wavelength. In addition, the structured mesh is chosen, and the mesh dimensions are selected in two distinct directions. In each model, all grids have a length of 2 (cm) and a height of 1 (cm). Afterwards, as an input of the software for all of the models, we define the time step as 0.001 (s). Moreover, the run time of every simulation is 30 (s). As mentioned before, our primary model is Schmitt model, and the flap properties is given in table2. For all simulations, the flume measures 15 meters in length and 0.65 meters in width, and water depth is equal to 0.335 (m). The flap is also located 7 meters from the flume’s inlet.

Finally, in order to compare the results, the capture factor is calculated for each simulation and compared to the primary model. It is worth mentioning that capture factor refers to the ratio of absorbed wave energy to the input wave energy.

According to primary model simulation and due to the decreasing horizontal velocity with depth, the wave crest has the highest velocity. Considering the fact that the wave’s orbital velocity causes the flap to move, the contact between the upper edge of the flap and the incident wave can enhance its performance. Additionally, the numerical model shows that the dynamic pressure decreases as depth increases, and the hydrostatic pressure increases as depth increases.

To determine the OSWEC design, it is imperative to understand the correlation between the capture factor, wave period, and wave height. Therefore, as it is shown in Figure8, we plot the change in capture factor over the variations in wave period and wave height in 3D and 2D. In this diagram, the first axis features changes in wave period, the second axis displays changes in wave height, and the third axis depicts changes in capture factor. According to our wave properties in the numerical model, the wave period and wave height range from 2 to 14 seconds and 2 to 8 meters, respectively. This is due to the fact that the flap does not oscillate if the wave height is less than 2 (m), and it does not reverse if the wave height is more than 8 (m). In addition, with wave periods more than 14 (s), the wavelength would be so long that it would violate the deep-water conditions, and with wave periods less than 2 (s), the flap would not oscillate properly due to the shortness of wavelength. The results of simulation are shown in Figure 8. As it can be perceived from Figure 8, in a constant wave period, the capture factor is in direct proportion to the wave height. It is because of the fact that waves with more height have more energy to rotate the flap. Besides, in a constant wave height, the capture factor increases when the wave period increases, until a given wave period value. However, the capture factor falls after this point. These results are expected since the flap’s angular displacement is not high in lower wave periods, while the oscillating motion of that is not fast enough to activate the power take-off system in very high wave periods.

As is shown in Figure 9, we plot the change in capture factor over the variations in wave period (s) and water depth (m) in 3D. As it can be seen in this diagram, the first axis features changes in water depth (m), the second axis depicts the wave period (s), and the third axis displays OSWEC’s capture factor. The wave period ranges from 0 to 10 seconds based on our wave properties, which have been adopted from Schmitt’s model, while water depth ranges from 0 to 0.5 meters according to the flume and flap dimensions and laboratory limitations. According to Figure9, for any specific water depth, the capture factor increases in a varying rate when the wave period increases, until a given wave period value. However, the capture factor falls steadily after this point. In fact, the maximum capture factor occurs when the wave period is around 6 seconds. This trend is expected since, in a specific water depth, the flap cannot oscillate properly when the wavelength is too short. As the wave period increases, the flap can oscillate more easily, and consequently its capture factor increases. However, the capture factor drops in higher wave periods because the wavelength is too large to move the flap. Furthermore, in a constant wave period, by changing the water depth, the capture factor does not alter. In other words, the capture factor does not depend on the water depth when it is around its maximum value.

3. Sensitivity Analysis

Based on previous studies, in addition to the flap design, the location of the flap relative to the water surface (freeboard) and its elevation relative to the flume bed (flap bottom elevation) play a significant role in extracting energy from the wave energy converter. This study measures the sensitivity of the model to various parameters related to the flap design including upper part width of the flap, lower part width of the flap, the freeboard, and the flap bottom elevation. Moreover, as a novel idea, we propose that the flap widths differ in the lower and upper parts. In Figure10, as an example, a flap with an upper thickness of 100 (mm) and a lower thickness of 50 (mm) and a flap with an upper thickness of 50 (mm) and a lower thickness of 100 (mm) are shown. The influence of such discrepancy between the widths of the upper and lower parts on the interaction between the wave and the flap, or in other words on the capture factor, is evaluated. To do so, other parameters are remained constant, such as the freeboard, the distance between the flap and the flume bed, and the wave properties.

In Figure11, models are simulated with distinct upper and lower widths. As it is clear in this figure, the first axis depicts the lower part width of the flap, the second axis indicates the upper part width of the flap, and the colors represent the capture factor values. Additionally, in order to consider a sufficient range of change, the flap thickness varies from half to double the value of the primary model for each part.

According to this study, the greater the discrepancy in these two parts, the lower the capture factor. It is on account of the fact that when the lower part of the flap is thicker than the upper part, and this thickness difference in these two parts is extremely conspicuous, the inertia against the motion is significant at zero degrees of rotation. Consequently, it is difficult to move the flap, which results in a low capture factor. Similarly, when the upper part of the flap is thicker than the lower part, and this thickness difference in these two parts is exceedingly noticeable, the inertia is so great that the flap can not reverse at the maximum degree of rotation. As the results indicate, the discrepancy can enhance the performance of the converter if the difference between these two parts is around 20%. As it is depicted in the Figure11, the capture factor reaches its own maximum amount, when the lower part thickness is from 5 to 6 (cm), and the upper part thickness is between 6 and 7 (cm). Consequently, as a result of this discrepancy, less material will be used, and therefore there will be less cost.

As illustrated in Figure12, this study examines the effects of freeboard (level difference between the flap top and water surface) and the flap bottom elevation (the distance between the flume bed and flap bottom) on the converter performance. In this diagram, the first axis demonstrates the freeboard and the second axis on the left side displays the flap bottom elevation, while the colors indicate the capture factor. In addition, the feasible range of freeboard is between -15 to 15 (cm) due to the limitation of the numerical model, so that we can take the wave slamming and the overtopping into consideration. Additionally, based on the Schmitt model and its scaled model of 1:40 of the base height, the flap bottom should be at least 9 (cm) high. Since the effect of surface waves is distributed over the depth of the flume, it is imperative to maintain a reasonable flap height exposed to incoming waves. Thus, the maximum flap bottom elevation is limited to 19 (cm). As the Figure12 pictures, at constant negative values of the freeboard, the capture factor is in inverse proportion with the flap bottom elevation, although slightly.

Furthermore, at constant positive values of the freeboard, the capture factor fluctuates as the flap bottom elevation decreases while it maintains an overall increasing trend. This is on account of the fact that increasing the flap bottom elevation creates turbulence flow behind the flap, which encumbers its rotation, as well as the fact that the flap surface has less interaction with the incoming waves. Furthermore, while keeping the flap bottom elevation constant, the capture factor increases by raising the freeboard. This is due to the fact that there is overtopping with adverse impacts on the converter performance when the freeboard is negative and the flap is under the water surface. Besides, increasing the freeboard makes the wave slam more vigorously, which improves the converter performance.

Adding ribs to the flap surface, as shown in Figure13, is a novel idea that is investigated in the next section. To achieve an optimized design for the proposed geometry of the flap, we determine the optimal number and dimensions of ribs based on the flap properties as our decision variables in the optimization process. As an example, Figure13 illustrates a flap with 3 ribs on each side with specific dimensions.

Figure14 shows the flow velocity field around the flap jointed to the flume bed. During the oscillation of the flap, the pressure on the upper and lower surfaces of the flap changes dynamically due to the changing angle of attack and the resulting change in the direction of fluid flow. As the flap moves upwards, the pressure on the upper surface decreases, and the pressure on the lower surface increases. Conversely, as the flap moves downwards, the pressure on the upper surface increases, and the pressure on the lower surface decreases. This results in a cyclic pressure variation around the flap. Under certain conditions, the pressure field around the flap can exhibit significant variations in magnitude and direction, forming vortices and other flow structures. These flow structures can affect the performance of the OSWEC by altering the lift and drag forces acting on the flap.

4. Design Optimization

We consider optimizing the design parameters of the flap of converter using a nature-based swarm optimization method, that fall in the category of metaheuristic algorithms [45]. Accordingly, we choose four state-of-the-art algorithms to perform an optimization study. Then, based on their performances to achieve the highest capture factor, one of them will be chosen to be combined with the Hill Climb algorithm to carry out a local search. Therefore, in the remainder of this section, we discuss the search process of each algorithm and visualize their performance and convergence curve as they try to find the best values for decision variables.

4.1. Metaheuristic Approaches

As the first considered algorithm, the Gray Wolf Optimizer (GWO) algorithm simulates the natural leadership and hunting performance of gray wolves which tend to live in colonies. Hunters must obey the alpha wolf, the leader, who is responsible for hunting. Then, the beta wolf is at the second level of the gray wolf hierarchy. A subordinate of alpha wolf, beta stands under the command of the alpha. At the next level in this hierarchy, there are the delta wolves. They are subordinate to the alpha and beta wolves. This category of wolves includes scouts, sentinels, elders, hunters, and caretakers. In this ranking, omega wolves are at the bottom, having the lowest level and obeying all other wolves. They are also allowed to eat the prey just after others have eaten. Despite the fact that they seem less important than others, they are really central to the pack survival. Since, it has been shown that without omega wolves, the entire pack would experience some problems like fighting, violence, and frustration. In this simulation, there are three primary steps of hunting including searching, surrounding, and finally attacking the prey. Mathematically model of gray wolves’ hunting technique and their social hierarchy are applied in determined by optimization. this study. As mentioned before, gray wolves can locate their prey and surround them. The alpha wolf also leads the hunt. Assuming that the alpha, beta, and delta have more knowledge about prey locations, we can mathematically simulate gray wolf hunting behavior. Hence, in addition to saving the top three best solutions obtained so far, we compel the rest of the search agents (also the omegas) to adjust their positions based on the best search agent. Encircling behavior can be mathematically modeled by the following equations: [46].(12)�→=|�→·��→(�)-�→(�)|(13)�→(�+1)=��→(�)-�→·�→(14)�→=2.�2→(15)�→=2�→·�1→-�→Where �→indicates the position vector of gray wolf, ��→ defines the vector of prey, t indicates the current iteration, and �→and �→are coefficient vectors. To force the search agent to diverge from the prey, we use �→ with random values greater than 1 or less than -1. In addition, C→ contains random values in the range [0,2], and �→ 1 and �2→ are random vectors in [0,1]. The second considered technique is the Moth Flame Optimizer (MFO) algorithm. This method revolves around the moths’ navigation mechanism, which is realized by positioning themselves and maintaining a fixed angle relative to the moon while flying. This effective mechanism helps moths to fly in a straight path. However, when the source of light is artificial, maintaining an angle with the light leads to a spiral flying path towards the source that causes the moth’s death [47]. In MFO algorithm, moths and flames are both solutions. The moths are actual search agents that fly in hyper-dimensional space by changing their position vectors, and the flames are considered pins that moths drop when searching the search space [48]. The problem’s variables are the position of moths in the space. Each moth searches around a flame and updates it in case of finding a better solution. The fitness value is the return value of each moth’s fitness (objective) function. The position vector of each moth is passed to the fitness function, and the output of the fitness function is assigned to the corresponding moth. With this mechanism, a moth never loses its best solution [49]. Some attributes of this algorithm are as follows:

  • •It takes different values to converge moth in any point around the flame.
  • •Distance to the flame is lowered to be eventually minimized.
  • •When the position gets closer to the flame, the updated positions around the flame become more frequent.

As another method, the Multi-Verse Optimizer is based on a multiverse theory which proposes there are other universes besides the one in which we all live. According to this theory, there are more than one big bang in the universe, and each big bang leads to the birth of a new universe [50]. Multi-Verse Optimizer (MVO) is mainly inspired by three phenomena in cosmology: white holes, black holes, and wormholes. A white hole has never been observed in our universe, but physicists believe the big bang could be considered a white hole [51]. Black holes, which behave completely in contrast to white holes, attract everything including light beams with their extremely high gravitational force [52]. In the multiverse theory, wormholes are time and space tunnels that allow objects to move instantly between any two corners of a universe (or even simultaneously from one universe to another) [53]. Based on these three concepts, mathematical models are designed to perform exploration, exploitation, and local search, respectively. The concept of white and black holes is implied as an exploration phase, while the concept of wormholes is considered as an exploitation phase by MVO. Additionally, each solution is analogous to a universe, and each variable in the solution represents an object in that universe. Furthermore, each solution is assigned an inflation rate, and the time is used instead of iterations. Following are the universe rules in MVO:

  • •The possibility of having white hole increases with the inflation rate.
  • •The possibility of having black hole decreases with the inflation rate.
  • •Objects tend to pass through black holes more frequently in universes with lower inflation rates.
  • •Regardless of inflation rate, wormholes may cause objects in universes to move randomly towards the best universe. [54]

Modeling the white/black hole tunnels and exchanging objects of universes mathematically was accomplished by using the roulette wheel mechanism. With every iteration, the universes are sorted according to their inflation rates, then, based on the roulette wheel, the one with the white hole is selected as the local extremum solution. This is accomplished through the following steps:

Assume that

(16)���=����1<��(��)����1≥��(��)

Where ��� represents the jth parameter of the ith universe, Ui indicates the ith universe, NI(Ui) is normalized inflation rate of the ith universe, r1 is a random number in [0,1], and j xk shows the jth parameter of the kth universe selected by a roulette wheel selection mechanism [54]. It is assumed that wormhole tunnels always exist between a universe and the best universe formed so far. This mechanism is as follows:(17)���=if�2<���:��+���×((���-���)×�4+���)�3<0.5��-���×((���-���)×�4+���)�3≥0.5����:���where Xj indicates the jth parameter of the best universe formed so far, TDR and WEP are coefficients, where Xj indicates the jth parameter of the best universelbjshows the lower bound of the jth variable, ubj is the upper bound of the jth variable, and r2, r3, and r4 are random numbers in [1], [54].

Finally, one of the newest optimization algorithms is WOA. The WOA algorithm simulates the movement of prey and the whale’s discipline when looking for their prey. Among several species, Humpback whales have a specific method of hunting [55]. Humpback whales can recognize the location of prey and encircle it before hunting. The optimal design position in the search space is not known a priori, and the WOA algorithm assumes that the best candidate solution is either the target prey or close to the optimum. This foraging behavior is called the bubble-net feeding method. Two maneuvers are associated with bubbles: upward spirals and double loops. A unique behavior exhibited only by humpback whales is bubble-net feeding. In fact, The WOA algorithm starts with a set of random solutions. At each iteration, search agents update their positions for either a randomly chosen search agent or the best solution obtained so far [56], [55]. When the best search agent is determined, the other search agents will attempt to update their positions toward that agent. It is important to note that humpback whales swim around their prey simultaneously in a circular, shrinking circle and along a spiral-shaped path. By using a mathematical model, the spiral bubble-net feeding maneuver is optimized. The following equation represents this behavior:(18)�→(�+1)=�′→·�bl·cos(2��)+�∗→(�)

Where:(19)�′→=|�∗→(�)-�→(�)|

X→(t+ 1) indicates the distance of the it h whale to the prey (best solution obtained so far),� is a constant for defining the shape of the logarithmic spiral, l is a random number in [−1, 1], and dot (.) is an element-by-element multiplication [55].

Comparing the four above-mentioned methods, simulations are run with 10 search agents for 400 iterations. In Figure 15, there are 20 plots the optimal values of different parameters in optimization algorithms. The five parameters of this study are freeboard, bottom elevations, number of ribs on the converter, rib thickness, and rib Height. The optimal value for each was found by optimization algorithms, naming WOA, MVO, MFO, and GWO. By looking through the first row, the freeboard parameter converges to its maximum possible value in the optimization process of GWO after 300 iterations. Similarly, MFO finds the same result as GWO. In contrast, the freeboard converges to its minimum possible value in MVO optimizing process, which indicates positioning the converter under the water. Furthermore, WOA found the optimal value of freeboard as around 0.02 after almost 200 iterations. In the second row, the bottom elevation is found at almost 0.11 (m) in all algorithms; however, the curves follow different trends in each algorithm. The third row shows the number of ribs, where results immediately reveal that it should be over 4. All algorithms coincide at 5 ribs as the optimal number in this process. The fourth row displays the trends of algorithms to find optimal rib thickness. MFO finds the optimal value early and sets it to around 0.022, while others find the same value in higher iterations. Finally, regarding the rib height, MVO, MFO, and GWO state that the optimal value is 0.06 meters, but WOA did not find a higher value than 0.039.

4.2. HCMVO Bi-level Approach

Despite several strong search characteristics of MVO and its high performance in various optimization problems, it suffers from a few deficiencies in local and global search mechanisms. For instance, it is trapped in the local optimum when wormholes stochastically generate many solutions near the best universe achieved throughout iterations, especially in solving complex multimodal problems with high dimensions [57]. Furthermore, MVO needs to be modified by an escaping strategy from the local optima to enhance the global search abilities. To address these shortages, we propose a fast and effective meta-algorithm (HCMVO) to combine MVO with a Random-restart hill-climbing local search. This meta-algorithm uses MVO on the upper level to develop global tracking and provide a range of feasible and proper solutions. The hill-climbing algorithm is designed to develop a comprehensive neighborhood search around the best-found solution proposed by the upper-level (MVO) when MVO is faced with a stagnation issue or falling into a local optimum. The performance threshold is formulated as follows.(20)Δ����THD=∑�=1�����TH��-����TH��-1�where BestTHDis the best-found solution per generation, andM is related to the domain of iterations to compute the average performance of MVO. If the proposed best solution by the local search is better than the initial one, the global best of MVO will be updated. HCMVO iteratively runs hill climbing when the performance of MVO goes down, each time with an initial condition to prepare for escaping such undesirable situations. In order to get a better balance between exploration and exploitation, the search step size linearly decreases as follows:(21)��=��-����Ma�iter��+1where iter and Maxiter are the current iteration and maximum number of evaluation, respectively. �� stands for the step size of the neighborhood search. Meanwhile, this strategy can improve the convergence rate of MVO compared with other algorithms.

Algorithm 1 shows the technical details of the proposed optimization method (HCMVO). The initial solution includes freeboard (�), bottom elevation (�), number of ribs (Nr), rib thickness (�), and rib height(�).

5. Conclusion

The high trend of diminishing worldwide energy resources has entailed a great crisis upon vulnerable societies. To withstand this effect, developing renewable energy technologies can open doors to a more reliable means, among which the wave energy converters will help the coastal residents and infrastructure. This paper set out to determine the optimized design for such devices that leads to the highest possible power output. The main goal of this research was to demonstrate the best design for an oscillating surge wave energy converter using a novel metaheuristic optimization algorithm. In this regard, the methodology was devised such that it argued the effects of influential parameters, including wave characteristics, WEC design, and interaction criteria.

To begin with, a numerical model was developed in Flow 3D software to simulate the response of the flap of a wave energy converter to incoming waves, followed by a validation study based upon a well-reputed experimental study to verify the accuracy of the model. Secondly, the hydrodynamics of the flap was investigated by incorporating the turbulence. The effect of depth, wave height, and wave period are also investigated in this part. The influence of two novel ideas on increasing the wave-converter interaction was then assessed: i) designing a flap with different widths in the upper and lower part, and ii) adding ribs on the surface of the flap. Finally, four trending single-objective metaheuristic optimization methods

Empty CellAlgorithm 1: Hill Climb Multiverse Optimization
01:procedure HCMVO
02:�=30,�=5▹���������������������������������
03:�=〈F1,B1,N,R,H1〉,…〈FN,B2,N,R,HN〉⇒lb1N⩽�⩽ubN
04:Initialize parameters�ER,�DR,�EP,Best�,���ite��▹Wormhole existence probability (WEP)
05:��=����(��)
06:��=Normalize the inflation rate��
07:for iter in[1,⋯,���iter]do
08:for�in[1,⋯,�]do
09:Update�EP,�DR,Black����Index=�
10:for���[1,⋯,�]��
11:�1=����()
12:if�1≤��(��)then
13:White HoleIndex=Roulette�heelSelection(-��)
14:�(Black HoleIndex,�)=��(White HoleIndex,�)
15:end if
16:�2=����([0,�])
17:if�2≤�EPthen
18:�3=����(),�4=����()
19:if�3<0.5then
20:�1=((��(�)-��(�))�4+��(�))
21:�(�,�)=Best�(�)+�DR�
22:else
23:�(�,�)=Best�(�)-�DR�
24:end if
25:end if
26:end for
27:end for
28:�HD=����([�1,�2,⋯,�Np])
29:Bes�TH�itr=����HD
30:ΔBestTHD=∑�=1�BestTII��-BestTII��-1�
31:ifΔBestTHD<��then▹Perform hill climbing local search
32:BestTHD=����-�lim��������THD
33:end if
34:end for
35:return�,BestTHD▹Final configuration
36:end procedure

The implementation details of the hill-climbing algorithm applied in HCMPA can be seen in Algorithm 2. One of the critical parameters isg, which denotes the resolution of the neighborhood search around the proposed global best by MVO. If we set a small step size for hill-climbing, the convergence speed will be decreased. On the other hand, a large step size reinforces the exploration ability. Still, it may reduce the exploitation ability and in return increase the act of jumping from a global optimum or surfaces with high-potential solutions. Per each decision variable, the neighborhood search evaluates two different direct searches, incremental or decremental. After assessing the generated solutions, the best candidate will be selected to iterate the search algorithm. It is noted that the hill-climbing algorithm should not be applied in the initial iteration of the optimization process due to the immense tendency for converging to local optima. Meanwhile, for optimizing largescale problems, hill-climbing is not an appropriate selection. In order to improve understanding of the proposed hybrid optimization algorithm’s steps, the flowchart of HCMVO is designed and can be seen in Figure 16.

Figure 17 shows the observed capture factor (which is the absorbed energy with respect to the available energy) by each optimization algorithm from iterations 1 to 400. The algorithms use ten search agents in their modified codes to find the optimal solutions. While GWO and MFO remain roughly constant after iterations 54 and 40, the other three algorithms keep improving the capture factor. In this case, HCMVO and MVO worked very well in the optimizing process with a capture factor obtained by the former as 0.594 and by the latter as 0.593. MFO almost found its highest value before the iteration 50, which means the exploration part of the algorithm works out well. Similarly, HCMVO does the same. However, it keeps finding the better solution during the optimization process until the last iteration, indicating the strong exploitation part of the algorithm. GWO reveals a weakness in exploration and exploitation because not only does it evoke the least capture factor value, but also the curve remains almost unchanged throughout 350 iterations.

Figure 18 illustrates complex interactions between the five optimization parameters and the capture factor for HCMVO (a), MPA (b), and MFO (c) algorithms. The first interesting observation is that there is a high level of nonlinear relationships among the setting parameters that can make a multi-modal search space. The dark blue lines represent the best-found configuration throughout the optimisation process. Based on both HCMVO (a) and MVO (b), we can infer that the dark blue lines concentrate in a specific range, showing the high convergence ability of both HCMVO and MVO. However, MFO (c) could not find the exact optimal range of the decision variables, and the best-found solutions per generation distribute mostly all around the search space.

Empty CellAlgorithm 1: Hill Climb Multiverse Optimization
01:procedure HCMVO
02:Initialization
03:Initialize the constraints��1�,��1�
04:�1�=Mi�1�+���1�/�▹Compute the step size,�is search resolution
05:So�1=〈�,�,�,�,�〉▹���������������
06:�������1=����So�1▹���������ℎ���������
07:Main loop
08:for iter≤���ita=do
09:���=���±��
10:while�≤���(Sol1)do
11:���=���+�,▹����ℎ���ℎ��������ℎ
12:fitness��iter=�������
13:t = t+1
14:end while
15:〈�����,������max〉=����������
16:���itev=���Inde�max▹�������ℎ�������������������������������ℎ�������
17:��=��-����Max��+1▹�����������������
18:end for
19:return���iter,����
20:end procedure

were utilized to illuminate the optimum values of the design parameters, and the best method was chosen to develop a new algorithm that performs both local and global search methods.

The correlation between hydrodynamic parameters and the capture factor of the converter was supported by the results. For any given water depth, the capture factor increases as the wave period increases, until a certain wave period value (6 seconds) is reached, after which the capture factor gradually decreases. It is expected since the flap cannot oscillate effectively when the wavelength is too short for a certain water depth. Conversely, when the wavelength is too long, the capture factor decreases. Furthermore, under a constant wave period, increasing the water depth does not affect the capture factor. Regarding the sensitivity analysis, the study found that increasing the flap bottom elevation causes turbulence flow behind the flap and limitation of rotation, which leads to less interaction with the incoming waves. Furthermore, while keeping the flap bottom elevation constant, increasing the freeboard improves the capture factor. Overtopping happens when the freeboard is negative and the flap is below the water surface, which has a detrimental influence on converter performance. Furthermore, raising the freeboard causes the wave impact to become more violent, which increases converter performance.

In the last part, we discussed the search process of each algorithm and visualized their performance and convergence curves as they try to find the best values for decision variables. Among the four selected metaheuristic algorithms, the Multi-verse Optimizer proved to be the most effective in achieving the best answer in terms of the WEC capture factor. However, the MVO needed modifications regarding its escape approach from the local optima in order to improve its global search capabilities. To overcome these constraints, we presented a fast and efficient meta-algorithm (HCMVO) that combines MVO with a Random-restart hill-climbing local search. On a higher level, this meta-algorithm employed MVO to generate global tracking and present a range of possible and appropriate solutions. Taken together, the results demonstrated that there is a significant degree of nonlinearity among the setup parameters that might result in a multimodal search space. Since MVO was faced with a stagnation issue or fell into a local optimum, we constructed a complete neighborhood search around the best-found solution offered by the upper level. In sum, the newly-developed algorithm proved to be highly effective for the problem compared to other similar optimization methods. The strength of the current findings may encourage future investigation on design optimization of wave energy converters using developed geometry as well as the novel approach.

CRediT authorship contribution statement

Erfan Amini: Conceptualization, Methodology, Validation, Data curation, Writing – original draft, Writing – review & editing, Visualization. Mahdieh Nasiri: Conceptualization, Methodology, Validation, Data curation, Writing – original draft, Writing – review & editing, Visualization. Navid Salami Pargoo: Writing – original draft, Writing – review & editing. Zahra Mozhgani: Conceptualization, Methodology. Danial Golbaz: Writing – original draft. Mehrdad Baniesmaeil: Writing – original draft. Meysam Majidi Nezhad: . Mehdi Neshat: Supervision, Conceptualization, Writing – original draft, Writing – review & editing, Visualization. Davide Astiaso Garcia: Supervision. Georgios Sylaios: Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This research has been carried out within ILIAD (Inte-grated Digital Framework for Comprehensive Maritime Data and Information Services) project that received funding from the European Union’s H2020 programme.

Data availability

Data will be made available on request.

References

원문보기
PDF View
Figure 15. Velocity distribution of impinging jet on a wall under different Reynolds numbers.

Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment

by Hongbo Mi 1,2, Chuan Wang 1,3, Xuanwen Jia 3,*, Bo Hu 2, Hongliang Wang 4, Hui Wang 3 and Yong Zhu 5

1College of Mechatronics Engineering, Hainan Vocational University of Science and Technology, Haikou 571126, China

2Department of Energy and Power Engineering, Tsinghua University, Beijing 100084, China

3College of Hydraulic Science and Engineering, Yangzhou University, Yangzhou 225009, China

4School of Aerospace and Mechanical Engineering/Flight College, Changzhou Institute of Technology, Changzhou 213032, China

5National Research Center of Pumps, Jiangsu University, Zhenjiang 212013, China

*Author to whom correspondence should be addressed.Sustainability2023, 15(6), 5159; https://doi.org/10.3390/su15065159

Received: 30 January 2023 / Revised: 4 March 2023 / Accepted: 10 March 2023 / Published: 14 March 2023(This article belongs to the Special Issue Advanced Technologies of Renewable Energy and Water Management for Sustainable Environment

Abstract

Due to their high efficiency, low heat loss and associated sustainability advantages, impinging jets have been used extensively in marine engineering, geotechnical engineering and other engineering practices. In this paper, the flow structure and impact characteristics of impinging jets with different Reynolds numbers and impact distances are systematically studied by Flow-3D based on PIV experiments. In the study, the relevant state parameters of the jets are dimensionlessly treated, obtaining not only the linear relationship between the length of the potential nucleation zone and the impinging distance, but also the linear relationship between the axial velocity and the axial distance in the impinging zone. In addition, after the jet impinges on the flat plate, the vortex action range caused by the wall-attached flow of the jet gradually decreases inward with the increase of the impinging distance. By examining the effect of Reynolds number Re on the hydraulic characteristics of the submerged impact jet, it can be found that the structure of the continuous submerged impact jet is relatively independent of the Reynolds number. At the same time, the final simulation results demonstrate the applicability of the linear relationship between the length of the potential core region and the impact distance. This study provides methodological guidance and theoretical support for relevant engineering practice and subsequent research on impinging jets, which has strong theoretical and practical significance.

Keywords: 

PIV; Flow-3D; impinging jet; hydraulic characteristics; impinging distance

원문 보기
Sustainability 15 05159 g001 550

Figure 1. Geometric model.

Sustainability 15 05159 g002 550

Figure 2. Model grid schematic.

Sustainability 15 05159 g003 550

Figure 3. (a) Schematic diagram of the experimental setup; (b) PIV images of vertical impinging jets with velocity fields.

Sustainability 15 05159 g004 550

Figure 4. (a) Velocity distribution verification at the outlet of the jet pipe; (b) Distribution of flow angle in the mid-axis of the jet [39].

Sustainability 15 05159 g005 550

Figure 5. Along-range distribution of the dimensionless axial velocity of the jet at different impact distances.Figure 6 shows the variation of H

Sustainability 15 05159 g006 550

Figure 6. Relationship between the distribution of potential core region and the impact height H/D.

Sustainability 15 05159 g007 550

Figure 7. The relationship between the potential core length 

Sustainability 15 05159 g008 550

Figure 8. Along-range distribution of the flow angle φ of the jet at different impact distances.

Sustainability 15 05159 g009 550

Figure 9. Velocity distribution along the axis of the jet at different impinging regions.

Sustainability 15 05159 g010 550

Figure 10. The absolute value distribution of slope under different impact distances.

Sustainability 15 05159 g011a 550
Sustainability 15 05159 g011b 550

Figure 11. Velocity distribution of impinging jet on wall under different impinging distances.

Sustainability 15 05159 g012 550

Figure 12. Along-range distribution of the dimensionless axial velocity of the jet at different Reynolds numbers.

Sustainability 15 05159 g013 550

Figure 13. Along-range distribution of the flow angle φ of the jet at different Reynolds numbers.

Sustainability 15 05159 g014 550

Figure 14. Velocity distribution along the jet axis at different Reynolds numbers.

Sustainability 15 05159 g015 550

Figure 15. Velocity distribution of impinging jet on a wall under different Reynolds numbers.

References

  1. Zhang, J.; Li, Y.; Zhang, Y.; Yang, F.; Liang, C.; Tan, S. Using a high-pressure water jet-assisted tunnel boring machine to break rock. Adv. Mech. Eng. 2020, 12, 1687814020962290. [Google Scholar] [CrossRef]
  2. Shi, X.; Zhang, G.; Xu, G.; Ma, Y.; Wu, X. Inactivating Microorganism on Medical Instrument Using Plasma Jet. High Volt. Eng. 2009, 35, 632–635. [Google Scholar]
  3. Gao, Y.; Han, P.; Wang, F.; Cao, J.; Zhang, S. Study on the Characteristics of Water Jet Breaking Coal Rock in a Drilling Hole. Sustainability 2022, 14, 8258. [Google Scholar] [CrossRef]
  4. Xu, W.; Wang, C.; Zhang, L.; Ge, J.; Zhang, D.; Gao, Z. Numerical study of continuous jet impinging on a rotating wall based on Wray—Agarwal turbulence model. J. Braz. Soc. Mech. Sci. Eng. 2022, 44, 433. [Google Scholar] [CrossRef]
  5. Hu, B.; Wang, C.; Wang, H.; Yu, Q.; Liu, J.; Zhu, Y.; Ge, J.; Chen, X.; Yang, Y. Numerical Simulation Study of the Horizontal Submerged Jet Based on the Wray—Agarwal Turbulence Model. J. Mar. Sci. Eng. 2022, 10, 1217. [Google Scholar] [CrossRef]
  6. Dahiya, A.K.; Bhuyan, B.K.; Kumar, S. Perspective study of abrasive water jet machining of composites—A review. J. Mech. Sci. Technol. 2022, 36, 213–224. [Google Scholar] [CrossRef]
  7. Abushanab, W.S.; Moustafa, E.B.; Harish, M.; Shanmugan, S.; Elsheikh, A.H. Experimental investigation on surface characteristics of Ti6Al4V alloy during abrasive water jet machining process. Alex. Eng. J. 2022, 61, 7529–7539. [Google Scholar] [CrossRef]
  8. Hu, B.; Wang, H.; Liu, J.; Zhu, Y.; Wang, C.; Ge, J.; Zhang, Y. A numerical study of a submerged water jet impinging on a stationary wall. J. Mar. Sci. Eng. 2022, 10, 228. [Google Scholar] [CrossRef]
  9. Peng, J.; Shen, H.; Xie, W.; Zhai, S.; Xi, G. Influence of flow fluctuation characteristics on flow and heat transfer in different regions. J. Drain. Irrig. Mach. Eng. 2022, 40, 826–833. [Google Scholar]
  10. Zhai, S.; Xie, F.; Yin, G.; Xi, G. Effect of gap ratio on vortex-induced vibration characteristics of different blunt bodies near-wall. J. Drain. Irrig. Mach. Eng. 2021, 39, 1132–1138. [Google Scholar]
  11. Lin, W.; Zhou, Y.; Wang, L.; Tao, L. PIV experiment and numerical simulation of trailing vortex structure of improved INTER-MIG impeller. J. Drain. Irrig. Mach. Eng. 2021, 39, 158–164. [Google Scholar]
  12. Han, B.; Yao, Z.; Tang, R.; Xu, H. On the supersonic impinging jet by laser Doppler velocimetry. Exp. Meas. Fluid Mech. 2002, 16, 99–103. [Google Scholar]
  13. Darisse, A.; Lemay, J.; Benaissa, A. LDV measurements of well converged third order moments in the far field of a free turbulent round jet. Exp. Therm. Fluid Sci. 2013, 44, 825–833. [Google Scholar] [CrossRef]
  14. Kumar, S.; Kumar, A. Effect of initial conditions on mean flow characteristics of a three dimensional turbulent wall jet. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 2021, 235, 6177–6190. [Google Scholar] [CrossRef]
  15. Tao, D.; Zhang, R.; Ying, C. Development and application of the pollutant diffusion testing apparatus based on the image analysis. J. Saf. Environ. 2016, 16, 247–251. [Google Scholar]
  16. Seo, H.; Kim, K.C. Experimental study on flow and turbulence characteristics of bubbly jet with low void fraction. Int. J. Multiph. Flow 2021, 142, 103738. [Google Scholar] [CrossRef]
  17. Wen, Q.; Sha, J.; Liu, Y. TR-PIV measurement of the turbulent submerged jet and POB analysis of the dynamic structure. J. Exp. Fluid Mech. 2014, 4, 16–24. [Google Scholar]
  18. Yang, Y.; Zhou, L.; Shi, W.; He, Z.; Han, Y.; Xiao, Y. Interstage difference of pressure pulsation in a three-stage electrical submersible pump. J. Petrol. Sci. Eng. 2021, 196, 107653. [Google Scholar] [CrossRef]
  19. Tang, S.; Zhu, Y.; Yuan, S. An improved convolutional neural network with an adaptable learning rate towards multi-signal fault diagnosis of hydraulic piston pump. Adv. Eng. Inform. 2021, 50, 101406. [Google Scholar] [CrossRef]
  20. Han, Y.; Song, X.; Li, K.; Yan, X. Hybrid modeling for submergence depth of the pumping well using stochastic configuration networks with random sampling. J. Petrol. Sci. Eng. 2022, 208, 109423. [Google Scholar] [CrossRef]
  21. Tang, S.; Zhu, Y.; Yuan, S. A novel adaptive convolutional neural network for fault diagnosis of hydraulic piston pump with acoustic images. Adv. Eng. Inform. 2022, 52, 101554. [Google Scholar] [CrossRef]
  22. Long, J.; Song, X.; Shi, J.; Chen, J. Optimization and CFD Analysis on Nozzle Exit Position of Two-phase Ejector. J. Refrig. 2022, 43, 39–45. [Google Scholar]
  23. Ni, Q.; Ruan, W. Optimization design of desilting jet pump parameters based on response surface model. J. Ship Mech. 2022, 26, 365–374. [Google Scholar]
  24. Zhang, K.; Zhu, X.; Ren, X.; Qiu, Q.; Shen, S. Numerical investigation on the effect of nozzle position for design of high performance ejector. Appl. Therm. Eng. 2017, 126, 594–601. [Google Scholar] [CrossRef]
  25. Fu, W.; Liu, Z.; Li, Y.; Wu, H.; Tang, Y. Numerical study for the influences of primary steam nozzle distance and mixing chamber throat diameter on steam ejector performance. Int. J. Therm. Sci. 2018, 132, 509–516. [Google Scholar] [CrossRef]
  26. Lucas, C.; Rusche, H.; Schroeder, A.; Koehler, J. Numerical investigation of a two-phase CO2 ejector. Int. J. Refrigeration 2014, 43, 154–166. [Google Scholar] [CrossRef]
  27. Ma, X.; Zhu, T.; Fu, Y.; Yan, Y.; Chen, W. Numerical simulation of rock breaking by abrasive water jet. J. Coast. Res. 2019, 93, 274–283. [Google Scholar] [CrossRef]
  28. He, L.; Liu, Y.; Shen, K.; Yang, X.; Ba, Q.; Xiong, W. Numerical research on the dynamic rock-breaking process of impact drilling with multi-nozzle water jets. J. Pet. Sci. Eng. 2021, 207, 109145. [Google Scholar] [CrossRef]
  29. Yu, Z.; Wang, Z.; Lei, C.; Zhou, Y.; Qiu, X. Numerical Simulation on Internal Flow Field of a Self-excited Oscillation Pulsed Jet Nozzle with Back-flow. Mech. Sci. Technol. Aerosp. Eng. 2022, 41, 998–1002. [Google Scholar]
  30. Huang, J.; Ni, F.; Gu, L. Numerical method of FLOW-3D for sediment erosion simulation. China Harb. Eng. 2019, 39, 6–11. [Google Scholar]
  31. Al Shaikhli, H.I.; Khassaf, S.I. Using of flow 3d as CFD materials approach in waves generation. Mater. Today Proc. 2022, 49, 2907–2911. [Google Scholar] [CrossRef]
  32. Kosaj, R.; Alboresha, R.S.; Sulaiman, S.O. Comparison Between Numerical Flow3d Software and Laboratory Data, For Sediment Incipient Motion. IOP Conf. Ser. Earth Environ. Sci. 2022, 961, 012031. [Google Scholar] [CrossRef]
  33. Du, C.; Liu, X.; Zhang, J.; Wang, B.; Chen, X.; Yu, X. Long-distance water hammer protection of pipeline after pump being first lowered and then rasied. J. Drain. Irrig. Mach. Eng. 2022, 40, 1248–1253, 1267. [Google Scholar]
  34. Gao, F.; Li, X.; Gao, Q. Experiment and numerical simulation on hydraulic characteristics of novel trapezoidal measuring weir. J. Drain. Irrig. Mach. Eng. 2022, 40, 1104–1111. [Google Scholar]
  35. Tu, A.; Nie, X.; Li, Y.; Li, H. Experimental and simulation study on water infiltration characteristics of layered red soil. J. Drain. Irrig. Mach. Eng. 2021, 39, 1243–1249. [Google Scholar]
  36. Chen, J.; Zeng, B.; Liu, L.; Tao, K.; Zhao, H.; Zhang, C.; Zhang, J.; Li, D. Investigating the anchorage performance of full-grouted anchor bolts with a modified numerical simulation method. Eng. Fail. Anal. 2022, 141, 106640. [Google Scholar] [CrossRef]
  37. Hu, B.; Yao, Y.; Wang, M.; Wang, C.; Liu, Y. Flow and Performance of the Disk Cavity of a Marine Gas Turbine at Varying Nozzle Pressure and Low Rotation Speeds: A Numerical Investigation. Machines 2023, 11, 68. [Google Scholar] [CrossRef]
  38. Yao, J.; Wang, X.; Zhang, S.; Xu, S.; Jin, B.; Ding, S. Orthogonal test of important parameters affecting hydraulic performance of negative pressure feedback jet sprinkler. J. Drain. Irrig. Mach. Eng. 2021, 39, 966–972. [Google Scholar]
  39. Wang, C.; Wang, X.; Shi, W.; Lu, W.; Tan, S.K.; Zhou, L. Experimental investigation on impingement of a submerged circular water jet at varying impinging angles and Reynolds numbers. Exp. Therm. Fluid Sci. 2017, 89, 189–198. [Google Scholar] [CrossRef]
  40. Speziale, C.G.; Thangam, S. Analysis of an RNG based turbulence model for separated flows. Int. J. Eng. Sci. 1992, 30, 1379–1388. [Google Scholar] [CrossRef]
  41. El Hassan, M.; Assoum, H.H.; Sobolik, V.; Vétel, J.; Abed-Meraim, K.; Garon, A.; Sakout, A. Experimental investigation of the wall shear stress and the vortex dynamics in a circular impinging jet. Exp. Fluids 2012, 52, 1475–1489. [Google Scholar] [CrossRef]
  42. Fairweather, M.; Hargrave, G. Experimental investigation of an axisymmetric, impinging turbulent jet. 1. Velocity field. Exp. Fluids 2002, 33, 464–471. [Google Scholar] [CrossRef]
  43. Ashforth-Frost, S.; Jambunathan, K. Effect of nozzle geometry and semi-confinement on the potential core of a turbulent axisymmetric free jet. Int. Commun. Heat Mass Transf. 1996, 23, 155–162. [Google Scholar] [CrossRef]
  44. Chen, M.; Huang, H.; Wang, D.; Lv, S.; Chen, Y. PIV tests for flow characteristics of impinging jet in a semi-closed circular pipe. J. Vib. Shock 2021, 40, 90–97, 113. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

      

MDPI and ACS Style

Mi, H.; Wang, C.; Jia, X.; Hu, B.; Wang, H.; Wang, H.; Zhu, Y. Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment. Sustainability 2023, 15, 5159. https://doi.org/10.3390/su15065159

AMA Style

Mi H, Wang C, Jia X, Hu B, Wang H, Wang H, Zhu Y. Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment. Sustainability. 2023; 15(6):5159. https://doi.org/10.3390/su15065159Chicago/Turabian Style

Mi, Hongbo, Chuan Wang, Xuanwen Jia, Bo Hu, Hongliang Wang, Hui Wang, and Yong Zhu. 2023. “Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment” Sustainability 15, no. 6: 5159. https://doi.org/10.3390/su15065159

Flow Hydrodynamics Influences Due to Flood Plain Sand Mining in a Meandering Channel

Flow Hydrodynamics Influences Due to Flood Plain Sand Mining in a Meandering Channel

Abstract

Flow hydrodynamics in the main channel due to floodplain sand mining is important for a better understanding of maintaining the natural habitat or the reliance between the flood plain and the main channel for the river’s long-term survival and also facilitates more effective river restoration engineering. Day by day anthropogenic stresses are increasing in the river corridor system, indiscriminate sand mining is one of them. In this study, a computational fluid dynamics (CFD)-based software Flow-3D hydro (renormalized group K-ε turbulence model used) is used to study the flow hydrodynamics of sinuous (sinuosity index = 1.25) channel 18 m long, 1 m width, and 0.3 m height with floodplain sand mining pit. Sand mining additionally increases the secondary current near the outer bank of the channel, therefore leading to scouring or erosion at the outer bank, as a result, rivers migrate laterally. The turbulence kinetic energy (TKE) is concentrated in the mining pit and near the inner bank. This study result can be used to understand the flow hydrodynamic of the river system due to the series of sand mining.

Keywords

  • Flow hydrodynamics
  • Turbulence modeling
  • Flow-3D
  • Sinuosity
  • Sand mining

References

  1. Best, J.: Anthropogenic stresses on the world’s big rivers. Nat. Geosci. 12(1), 7–21 (2019)CrossRef CAS Google Scholar 
  2. Bagnold, R.A.: Some Aspects of the Shape of River Meanders. US Government Printing Office (1960)Google Scholar 
  3. Kondolf, G.M.: Freshwater Gravel Mining and Dredging Issues: White Paper. Washington Department of Fish and Wildlife (2002)Google Scholar 
  4. Molnár, P., Ramírez, J.A.: Energy dissipation theories and optimal channel characteristics of river networks. Water Resour. Res. 34(7), 1809–1818 (1998)CrossRef Google Scholar 
  5. Padmalal, D., Maya, K.: Sand Mining: Environmental Impacts and Selected Case Studies. Springer (2014)Google Scholar 
  6. Hübler, M., Pothen, F.: Can smart policies solve the sand mining problem? PLoS ONE 16(4), e0248882 (2021)CrossRef Google Scholar 
  7. Khan, S., Sugie, A.: Sand mining and its social impacts on local society in rural Bangladesh: a case study of a village in Tangail district. J. Urban Reg. Stud. Contemp. India 2(1), 1–11 (2015)Google Scholar 
  8. Daneshfaraz, R. et al.: The experimental study of the effects of river mining holes on the bridge piers. Iranian J. Soil Water Res. 50(7), 1619–1633 (2019)Google Scholar 
  9. Hackney, C. R., Darby, S. E., Parsons, D. R., Leyland, J., Best, J. L., Aalto, R., … & Houseago, R. C.: River bank instability from unsustainable sand mining in the lower Mekong River. Nat. Sustain. 3(3), 217–225 (2020)Google Scholar 
  10. Callander, R.A.: River meandering. Annu. Rev. Fluid Mech. 10(1), 129–158 (1978)CrossRef Google Scholar 
  11. Koehnken, L., Rintoul, M.: Impacts of sand mining on ecosystem structure, process and biodiversity in rivers. World Wildlife Fund International (2018)Google Scholar 
  12. Gavriletea, M.D.: Environmental impacts of sand exploitation. Analysis of sand market. Sustainability 9(7), 1118 (2017)Google Scholar 
  13. Koehnken, L., et al.: Impacts of riverine sand mining on freshwater ecosystems: a review of the scientific evidence and guidance for future research. River Res. Appl. 36(3), 362–370 (2020)Google Scholar 
  14. Myers, W.R.C.: Momentum transfer in a compound channel. J. Hydraul. Res. 16(2), 139–150 (1978)CrossRef Google Scholar 
  15. Rajaratnam, N., Ahmadi, R.M.: Interaction between main channel and flood-plain flows. J. Hydraul. Div. 105(5), 573–588 (1979)CrossRef Google Scholar 
  16. Sellin, R.H.J.: A laboratory investigation into the interaction between the flow in the channel of a river and that over its flood plain. La Houille Blanche 7, 793–802 (1964)CrossRef Google Scholar 
  17. Karami, H., et al.: Verification of numerical study of scour around spur dikes using experimental data. Water Environ. J. 28(1), 124–134 (2014)Google Scholar 
  18. Bathurst, J.C., et al.: Overbank sediment deposition patterns for straight and meandering flume channels. Earth Surf. Proc. Land. 27(6), 659–665 (2002)CrossRef Google Scholar 
  19. Xu, D., Bai, Y.: Experimental study on the bed topography evolution in alluvial meandering rivers with various sinuousnesses. J. Hydro-Environ. Res. 7(2), 92–102 (2013)CrossRef Google Scholar 
  20. Priego-Hernández, G.A., Rivera-Trejo, F.: Secondary currents: measurement and analysis. Atmósfera 29(1), 23–34 (2016)Google Scholar 
  21. Alshamani, K.M.M.: Correlations among turbulent shear stress, turbulent kinetic energy, and axial turbulence intensity. AIAA J. 16(8), 859–861 (1978)CrossRef Google Scholar 
  22. Biron, P.M., et al.: Comparing different methods of bed shear stress estimates in simple and complex flow fields. Earth Surface Process. Landforms: J. British Geomorphol. Res. Group 29(11), 1403–1415 (2004)Google Scholar 
  23. Clark, L.A., Theresa, M.W.: Boundary Shear Stress Along Vegetated Streambanks (2007)Google Scholar 
  24. Kim, S.-C., et al.: Estimating bottom stress in tidal boundary layer from acoustic Doppler velocimeter data. J. Hydraul. Eng. 126(6), 399–406 (2000)CrossRef Google Scholar 

  1. Home  
  2. Sustainable Environment  
  3. Conference paper

Flow Hydrodynamics Influences Due to Flood Plain Sand Mining in a Meandering Channel

  • 14 Accesses

Abstract

Flow hydrodynamics in the main channel due to floodplain sand mining is important for a better understanding of maintaining the natural habitat or the reliance between the flood plain and the main channel for the river’s long-term survival and also facilitates more effective river restoration engineering. Day by day anthropogenic stresses are increasing in the river corridor system, indiscriminate sand mining is one of them. In this study, a computational fluid dynamics (CFD)-based software Flow-3D hydro (renormalized group K-ε turbulence model used) is used to study the flow hydrodynamics of sinuous (sinuosity index = 1.25) channel 18 m long, 1 m width, and 0.3 m height with floodplain sand mining pit. Sand mining additionally increases the secondary current near the outer bank of the channel, therefore leading to scouring or erosion at the outer bank, as a result, rivers migrate laterally. The turbulence kinetic energy (TKE) is concentrated in the mining pit and near the inner bank. This study result can be used to understand the flow hydrodynamic of the river system due to the series of sand mining.

Keywords

  • Flow hydrodynamics
  • Turbulence modeling
  • Flow-3D
  • Sinuosity
  • Sand mining

This is a preview of subscription content, access via your institution.

References

  1. Best, J.: Anthropogenic stresses on the world’s big rivers. Nat. Geosci. 12(1), 7–21 (2019)CrossRef CAS Google Scholar 
  2. Bagnold, R.A.: Some Aspects of the Shape of River Meanders. US Government Printing Office (1960)Google Scholar 
  3. Kondolf, G.M.: Freshwater Gravel Mining and Dredging Issues: White Paper. Washington Department of Fish and Wildlife (2002)Google Scholar 
  4. Molnár, P., Ramírez, J.A.: Energy dissipation theories and optimal channel characteristics of river networks. Water Resour. Res. 34(7), 1809–1818 (1998)CrossRef Google Scholar 
  5. Padmalal, D., Maya, K.: Sand Mining: Environmental Impacts and Selected Case Studies. Springer (2014)Google Scholar 
  6. Hübler, M., Pothen, F.: Can smart policies solve the sand mining problem? PLoS ONE 16(4), e0248882 (2021)CrossRef Google Scholar 
  7. Khan, S., Sugie, A.: Sand mining and its social impacts on local society in rural Bangladesh: a case study of a village in Tangail district. J. Urban Reg. Stud. Contemp. India 2(1), 1–11 (2015)Google Scholar 
  8. Daneshfaraz, R. et al.: The experimental study of the effects of river mining holes on the bridge piers. Iranian J. Soil Water Res. 50(7), 1619–1633 (2019)Google Scholar 
  9. Hackney, C. R., Darby, S. E., Parsons, D. R., Leyland, J., Best, J. L., Aalto, R., … & Houseago, R. C.: River bank instability from unsustainable sand mining in the lower Mekong River. Nat. Sustain. 3(3), 217–225 (2020)Google Scholar 
  10. Callander, R.A.: River meandering. Annu. Rev. Fluid Mech. 10(1), 129–158 (1978)CrossRef Google Scholar 
  11. Koehnken, L., Rintoul, M.: Impacts of sand mining on ecosystem structure, process and biodiversity in rivers. World Wildlife Fund International (2018)Google Scholar 
  12. Gavriletea, M.D.: Environmental impacts of sand exploitation. Analysis of sand market. Sustainability 9(7), 1118 (2017)Google Scholar 
  13. Koehnken, L., et al.: Impacts of riverine sand mining on freshwater ecosystems: a review of the scientific evidence and guidance for future research. River Res. Appl. 36(3), 362–370 (2020)Google Scholar 
  14. Myers, W.R.C.: Momentum transfer in a compound channel. J. Hydraul. Res. 16(2), 139–150 (1978)CrossRef Google Scholar 
  15. Rajaratnam, N., Ahmadi, R.M.: Interaction between main channel and flood-plain flows. J. Hydraul. Div. 105(5), 573–588 (1979)CrossRef Google Scholar 
  16. Sellin, R.H.J.: A laboratory investigation into the interaction between the flow in the channel of a river and that over its flood plain. La Houille Blanche 7, 793–802 (1964)CrossRef Google Scholar 
  17. Karami, H., et al.: Verification of numerical study of scour around spur dikes using experimental data. Water Environ. J. 28(1), 124–134 (2014)Google Scholar 
  18. Bathurst, J.C., et al.: Overbank sediment deposition patterns for straight and meandering flume channels. Earth Surf. Proc. Land. 27(6), 659–665 (2002)CrossRef Google Scholar 
  19. Xu, D., Bai, Y.: Experimental study on the bed topography evolution in alluvial meandering rivers with various sinuousnesses. J. Hydro-Environ. Res. 7(2), 92–102 (2013)CrossRef Google Scholar 
  20. Priego-Hernández, G.A., Rivera-Trejo, F.: Secondary currents: measurement and analysis. Atmósfera 29(1), 23–34 (2016)Google Scholar 
  21. Alshamani, K.M.M.: Correlations among turbulent shear stress, turbulent kinetic energy, and axial turbulence intensity. AIAA J. 16(8), 859–861 (1978)CrossRef Google Scholar 
  22. Biron, P.M., et al.: Comparing different methods of bed shear stress estimates in simple and complex flow fields. Earth Surface Process. Landforms: J. British Geomorphol. Res. Group 29(11), 1403–1415 (2004)Google Scholar 
  23. Clark, L.A., Theresa, M.W.: Boundary Shear Stress Along Vegetated Streambanks (2007)Google Scholar 
  24. Kim, S.-C., et al.: Estimating bottom stress in tidal boundary layer from acoustic Doppler velocimeter data. J. Hydraul. Eng. 126(6), 399–406 (2000)CrossRef Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Civil Engineering, Indian Institute of Technology Guwahati, Guwahati, Assam, IndiaO. P. Maurya, K. K. Nandi, S. Modalavalasa & S. Dutta

Corresponding author

Correspondence to O. P. Maurya .

Editor information

Editors and Affiliations

  1. Centre for the Environment, Indian Institute of Technology Guwahati, Guwahati, IndiaDeepmoni Deka
  2. Department of Chemical engineering, Indian Institute of Technology Guwahati, Guwahati, IndiaSubrata Kumar Majumder
  3. Department of Chemical engineering, Indian Institute of Technology Guwahati, Guwahati, IndiaMihir Kumar Purkait
Figure 5 A schematic of the water model of reactor URO 200.

Physical and Numerical Modeling of the Impeller Construction Impact on the Aluminum Degassing Process

알루미늄 탈기 공정에 미치는 임펠러 구성의 물리적 및 수치적 모델링

Kamil Kuglin,1 Michał Szucki,2 Jacek Pieprzyca,3 Simon Genthe,2 Tomasz Merder,3 and Dorota Kalisz1,*

Mikael Ersson, Academic Editor

Author information Article notes Copyright and License information Disclaimer

Associated Data

Data Availability Statement

Go to:

Abstract

This paper presents the results of tests on the suitability of designed heads (impellers) for aluminum refining. The research was carried out on a physical model of the URO-200, followed by numerical simulations in the FLOW 3D program. Four design variants of impellers were used in the study. The degree of dispersion of the gas phase in the model liquid was used as a criterion for evaluating the performance of each solution using different process parameters, i.e., gas flow rate and impeller speed. Afterward, numerical simulations in Flow 3D software were conducted for the best solution. These simulations confirmed the results obtained with the water model and verified them.

Keywords: aluminum, impeller construction, degassing process, numerical modeling, physical modeling

Go to:

1. Introduction

Constantly increasing requirements concerning metallurgical purity in terms of hydrogen content and nonmetallic inclusions make casting manufacturers use effective refining techniques. The answer to this demand is the implementation of the aluminum refining technique making use of a rotor with an original design guaranteeing efficient refining [1,2,3,4]. The main task of the impeller (rotor) is to reduce the contamination of liquid metal (primary and recycled aluminum) with hydrogen and nonmetallic inclusions. An inert gas, mainly argon or a mixture of gases, is introduced through the rotor into the liquid metal to bring both hydrogen and nonmetallic inclusions to the metal surface through the flotation process. Appropriately and uniformly distributed gas bubbles in the liquid metal guarantee achieving the assumed level of contaminant removal economically. A very important factor in deciding about the obtained degassing effect is the optimal rotor design [5,6,7,8]. Thanks to the appropriate geometry of the rotor, gas bubbles introduced into the liquid metal are split into smaller ones, and the spinning movement of the rotor distributes them throughout the volume of the liquid metal bath. In this solution impurities in the liquid metal are removed both in the volume and from the upper surface of the metal. With a well-designed impeller, the costs of refining aluminum and its alloys can be lowered thanks to the reduced inert gas and energy consumption (optimal selection of rotor rotational speed). Shorter processing time and a high degree of dehydrogenation decrease the formation of dross on the metal surface (waste). A bigger produced dross leads to bigger process losses. Consequently, this means that the choice of rotor geometry has an indirect impact on the degree to which the generated waste is reduced [9,10].

Another equally important factor is the selection of process parameters such as gas flow rate and rotor speed [11,12]. A well-designed gas injection system for liquid metal meets two key requirements; it causes rapid mixing of the liquid metal to maintain a uniform temperature throughout the volume and during the entire process, to produce a chemically homogeneous metal composition. This solution ensures effective degassing of the metal bath. Therefore, the shape of the rotor, the arrangement of the nozzles, and their number are significant design parameters that guarantee the optimum course of the refining process. It is equally important to complete the mixing of the metal bath in a relatively short time, as this considerably shortens the refining process and, consequently, reduces the process costs. Another important criterion conditioning the implementation of the developed rotor is the generation of fine diffused gas bubbles which are distributed throughout the metal volume, and whose residence time will be sufficient for the bubbles to collide and adsorb the contaminants. The process of bubble formation by the spinning rotors differs from that in the nozzles or porous molders. In the case of a spinning rotor, the shear force generated by the rotor motion splits the bubbles into smaller ones. Here, the rotational speed, mixing force, surface tension, and fluid density have a key effect on the bubble size. The velocity of the bubbles, which depends mainly on their size and shape, determines their residence time in the reactor and is, therefore, very important for the refining process, especially since gas bubbles in liquid aluminum may remain steady only below a certain size [13,14,15].

The impeller designs presented in the article were developed to improve the efficiency of the process and reduce its costs. The impellers used so far have a complicated structure and are very pricey. The success of the conducted research will allow small companies to become independent of external supplies through the possibility of making simple and effective impellers on their own. The developed structures were tested on the water model. The results of this study can be considered as pilot.

Go to:

2. Materials and Methods

Rotors were realized with the SolidWorks computer design technique and a 3D printer. The developed designs were tested on a water model. Afterward, the solution with the most advantageous refining parameters was selected and subjected to calculations with the Flow3D package. As a result, an impeller was designed for aluminum refining. Its principal lies in an even distribution of gas bubbles in the entire volume of liquid metal, with the largest possible participation of the bubble surface, without disturbing the metal surface. This procedure guarantees the removal of gaseous, as well as metallic and nonmetallic, impurities.

2.1. Rotor Designs

The developed impeller constructions, shown in Figure 1, Figure 2, Figure 3 and Figure 4, were printed on a 3D printer using the PLA (polylactide) material. The impeller design models differ in their shape and the number of holes through which the inert gas flows. Figure 1, Figure 2 and Figure 3 show the same impeller model but with a different number of gas outlets. The arrangement of four, eight, and 12 outlet holes was adopted in the developed design. A triangle-shaped structure equipped with three gas outlet holes is presented in Figure 4.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g001.jpg

Figure 1

A 3D model—impeller with four holes—variant B4.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g002.jpg

Figure 2

A 3D model—impeller with eight holes—variant B8.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g003.jpg

Figure 3

A 3D model—impeller with twelve holes—variant B12.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g004.jpg

Figure 4

A 3D model—‘red triangle’ impeller with three holes—variant RT3.

2.2. Physical Models

Investigations were carried out on a water model of the URO 200 reactor of the barbotage refining process (see Figure 5).

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g005.jpg

Figure 5

A schematic of the water model of reactor URO 200.

The URO 200 reactor can be classified as a cyclic reactor. The main element of the device is a rotor, which ends the impeller. The whole system is attached to a shaft via which the refining gas is supplied. Then, the shaft with the rotor is immersed in the liquid metal in the melting pot or the furnace chamber. In URO 200 reactors, the refining process lasts 600 s (10 min), the gas flow rate that can be obtained ranges from 5 to 20 dm3·min−1, and the speed at which the rotor can move is 0 to 400 rpm. The permissible quantity of liquid metal for barbotage refining is 300 kg or 700 kg [8,16,17]. The URO 200 has several design solutions which improve operation and can be adapted to the existing equipment in the foundry. These solutions include the following [8,16]:

  • URO-200XR—used for small crucible furnaces, the capacity of which does not exceed 250 kg, with no control system and no control of the refining process.
  • URO-200SA—used to service several crucible furnaces of capacity from 250 kg to 700 kg, fully automated and equipped with a mechanical rotor lift.
  • URO-200KA—used for refining processes in crucible furnaces and allows refining in a ladle. The process is fully automated, with a hydraulic rotor lift.
  • URO-200KX—a combination of the XR and KA models, designed for the ladle refining process. Additionally, refining in heated crucibles is possible. The unit is equipped with a manual hydraulic rotor lift.
  • URO-200PA—designed to cooperate with induction or crucible furnaces or intermediate chambers, the capacity of which does not exceed one ton. This unit is an integral part of the furnace. The rotor lift is equipped with a screw drive.

Studies making use of a physical model can be associated with the observation of the flow and circulation of gas bubbles. They require meeting several criteria regarding the similarity of the process and the object characteristics. The similarity conditions mainly include geometric, mechanical, chemical, thermal, and kinetic parameters. During simulation of aluminum refining with inert gas, it is necessary to maintain the geometric similarity between the model and the real object, as well as the similarity related to the flow of liquid metal and gas (hydrodynamic similarity). These quantities are characterized by the Reynolds, Weber, and Froude numbers. The Froude number is the most important parameter characterizing the process, its magnitude is the same for the physical model and the real object. Water was used as the medium in the physical modeling. The factors influencing the choice of water are its availability, relatively low cost, and kinematic viscosity at room temperature, which is very close to that of liquid aluminum.

The physical model studies focused on the flow of inert gas in the form of gas bubbles with varying degrees of dispersion, particularly with respect to some flow patterns such as flow in columns and geysers, as well as disturbance of the metal surface. The most important refining parameters are gas flow rate and rotor speed. The barbotage refining studies for the developed impeller (variants B4, B8, B12, and RT3) designs were conducted for the following process parameters:

  • Rotor speed: 200, 300, 400, and 500 rpm,
  • Ideal gas flow: 10, 20, and 30 dm3·min−1,
  • Temperature: 293 K (20 °C).

These studies were aimed at determining the most favorable variants of impellers, which were then verified using the numerical modeling methods in the Flow-3D program.

2.3. Numerical Simulations with Flow-3D Program

Testing different rotor impellers using a physical model allows for observing the phenomena taking place while refining. This is a very important step when testing new design solutions without using expensive industrial trials. Another solution is modeling by means of commercial simulation programs such as ANSYS Fluent or Flow-3D [18,19]. Unlike studies on a physical model, in a computer program, the parameters of the refining process and the object itself, including the impeller design, can be easily modified. The simulations were performed with the Flow-3D program version 12.03.02. A three-dimensional system with the same dimensions as in the physical modeling was used in the calculations. The isothermal flow of liquid–gas bubbles was analyzed. As in the physical model, three speeds were adopted in the numerical tests: 200, 300, and 500 rpm. During the initial phase of the simulations, the velocity field around the rotor generated an appropriate direction of motion for the newly produced bubbles. When the required speed was reached, the generation of randomly distributed bubbles around the rotor was started at a rate of 2000 per second. Table 1 lists the most important simulation parameters.

Table 1

Values of parameters used in the calculations.

ParameterValueUnit
Maximum number of gas particles1,000,000–
Rate of particle generation20001·s−1
Specific gas constant287.058J·kg−1·K−1
Atmospheric pressure1.013 × 105Pa
Water density1000kg·m−3
Water viscosity0.001kg·m−1·s−1
Boundary condition on the wallsNo-slip–
Size of computational cell0.0034m

Open in a separate window

In the case of the CFD analysis, the numerical solutions require great care when generating the computational mesh. Therefore, computational mesh tests were performed prior to the CFD calculations. The effect of mesh density was evaluated by taking into account the velocity of water in the tested object on the measurement line A (height of 0.065 m from the bottom) in a characteristic cross-section passing through the object axis (see Figure 6). The mesh contained 3,207,600, 6,311,981, 7,889,512, 11,569,230, and 14,115,049 cells.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g006.jpg

Figure 6

The velocity of the water depending on the size of the computational grid.

The quality of the generated computational meshes was checked using the criterion skewness angle QEAS [18]. This criterion is described by the following relationship:

QEAS=max{βmax−βeq180−βeq,βeq−βminβeq},

(1)

where βmax, βmin are the maximal and minimal angles (in degrees) between the edges of the cell, and βeq is the angle corresponding to an ideal cell, which for cubic cells is 90°.

Normalized in the interval [0;1], the value of QEAS should not exceed 0.75, which identifies the permissible skewness angle of the generated mesh. For the computed meshes, this value was equal to 0.55–0.65.

Moreover, when generating the computational grids in the studied facility, they were compacted in the areas of the highest gradients of the calculated values, where higher turbulence is to be expected (near the impeller). The obtained results of water velocity in the studied object at constant gas flow rate are shown in Figure 6.

The analysis of the obtained water velocity distributions (see Figure 6) along the line inside the object revealed that, with the density of the grid of nodal points, the velocity changed and its changes for the test cases of 7,889,512, 11,569,230, and 14,115,049 were insignificant. Therefore, it was assumed that a grid containing not less than 7,900,000 (7,889,512) cells would not affect the result of CFD calculations.

A single-block mesh of regular cells with a size of 0.0034 m was used in the numerical calculations. The total number of cells was approximately 7,900,000 (7,889,512). This grid resolution (see Figure 7) allowed the geometry of the system to be properly represented, maintaining acceptable computation time (about 3 days on a workstation with 2× CPU and 12 computing cores).

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g007.jpg

Figure 7

Structured equidistant mesh used in numerical calculations: (a) mesh with smoothed, surface cells (the so-called FAVOR method) used in Flow-3D; (b) visualization of the applied mesh resolution.

The calculations were conducted with an explicit scheme. The timestep was selected by the program automatically and controlled by stability and convergence. From the moment of the initial velocity field generation (start of particle generation), it was 0.0001 s.

When modeling the degassing process, three fluids are present in the system: water, gas supplied through the rotor head (impeller), and the surrounding air. Modeling such a multiphase flow is a numerically very complex issue. The necessity to overcome the liquid backpressure by the gas flowing out from the impeller leads to the formation of numerical instabilities in the volume of fluid (VOF)-based approach used by Flow-3D software. Therefore, a mixed description of the analyzed flow was used here. In this case, water was treated as a continuous medium, while, in the case of gas bubbles, the discrete phase model (DPM) model was applied. The way in which the air surrounding the system was taken into account is later described in detail.

The following additional assumptions were made in the modeling:

  • —The liquid phase was considered as an incompressible Newtonian fluid.
  • —The effect of chemical reactions during the refining process was neglected.
  • —The composition of each phase (gas and liquid) was considered homogeneous; therefore, the viscosity and surface tension were set as constants.
  • —Only full turbulence existed in the liquid, and the effect of molecular viscosity was neglected.
  • —The gas bubbles were shaped as perfect spheres.
  • —The mutual interaction between gas bubbles (particles) was neglected.

2.3.1. Modeling of Liquid Flow 

The motion of the real fluid (continuous medium) is described by the Navier–Stokes Equation [20].

dudt=−1ρ∇p+ν∇2u+13ν∇(∇⋅ u)+F,

(2)

where du/dt is the time derivative, u is the velocity vector, t is the time, and F is the term accounting for external forces including gravity (unit components denoted by X, Y, Z).

In the simulations, the fluid flow was assumed to be incompressible, in which case the following equation is applicable:

∂u∂t+(u⋅∇)u=−1ρ∇p+ν∇2u+F.

(3)

Due to the large range of liquid velocities during flows, the turbulence formation process was included in the modeling. For this purpose, the k–ε model turbulence kinetic energy k and turbulence dissipation ε were the target parameters, as expressed by the following equations [21]:

∂(ρk)∂t+∂(ρkvi)∂xi=∂∂xj[(μ+μtσk)⋅∂k∂xi]+Gk+Gb−ρε−Ym+Sk,

(4)

∂(ρε)∂t+∂(ρεui)∂xi=∂∂xj[(μ+μtσε)⋅∂k∂xi]+C1εεk(Gk+G3εGb)+C2ερε2k+Sε,

(5)

where ρ is the gas density, σκ and σε are the Prandtl turbulence numbers, k and ε are constants of 1.0 and 1.3, and Gk and Gb are the kinetic energy of turbulence generated by the average velocity and buoyancy, respectively.

As mentioned earlier, there are two gas phases in the considered problem. In addition to the gas bubbles, which are treated here as particles, there is also air, which surrounds the system. The boundary of phase separation is in this case the free surface of the water. The shape of the free surface can change as a result of the forming velocity field in the liquid. Therefore, it is necessary to use an appropriate approach to free surface tracking. The most commonly used concept in liquid–gas flow modeling is the volume of fluid (VOF) method [22,23], and Flow-3D uses a modified version of this method called TrueVOF. It introduces the concept of the volume fraction of the liquid phase fl. This parameter can be used for classifying the cells of a discrete grid into areas filled with liquid phase (fl = 1), gaseous phase, or empty cells (fl = 0) and those through which the phase separation boundary (fl ∈ (0, 1)) passes (free surface). To determine the local variations of the liquid phase fraction, it is necessary to solve the following continuity equation:

dfldt=0.

(6)

Then, the fluid parameters in the region of coexistence of the two phases (the so-called interface) depend on the volume fraction of each phase.

ρ=flρl+(1−fl)ρg,

(7)

ν=flνl+(1−fl)νg,

(8)

where indices l and g refer to the liquid and gaseous phases, respectively.

The parameter of fluid velocity in cells containing both phases is also determined in the same way.

u=flul+(1−fl)ug.

(9)

Since the processes taking place in the surrounding air can be omitted, to speed up the calculations, a single-phase, free-surface model was used. This means that no calculations were performed in the gas cells (they were treated as empty cells). The liquid could fill them freely, and the air surrounding the system was considered by the atmospheric pressure exerted on the free surface. This approach is often used in modeling foundry and metallurgical processes [24].

2.3.2. Modeling of Gas Bubble Flow 

As stated, a particle model was used to model bubble flow. Spherical particles (gas bubbles) of a given size were randomly generated in the area marked with green in Figure 7b. In the simulations, the gas bubbles were assumed to have diameters of 0.016 and 0.02 m corresponding to the gas flow rates of 10 and 30 dm3·min−1, respectively.

Experimental studies have shown that, as a result of turbulent fluid motion, some of the bubbles may burst, leading to the formation of smaller bubbles, although merging of bubbles into larger groupings may also occur. Therefore, to be able to observe the behavior of bubbles of different sizes (diameter), the calculations generated two additional particle types with diameters twice smaller and twice larger, respectively. The proportion of each species in the system was set to 33.33% (Table 2).

Table 2

Data assumed for calculations.

NoRotor Speed (Rotational Speed)
rpm		Bubbles Diameter
m		Corresponding Gas Flow Rate
dm3·min−1		NoRotor Speed (Rotational Speed)
rpm		Bubbles Diameter
m		Corresponding Gas Flow Rate
dm3·min−1
A2000.01610D2000.0230
0.0080.01
0.0320.04
B3000.01610E3000.0230
0.0080.01
0.0320.04
C5000.01610F5000.0230
0.0080.01
0.0320.04

Open in a separate window

The velocity of the particle results from the generated velocity field (calculated from Equation (3) in the liquid ul around it and its velocity resulting from the buoyancy force ub. The effect of particle radius r on the terminal velocity associated with buoyancy force can be determined according to Stokes’ law.

ub=29 (ρg−ρl)μlgr2,

(10)

where g is the acceleration (9.81).

The DPM model was used for modeling the two-phase (water–air) flow. In this model, the fluid (water) is treated as a continuous phase and described by the Navier–Stokes equation, while gas bubbles are particles flowing in the model fluid (discrete phase). The trajectories of each bubble in the DPM system are calculated at each timestep taking into account the mass forces acting on it. Table 3 characterizes the DPM model used in our own research [18].

Table 3

Characteristic of the DPM model.

MethodEquations
Euler–LagrangeBalance equation:
dugdt=FD(u−ug)+g(ϱg−ϱ)ϱg+F.
FD (u − up) denotes the drag forces per mass unit of a bubble, and the expression for the drag coefficient FD is of the form
FD=18μCDReϱ⋅gd2g24.
The relative Reynolds number has the form
Re≡ρdg|ug−u|μ.
On the other hand, the force resulting from the additional acceleration of the model fluid has the form
F=12dρdtρg(u−ug),
where ug is the gas bubble velocity, u is the liquid velocity, dg is the bubble diameter, and CD is the drag coefficient.

Open in a separate window

Go to:

3. Results and Discussion

3.1. Calculations of Power and Mixing Time by the Flowing Gas Bubbles

One of the most important parameters of refining with a rotor is the mixing power induced by the spinning rotor and the outflowing gas bubbles (via impeller). The mixing power of liquid metal in a ladle of height (h) by gas injection can be determined from the following relation [15]:

pgVm=ρ⋅g⋅uB,

(11)

where pg is the mixing power, Vm is the volume of liquid metal in the reactor, ρ is the density of liquid aluminum, and uB is the average speed of bubbles, given below.

uB=n⋅R⋅TAc⋅Pm⋅t,

(12)

where n is the number of gas moles, R is the gas constant (8.314), Ac is the cross-sectional area of the reactor vessel, T is the temperature of liquid aluminum in the reactor, and Pm is the pressure at the middle tank level. The pressure at the middle level of the tank is calculated by a function of the mean logarithmic difference.

Pm=(Pa+ρ⋅g⋅h)−Paln(Pa+ρ⋅g⋅h)Pa,

(13)

where Pa is the atmospheric pressure, and h is the the height of metal in the reactor.

Themelis and Goyal [25] developed a model for calculating mixing power delivered by gas injection.

pg=2Q⋅R⋅T⋅ln(1+m⋅ρ⋅g⋅hP),

(14)

where Q is the gas flow, and m is the mass of liquid metal.

Zhang [26] proposed a model taking into account the temperature difference between gas and alloy (metal).

pg=QRTgVm[ln(1+ρ⋅g⋅hPa)+(1−TTg)],

(15)

where Tg is the gas temperature at the entry point.

Data for calculating the mixing power resulting from inert gas injection into liquid aluminum are given below in Table 4. The design parameters were adopted for the model, the parameters of which are shown in Figure 5.

Table 4

Data for calculating mixing power introduced by an inert gas.

ParameterValueUnit
Height of metal column0.7m
Density of aluminum2375kg·m−3
Process duration20s
Gas temperature at the injection site940K
Cross-sectional area of ladle0.448m2
Mass of liquid aluminum546.25kg
Volume of ladle0.23M3
Temperature of liquid aluminum941.15K

Open in a separate window

Table 5 presents the results of mixing power calculations according to the models of Themelis and Goyal and of Zhang for inert gas flows of 10, 20, and 30 dm3·min−1. The obtained calculation results significantly differed from each other. The difference was an order of magnitude, which indicates that the model is highly inaccurate without considering the temperature of the injected gas. Moreover, the calculations apply to the case when the mixing was performed only by the flowing gas bubbles, without using a rotor, which is a great simplification of the phenomenon.

Table 5

Mixing power calculated from mathematical models.

Mathematical ModelMixing Power (W·t−1)
for a Given Inert Gas Flow (dm3·min−1)
102030
Themelis and Goyal11.4923.3335.03
Zhang0.821.662.49

Open in a separate window

The mixing time is defined as the time required to achieve 95% complete mixing of liquid metal in the ladle [27,28,29,30]. Table 6 groups together equations for the mixing time according to the models.

Table 6

Models for calculating mixing time.

AuthorsModelRemarks
Szekely [31]τ=800ε−0.4ε—W·t−1
Chiti and Paglianti [27]τ=CVQlV—volume of reactor, m3
Ql—flow intensity, m3·s−1
Iguchi and Nakamura [32]τ=1200⋅Q−0.4D1.97h−1.0υ0.47υ—kinematic viscosity, m2·s−1
D—diameter of ladle, m
h—height of metal column, m
Q—liquid flow intensity, m3·s−1

Open in a separate window

Figure 8 and Figure 9 show the mixing time as a function of gas flow rate for various heights of the liquid column in the ladle and mixing power values.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g008.jpg

Figure 8

Mixing time as a function of gas flow rate for various heights of the metal column (Iguchi and Nakamura model).

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g009.jpg

Figure 9

Mixing time as a function of mixing power (Szekly model).

3.2. Determining the Bubble Size

The mechanisms controlling bubble size and mass transfer in an alloy undergoing refining are complex. Strong mixing conditions in the reactor promote impurity mass transfer. In the case of a spinning rotor, the shear force generated by the rotor motion separates the bubbles into smaller bubbles. Rotational speed, mixing force, surface tension, and liquid density have a strong influence on the bubble size. To characterize the kinetic state of the refining process, parameters k and A were introduced. Parameters k, A, and uB can be calculated using the below equations [33].

k=2D⋅uBdB⋅π−−−−−−√,

(16)

A=6Q⋅hdB⋅uB,

(17)

uB=1.02g⋅dB,−−−−−√

(18)

where D is the diffusion coefficient, and dB is the bubble diameter.

After substituting appropriate values, we get

dB=3.03×104(πD)−2/5g−1/5h4/5Q0.344N−1.48.

(19)

According to the last equation, the size of the gas bubble decreases with the increasing rotational speed (see Figure 10).

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g010.jpg

Figure 10

Effect of rotational speed on the bubble diameter.

In a flow of given turbulence intensity, the diameter of the bubble does not exceed the maximum size dmax, which is inversely proportional to the rate of kinetic energy dissipation in a viscous flow ε. The size of the gas bubble diameter as a function of the mixing energy, also considering the Weber number and the mixing energy in the negative power, can be determined from the following equations [31,34]:

  • —Sevik and Park:

dBmax=We0.6kr⋅(σ⋅103ρ⋅10−3)0.6⋅(10⋅ε)−0.4⋅10−2.

(20)

  • —Evans:

dBmax=⎡⎣Wekr⋅σ⋅1032⋅(ρ⋅10−3)13⎤⎦35 ⋅(10⋅ε)−25⋅10−2.

(21)

The results of calculating the maximum diameter of the bubble dBmax determined from Equation (21) are given in Table 7.

Table 7

The results of calculating the maximum diameter of the bubble using Equation (21).

ModelMixing Energy
ĺ (m2·s−3)		Weber Number (Wekr)
0.591.01.2
Zhang and Taniguchi
dmax		0.10.01670.02300.026
0.50.00880.01210.013
1.00.00670.00910.010
1.50.00570.00780.009
Sevik and Park
dBmax		0.10.2650.360.41
0.50.1390.190.21
1.00.1060.140.16
1.50.0900.120.14
Evans
dBmax		0.10.2470.3400.38
0.50.1300.1780.20
1.00.0980.1350.15
1.50.0840.1150.13

Open in a separate window

3.3. Physical Modeling

The first stage of experiments (using the URO-200 water model) included conducting experiments with impellers equipped with four, eight, and 12 gas outlets (variants B4, B8, B12). The tests were carried out for different process parameters. Selected results for these experiments are presented in Figure 11, Figure 12, Figure 13 and Figure 14.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g011.jpg

Figure 11

Impeller variant B4—gas bubbles dispersion registered for a gas flow rate of 10 dm3·min−1 and rotor speed of (a) 200, (b) 300, (c) 400, and (d) 500 rpm.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g012.jpg

Figure 12

Impeller variant B8—gas bubbles dispersion registered for a gas flow rate of 10 dm3·min−1 and rotor speed of (a) 200, (b) 300, (c) 400, and (d) 500 rpm.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g013.jpg

Figure 13

Gas bubble dispersion registered for different processing parameters (impeller variant B12).

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g014.jpg

Figure 14

Gas bubble dispersion registered for different processing parameters (impeller variant RT3).

The analysis of the refining variants presented in Figure 11, Figure 12, Figure 13 and Figure 14 reveals that the proposed impellers design model is not useful for the aluminum refining process. The number of gas outlet orifices, rotational speed, and flow did not affect the refining efficiency. In all the variants shown in the figures, very poor dispersion of gas bubbles was observed in the object. The gas bubble flow had a columnar character, and so-called dead zones, i.e., areas where no inert gas bubbles are present, were visible in the analyzed object. Such dead zones were located in the bottom and side zones of the ladle, while the flow of bubbles occurred near the turning rotor. Another negative phenomenon observed was a significant agitation of the water surface due to excessive (rotational) rotor speed and gas flow (see Figure 13, cases 20; 400, 30; 300, 30; 400, and 30; 500).

Research results for a ‘red triangle’ impeller equipped with three gas supply orifices (variant RT3) are presented in Figure 14.

In this impeller design, a uniform degree of bubble dispersion in the entire volume of the modeling fluid was achieved for most cases presented (see Figure 14). In all tested variants, single bubbles were observed in the area of the water surface in the vessel. For variants 20; 200, 30; 200, and 20; 300 shown in Figure 14, the bubble dispersion results were the worst as the so-called dead zones were identified in the area near the bottom and sidewalls of the vessel, which disqualifies these work parameters for further applications. Interestingly, areas where swirls and gas bubble chains formed were identified only for the inert gas flows of 20 and 30 dm3·min−1 and 200 rpm in the analyzed model. This means that the presented model had the best performance in terms of dispersion of gas bubbles in the model liquid. Its design with sharp edges also differed from previously analyzed models, which is beneficial for gas bubble dispersion, but may interfere with its suitability in industrial conditions due to possible premature wear.

3.4. Qualitative Comparison of Research Results (CFD and Physical Model)

The analysis (physical modeling) revealed that the best mixing efficiency results were obtained with the RT3 impeller variant. Therefore, numerical calculations were carried out for the impeller model with three outlet orifices (variant RT3). The CFD results are presented in Figure 15 and Figure 16.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g015.jpg

Figure 15

Simulation results of the impeller RT3, for given flows and rotational speeds after a time of 1 s: simulation variants (a) A, (b) B, (c) C, (d) D, (e) E, and (f) F.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g016.jpg

Figure 16

Simulation results of the impeller RT3, for given flows and rotational speeds after a time of 5.4 s.: simulation variants (a) A, (b) B, (c) C, (d) D, (e) E, and (f) F.

CFD results are presented for all analyzed variants (impeller RT3) at two selected calculation timesteps of 1 and 5.40 s. They show the velocity field of the medium (water) and the dispersion of gas bubbles.

Figure 15 shows the initial refining phase after 1 s of the process. In this case, the gas bubble formation and flow were observed in an area close to contact with the rotor. Figure 16 shows the phase when the dispersion and flow of gas bubbles were advanced in the reactor area of the URO-200 model.

The quantitative evaluation of the obtained results of physical and numerical model tests was based on the comparison of the degree of gas dispersion in the model liquid. The degree of gas bubble dispersion in the volume of the model liquid and the areas of strong turbulent zones formation were evaluated during the analysis of the results of visualization and numerical simulations. These two effects sufficiently characterize the required course of the process from the physical point of view. The known scheme of the below description was adopted as a basic criterion for the evaluation of the degree of dispersion of gas bubbles in the model liquid.

  • Minimal dispersion—single bubbles ascending in the region of their formation along the ladle axis; lack of mixing in the whole bath volume.
  • Accurate dispersion—single and well-mixed bubbles ascending toward the bath mirror in the region of the ladle axis; no dispersion near the walls and in the lower part of the ladle.
  • Uniform dispersion—most desirable; very good mixing of fine bubbles with model liquid.
  • Excessive dispersion—bubbles join together to form chains; large turbulence zones; uneven flow of gas.

The numerical simulation results give a good agreement with the experiments performed with the physical model. For all studied variants (used process parameters), the single bubbles were observed in the area of water surface in the vessel. For variants presented in Figure 13 (200 rpm, gas flow 20 and dm3·min−1) and relevant examples in numerical simulation Figure 16, the worst bubble dispersion results were obtained because the dead zones were identified in the area near the bottom and sidewalls of the vessel, which disqualifies these work parameters for further use. The areas where swirls and gas bubble chains formed were identified only for the inert gas flows of 20 and 30 dm3·min−1 and 200 rpm in the analyzed model (physical model). This means that the presented impeller model had the best performance in terms of dispersion of gas bubbles in the model liquid. The worst bubble dispersion results were obtained because the dead zones were identified in the area near the bottom and side walls of the vessel, which disqualifies these work parameters for further use.

Figure 17 presents exemplary results of model tests (CFD and physical model) with marked gas bubble dispersion zones. All variants of tests were analogously compared, and this comparison allowed validating the numerical model.

An external file that holds a picture, illustration, etc. Object name is materials-15-05273-g017.jpg

Figure 17

Compilations of model research results (CFD and physical): A—single gas bubbles formed on the surface of the modeling liquid, B—excessive formation of gas chains and swirls, C—uniform distribution of gas bubbles in the entire volume of the tank, and D—dead zones without gas bubbles, no dispersion. (a) Variant B; (b) variant F.

It should be mentioned here that, in numerical simulations, it is necessary to make certain assumptions and simplifications. The calculations assumed three particle size classes (Table 2), which represent the different gas bubbles that form due to different gas flow rates. The maximum number of particles/bubbles (Table 1) generated was assumed in advance and related to the computational capabilities of the computer. Too many particles can also make it difficult to visualize and analyze the results. The size of the particles, of course, affects their behavior during simulation, while, in the figures provided in the article, the bubbles are represented by spheres (visualization of the results) of the same size. Please note that, due to the adopted Lagrangian–Eulerian approach, the simulation did not take into account phenomena such as bubble collapse or fusion. However, the obtained results allow a comprehensive analysis of the behavior of gas bubbles in the system under consideration.

The comparative analysis of the visualization (quantitative) results obtained with the water model and CFD simulations (see Figure 17) generated a sufficient agreement from the point of view of the trends. A precise quantitative evaluation is difficult to perform because of the lack of a refraction compensating system in the water model. Furthermore, in numerical simulations, it is not possible to determine the geometry of the forming gas bubbles and their interaction with each other as opposed to the visualization in the water model. The use of both research methods is complementary. Thus, a direct comparison of images obtained by the two methods requires appropriate interpretation. However, such an assessment gives the possibility to qualitatively determine the types of the present gas bubble dispersion, thus ultimately validating the CFD results with the water model.

A summary of the visualization results for impellers RT3, i.e., analysis of the occurring gas bubble dispersion types, is presented in Table 8.

Table 8

Summary of visualization results (impeller RT3)—different types of gas bubble dispersion.

No Exp.ABCDEF
Gas flow rate, dm3·min−11030
Impeller speed, rpm200300500200300500
Type of dispersionAccurateUniformUniform/excessiveMinimalExcessiveExcessive

Open in a separate window

Tests carried out for impeller RT3 confirmed the high efficiency of gas bubble distribution in the volume of the tested object at a low inert gas flow rate of 10 dm3·min−1. The most optimal variant was variant B (300 rpm, 10 dm3·min−1). However, the other variants A and C (gas flow rate 10 dm3·min−1) seemed to be favorable for this type of impeller and are recommended for further testing. The above process parameters will be analyzed in detail in a quantitative analysis to be performed on the basis of the obtained efficiency curves of the degassing process (oxygen removal). This analysis will give an unambiguous answer as to which process parameters are the most optimal for this type of impeller; the results are planned for publication in the next article.

It should also be noted here that the high agreement between the results of numerical calculations and physical modelling prompts a conclusion that the proposed approach to the simulation of a degassing process which consists of a single-phase flow model with a free surface and a particle flow model is appropriate. The simulation results enable us to understand how the velocity field in the fluid is formed and to analyze the distribution of gas bubbles in the system. The simulations in Flow-3D software can, therefore, be useful for both the design of the impeller geometry and the selection of process parameters.

Go to:

4. Conclusions

The results of experiments carried out on the physical model of the device for the simulation of barbotage refining of aluminum revealed that the worst results in terms of distribution and dispersion of gas bubbles in the studied object were obtained for the black impellers variants B4, B8, and B12 (multi-orifice impellers—four, eight, and 12 outlet holes, respectively).

In this case, the control of flow, speed, and number of gas exit orifices did not improve the process efficiency, and the developed design did not meet the criteria for industrial tests. In the case of the ‘red triangle’ impeller (variant RT3), uniform gas bubble dispersion was achieved throughout the volume of the modeling fluid for most of the tested variants. The worst bubble dispersion results due to the occurrence of the so-called dead zones in the area near the bottom and sidewalls of the vessel were obtained for the flow variants of 20 dm3·min−1 and 200 rpm and 30 dm3·min−1 and 200 rpm. For the analyzed model, areas where swirls and gas bubble chains were formed were found only for the inert gas flow of 20 and 30 dm3·min−1 and 200 rpm. The model impeller (variant RT3) had the best performance compared to the previously presented impellers in terms of dispersion of gas bubbles in the model liquid. Moreover, its design differed from previously presented models because of its sharp edges. This can be advantageous for gas bubble dispersion, but may negatively affect its suitability in industrial conditions due to premature wearing.

The CFD simulation results confirmed the results obtained from the experiments performed on the physical model. The numerical simulation of the operation of the ‘red triangle’ impeller model (using Flow-3D software) gave good agreement with the experiments performed on the physical model. This means that the presented model impeller, as compared to other (analyzed) designs, had the best performance in terms of gas bubble dispersion in the model liquid.

In further work, the developed numerical model is planned to be used for CFD simulations of the gas bubble distribution process taking into account physicochemical parameters of liquid aluminum based on industrial tests. Consequently, the obtained results may be implemented in production practice.

Go to:

Funding Statement

This paper was created with the financial support grants from the AGH-UST, Faculty of Foundry Engineering, Poland (16.16.170.654 and 11/990/BK_22/0083) for the Faculty of Materials Engineering, Silesian University of Technology, Poland.

Go to:

Author Contributions

Conceptualization, K.K. and D.K.; methodology, J.P. and T.M.; validation, M.S. and S.G.; formal analysis, D.K. and T.M.; investigation, J.P., K.K. and S.G.; resources, M.S., J.P. and K.K.; writing—original draft preparation, D.K. and T.M.; writing—review and editing, D.K. and T.M.; visualization, J.P., K.K. and S.G.; supervision, D.K.; funding acquisition, D.K. and T.M. All authors have read and agreed to the published version of the manuscript.

Go to:

Institutional Review Board Statement

Not applicable.

Go to:

Informed Consent Statement

Not applicable.

Go to:

Data Availability Statement

Data are contained within the article.

Go to:

Conflicts of Interest

The authors declare no conflict of interest.

Go to:

Footnotes

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Go to:

References

1. Zhang L., Xuewei L., Torgerson A.T., Long M. Removal of Impurity Elements from Molten Aluminium: A Review. Miner. Process. Extr. Metall. Rev. 2011;32:150–228. doi: 10.1080/08827508.2010.483396. [CrossRef] [Google Scholar]

2. Saternus M. Impurities of liquid aluminium-methods on their estimation and removal. Met. Form. 2015;23:115–132. [Google Scholar]

3. Żak P.L., Kalisz D., Lelito J., Gracz B., Szucki M., Suchy J.S. Modelling of non-metallic particle motion process in foundry alloys. Metalurgija. 2015;54:357–360. [Google Scholar]

4. Kalisz D., Kuglin K. Efficiency of aluminum oxide inclusions rmoval from liquid steel as a result of collisions and agglomeration on ceramic filters. Arch. Foundry Eng. 2020;20:43–48. [Google Scholar]

5. Kuglin K., Kalisz D. Evaluation of the usefulness of rotors for aluminium refining. IOP Conf. Ser. Mater. Sci. Eng. 2021;1178:012036. doi: 10.1088/1757-899X/1178/1/012036. [CrossRef] [Google Scholar]

6. Saternus M., Merder T. Physical modeling of the impeller construction impact o the aluminium refining process. Materials. 2022;15:575. doi: 10.3390/ma15020575. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

7. Saternus M., Merder T. Physical modelling of aluminum refining process conducted in batch reactor with rotary impeller. Metals. 2018;8:726. doi: 10.3390/met8090726. [CrossRef] [Google Scholar]

8. Saternus M., Merder T., Pieprzyca J. The influence of impeller geometry on the gas bubbles dispersion in uro-200 reactor—RTD curves. Arch. Metall. Mater. 2015;60:2887–2893. doi: 10.1515/amm-2015-0461. [CrossRef] [Google Scholar]

9. Hernández-Hernández M., Camacho-Martínez J., González-Rivera C., Ramírez-Argáez M.A. Impeller design assisted by physical modeling and pilot plant trials. J. Mater. Process. Technol. 2016;236:1–8. doi: 10.1016/j.jmatprotec.2016.04.031. [CrossRef] [Google Scholar]

10. Mancilla E., Cruz-Méndez W., Garduño I.E., González-Rivera C., Ramírez-Argáez M.A., Ascanio G. Comparison of the hydrodynamic performance of rotor-injector devices in a water physical model of an aluminum degassing ladle. Chem. Eng. Res. Des. 2017;118:158–169. doi: 10.1016/j.cherd.2016.11.031. [CrossRef] [Google Scholar]

11. Michalek K., Socha L., Gryc K., Tkadleckova M., Saternus M., Pieprzyca J., Merder T. Modelling of technological parameters of aluminium melt refining in the ladle by blowing of inert gas through the rotating impeller. Arch. Metall. Mater. 2018;63:987–992. [Google Scholar]

12. Walek J., Michalek K., Tkadlecková M., Saternus M. Modelling of Technological Parameters of Aluminium Melt Refining in the Ladle by Blowing of Inert Gas through the Rotating Impeller. Metals. 2021;11:284. doi: 10.3390/met11020284. [CrossRef] [Google Scholar]

13. Michalek K., Gryc K., Moravka J. Physical modelling of bath homogenization in argon stirred ladle. Metalurgija. 2009;48:215–218. [Google Scholar]

14. Michalek K. The Use of Physical Modeling and Numerical Optimization for Metallurgical Processes. VSB; Ostrawa, Czech Republic: 2001. [Google Scholar]

15. Chen J., Zhao J. Light Metals. TMS; Warrendale, PA, USA: 1995. Bubble distribution in a melt treatment water model; pp. 1227–1231. [Google Scholar]

16. Saternus M. Model Matematyczny do Sterowania Procesem Rafinacji Ciekłych Stopów Aluminium Przy Zastosowaniu URO-200. Katowice, Poland: 2004. Research Project Nr 7 T08B 019 21. [Google Scholar]

17. Pietrewicz L., Wężyk W. Urządzenia do rafinacji gazowej typu URO-200 sześć lat produkcji i doświadczeń; Proceedings of the Aluminum Conference; Zakopane, Poland. 12–16 October 1998. [Google Scholar]

18. Flow3d User’s Guide. Flow Science, Inc.; Santa Fe, NM, USA: 2020. [Google Scholar]

19. Sinelnikov V., Szucki M., Merder T., Pieprzyca J., Kalisz D. Physical and numerical modeling of the slag splashing process. Materials. 2021;14:2289. doi: 10.3390/ma14092289. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

20. White F. Fluid Mechanics. McGraw-Hill; New York, NY, USA: 2010. (McGraw-Hill Series in Mechanical Engineering). [Google Scholar]

21. Yang Z., Yang L., Cheng T., Chen F., Zheng F., Wang S., Guo Y. Fluid Flow Characteristic of EAF Molten Steel with Different Bottom-Blowing Gas Flow Rate Distributions. ISIJ. 2020;60:1957–1967. doi: 10.2355/isijinternational.ISIJINT-2019-794. [CrossRef] [Google Scholar]

22. Nichols B.D., Hirt C.W. Methods for calculating multi-dimensional, transient free surface flows past bodies; Proceedings of the First International Conference on Numerical Ship Hydrodynamics; Gaithersburg, MD, USA. 20–22 October 1975. [Google Scholar]

23. Hirt C.W., Nichols B.D. Volume of Fluid (VOF) Method for the Dynamics of Free Boundaries. J. Comput. Phys. 1981;39:201–255. doi: 10.1016/0021-9991(81)90145-5. [CrossRef] [Google Scholar]

24. Szucki M., Suchy J.S., Lelito J., Malinowski P., Sobczyk J. Application of the lattice Boltzmann method for simulation of the mold filling process in the casting industry. Heat Mass Transf. 2017;53:3421–3431. doi: 10.1007/s00231-017-2069-5. [CrossRef] [Google Scholar]

25. Themelis N.J., Goyal P. Gas injection in steelmaking. Candian Metall. Trans. 1983;22:313–320. [Google Scholar]

26. Zhang L., Jing X., Li Y., Xu Z., Cai K. Mathematical model of decarburization of ultralow carbon steel during RH treatment. J. Univ. Sci. Technol. Beijing. 1997;4:19–23. [Google Scholar]

27. Chiti F., Paglianti A., Bujalshi W. A mechanistic model to estimate powder consumption and mixing time in aluminium industries. Chem. Eng. Res. Des. 2004;82:1105–1111. doi: 10.1205/cerd.82.9.1105.44156. [CrossRef] [Google Scholar]

28. Bouaifi M., Roustan M. Power consumption, mixing time and homogenization energy in dual-impeller agitated gas-liquid reactors. Chem. Eng. Process. 2011;40:87–95. doi: 10.1016/S0255-2701(00)00128-8. [CrossRef] [Google Scholar]

29. Kang J., Lee C.H., Haam S., Koo K.K., Kim W.S. Studies on the overall oxygen transfer rate and mixing time in pilot-scale surface aeration vessel. Environ. Technol. 2001;22:1055–1068. doi: 10.1080/09593332208618215. [PubMed] [CrossRef] [Google Scholar]

30. Moucha T., Linek V., Prokopov E. Gas hold-up, mixing time and gas-liquid volumetric mass transfer coefficient of various multiple-impeller configurations: Rushton turbine, pitched blade and techmix impeller and their combinations. Chem. Eng. Sci. 2003;58:1839–1846. doi: 10.1016/S0009-2509(02)00682-6. [CrossRef] [Google Scholar]

31. Szekely J. Flow phenomena, mixing and mass transfer in argon-stirred ladles. Ironmak. Steelmak. 1979;6:285–293. [Google Scholar]

32. Iguchi M., Nakamura K., Tsujino R. Mixing time and fluid flow phenomena in liquids of varying kinematic viscosities agitated by bottom gas injection. Metall. Mat. Trans. 1998;29:569–575. doi: 10.1007/s11663-998-0091-1. [CrossRef] [Google Scholar]

33. Hjelle O., Engh T.A., Rasch B. Removal of Sodium from Aluminiummagnesium Alloys by Purging with Cl2. Aluminium-Verlag GmbH; Dusseldorf, Germany: 1985. pp. 343–360. [Google Scholar]

34. Zhang L., Taniguchi S. Fundamentals of inclusion removal from liquid steel by bubble flotation. Int. Mat. Rev. 2000;45:59–82. doi: 10.1179/095066000101528313. [CrossRef] [Google Scholar]

원문 보기
Figure 1: Mold drawings

3D Flow and Temperature Analysis of Filling a Plutonium Mold

플루토늄 주형 충전의 3D 유동 및 온도 분석

Authors: Orenstein, Nicholas P. [1]

Publication Date:2013-07-24
Research Org.: Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.: DOE/LANL
OSTI Identifier: 1088904
Report Number(s): LA-UR-13-25537
DOE Contract Number: AC52-06NA25396
Resource Type: Technical Report
Country of Publication: United States
Language: English
Subject: Engineering(42); Materials Science(36); Radiation Chemistry, Radiochemistry, & Nuclear Chemistry(38)

Introduction

The plutonium foundry at Los Alamos National Laboratory casts products for various special nuclear applications. However, plutonium’s radioactivity, material properties, and security constraints complicate the ability to perform experimental analysis of mold behavior. The Manufacturing Engineering and Technologies (MET-2) group previously developed a graphite mold to vacuum cast small plutonium disks to be used by the Department of Homeland Security as point sources for radiation sensor testing.

A two-stage pouring basin consisting of a funnel and an angled cavity directs the liquid into a vertical runner. A stack of ten disk castings connect to the runner by horizontal gates. Volumetric flow rates were implemented to limit overflow into the funnel and minimize foundry returns. Models using Flow-3D computational fluid dynamics software are employed here to determine liquid Pu flow paths, optimal pour regimes, temperature changes, and pressure variations.

Setup

Hardcopy drawings provided necessary information to create 3D .stl models for import into Flow-3D (Figs. 1 and 2). The mesh was refined over several iterations to isolate the disk cavities, runner, angled cavity, funnel, and input pour. The final flow and mold-filling simulation utilizes a fine mesh with ~5.5 million total cells. For the temperature study, the mesh contained 1/8 as many cells to reduce computational time and set temperatures to 850 °C for the molten plutonium and 500 °C for the solid graphite mold components (Fig. 3).

Flow-3D solves mass continuity and Navier-Stokes momentum equations over the structured rectangular grid model using finite difference and finite volume numerical algorithms. The solver includes terms in the momentum equation for body and viscous accelerations and uses convective heat transfer.

Simulation settings enabled Flow-3D physics calculations for gravity at 980.665 cm/s 2 in the negative Z direction (top of mold to bottom); viscous, turbulent, incompressible flow using dynamically-computed Renormalized Group Model turbulence calculations and no-slip/partial slip wall shear, and; first order, full energy equation heat transfer.

Mesh boundaries were all set to symmetric boundary conditions except for the Zmin boundary set to outflow and the Zmax boundary set to a volume flow. Vacuum casting conditions and the high reactivity of remaining air molecules with Pu validate the assumption of an initially fluidless void.

Results

The flow follows a unique three-dimensional path. The mold fills upwards with two to three disks receiving fluid in a staggered sequence. Figures 5-9 show how the fluid fills the cavity, and Figure 7 includes the color scale for pressure levels in these four figures. The narrow gate causes a high pressure region which forces the fluid to flow down the cavity centerline.

It proceeds to splash against the far wall and then wrap around the circumference back to the gate (Figs. 5 and 6). Flow in the angled region of the pouring basin cascades over the bottom ledge and attaches to the far wall of the runner, as seen in Figure 7.

This channeling becomes less pronounced as fluid volume levels increase. Finally, two similar but non-uniform depressed regions form about the centerline. These regions fill from their perimeter and bottom until completion (Fig. 8). Such a pattern is counter, for example, to a steady scenario in which a circle of molten Pu encompassing the entire bottom surface rises as a growing cylinder.

Cavity pressure becomes uniform when the cavity is full. Pressure levels build in the rising well section of the runner, where impurities were found to settle in actual casting. Early test simulations optimized the flow as three pours so that the fluid would never overflow to the funnel, the cavities would all fill completely, and small amounts of fluid would remain as foundry returns in the angled cavity.

These rates and durations were translated to the single 2.7s pour at 100 cm 3 per second used here. Figure 9 shows anomalous pressure fluctuations which occurred as the cavities became completely filled. Multiple simulations exhibited a rapid change in pressure from positive to negative and back within the newly-full disk and surrounding, already-full disks.

The time required to completely fill each cavity is plotted in Figure 10. Results show negligible temperature change within the molten Pu during mold filling and, as seen in Figure 11, at fill completion.

Figure 1: Mold drawings
Figure 1: Mold drawings
Figure 2: Mold Assembly
Figure 2: Mold Assembly
Figure 4: Actual mold and cast Pu
Figure 4: Actual mold and cast Pu
Figure 5: Bottom cavity filling from runner
Figure 5: Bottom cavity filling from runner
Figure 6: Pouring and filling
Figure 6: Pouring and filling
Figure 8: Edge detection of cavity fill geometry. Two similar depressed areas form about the centerline. Top cavity shown; same pressure scale as other figures
Figure 8: Edge detection of cavity fill geometry. Two similar depressed areas form about the centerline. Top cavity shown; same pressure scale as other figures
Figure 10: Cavity fill times,from first fluid contact with pouring basin, Figure 11:Fluid temperature remains essentially constant
Figure 10: Cavity fill times,from first fluid contact with pouring basin, Figure 11:Fluid temperature remains essentially constant

Conclusions

Non-uniform cavity filling could cause crystal microstructure irregularities during solidification. However, the small temperature changes seen – due to large differences in specific heat between Pu and graphite – over a relatively short time make such problems unlikely in this case.

In the actual casting, cooling required approximately ten minutes. This large difference in time scales further reduces the chance for temperature effects in such a superheated scenario. Pouring basin emptying decreases pressure at the gate which extends fill time of the top two cavities.

The bottom cavity takes longer to fill because fluid must first enter the runner and fill the well. Fill times continue linearly until the top two cavities. The anomalous pressure fluctuations may be due to physical attempts by the system to reach equilibrium, but they are more likely due to numerical errors in the Flow3D solver.

Unsuccessful tests were performed to remove them by halving fluid viscosity. The fine mesh reduced, but did not eliminate, the extent of the fluctuations. Future work is planned to study induction and heat transfer in the full Pu furnace system, including quantifying temporal lag of the cavity void temperature to the mold wall temperature during pre-heat and comparing heat flux levels between furnace components during cool-down.

Thanks to Doug Kautz for the opportunity to work with MET-2 and for assigning an interesting unclassified project. Additional thanks to Mike Bange for CFD guidance, insight of the project’s history, and draft review.

원본 보기
Figure 1.| Physical models of the vertical drop, backdrop and stepped drop developed in the Technical University of Lisbon.

Numerical modelling of air-water flows in sewer drops

하수구 방울의 공기-물 흐름 수치 모델링

Paula Beceiro (corresponding author)
Maria do Céu Almeida
Hydraulic and Environment Department (DHA), National Laboratory for Civil Engineering, Avenida do Brasil 101, 1700-066 Lisbon, Portugal
E-mail: pbeceiro@lnec.pt
Jorge Matos
Department of Civil Engineering, Arquitecture and Geosources,
Technical University of Lisbon (IST), Avenida Rovisco Pais 1, 1049-001 Lisbon, Portugal

ABSTRACT

물 흐름에 용존 산소(DO)의 존재는 해로운 영향의 발생을 방지하는 데 유익한 것으로 인식되는 호기성 조건을 보장하는 중요한 요소입니다.

하수도 시스템에서 흐르는 폐수에 DO를 통합하는 것은 공기-액체 경계면 또는 방울이나 접합부와 같은 특이점의 존재로 인해 혼입된 공기를 통한 연속 재방출의 영향을 정량화하기 위해 광범위하게 조사된 프로세스입니다. 공기 혼입 및 후속 환기를 향상시키기 위한 하수구 드롭의 위치는 하수구의 호기성 조건을 촉진하는 효과적인 방법입니다.

본 논문에서는 수직 낙하, 배경 및 계단식 낙하를 CFD(전산유체역학) 코드 FLOW-3D®를 사용하여 모델링하여 이러한 유형의 구조물의 존재로 인해 발생하는 난류로 인한 공기-물 흐름을 평가했습니다. 이용 가능한 실험적 연구에 기초한 수력학적 변수의 평가와 공기 혼입의 분석이 수행되었습니다.

이러한 구조물에 대한 CFD 모델의 결과는 Soares(2003), Afonso(2004) 및 Azevedo(2006)가 개발한 해당 물리적 모델에서 얻은 방류, 압력 헤드 및 수심의 측정을 사용하여 검증되었습니다.

유압 거동에 대해 매우 잘 맞았습니다. 수치 모델을 검증한 후 공기 연행 분석을 수행했습니다.

The presence of dissolved oxygen (DO) in water flows is an important factor to ensure the aerobic conditions recognised as beneficial to prevent the occurrence of detrimental effects. The incorporation of DO in wastewater flowing in sewer systems is a process widely investigated in order to quantify the effect of continuous reaeration through the air-liquid interface or air entrained due the presence of singularities such as drops or junctions. The location of sewer drops to enhance air entrainment and subsequently reaeration is an effective practice to promote aerobic conditions in sewers. In the present paper, vertical drops, backdrops and stepped drop was modelled using the computational fluid dynamics (CFD) code FLOW-3D® to evaluate the air-water flows due to the turbulence induced by the presence of this type of structures. The assessment of the hydraulic variables and an analysis of the air entrainment based in the available experimental studies were carried out. The results of the CFD models for these structures were validated using measurements of discharge, pressure head and water depth obtained in the corresponding physical models developed by Soares (2003), Afonso (2004) and Azevedo (2006). A very good fit was obtained for the hydraulic behaviour. After validation of numerical models, analysis of the air entrainment was carried out.

Key words | air entrainment, computational fluid dynamics (CFD), sewer drops

Figure 1.| Physical models of the vertical drop, backdrop and stepped drop developed in the Technical University of Lisbon.
Figure 1.| Physical models of the vertical drop, backdrop and stepped drop developed in the Technical University of Lisbon.
Figure 3. Comparison between the experimental and numerical pressure head along of the invert of the outlet pipe.
Figure 3. Comparison between the experimental and numerical pressure head along of the invert of the outlet pipe.
Figure 4. Average void fraction along the longitudinal axis of the outlet pipe for the lower discharges in the vertical drop and backdrop.
Figure 4. Average void fraction along the longitudinal axis of the outlet pipe for the lower discharges in the vertical drop and backdrop.

REFERENCES

Afonso, J. Dissipação de energia e rearejamento em quedas em colectores. M.Sc. Thesis, UTL/IST, Lisboa, Portugal.
Almeida, M. C., Butler, D. & Matos, J. S. Reaeration by sewer drops. In: 8th Int. Conf. on Urban Storm Drainage, Sydney, Australia.
Azevedo, R. I. Transferência de oxigénio em quedas guiadas em colectores. M.Sc. Thesis, IST, Lisboa, Portugal.
Beceiro, P., Almeida, M. C. & Matos, J. Numerical Modelling of air-water flows in a vertical drop and a backdrop. In: 3rd IAHR Europe Congress, Porto, Portugal.
Bombardelli, F. A., Meireles, I. & Matos, J. S. Laboratory measurements and multi-block numerical simulations of the mean flow and turbulence in the non-aerated skimming flow region of step stepped spillways. Environ. Fluid Mech. 11 (3), 263–288.
Brethour, J. M. & Hirt, C. W. Drift Model for TwoComponent Flows. Flow Science, Inc., Los Alamos, NM, USA.
Chamani, M. R. Jet Flow on Stepped Spillways and Drops. M.Sc. Thesis, University of Alberta, Alberta, Canada.
Chanson, H. Air Bubble Entrainment in Free-Surface Turbulent Shear Flow. Academic Press Inc., California, USA.
Chanson, H. Air bubble entrainment in open channels: flow structure and bubble size distribution. Int. J. Multiphase 23 (1), 193–203.
Chanson, H. Hydraulics of aerated flows: qui pro quo? Journal of Hydraulic Research 51 (3), 223–243.
Dufresne, M., Vazques, J., Terfous, A., Ghenaim, A. & Poulet, J. Experimental investigation and CFD modelling of flow, sedimentation, and solids separation in a combined sewer detention tank. Computer and Fluids 38, 1042–1049.
Durve, A. P. & Patwardhan, A. W. Numerical and experimental investigation of onset of gas entrainment phenomenon. Chemical Engineering Science 73, 140–150.
Felder, S. & Chanson, H. Air–water flows and free-surface profiles on a non-uniform stepped chute. Journal of Hydraulic Research 52 (2), 253–263.
Flow Science FLOW-3D User’s Manuals Version 10.0. Vol.1/2. Flow Science Inc., Los Alamos, NM, USA.
Granata, F., Marinis, G., Gargano, R. & Hager, W. H. Energy loss in circular drop manholes. In: 33rd IAHR Congress: Water Engineering for Sustainable Environment, British
Columbia, Vancouver, Canada. Hirt, C. W. Modeling Turbulent Entrainment of air at A Free Surface. Flow Science Inc., Los Alamos, NM, USA.
Hirt, C. W. & Nichols, B. D. Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics 39, 201–225.
Hirt, C. W. & Sicilian, J. M. A porosity technique for the definition of obstacles in rectangular cell meshes. In: Proc. 4th Int, Conf. Ship Hydro., National Academy of Science, Washington, DC, USA.
Isfahani, A. H. G. & Brethour, J. On the Implementation of Two-Equation Turbulence Models in FLOW-3D. Flow Science Inc., Los Alamos, NM, USA.
Kouyi, G. L., Bret, P., Didier, J. M., Chocat, B. & Billat, C. The use of CFD modelling to optimise measurement of overflow rates in a downstream-controlled dual-overflow structure. Water Science and Technology 64 (2), 521–527.
Lopes, P., Leandro, J., Carvalho, R. F., Páscoa, P. & Martins, R. Numerical and experimental investigation of a gully under surcharge conditions. Urban Water Journal 12 (6), 468–476.
Martins, R., Leandro, J. & Carvalho, R. F. Characterization of the hydraulic performance of a gully under drainage conditions. Water Science and Technology 69 (12), 2423–2430.
Matias, N., Nielsel, A. H., Vollertsen, J., Ferreira, F. & Matos, J. S. Reaeration and hydrogen sulfide release at drop structures. In: 8th International Conference on Sewer Processes and Networks (SPN8), Rotterdam, Netherlands.
Matos, J. S. & Sousa, E. R. Prediction of dissolved oxygen concentration along sanitary sewers. Water Science and Technology 34 (5–6), 525–532.
Mignot, E., Bonakdari, H., Knothe, P., Lipeme Kouyi, G., Bessette, A., Rivière, N. & Bertrand-Krajewski, J. L. Experiments and 3D simulations of flow structures in junctions and of their influence on location of flowmeters. In: 12th International Conference on Urban Drainage, Porto Alegre, Brazil.
Ozmen-Cagatay, H. & Kocaman, S. Dam-break flow in the presence of obstacle: experiment and CFD Simulation. Engineering Applications of Computational Fluid Mechanics 5 (4), 541–552.
Shojaee Fard, M. H. & Boyaghchi, F. A. Studies of the influence of various blade outlet angles in a centrifugal pump when handling viscous fluids. American Journal of Applied Sciences 4 (9), 718–724.
Soares, A. Rearejamento em Quedas em Colectores de Águas Residuais. M.Sc. Thesis, FCTUC, Coimbra, Portugal.
Sousa, C. M. & Lopes, R. R. Hidráulica e rearejamento em quedas verticais em colectores. Estudo Experimental. Research Report, UTL/IST, Lisboa, Portugal.
Sousa, V., Meireles, I., Matos, J. & Almeida, M. C. Numerical modelling of air-water flow in a vertical drop manhole. In: 7th International Conference on Sewer Processes and Networks (SPN7), Shefield, UK.
Stovin, V., Guymer, I. & Lau, S. D. Approaches to validating a 3D CFD manhole model. In: 11th International Conference on Urban Drainage, Edinburgh, Scotland, UK.
Tota, P. V. Turbulent Flow Over A Backward-Facing Step Using the RNG Model. Flow Science Inc., Los Alamos, NM, USA.
Valero, D. & García-Bartual, R. Calibration of an air entrainment model for CFD spillway applications. In: Advances in Hydroinformatics. Springer, Singapore, pp. 571–582.
Versteeg, H. K. & Malalasekera, W. An Introduction to Computational Fluid Dynamics. The Finite Volume Method. Longman Group limited, England.
Yang, Y., Yang, J., Zuo, J., Li, Y., He, S., Yang, X. & Zhang, K. Study on two operating conditions of a full-scale oxidation ditch for optimization of energy consumption and effluent quality by using CFD model. Water Research 45 (11), 3439–3452.
Zhai, A. J., Zhang, Z., Zhang, W. & Chen, Q. Y. Evaluation of various turbulence models in predicting airflow and turbulence in enclosed environments by CFD: part 1— summary of prevalent Turbulence models. HVAC&R Research 13 (6), 853–870.
Zhao, C., Zhu, D. Z. & Rajaratnam, N. Computational and experimental study of surcharged flow at a 90W combining sewer junction. Journal of Hydraulic Engineering 134 (6), 688–700.

원문보기
Fig. 1 Geometrical 3D model of Caisson

환기실에서 의도된 삼중수소 방출 후 삼중수소 거동 시뮬레이션

Simulation of Tritium Behavior after Intended Tritium Release in Ventilated Room


Yasunori IWAI, Takumi HAYASHI, Toshihiko YAMANISHI, Kazuhiro KOBAYASHI & Masataka NISHI

Abstract

일본원자력연구소(JAERI) 산하 삼중수소공정연구소(TPL)에서는 핵융합로의 안전성 확인 및 강화를 위해 12m3의 대형 밀폐용기(Caisson)로 삼중수소 안전 연구(CATS)용 케이슨 조립체를 제작하여 추정 삼중수소 누출 이벤트가 발생해야 하는 경우 삼중수소 거동. 본 연구의 주요 목적 중 하나는 환기실에서 삼중수소 누출 사건이 발생한 후 삼중수소 거동을 예측하기 위한 시뮬레이션 방법을 확립하는 것입니다.

RNG 모델은 허용 가능한 엔지니어링 정밀도로 50m3/h 환기 케이슨에서 맴돌이 흐름 계산에 유효한 것으로 밝혀졌습니다. 의도된 삼중수소 방출 후 계산된 초기 및 제거 삼중수소 농도 이력은 50m3/h 환기 케이슨에서 실험 관찰과 일치했습니다.

환기실의 삼중수소 수송에는 벽 근처의 흐름이 중요한 역할을 하는 것으로 밝혀졌다. 한편, 3,000m3의 삼중수소 취급실에서 의도적으로 방출된 삼중수소 거동은 미일 협력하에 실험적으로 조사되었습니다. 동일한 방법으로 계산된 삼중수소 농도 이력은 실험적 관찰과 일치하였으며, 이는 현재 개발된 방법이 삼중수소 취급실의 실제 규모에 적용될 수 있음을 입증한다.

At the Tritium Process Laboratory (TPL) at the Japan Atomic Energy Research Institute (JAERI), Caisson Assembly for Tritium Safety study (CATS) with 12 m3 of large airtight vessel (Caisson) was fabricated for confirmation and enhancement of fusion reactor safety to estimate tritium behavior in the case where a tritium leak event should happen. One of the principal objectives of the present studies is the establishment of simulation method to predict the tritium behavior after the tritium leak event should happen in a ventilated room. The RNG model was found to be valid for eddy flow calculation in the 50m3/h ventilated Caisson with acceptable engineering precision. The calculated initial and removal tritium concentration histories after intended tritium release were consistent with the experimental observations in the 50 m3/h ventilated Caisson. It is found that the flow near a wall plays an important role for the tritium transport in the ventilated room. On the other hand, tritium behavior intentionally released in the 3,000 m3 of tritium handling room was investigated experimentally under a US-Japan collaboration. The tritium concentration history calculated with the same method was consistent with the experimental observations, which proves that the present developed method can be applied to the actual scale of tritium handling room.

KEYWORDS: 

Fig. 1 Geometrical 3D model of Caisson
Fig. 1 Geometrical 3D model of Caisson
Fig. 2 Geometrical 3D model of "main cell" of TSTA
Fig. 2 Geometrical 3D model of “main cell” of TSTA

REFERENCES

(1) Los Alamos National Laboratory: Final Safety Analysis Report of Tritium Systems Test Assembly at the Los Alamos National Laboratory, TSTA-SAR, (1996).
(2) Naruse, Y., Matsuda, Y., Tanaka, K.: Fusion Eng. Des., 12, 293 (1990).
(3) Schira, P., Hutter, E., Jourdan, G., Penzhone, R.: Fu-VOL. 38, NO. 1, JANUARY 2001 sion Eng. Des., 18, 19 (1991).
(4) Bartlit, J. R., Anderson, J. L., Jalbert, R. A., Carl-son, R. V., Okuno, K., Ide, T., Fukui, H., Enoeda, M., Naruse, Y.: Proc. 13th SOFE, Knoxville, TN., U.S.A., 798 (1989).
(5) Hayashi, T., Kobayashi, K., Iwai, Y., Yamanishi, T., Nishi, M., Okuno, K., Carlson, R. V., Willms, R. S., Hyatt, D., Roybal, B.: Fusion Thecnol., 34, 521 (1998).
(6) Hayashi, T., Kobayashi, K., Iwai, Y., Yamada, M., Suzuki, T., O’hira, S., Nakamura, H., Shu, W., Yama-nishi, T., Kawamura, Y., Isobe, K., Konishi, S., Nishi, M.: Submitted to Fusion Eng. Des. (1999).
(7) Yakhot, V., Orgazag, S. A.: J. Sci. Comput., 1, 3 (1986).
(8) Hirt, C. W., Cook, J. L.: J. Comp. Phys., 10, 324 (1972).
(9) Daiguji, H., Miyake, Y., Yoshizawa, A.: “Computa-tional Fluid Dynamics of Turbulent Flow-Models and Numerical Methods”, Univ. of Tokyo Press, Tokyo, 183 (1998), [in Japanese].
(10) Hinze, J. 0.: “Turbulence”, McGraw-Hill, New York, 227 (1959).
(11) Launder, B. E., Spalding, D. B.: “Mathematical Models of Turbulence”, Academics, London, (1972).
(12) Jones, W. P., Launder, B. E.: Int. J. Heat Mass Trans-fer, 15, 301 (1972).
(13) Jones, W. P., Launder, B. E.: Int. J. Heat Mass Trans-fer, 16, 1119 (1973).
(14) Hanjalic, K., Launder, B. E.: J. Fluid Mech., 52, 609 (1972).
(15) Harlow, F. W., Nakayama, P. I.: Phys. Fluids, 10, 2303 (1967).
(16) Nakayama, P. I.: 8th Aerospace Science Meeting, AIAA paper No. 70-3, (1970).
(17) Boussinesq, J.: “Theorie Analytique de la chaleur”, Ganthier-Villars, Paris, 157 (1903).
(18) Hashimoto, K.: “Chemical Reaction Engineering”, (1st ed.), Baifuukan, Tokyo, 173 (1979), [in Japanese].
(19) Iwai, Y., Hayashi, T., Kobayashi, K., O’hira, S., Nishi, M.: Submitted to Fusion Eng. Des. (2000).

원본보기
Figura 7. Influencia del modelo de turbulencia. Qmodelo=27.95l/s.

Flow-3D를 사용하여 전산유체역학(CFD)을 적용한 빠른 단계의 플러시 유동 수치 모델링

Numerical Modeling of Flush Flow in a Rapid Step Applying Computational Fluid Dynamics (CFD) Using Flow-3D.

레브 폴리텍. (Quito) [온라인]. 2018, vol.41, n.2, pp.53-64. ISSN 2477-8990.

이 프로젝트의 주요 목표는 FLOW-3D를 사용하여 계단식 여수로에서 스키밍 흐름의 수치 모델링을 개발하는 것입니다. 이러한 구조의 설계는 물리적 모델링에서 얻은 경험적 표현과 CFD 코드를 지원하는 계단식 여수로를 통한 흐름의 수치 모델링에서 보완 연구를 기반으로 합니다. 수치 모델은 균일한 영역의 유속과 계단 여수로의 마찰 계수를 추정하는 데 사용됩니다(ϴ = 45º, Hd=4.61m). 흐름에 대한 자동 통기의 표현은 복잡하므로 프로그램은 공기 연행 모델을 사용하여 특정 제한이 있는 솔루션에 근접합니다.

The main objective of this project is to develop the numerical modeling of the skimming flow in a stepped spillway using FLOW-3D. The design of these structures is based on the use of empirical expressions obtained from physical modeling and complementary studies in the numerical modeling of flow over the stepped spillway with support of CFD code. The numerical model is used to estimate the flow velocity in the uniform region and the friction coefficient of the stepped spillway (ϴ = 45º, Hd=4.61m). The representation of auto aeration a flow is complex, so the program approximates the solution with certain limitations, using an air entrainment model; drift flux model and turbulence model k-ԑ RNG. The results obtained with numerical modeling and physical modeling at the beginning of natural auto aeration of flow and depth of the biphasic flow in the uniform region presents deviations above to 10% perhaps the flow is highly turbulent.

Keywords : Stepped spillway; skimming flow; air entrainment; drift flux; numerical modeling; FLOW-3D.

Keywords : 계단식 여수로; 스키밍 흐름; 공기 연행; 드리프트 플럭스; 수치 모델링; 흐름-3D.· 

스페인어로 된 초록 · 스페인어 로 된 텍스트 · 스페인어로 된 텍스트( pdf ) 

Figure 1. Grazing flow over a rapid step.
Figure 1. Grazing flow over a rapid step.
Figura 2. Principales regiones existentes en un flujo rasante.
Figura 2. Principales regiones existentes en un flujo rasante.
Figure 3. Dimensions of the El Batán stepped rapid.
Figure 3. Dimensions of the El Batán stepped rapid.
Figure 4. 3D physical model of the El Batán stepped rapid
Figure 4. 3D physical model of the El Batán stepped rapid
Figura 7. Influencia del modelo de turbulencia. Qmodelo=27.95l/s.
Figura 7. Influencia del modelo de turbulencia. Qmodelo=27.95l/s.

REFERENCIAS

ARAGUA. (2013). “Modelación numérica y experimental de flujos aire-agua
en caídas en colectores.”, Laboratório Nacional de Engenharia Civil, I.
P. Av do Brasil 101 • 1700-066 Lisboa.
Bombardelli, F.A., Meireles, I. and Matos, J., (2010), “Laboratory
measurement and multi-block numerical simulations of the mean flow
and turbulence in the non-aerated skimming flow region of steep stepped
spillways”, Environ Fluid Mechanics.
Castro M. (2015) “Análisis Dimensional y Modelación física en Hidráulica”.
Escuela Politécnica Nacional. Quito Ecuador. 50 p.
Chanson H., D. B. Bung., J. Matos (2015). “Stepped spillways and cascades”.
IAHR Monograph. School of Civil Engineering, University of
Queensland, Brisbane, Australia.
Chanson H. (1993). “Stepped Spillway Flows and Air Entrainment.” Can. Jl
of Civil Eng., Vol. 20, No. 3, June, pp. 422-435 (ISSN 0315-1468).
CIERHI, EPN TECH, (2016). “Estudio experimental en modelo físico de las
rápidas con perfil escalonado y liso de la quebrada el Batán Fase I y Fase
II”, Escuela Politécnica Nacional, Quito Ecuador.
Fernández Oro J. M. (2012)., “Técnicas Numéricas en Ingeniería de Fluidos:
Introducción a la Dinámica de Fluidos Computacional (CFD) por el
Método de Volúmenes Finitos”. Barcelona: Reverté.
Flow Science, Inc. (2012). “FLOW 3D 10.1.0 Documentation Release.
Manual de Usuario”, Los Alamos National Laboratory. Santa Fe, New
México
Khatsuria, R.M., (2005)., “Hydraulics of Spillways and Energy Dissipators”.
Department of Civil and Environmental Engineering Georgia Institute
of Technology Atlanta.
Lucio I., Matos J., Meireles I. (2015). “Stepped spillway flow over small
embankment dams: some computational experiments”. 15th FLOW-3D
European users conference.
Mohammad S., Jalal A. and Michael P., (2012). “Numerical Computation of
Inception Point Location for Steeply Sloping Stepped Spillways” 9th
International Congress on Civil Engineering. Isfahan University of
Technology (IUT), Isfahan, Iran
Pfister M., Chanson H., (2013), “Scale Effects in Modelling Two-phase Airwater Flows”, Proceedings of 2013 IAHR World Congress.
Sarfaraz, M. and Attari, J. (2011), “Numerical Simulation of Uniform Flow
Region over a Steeply Sloping Stepped Spillway”, 6th National
Congress on Civil Engineering, Semnan University, Semnan, Iran.
Valero, D., Bung, D., (2015), “Hybrid investigation of air transport processes
in moderately sloped stepped spillway flows”, E-proceedings of the 36th
IAHR World Congress 28 June – 3 July, 2015, The Hague, the Netherlands.

원문보기

Figure 2. Different PKW Types.

A review of Piano Key Weir as a superior alternative for dam rehabilitation

댐 복구를 위한 우수한 대안으로서의 Piano Key Weir에 대한 검토

Amiya Abhash &

K. K. Pandey

Pages 541-551 | Received 03 Mar 2020, Accepted 07 May 2020, Published online: 21 May 2020

ABSTRACT

Dams fall in ‘installations containing dangerous forces’ because of their massive impact on the environment and civilian life and property as per International humanitarian law. As such, it becomes vital for hydraulic engineers to refurbish various solutions for dam rehabilitation. This paper presents a review of a new type of weir installation called Piano Key Weir (PKW), which is becoming popular around the world for its higher spillway capacity both for existing and new dam spillway installations. This paper reviews the geometry along with structural integrity, discharging capacity, economic aspects, aeration requirements, sediment transport and erosion aspects of Piano Key Weir (PKW) as compared with other traditional spillway structures and alternatives from literature. The comparison with other alternatives shows PKW to be an excellent alternative for dam risk mitigation owing to its high spillway capabilities and economy, along with its use in both existing and new hydraulic structures.

댐은 국제 인도법에 따라 환경과 민간인 생활 및 재산에 막대한 영향을 미치기 때문에 ‘위험한 힘을 포함하는 시설물’에 속합니다. 따라서 유압 엔지니어는 댐 복구를 위한 다양한 솔루션을 재정비해야 합니다.

이 백서에서는 PKW(Piano Key Weir)라는 새로운 유형의 둑 설치에 대한 검토를 제공합니다. PKW는 기존 및 신규 댐 방수로 설치 모두에서 더 높은 방수로 용량으로 전 세계적으로 인기를 얻고 있습니다.

이 백서에서는 구조적 무결성, 배출 용량, 경제적 측면, 폭기 요구 사항, 퇴적물 운반 및 PKW(Piano Key Weir)의 침식 측면과 함께 다른 전통적인 여수로 구조 및 문헌의 대안과 비교하여 기하학을 검토합니다.

다른 대안과의 비교는 PKW가 높은 여수로 기능과 경제성으로 인해 댐 위험 완화를 위한 탁월한 대안이며 기존 및 새로운 수력 구조물 모두에 사용됨을 보여줍니다.

KEYWORDS: 

Figure 2. Different PKW Types.
Figure 2. Different PKW Types.

References

  • Anderson, R., and Tullis, B. (2011). Influence of Piano Key Weir geometry on discharge. Proc. Int. Conf. Labyrinth and Piano Key Weirs Liège B. Liège, Belgium. [Crossref], [Google Scholar]
  • Anderson, R., and Tullis, B. (2012a). “Piano key weir hydraulics and labyrinth weir comparison”. J. Irrig. Drain. Eng., 139(3), 246–253. doi:https://doi.org/10.1061/(ASCE)IR.1943-4774.0000530 [Crossref], [Web of Science ®], [Google Scholar]
  • Anderson, R., and Tullis, B. (2012b). “Piano key weir: Reservoir versus channel application”. J. Irrig. Drain. Eng., 138(8), 773–776. doi:https://doi.org/10.1061/(ASCE)IR.1943-4774.0000464 [Crossref], [Web of Science ®], [Google Scholar]
  • Anderson, R.M. 2011. Piano key weir head discharge relationships, M.S. Thesis, Utah State University, Logan, Utah. [Google Scholar]
  • Bashiri, H., Dewals, B., Pirotton, M., Archambeau, P., and Erpicum, S. (2016). “Towards a new design equation for piano key weirs discharge capacity.” Proc. of the 6th International Symposium on Hydraulic Structures. Portland, USA. [Google Scholar]
  • Bianucci, S.P., Sordo Ward, Á.F., Pérez Díaz, J.I., García-Palacios, J.H., Mediero Orduña, L.J., and Garrote de Marcos, L. (2013). “Risk-based methodology for parameter calibration of a reservoir flood control model”. Natl. Hazard Earth Syst. Sci., 13(4), 965–981. doi:https://doi.org/10.5194/nhess-13-965-2013 [Crossref], [Web of Science ®], [Google Scholar]
  • Blancher, B., Montarros, F., and Laugier, F. (2011). Hydraulic comparison between Piano Key Weirs and labyrinth spillways. Proc. Int. Conf. Labyrinth and Piano Key Weirs Liège B. Liège, Belgium. [Crossref], [Google Scholar]
  • Botha, A., Fitz, I., Moore, A., Mulder, F., and Van Deventer, N. 2013. “Application of the Piano Key Weir spillway in the Republic of South Africa”. Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs, Chatou, Paris, France, 20–22, 185. [Crossref], [Google Scholar]
  • Chahartaghi, M.K., Nazari, S., and Shooshtari, M.M. 2019. “Experimental and numerical simulation of arced trapezoidal Piano Key Weirs”. Flow Meas. Instrum., 68, 101576. doi:https://doi.org/10.1016/j.flowmeasinst.2019.101576 [Crossref], [Web of Science ®], [Google Scholar]
  • Chi Hien, T., Thanh Son, H., and Ho Ta Khanh, M. (2006). Results of some ‘piano keys’ weir hydraulic model tests in Vietnam. Proc., 22nd Int. Congress of Large Dams, Question 87, Response 39, International Commission on Large Dams (ICOLD). Barcelona, Spain. [Google Scholar]
  • Cicero, G., Barcouda, M., Luck, M., and Vettori, E. (2011). Study of a piano key morning glory to increase the spillway capacity of the Bage dam. Proc. Int. Conf. Labyrinth Piano Key Weirs-PKW2011, Taylor & Francis, London. [Crossref], [Google Scholar]
  • Cicero, G., De Miranda, D., and Luck, M. (2012). “Assessment of the code Wolf 1D PKW for predicting the hydraulic behaviour of PK-Weirs.” Congrès SHF-33èmes journées de l’hydraulique “Grands aménagements hydrauliques 2012”, Paris, France. [Google Scholar]
  • Cicero, G., and Delisle, J. (2013). “Discharge characteristics of Piano Key weirs under submerged flow”. Labyrinth and Piano Key Weirs II–PKW 2013, 101–109. [Crossref], [Google Scholar]
  • Cicero, G., Delisle, J., Lefebvre, V., and Vermeulen, J. (2013). “Experimental and numerical study of the hydraulic performance of a trapezoidal Piano Key weir.” Labyrinth and Piano Key Weirs II: Proceedings of the Second International Workshop on Labyrinth and Piano key weirs, Chatou, Paris, France, 20–22, 265. [Crossref], [Google Scholar]
  • Cicéro, G., Guene, C., Luck, M., Pinchard, T., Lochu, A., and Brousse, P. (2010). “Experimental optimization of a Piano Key Weir to increase the spillway capacity of the Malarce dam.” 1st IAHR European Congress, Edinbourgh, Mai 4–6, 2010. [Google Scholar]
  • Crookston, B., Anderson, R., and Tullis, B. (2018). “Free-flow discharge estimation method for Piano Key weir geometries.” J. Hydro. Environ. Res., 19, 160–167. doi:https://doi.org/10.1016/j.jher.2017.10.003 [Crossref], [Web of Science ®], [Google Scholar]
  • Das Singhal, G., and Sharma, N. 2011. “Rehabilitation of Sawara Kuddu Hydroelectric Project–Model studies of Piano Key Weir in India”. Proc. Int. Workshop on Labyrinths and Piano Key Weirs PKW 2011. Taylor & Francis, London. [Crossref], [Google Scholar]
  • Denys, F., Basson, G., and Strasheim, J. (2017). Fluid Structure Interaction of Piano Key Weirs. Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam. [Crossref], [Google Scholar]
  • Eichenberger, P. (2013). “The first commercial piano key weir in Switzerland.” Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 227. [Crossref], [Google Scholar]
  • Erpicum, S., Laugier, F., Pfister, M., Pirotton, M., Cicero, G.-M., and Schleiss, A.J. 2013. Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, CRC Press. [Crossref], [Google Scholar]
  • Erpicum, S., Machiels, O., Dewals, B., Pirotton, M., and Archambeau, P. (2012). “Numerical and physical hydraulic modelling of Piano Key Weirs.” Proceedings of the 4th Int. Conf. on Water Resources and Renewable Energy Development in Asia. Chiang Mai, Thailande. [Google Scholar]
  • Erpicum, S., Nagel, V., and Laugier, F. (2011). “Piano Key Weir design study at Raviege dam”. Labyrinth and Piano Key Weirs–PKW 2011, 43–50. [Crossref], [Google Scholar]
  • Ervine, D., and Elsawy, E. (1975). “The effect of a falling nappe on river aeration.” Proc. 16th IAHR Congress, Sao Paulo, Brazil. [Google Scholar]
  • Falvey, H.T. 1980. “Air-water flow in hydraulic structures”. NASA STI/Recon Technical Report N, 81. [Google Scholar]
  • Gabriel-Martin, I., Sordo-Ward, A., Garrote, L., and Castillo, L.G. (2017). “Influence of initial reservoir level and gate failure in dam safety analysis. Stochastic approach.” J. Hydrol., 550, 669–684. doi:https://doi.org/10.1016/j.jhydrol.2017.05.032 [Crossref], [Web of Science ®], [Google Scholar]
  • Gebhardt, M., Herbst, J., Merkel, J., and Belzner, F. (2019). “Sedimentation at labyrinth weirs–an experimental study of the self-cleaning process”. J. Hydraulic Res., 57(4), 579–590. doi:https://doi.org/10.1080/00221686.2018.1494053 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Hu, H., Qian, Z., Yang, W., Hou, D., and Du, L. (2018). “Numerical study of characteristics and discharge capacity of piano key weirs.” Flow Meas. Instrum., 62, 27–32. doi:https://doi.org/10.1016/j.flowmeasinst.2018.05.004 [Crossref], [Web of Science ®], [Google Scholar]
  • Javaheri, A., and Kabiri-Samani, A. (2012). “Threshold submergence of flow over PK weirs”. Int. J. Civil Geol. Eng., 6, 46–49. [Google Scholar]
  • Jayatillake, H., and Perera, K. (2013). “Design of a Piano-Key Weir for Giritale Dam spillway in Sri Lanka.” Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 151. [Crossref], [Google Scholar]
  • Jayatillake, H., and Perera, K. (2017). “Adoption of a type D Piano Key Weir spillway with tapered noses at Rambawa Tank, Sri Lanka.” Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam. [Crossref], [Google Scholar]
  • Jüstrich, S., Pfister, M., and Schleiss, A.J. (2016). “Mobile riverbed scour downstream of a Piano Key weir”. J. Hydraulic Eng., 142(11), 04016043. doi:https://doi.org/10.1061/(ASCE)HY.1943-7900.0001189 [Crossref], [Google Scholar]
  • Kabiri-Samani, A., and Javaheri, A. (2012). “Discharge coefficients for free and submerged flow over Piano Key weirs”. J. Hydraulic Res., 50(1), 114–120. doi:https://doi.org/10.1080/00221686.2011.647888 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Karimi, M., Attari, J., Saneie, M., and Jalili Ghazizadeh, M.R. (2018). “Side weir flow characteristics: comparison of piano key, labyrinth, and linear types”. J. Hydraulic Eng., 144(12), 04018075. doi:https://doi.org/10.1061/(ASCE)HY.1943-7900.0001539 [Crossref], [Google Scholar]
  • Karimi, M., Attari, J., Saneie, M., and Jalili-Ghazizadeh, M. (2017). “Experimental study of discharge coefficient of a piano key side weir.” Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017). Proceedings of the Third International Workshop on Labyrinth and Piano key weirs 2017, Qui Nhon, Vietnam, 22–24. [Crossref], [Google Scholar]
  • Khanh, M.H.T. (2013). “The Piano Key Weirs: 15 years of Research & Development–Prospect.” Labyrinth and piano key weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 3. [Crossref], [Google Scholar]
  • Khanh, M.H.T. (2017). “History and development of Piano Key Weirs in Vietnam from 2004 to 2016.” Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam. [Google Scholar]
  • Khanh, M.H.T., Hien, T.C., and Hai, N.T. (2011). “Main results of the PK weir model tests in Vietnam (2004 to 2010).” Labyrinth and Piano Key Weirs, 191. Liège, Belgium. [Crossref], [Google Scholar]
  • Khassaf, S.I., Aziz, L.J., and Elkatib, Z.A. (2016). “Hydraulic behavior of piano key weir type B under free flow conditions”. Int. J. Sci. Technol. Res., 5(3), 158–163. [Google Scholar]
  • Khassaf, S.I., and Al-Baghdadi, M.B. (2015). “Experimental study of non-rectangular piano key weir discharge coefficient”. J. Homepage, 6(5), 425–436. [Google Scholar]
  • Khassaf, S.I., and Al-Baghdadi, M.B.N. (2018). “Experimental investigation of submerged flow over piano key weir”. Int. J. Energy Environ., 9(3), 249–260. [Google Scholar]
  • Kwon, -H.-H., and Moon, Y.-I. (2006). “Improvement of overtopping risk evaluations using probabilistic concepts for existing dams”. Stochastic Environ. Res. Risk Assess., 20(4), 223. doi:https://doi.org/10.1007/s00477-005-0017-2 [Crossref], [Web of Science ®], [Google Scholar]
  • Laugier, F. (2007). “Design and construction of the first Piano Key Weir spillway at Goulours dam”. Int. J. Hydropower Dams, 14(5), 94. [Google Scholar]
  • Laugier, F., Lochu, A., Gille, C., Leite Ribeiro, M., and Boillat, J.-L. (2009). “Design and construction of a labyrinth PKW spillway at Saint-Marc dam, France”. Hydropower Dams, 16(LCH–ARTICLE–2009–023), 100–107. [Google Scholar]
  • Laugier, F., Pralong, J., and Blancher, B. (2011). “Influence of structural thickness of sidewalls on PKW spillway discharge capacity.” Proc. Intl Workshop on Labyrinths and Piano Key Weirs PKW 2011. Liège, Belgium. [Crossref], [Google Scholar]
  • Le Blanc, M., Spinazzola, U., and Kocahan, H. (2011). “Labyrinth fusegate applications on free overflow spillways–Overview of recent projects.” Labyrinth and Piano Key Weirs, 261, Liège, Belgium. [Crossref], [Google Scholar]
  • Leite Ribeiro, M., Bieri, M., Boillat, J.-L., Schleiss, A., Delorme, F., and Laugier, F. (2009). “Hydraulic capacity improvement of existing spillways–design of a piano key weirs.” Proc. (on CD) of the 23rd Congress of the Int. Commission on Large Dams CIGB-ICOLD. Brasilia, Brazil. [Google Scholar]
  • Leite Ribeiro, M., Bieri, M., Boillat, J.-L., Schleiss, A., Singhal, G., and Sharma, N. (2011). “Discharge capacity of piano key weirs”. J. Hydraulic Eng., 138(2), 199–203. doi:https://doi.org/10.1061/(ASCE)HY.1943-7900.0000490 [Crossref], [Google Scholar]
  • Lempérière, F., and Ouamane, A. (2003). “The Piano Keys weir: a new cost-effective solution for spillways”. Int. J. Hydropower Dams, 10(5), 144–149. [Google Scholar]
  • Lempérière, F., and Vigny, J. (2011). “General comments on labyrinth and Piano Keys Weirs–The future”. Labyrinth and Piano Key weirs–PKW 2011, 289–294. [Crossref], [Google Scholar]
  • Lempérière, F., Vigny, J., and Ouamane, A. (2011). General comments on Labyrinth and Piano Key Weirs: The past and present. Proc. Intl. Conf. Labyrinth and Piano Key Weirs, Liège B. Liège, Belgium. [Crossref], [Google Scholar]
  • Lewin, J., Ballard, G., and Bowles, D.S. (2003). “Spillway gate reliability in the context of overall dam failure risk.” USSD Annual Lecture, Charleston, South Carolina. [Google Scholar]
  • Lodomez, M., Pirotton, M., Dewals, B., Archambeau, P., and Erpicum, S. (2017). “Could piano key weirs be subject to nappe oscillations?” Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam [Crossref], [Google Scholar]
  • Machiels, O., Erpicum, S., Archambeau, P., Dewals, B., and Pirotton, M. (2009). “Large scale experimental study of piano key weirs.” Proc. 33rd IAHR Congress: Water Engineering for a Sustainable Environment, IAHR. Vancouver, Canada [Google Scholar]
  • Machiels, O., Erpicum, S., Archambeau, P., Dewals, B., and Pirotton, M. (2011a). “Piano Key Weir preliminary design method–Application to a new dam project.” Proc. Int. Conf. Labyrinth and Piano Key Weirs Liège B. Liège, Belgium. [Crossref], [Google Scholar]
  • Machiels, O., Erpicum, S., Dewals, B., Archambeau, P., and Pirotton, M. (2010). “Piano Key Weirs: The experimental study of an efficient solution for rehabilitation”. WIT Trans. Ecol., 133, 95–106. [Crossref], [Google Scholar]
  • Machiels, O., Erpicum, S., Dewals, B.J., Archambeau, P., and Pirotton, M. (2011b). “Experimental observation of flow characteristics over a Piano Key Weir”. J Hydraulic Res, 49(3), 359–366. doi:https://doi.org/10.1080/00221686.2011.567761 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Machiels, O., Pirotton, M., Pierre, A., Dewals, B., and Erpicum, S. (2014). “Experimental parametric study and design of Piano Key Weirs”. J. Hydraulic Res., 52(3), 326–335. doi:https://doi.org/10.1080/00221686.2013.875070 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Mehboudi, A., Attari, J., and Hosseini, S. (2016). “Experimental study of discharge coefficient for trapezoidal piano key weirs.” Flow Meas. Instrum., 50, 65–72. doi:https://doi.org/10.1016/j.flowmeasinst.2016.06.005 [Crossref], [Web of Science ®], [Google Scholar]
  • Micovic, Z., Hartford, D.N., Schaefer, M.G., and Barker, B.L. (2016). “A non-traditional approach to the analysis of flood hazard for dams”. Stochastic Environ. Res. Risk Assess., 30(2), 559–581. doi:https://doi.org/10.1007/s00477-015-1052-2 [Crossref], [Web of Science ®], [Google Scholar]
  • Monjezi, R., Heidarnejad, M., Masjedi, A., Purmohammadi, M.H., and Kamanbedast, A. (2018). “Laboratory investigation of the discharge coefficient of flow in arced labyrinth weirs with triangular plans.” Flow Meas. Instrum., 64, 64–70. doi:https://doi.org/10.1016/j.flowmeasinst.2018.10.011 [Crossref], [Web of Science ®], [Google Scholar]
  • Noseda, M., Stojnic, I., Pfister, M., and Schleiss, A.J. (2019). “Upstream Erosion and sediment passage at piano key weirs”. J. Hydraulic Eng., 145(8), 04019029. doi:https://doi.org/10.1061/(ASCE)HY.1943-7900.0001616 [Crossref], [Google Scholar]
  • Oertel, M. (2015). “Discharge coefficients of piano key weirs from experimental and numerical modelS.” E= proceedings of the 36th IAHR world congress. 28 June – 3 July, The Hague, The Netherlands. [Google Scholar]
  • Ouamane, A. (2011). Nine years of study of the Piano Key Weir in the university laboratory of Biskra “lessons and reflections”. Proc. Int. Conf. Labyrinth Piano Key Weirs-PKW2011, Taylor & Francis, London. [Crossref], [Google Scholar]
  • Ouamane, A., Debabeche, M., Lempérière, F., and Vigny, J. (2017). Twenty years of research in Biskra University for Labyrinths and Piano Key Weirs and associated fuse plugs. Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam. [Crossref], [Google Scholar]
  • Ouamane, A., and Lempérière, F. (2006). Design of a new economic shape of weir. Proc. Int. Symp. on Dams in the Societies of the 21st Century. Barcelona, Spain. [Crossref], [Google Scholar]
  • Patev, R., and Putcha, C. (2005). “Development of fault trees for risk assessment of dam gates and associated operating equipment”. Int. J. Modell. Simul., 25(3), 190–201. doi:https://doi.org/10.1080/02286203.2005.11442336 [Taylor & Francis Online], [Google Scholar]
  • Paxson, G., Tullis, B., and Hertel, D. 2013. “Comparison of Piano Key Weirs with labyrinth and gated spillways: Hydraulics, cost, constructability and operations”. Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 123–130. [Crossref], [Google Scholar]
  • Pfister, M., Capobianco, D., Tullis, B., and Schleiss, A.J. (2013). “Debris-blocking sensitivity of piano key weirs under reservoir-type approach flow”. J. Hydraulic Eng., 139(11), 1134–1141. doi:https://doi.org/10.1061/(ASCE)HY.1943-7900.0000780 [Crossref], [Google Scholar]
  • Phillips, M., and Lesleighter, E. 2013. “Piano Key Weir spillway: Upgrade option for a major dam”. Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 159–168. [Crossref], [Google Scholar]
  • Pinchard, T., Boutet, J., and Cicero, G. (2011). “Spillway capacity upgrade at Malarce dam: design of an additional Piano Key Weir spillway.” Proc. Int. Workshop on Labyrinths and Piano Key Weirs PKW. Liège, Belgium. [Crossref], [Google Scholar]
  • Pralong, J., J. Vermeulen, B. Blancher, F. Laugier, S. Erpicum, O. Machiels, M. Pirotton, J.-L. Boillat, M. Leite Ribeiro and A. Schleiss (2011). “A naming convention for the piano key weirs geometrical parameters.” Labyrinth and piano key weirs, 271–278. [Crossref], [Google Scholar]
  • Ribeiro, M.L., Boillat, J.-L., Schleiss, A., Laugier, F., and Albalat, C. (2007). “Rehabilitation of St-Marc dam.” Experimental optimization of a piano key weir. Proc. of 32nd Congress of IAHR, Vince, Italy. [Google Scholar]
  • Ribeiro, M.L., Pfister, M., and Schleiss, A.J. (2013). “Overview of Piano Key weir prototypes and scientific model investigations”. Labyrinth and Piano Key Weirs II, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 273. [Crossref], [Google Scholar]
  • Ribeiro, M.L., Pfister, M., Schleiss, A.J., and Boillat, J.-L. (2012). “Hydraulic design of A-type piano key weirs”. J. Hydraulic Res., 50(4), 400–408. doi:https://doi.org/10.1080/00221686.2012.695041 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Ribi, J., Spahni, B., Dorthe, D., and Pfister, M. (2017). Piano Key Weir as overflow on sedimentation basin of wastewater treatment plant. Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam [Crossref], [Google Scholar]
  • Schleiss, A. (2011). “From labyrinth to piano key weirs: a historical review.” Proc. Int. Conf. Labyrinth and Piano Key Weirs Liège B. Liège, Belgium. [Crossref], [Google Scholar]
  • Sharma, N., and Tiwari, H. (2013). “Experimental study on vertical velocity and submergence depth near Piano Key Weir.” Labyrinth and Piano Key Weirs II-PKW, Proceedings of the Second International Workshop on Labyrinth and Piano key weirs 2013, Chatou, Paris, France, 20–22, 93–100. [Crossref], [Google Scholar]
  • Tiwari, H. (2016). Experimental Study of Turbulence Characteristics Near Piano Key Weir. PhD, Indian Institute of Technology Roorkee. [Google Scholar]
  • Tiwari, H., and Sharma, N. 2017. “Empirical and Mathematical Modeling of Head and Discharge Over Piano Key Weir”. Development of Water Resources in India. Springer, Cham. 341–354. https://doi.org/10.1007/978-3-319-55125-8_29 [Crossref], [Google Scholar]
  • Valley, P., and Blancher, B. (2017). Construction and testing of two Piano Key Weirs at Charmines dam. Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), Feb 22–24, 2017, CRC Press, Qui Nhon, Vietnam. [Crossref], [Google Scholar]
  • Vermeulen, J., Lassus, C., and Pinchard, T. (2017). Design of a Piano Key Weir aeration network. Labyrinth and Piano Key Weirs III: Proceedings of the 3rd International Workshop on Labyrinth and Piano Key Weirs (PKW 2017), February 22- 24,2017, Qui Nhon, Vietnam, CRC Press. [Crossref], [Google Scholar]
  • Vermeulen, J., Laugier, F., Faramond, L., and Gille, C. (2011). “Lessons learnt from design and construction of EDF first Piano Key Weirs”. Labyrinth and Piano Key weirs-PKW 2011, 215–224. [Crossref], [Google Scholar]
원문보기
Sketch of approach channel and spillway of the Kamal-Saleh dam

CFD modeling of flow pattern in spillway’s approach channel

Sustainable Water Resources Management volume 1, pages245–251 (2015)Cite this article

Abstract

Analysis of behavior and hydraulic characteristics of flow over the dam spillway is a complicated task that takes lots of money and time in water engineering projects planning. To model those hydraulic characteristics, several methods such as physical and numerical methods can be used. Nowadays, by utilizing new methods in computational fluid dynamics (CFD) and by the development of fast computers, the numerical methods have become accessible for use in the analysis of such sophisticated flows. The CFD softwares have the capability to analyze two- and three-dimensional flow fields. In this paper, the flow pattern at the guide wall of the Kamal-Saleh dam was modeled by Flow 3D. The results show that the current geometry of the left wall causes instability in the flow pattern and making secondary and vortex flow at beginning approach channel. This shape of guide wall reduced the performance of weir to remove the peak flood discharge.

댐 여수로 흐름의 거동 및 수리학적 특성 분석은 물 공학 프로젝트 계획에 많은 비용과 시간이 소요되는 복잡한 작업입니다. 이러한 수력학적 특성을 모델링하기 위해 물리적, 수치적 방법과 같은 여러 가지 방법을 사용할 수 있습니다. 요즘에는 전산유체역학(CFD)의 새로운 방법을 활용하고 빠른 컴퓨터의 개발로 이러한 정교한 흐름의 해석에 수치 방법을 사용할 수 있게 되었습니다. CFD 소프트웨어에는 2차원 및 3차원 유동장을 분석하는 기능이 있습니다. 본 논문에서는 Kamal-Saleh 댐 유도벽의 흐름 패턴을 Flow 3D로 모델링하였다. 결과는 왼쪽 벽의 현재 형상이 흐름 패턴의 불안정성을 유발하고 시작 접근 채널에서 2차 및 와류 흐름을 만드는 것을 보여줍니다. 이러한 형태의 안내벽은 첨두방류량을 제거하기 위해 둑의 성능을 저하시켰다.

Introduction

Spillways are one of the main structures used in the dam projects. Design of the spillway in all types of dams, specifically earthen dams is important because the inability of the spillway to remove probable maximum flood (PMF) discharge may cause overflow of water which ultimately leads to destruction of the dam (Das and Saikia et al. 2009; E 2013 and Novak et al. 2007). So study on the hydraulic characteristics of this structure is important. Hydraulic properties of spillway including flow pattern at the entrance of the guide walls and along the chute. Moreover, estimating the values of velocity and pressure parameters of flow along the chute is very important (Chanson 2004; Chatila and Tabbara 2004). The purpose of the study on the flow pattern is the effect of wall geometry on the creation transverse waves, flow instability, rotating and reciprocating flow through the inlet of spillway and its chute (Parsaie and Haghiabi 2015a, b; Parsaie et al. 2015; Wang and Jiang 2010). The purpose of study on the values of velocity and pressure is to calculate the potential of the structure to occurrence of phenomena such as cavitation (Fattor and Bacchiega 2009; Ma et al. 2010). Sometimes, it can be seen that the spillway design parameters of pressure and velocity are very suitable, but geometry is considered not suitable for conducting walls causing unstable flow pattern over the spillway, rotating flows at the beginning of the spillway and its design reduced the flood discharge capacity (Fattor and Bacchiega 2009). Study on spillway is usually conducted using physical models (Su et al. 2009; Suprapto 2013; Wang and Chen 2009; Wang and Jiang 2010). But recently, with advances in the field of computational fluid dynamics (CFD), study on hydraulic characterist–ics of this structure has been done with these techniques (Chatila and Tabbara 2004; Zhenwei et al. 2012). Using the CFD as a powerful technique for modeling the hydraulic structures can reduce the time and cost of experiments (Tabbara et al. 2005). In CFD field, the Navier–Stokes equation is solved by powerful numerical methods such as finite element method and finite volumes (Kim and Park 2005; Zhenwei et al. 2012). In order to obtain closed-form Navier–Stokes equations turbulence models, such k − ε and Re-Normalisation Group (RNG) models have been presented. To use the technique of computational fluid dynamics, software packages such as Fluent and Flow 3D, etc., are provided. Recently, these two software packages have been widely used in hydraulic engineering because the performance and their accuracy are very suitable (Gessler 2005; Kim 2007; Kim et al. 2012; Milési and Causse 2014; Montagna et al. 2011). In this paper, to assess the flow pattern at Kamal-Saleh guide wall, numerical method has been used. All the stages of numerical modeling were conducted in the Flow 3D software.

Materials and methods

Firstly, a three-dimensional model was constructed according to two-dimensional map that was prepared for designing the spillway. Then a small model was prepared with scale of 1:80 and entered into the Flow 3D software; all stages of the model construction was conducted in AutoCAD 3D. Flow 3D software numerically solved the Navier–Stokes equation by finite volume method. Below is a brief reference on the equations that used in the software. Figure 1 shows the 3D sketch of Kamal-Saleh spillway and Fig. 2 shows the uploading file of the Kamal-Saleh spillway in Flow 3D software.

figure 1
Fig. 1
figure 2
Fig. 2

Review of the governing equations in software Flow 3D

Continuity equation at three-dimensional Cartesian coordinates is given as Eq (1).

vf∂ρ∂t+∂∂x(uAx)+∂∂x(vAy)+∂∂x(wAz)=PSORρ,vf∂ρ∂t+∂∂x(uAx)+∂∂x(vAy)+∂∂x(wAz)=PSORρ,

(1)

where u, v, z are velocity component in the x, y, z direction; A x, A y, A z cross-sectional area of the flow; ρ fluid density; PSOR the source term; v f is the volume fraction of the fluid and three-dimensional momentum equations given in Eq (2).

∂u∂t+1vf(uAx∂u∂x+vAy∂u∂y+wAz∂u∂z)=−1ρ∂P∂x+Gx+fx∂v∂t+1vf(uAx∂v∂x+vAy∂v∂y+wAz∂v∂z)=−1ρ∂P∂y+Gy+fy∂w∂t+1vf(uAx∂w∂x+vAy∂w∂y+wAz∂w∂z)=−1ρ∂P∂y+Gz+fz,∂u∂t+1vf(uAx∂u∂x+vAy∂u∂y+wAz∂u∂z)=−1ρ∂P∂x+Gx+fx∂v∂t+1vf(uAx∂v∂x+vAy∂v∂y+wAz∂v∂z)=−1ρ∂P∂y+Gy+fy∂w∂t+1vf(uAx∂w∂x+vAy∂w∂y+wAz∂w∂z)=−1ρ∂P∂y+Gz+fz,

(2)

where P is the fluid pressure; G x, G y, G z the acceleration created by body fluids; f x, f y, f z viscosity acceleration in three dimensions and v f is related to the volume of fluid, defined by Eq. (3). For modeling of free surface profile the VOF technique based on the volume fraction of the computational cells has been used. Since the volume fraction F represents the amount of fluid in each cell, it takes value between 0 and 1.

∂F∂t+1vf[∂∂x(FAxu)+∂∂y(FAyv)+∂∂y(FAzw)]=0∂F∂t+1vf[∂∂x(FAxu)+∂∂y(FAyv)+∂∂y(FAzw)]=0

(3)

Turbulence models

Flow 3D offers five types of turbulence models: Prantl mixing length, k − ε equation, RNG models, Large eddy simulation model. Turbulence models that have been proposed recently are based on Reynolds-averaged Navier–Stokes equations. This approach involves statistical methods to extract an averaged equation related to the turbulence quantities.

Steps of solving a problem in Flow 3D software

(1) Preparing the 3D model of spillway by AutoCAD software. (2) Uploading the file of 3D model in Flow 3D software and defining the problem in the software and checking the final mesh. (3) Choosing the basic equations that should be solved. (4) Defining the characteristics of fluid. (5) Defining the boundary conditions; it is notable that this software has a wide range of boundary conditions. (6) Initializing the flow field. (7) Adjusting the output. (8) Adjusting the control parameters, choice of the calculation method and solution formula. (9) Start of calculation. Figure 1 shows the 3D model of the Kamal-Saleh spillway; in this figure, geometry of the left and right guide wall is shown.

Figure 2 shows the uploading of the 3D spillway dam in Flow 3D software. Moreover, in this figure the considered boundary condition in software is shown. At the entrance and end of spillway, the flow rate or fluid elevation and outflow was considered as BC. The bottom of spillway was considered as wall and left and right as symmetry.

Model calibration

Calibration of the Flow 3D for modeling the effect of geometry of guide wall on the flow pattern is included for comparing the results of Flow 3D with measured water surface profile. Calibration the Flow 3D software could be conducted in two ways: first, changing the value of upstream boundary conditions is continued until the results of water surface profile of the Flow 3D along the spillway successfully covered the measurement water surface profile; second is the assessment the mesh sensitivity. Analyzing the size of mesh is a trial-and-error process where the size of mesh is evaluated form the largest to the smallest. With fining the size of mesh the accuracy of model is increased; whereas, the cost of computation is increased. In this research, the value of upstream boundary condition was adjusted with measured data during the experimental studies on the scaled model and the mesh size was equal to 1 × 1 × 1 cm3.

Results and discussion

The behavior of water in spillway is strongly affected by the flow pattern at the entrance of the spillway, the flow pattern formation at the entrance is affected by the guide wall, and choice of an optimized form for the guide wall has a great effect on rising the ability of spillway for easy passing the PMF, so any nonuniformity in flow in the approach channel can cause reduction of spillway capacity, reduction in discharge coefficient of spillway, and even probability of cavitation. Optimizing the flow guiding walls (in terms of length, angle and radius) can cause the loss of turbulence and flow disturbances on spillway. For this purpose, initially geometry proposed for model for the discharge of spillway dam, Kamal-Saleh, 80, 100, and 120 (L/s) were surveyed. These discharges of flow were considered with regard to the flood return period, 5, 100 and 1000 years. Geometric properties of the conducting guidance wall are given in Table 1.Table 1 Characteristics and dimensions of the guidance walls tested

Full size table

Results of the CFD simulation for passing the flow rate 80 (L/s) are shown in Fig. 3. Figure 3 shows the secondary flow and vortex at the left guide wall.

figure 3
Fig. 3

For giving more information about flow pattern at the left and right guide wall, Fig. 4 shows the flow pattern at the right side guide wall and Fig. 5 shows the flow pattern at the left side guide wall.

figure 4
Fig. 4
figure 5
Fig. 5

With regard to Figs. 4 and 5 and observing the streamlines, at discharge equal to 80 (L/s), the right wall has suitable performance but the left wall has no suitable performance and the left wall of the geometric design creates a secondary and circular flow, and vortex motion in the beginning of the entrance of spillway that creates cross waves at the beginning of spillway. By increasing the flow rate (Q = 100 L/s), at the inlet spillway secondary flows and vortex were removed, but the streamline is severely distorted. Results of the guide wall performances at the Q = 100 (L/s) are shown in Fig. 6.

figure 6
Fig. 6

Also more information about the performance of each guide wall can be derived from Figs. 7 and 8. These figures uphold that the secondary and vortex flows were removed, but the streamlines were fully diverted specifically near the left side guide wall.

figure 7
Fig. 7
figure 8
Fig. 8

As mentioned in the past, these secondary and vortex flows and diversion in streamline cause nonuniformity and create cross wave through the spillway. Figure 9 shows the cross waves at the crest of the spillway.

figure 9
Fig. 9

The performance of guide walls at the Q = 120 (L/s) also was assessed. The result of simulation is shown in Fig. 10. Figures 11 and 12 show a more clear view of the streamlines near to right and left side guide wall, respectively. As seen in Fig. 12, the left side wall still causes vortex flow and creation of and diversion in streamline.

figure 10
Fig. 10
figure 11
Fig. 11
figure 12
Fig. 12

The results of the affected left side guide wall shape on the cross wave creation are shown in Fig. 13. As seen from Fig. 3, the left side guide wall also causes cross wave at the spillway crest.

figure 13
Fig. 13

As can be seen clearly in Figs. 9 and 13, by moving from the left side to the right side of the spillway, the cross waves and the nonuniformity in flow is removed. By reviewing Figs. 9 and 13, it is found that the right side guide wall removes the cross waves and nonuniformity. With this point as aim, a geometry similar to the right side guide wall was considered instead of the left side guide wall. The result of simulation for Q = 120 (L/s) is shown in Fig. 14. As seen from this figure, the proposed geometry for the left side wall has suitable performance smoothly passing the flow through the approach channel and spillway.

figure 14
Fig. 14

More information about the proposed shape for the left guide wall is shown in Fig. 15. As seen from this figure, this shape has suitable performance for removing the cross waves and vortex flows.

figure 15
Fig. 15

Figure 16 shows the cross section of flow at the crest of spillway. As seen in this figure, the proposed shape for the left side guide wall is suitable for removing the cross waves and secondary flows.

figure 16
Fig. 16

Conclusion

Analysis of behavior and hydraulic properties of flow over the spillway dam is a complicated task which is cost and time intensive. Several techniques suitable to the purposes of study have been undertaken in this research. Physical modeling, usage of expert experience, usage of mathematical models on simulation flow in one-dimensional, two-dimensional and three-dimensional techniques, are some of the techniques utilized to study this phenomenon. The results of the modeling show that the CFD technique is a suitable tool for simulating the flow pattern in the guide wall. Using this tools helps the designer for developing the optimal shape for hydraulic structure which the flow pattern through them are important.

References

  • Chanson H (2004) 19—Design of weirs and spillways. In: Chanson H (ed) Hydraulics of open channel flow, 2nd edn. Butterworth-Heinemann, Oxford, pp 391–430Chapter Google Scholar 
  • Chatila J, Tabbara M (2004) Computational modeling of flow over an ogee spillway. Comput Struct 82:1805–1812Article Google Scholar 
  • Das MM, Saikia MD (2009) Irrigation and water power engineering. PHI Learning, New DelhiGoogle Scholar 
  • E, Department Of Army: U.S. Army Corps (2013) Hydraulic Design of Spillways. BiblioBazaar, CharlestonGoogle Scholar 
  • Fattor C, Bacchiega J (2009) Design conditions for morning-glory spillways: application to potrerillos dam spillway. Adv Water Res Hydraul Eng Springer, Berlin, pp 2123–2128Google Scholar 
  • Gessler D (2005) CFD modeling of spillway performance. Impacts Glob Clim Change. doi:10.1061/40792(173)398
  • Kim D-G (2007) Numerical analysis of free flow past a sluice gate. KSCE J Civ Eng 11:127–132Article Google Scholar 
  • Kim D, Park J (2005) Analysis of flow structure over ogee-spillway in consideration of scale and roughness effects by using CFD model. KSCE J Civ Eng 9:161–169Article Google Scholar 
  • Kim S, Yu K, Yoon B, Lim Y (2012) A numerical study on hydraulic characteristics in the ice Harbor-type fishway. KSCE J Civ Eng 16:265–272Article Google Scholar 
  • Ma X-D, Dai G-Q, Yang Q, Li G-J, Zhao L (2010) Analysis of influence factors of cavity length in the spillway tunnel downstream of middle gate chamber outlet with sudden lateral enlargement and vertical drop aerator. J Hydrodyn Ser B 22:680–686Article Google Scholar 
  • Milési G, Causse S (2014) 3D numerical modeling of a side-channel spillway. In: Gourbesville P, Cunge J, Caignaert G (eds) Advances in hydroinformatics. Springer, Singapore, pp 487–498Chapter Google Scholar 
  • Montagna F, Bellotti G, Di Risio M (2011) 3D numerical modeling of landslide-generated tsunamis around a conical island. Nat Hazards 58:591–608Article Google Scholar 
  • Novak P, Moffat AIB, Nalluri C, Narayanan R (2007) Hydraulic structures. Taylor & Francis, LondonGoogle Scholar 
  • Parsaie A, Haghiabi A (2015a) Computational modeling of pollution transmission in rivers. Appl Water Sci. doi:10.1007/s13201-015-0319-6
  • Parsaie A, Haghiabi A (2015b) The effect of predicting discharge coefficient by neural network on increasing the numerical modeling accuracy of flow over side weir. Water Res Manag 29:973–985Article Google Scholar 
  • Parsaie A, Yonesi H, Najafian S (2015) Predictive modeling of discharge in compound open channel by support vector machine technique. Model Earth Syst Environ 1:1–6Article Google Scholar 
  • Su P-L, Liao H-S, Qiu Y, Li CJ (2009) Experimental study on a new type of aerator in spillway with low Froude number and mild slope flow. J Hydrodyn Ser B 21:415–422Article Google Scholar 
  • Suprapto M (2013) Increase spillway capacity using Labyrinth Weir. Procedia Eng 54:440–446Article Google Scholar 
  • Tabbara M, Chatila J, Awwad R (2005) Computational simulation of flow over stepped spillways. Comput Struct 83:2215–2224Article Google Scholar 
  • Wang J, Chen H (2009) Experimental study of elimination of vortices along guide wall of bank spillway. Adv Water Res Hydraul Eng Springer, Berlin, pp 2059–2063Google Scholar 
  • Wang Y, Jiang C (2010) Investigation of the surface vortex in a spillway tunnel intake. Tsinghua Sci Technol 15:561–565Article Google Scholar 
  • Zhenwei MU, Zhiyan Z, Tao Z (2012) Numerical simulation of 3-D flow field of spillway based on VOF method. Procedia Eng 28:808–812Article Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Water Engineering, Lorestan University, Khorram Abad, IranAbbas Parsaie, Amir Hamzeh Haghiabi & Amir Moradinejad

Corresponding author

Correspondence to Abbas Parsaie.

Reprints and Permissions

About this article

Cite this article

Parsaie, A., Haghiabi, A.H. & Moradinejad, A. CFD modeling of flow pattern in spillway’s approach channel. Sustain. Water Resour. Manag. 1, 245–251 (2015). https://doi.org/10.1007/s40899-015-0020-9

Download citation

  • Received28 April 2015
  • Accepted28 August 2015
  • Published15 September 2015
  • Issue DateSeptember 2015
  • DOIhttps://doi.org/10.1007/s40899-015-0020-9

Share this article

Anyone you share the following link with will be able to read this content:Get shareable link

Provided by the Springer Nature SharedIt content-sharing initiative

Keywords

  • Approach channel
  • Kamal-Saleh dam
  • Guide wall
  • Flow pattern
  • Numerical modeling
  • Flow 3D software
    원문 보기
    Figure 10. Flow distribution at the approach channel in PMF based on revised plan design. A. Hydarulic model test; B. Numerical simulation; C. Section view.

    Improvement of hydraulic stability for spillway using CFD model

    Hydraulic model test was used to analyze the rapidly varied flow on the spillway. But, it has some shortcomings such as error of scale effect and expensive costs. Recently, through the development of three dimensional computational fluid dynamics (CFD), rapidly varied flow and turbulence can be simulated. In this study, the applicability of CFD model to simulate flow on the spillway was reviewed. The Karian dam in Indonesia was selected as the study area. The FLOW-3d model, which is well known to simulate a flow having a free surface, was used to analyze flow. The flow stability in approach channel was investigated with the initial plan design, and the results showed that the flow in approach channel is unstable in the initial plan design. To improve flow stability in the spillway, therefore, the revised plan design was formulated. The appropriateness of the revised design was examined by a numerical modeling. The results showed that the flow in spillway is stable in the revised design.

    여수로의 급격하게 변화하는 흐름을 분석하기 위해 수리학적 모델 테스트를 사용했습니다. 그러나 스케일 효과의 오차와 고가의 비용 등의 단점이 있다. 최근에는 3차원 전산유체역학(CFD)의 발달로 급변하는 유동과 난류를 모사할 수 있다. 본 연구에서는 여수로의 흐름을 시뮬레이션하기 위한 CFD 모델의 적용 가능성을 검토했습니다. 인도네시아의 Karian 댐이 연구 지역으로 선정되었습니다. 자유표면을 갖는 유동을 모의하는 것으로 잘 알려진 FLOW-3d 모델을 유동해석에 사용하였다. 접근수로의 흐름 안정성은 초기 계획설계와 함께 조사한 결과 초기 계획설계에서 접근수로의 흐름이 불안정한 것으로 나타났다. 따라서 방수로의 흐름 안정성을 향상시키기 위해 수정된 계획 설계가 공식화되었습니다. 수정된 설계의 적합성을 수치모델링을 통해 검토하였다. 결과는 수정된 설계에서 여수로의 흐름이 안정적이라는 것을 보여주었습니다.

    Key words

    Spillway, FLOW-3D, approach channel, flow stability, numerical modeling, hydraulic model test.

    Figure 6. Two dimensional flow velocity distribution at the approach channel (Flow velocity distribution at depth EL. 68.12 m).
    Figure 6. Two dimensional flow velocity distribution at the approach channel (Flow velocity distribution at depth EL. 68.12 m).
    Figure 7. Flow distribution at the approach channel in PMF. A. Hydraulic model test; B. Numerial simulatio C. Cross section view.
    Figure 7. Flow distribution at the approach channel in PMF. A. Hydraulic model test; B. Numerial simulatio C. Cross section view.
    Figure 8. Revised approach channel section. A. Initial plan design; B. Revised plan design.
    Figure 8. Revised approach channel section. A. Initial plan design; B. Revised plan design.
    Figure 9. Two dimensional flow velocity distribution at the approach channel based on revised plan design (Flow velocity distribution at depth EL. 68.12 m).
    Figure 9. Two dimensional flow velocity distribution at the approach channel based on revised plan design (Flow velocity distribution at depth EL. 68.12 m).
    Figure 10. Flow distribution at the approach channel in PMF based on revised plan design. A. Hydarulic model test; B. Numerical simulation; C. Section view.
    Figure 10. Flow distribution at the approach channel in PMF based on revised plan design. A. Hydarulic model test; B. Numerical simulation; C. Section view.

    REFERENCES

    Betts PL (1979). A variation principle in terms of stream function for free
    surface flows and its application to finite element method. Comp.
    Fluids, 7(2): 145-153.
    Cassidy JJ (1965). Irrotational flow over spillways of finite height. J.
    Eng. Mech. Div. ASCE., 91(6): 155-173.
    Flow Science (2002). FLOW-3D -Theory manual. Los Alamos, NM.
    Guo Y, Wen X, Wu C, Fang D (1998). Numerical modeling of spillway
    flow with free drop and initially unknown discharge. J. Hydraulic Res.
    IAHR, 36(5): 785-801.
    Ho DKH, Donohoo SM (2001). Investigation of spillway behavior under
    increased maximum flood by computational fluid dynamics technique.
    Proceeding 14
    th Australasian Fluid Mech. Conference, Adelaide
    University, Adelaide, Australia, pp. 10-14.
    Ikegawa M, Washizu K (1973). Finite element method applied to
    analysis of flow over a spillway crest. Int. J. Numerical Methods Eng.,
    6: 179-189.
    Kim DG, Park JH (2005). Analysis of flow structure over ogee-spillway
    in consideration of scale and roughness effects by using CFD model.
    J. Civil Eng. KSCE., pp. 161-169.
    KRA, KWATER (2006). Feasibility study and detail design of the Karian
    dam project. Indonesia.
    Li W, Xie Q, Chen CJ (1989). Finite analytic solution of flow over
    spillways, J. Eng. Mech. ASCE, 115(2): 2645-2648.
    Olsen NR, Kjellesvig HM (1998).Three-dimensional numerical flow
    modeling for estimation of spillway capacity. J. Hydraulic Res. IAHR.,
    36(5): 775-784.
    Savage BM, Johnson MC (2001). Flow over ogee spillway: Physical and
    numerical model case study. J. Hydraulic Eng. ASCE., 127(8): 640-
    649.
    Tabbara M, Chatial J, Awwad R (2005). Computational simulation of
    flow over stepped spillways. Comput. Structure, 83: 2215-2224.

    PDF VIEW
    Numerical analysis of energy dissipator options using computational fluid dynamics modeling — a case study of Mirani Dam

    전산 유체 역학 모델링을 사용한 에너지 소산자 옵션의 수치적 해석 — Mirani 댐의 사례 연구

    Arabian Journal of Geosciences volume 15, Article number: 1614 (2022) Cite this article

    Abstract

    이 연구에서 FLOW 3D 전산 유체 역학(CFD) 소프트웨어를 사용하여 파키스탄 Mirani 댐 방수로에 대한 에너지 소산 옵션으로 미국 매립지(USBR) 유형 II 및 USBR 유형 III 유역의 성능을 추정했습니다. 3D Reynolds 평균 Navier-Stokes 방정식이 해결되었으며, 여기에는 여수로 위의 자유 표면 흐름을 캡처하기 위해 공기 유입, 밀도 평가 및 드리프트-플럭스에 대한 하위 그리드 모델이 포함되었습니다. 본 연구에서는 5가지 모델을 고려하였다. 첫 번째 모델에는 길이가 39.5m인 USBR 유형 II 정수기가 있습니다. 두 번째 모델에는 길이가 44.2m인 USBR 유형 II 정수기가 있습니다. 3번째와 4 번째모델에는 길이가 각각 48.8m인 USBR 유형 II 정수조와 39.5m의 USBR 유형 III 정수조가 있습니다. 다섯 번째 모델은 네 번째 모델과 동일하지만 마찰 및 슈트 블록 높이가 0.3m 증가했습니다. 최상의 FLOW 3D 모델 조건을 설정하기 위해 메쉬 민감도 분석을 수행했으며 메쉬 크기 0.9m에서 최소 오차를 산출했습니다. 세 가지 경계 조건 세트가 테스트되었으며 최소 오류를 제공하는 세트가 사용되었습니다. 수치적 검증은 USBR 유형 II( L = 48.8m), USBR 유형 III( L = 35.5m) 및 USBR 유형 III 의 물리적 모델 에너지 소산을 0.3m 블록 단위로 비교하여 수행되었습니다( L= 35.5m). 통계 분석 결과 평균 오차는 2.5%, RMSE(제곱 평균 제곱근 오차) 지수는 3% 미만이었습니다. 수리학적 및 경제성 분석을 바탕으로 4 번째 모델이 최적화된 에너지 소산기로 밝혀졌습니다. 흡수된 에너지 백분율 측면에서 물리적 모델과 수치적 모델 간의 최대 차이는 5% 미만인 것으로 나타났습니다.

    In this study, the FLOW 3D computational fluid dynamics (CFD) software was used to estimate the performance of the United States Bureau of Reclamation (USBR) type II and USBR type III stilling basins as energy dissipation options for the Mirani Dam spillway, Pakistan. The 3D Reynolds-averaged Navier–Stokes equations were solved, which included sub-grid models for air entrainment, density evaluation, and drift–flux, to capture free-surface flow over the spillway. Five models were considered in this research. The first model has a USBR type II stilling basin with a length of 39.5 m. The second model has a USBR type II stilling basin with a length of 44.2 m. The 3rd and 4th models have a USBR type II stilling basin with a length of 48.8 m and a 39.5 m USBR type III stilling basin, respectively. The fifth model is identical to the fourth, but the friction and chute block heights have been increased by 0.3 m. To set up the best FLOW 3D model conditions, mesh sensitivity analysis was performed, which yielded a minimum error at a mesh size of 0.9 m. Three sets of boundary conditions were tested and the set that gave the minimum error was employed. Numerical validation was done by comparing the physical model energy dissipation of USBR type II (L = 48.8 m), USBR type III (L =35.5 m), and USBR type III with 0.3-m increments in blocks (L = 35.5 m). The statistical analysis gave an average error of 2.5% and a RMSE (root mean square error) index of less than 3%. Based on hydraulics and economic analysis, the 4th model was found to be an optimized energy dissipator. The maximum difference between the physical and numerical models in terms of percentage energy absorbed was found to be less than 5%.

    Keywords

    • Numerical modeling
    • Spillway
    • Hydraulic jump
    • Energy dissipation
    • FLOW 3D

    References

    • Abbasi S, Fatemi S, Ghaderi A, Di Francesco S (2021) The effect of geometric parameters of the antivortex on a triangular labyrinth side weir. Water (Switzerland) 13(1). https://doi.org/10.3390/w13010014
    • Amorim JCC, Amante RCR, Barbosa VD (2015) Experimental and numerical modeling of flow in a stilling basin. Proceedings of the 36th IAHR World Congress 28 June–3 July, the Hague, the Netherlands, 1, 1–6
    • Asaram D, Deepamkar G, Singh G, Vishal K, Akshay K (2016) Energy dissipation by using different slopes of ogee spillway. Int J Eng Res Gen Sci 4(3):18–22Google Scholar 
    • Boes RM, Hager WH (2003) Hydraulic design of stepped spillways. J Hydraul Eng 129(9):671–679. https://doi.org/10.1061/(ASCE)0733-9429(2003)129:9(671)Article Google Scholar 
    • Celik IB, Ghia U, Roache PJ, Freitas CJ, Coleman H, Raad PE (2008) Procedure for estimation and reporting of uncertainty due to discretization in CFD applications. J Fluids Eng Trans ASME 130(7):0780011–0780014. https://doi.org/10.1115/1.2960953Article Google Scholar 
    • Chen Q, Dai G, Liu H (2002) Volume of fluid model for turbulence numerical simulation of stepped spillway overflow. J Hydraul Eng 128(7):683–688. 10.1061/共ASCE兲0733-9429共2002兲128:7共683兲 CE
    • Damiron R (2015) CFD modelling of dam spillway aerator. Lund University Sweden
    • Dunlop SL, Willig IA, Paul GE (2016) Cabinet Gorge Dam spillway modifications for TDG abatement – design evolution and field performance. 6th International Symposium on Hydraulic Structures: Hydraulic Structures and Water System Management, ISHS 2016, 3650628160, 460–470. 10.15142/T3650628160853
    • Fleit G, Baranya S, Bihs H (2018) CFD modeling of varied flow conditions over an ogee-weir. Period Polytech Civ Eng 62(1):26–32. https://doi.org/10.3311/PPci.10821Article Google Scholar 
    • Frizell KW, Frizell KH (2015) Guidelines for hydraulic design of stepped spillways. Hydraulic Laboratory Report HL-2015-06, May
    • Ghaderi A, Abbasi S (2021) Experimental and numerical study of the effects of geometric appendance elements on energy dissipation over stepped spillway. Water (Switzerland) 13(7). https://doi.org/10.3390/w13070957
    • Ghaderi A, Dasineh M, Aristodemo F, Ghahramanzadeh A (2020) Characteristics of free and submerged hydraulic jumps over different macroroughnesses. J Hydroinform 22(6):1554–1572. https://doi.org/10.2166/HYDRO.2020.298Article Google Scholar 
    • Güven A, Mahmood AH (2021) Numerical investigation of flow characteristics over stepped spillways. Water Sci Technol Water Supply 21(3):1344–1355. https://doi.org/10.2166/ws.2020.283Article Google Scholar 
    • Herrera-Granados O, Kostecki SW (2016) Numerical and physical modeling of water flow over the ogee weir of the new Niedów barrage. J Hydrol Hydromech 64(1):67–74. https://doi.org/10.1515/johh-2016-0013Article Google Scholar 
    • Ho DKH, Riddette KM (2010) Application of computational fluid dynamics to evaluate hydraulic performance of spillways in australia. Aust J Civ Eng 6(1):81–104. https://doi.org/10.1080/14488353.2010.11463946Article Google Scholar 
    • Kocaer Ö, Yarar A (2020) Experimental and numerical investigation of flow over ogee spillway. Water Resour Manag 34(13):3949–3965. https://doi.org/10.1007/s11269-020-02558-9Article Google Scholar 
    • Kumcu SY (2017) Investigation of flow over spillway modeling and comparison between experimental data and CFD analysis. KSCE J Civ Eng 21(3):994–1003. https://doi.org/10.1007/s12205-016-1257-zArticle Google Scholar 
    • Li S, Li Q, Yang J (2019) CFD modelling of a stepped spillway with various step layouts. Math Prob Eng 2019:1–12. https://doi.org/10.1155/2019/6215739Article Google Scholar 
    • Muthukumaran N, Prince Arulraj G (2020) Experimental investigation on augmenting the discharge over ogee spillways with nanocement. Civ Eng Archit 8(5):838–845. https://doi.org/10.13189/cea.2020.080511Article Google Scholar 
    • Naderi V, Farsadizadeh D, Lin C, Gaskin S (2019) A 3D study of an air-core vortex using HSPIV and flow visualization. Arab J Sci Eng 44(10):8573–8584. https://doi.org/10.1007/s13369-019-03764-3Article Google Scholar 
    • Nangare PB, Kote AS (2017) Experimental investigation of an ogee stepped spillway with plain and slotted roller bucket for energy dissipation. Int J Civ Eng Technol 8(8):1549–1555Google Scholar 
    • Parsaie A, Moradinejad A, Haghiabi AH (2018) Numerical modeling of flow pattern in spillway approach channel. Jordan J Civ Eng 12(1):1–9Google Scholar 
    • Pasbani Khiavi M, Ali Ghorbani M, Yusefi M (2021) Numerical investigation of the energy dissipation process in stepped spillways using finite volume method. J Irrig Water Eng 11(4):22–37Google Scholar 
    • Peng Y, Zhang X, Yuan H, Li X, Xie C, Yang S, Bai Z (2019) Energy dissipation in stepped spillways with different horizontal face angles. Energies 12(23). https://doi.org/10.3390/en12234469
    • Raza A, Wan W, Mehmood K (2021) Stepped spillway slope effect on air entrainment and inception point location. Water (Switzerland) 13(10). https://doi.org/10.3390/w13101428
    • Reeve DE, Zuhaira AA, Karunarathna H (2019) Computational investigation of hydraulic performance variation with geometry in gabion stepped spillways. Water Sci Eng 12(1):62–72. https://doi.org/10.1016/j.wse.2019.04.002Article Google Scholar 
    • Rice CE, Kadavy KC (1996) Model study of a roller compacted concrete stepped spillway. J Hydraul Eng 122(6):292–297. https://doi.org/10.1061/(ASCE)0733-9429(1996)122:6(292)Article Google Scholar 
    • Rong Y, Zhang T, Peng L, Feng P (2019) Three-dimensional numerical simulation of dam discharge and flood routing in Wudu reservoir. Water (Switzerland) 11(10). https://doi.org/10.3390/w11102157
    • Saqib N, Akbar M, Pan H, Ou G, Mohsin M, Ali A, Amin A (2022) Numerical analysis of pressure profiles and energy dissipation across stepped spillways having curved risers. Appl Sci 12(448):1–18Google Scholar 
    • Saqib N, Ansari K, Babar M (2021) Analysis of pressure profiles and energy dissipation across stepped spillways having curved treads using computational fluid dynamics. Intl Conf Adv Mech Eng :1–10
    • Saqib Nu, Akbar M, Huali P, Guoqiang O (2022) Numerical investigation of pressure profiles and energy dissipation across the stepped spillway having curved treads using FLOW 3D. Arab J Geosci 15(1):1363–1400. https://doi.org/10.1007/s12517-022-10505-8Article Google Scholar 
    • Sarkardeh H, Marosi M, Roshan R (2015) Stepped spillway optimization through numerical and physical modeling. Int J Energy Environ 6(6):597–606Google Scholar 
    • Serafeim A, Avgeris V, Hrissanthou V (2015) Experimental and numerical modeling of flow over a spillway. Eur Water Publ 14(2015):55–59. https://doi.org/10.15224/978-1-63248-042-2-11Article Google Scholar 
    • Sorensen RM (1986) Stepped spillway model investigation. J Hydraul Eng I(12):1461–1472. https://ascelibrary.org/doi/full/10.1061/%28ASCE%290733-
    • Tabbara M, Chatila J, Awwad R (2005) Computational simulation of flow over stepped spillways. Comput Struct 83(27):2215–2224. https://doi.org/10.1016/j.compstruc.2005.04.005Article Google Scholar 
    • Valero D, Bung DB, Crookston BM, Matos J (2016) Numerical investigation of USBR type III stilling basin performance downstream of smooth and stepped spillways. 6th International Symposium on Hydraulic Structures: Hydraulic Structures and Water System Management, ISHS 2016, 3406281608, 635–646. https://doi.org/10.15142/T340628160853
    • Versteeg H, Malalasekera W (1979) An introduction to computational fluid mechanics. (Vol. 2). https://doi.org/10.1016/0010-4655(80)90010-7
    • WAPDA model studies cell, IRI Lahore (2003) Mirani Dam Project hydraulic model studies for the spillway. November 2003
    • Yakhot V, Orszag S (1986) Renormalization group analysis of turbulence. I. Basic theory. J Sci Comput 1(1):3–51Article Google Scholar 
    원문보기
    Effect of tailwater depth on non-cohesive earth dam failure due to overtopping

    Effect of tailwater depth on non-cohesive earth dam failure due to overtopping

    범람으로 인한 비점착성 흙댐 붕괴에 대한 테일워터 깊이의 영향

    ShaimaaAmanaMohamedAbdelrazek RezkbRabieaNasrc

    Abstract

    본 연구에서는 범람으로 인한 토사댐 붕괴에 대한 테일워터 깊이의 영향을 실험적으로 조사하였다. 테일워터 깊이의 네 가지 다른 값을 검사합니다. 각 실험에 대해 댐 수심 측량 프로파일의 진화, 고장 기간, 침식 체적 및 유출 수위곡선을 관찰하고 기록합니다.

    결과는 tailwater 깊이를 늘리면 고장 시간이 최대 57% 감소하고 상대적으로 침식된 마루 높이가 최대 77.6% 감소한다는 것을 보여줍니다. 또한 상대 배수 깊이가 3, 4, 5인 경우 누적 침식 체적의 감소는 각각 23, 36.5 및 75%인 반면 최대 유출량의 감소는 각각 7, 14 및 17.35%입니다.

    실험 결과는 침식 과정을 복제할 때 Flow 3D 소프트웨어의 성능을 평가하는 데 활용됩니다. 수치 모델은 비응집성 흙댐의 침식 과정을 성공적으로 시뮬레이션합니다.

    The influence of tailwater depth on earth dam failure due to overtopping is investigated experimentally in this work. Four different values of tailwater depths are examined. For each experiment, the evolution of the dam bathymetry profile, the duration of failure, the eroded volume, and the outflow hydrograph are observed and recorded. The results reveal that increasing the tailwater depth reduces the time of failure by up to 57% and decreases the relative eroded crest height by up to 77.6%. In addition, for relative tailwater depths equal to 3, 4, and 5, the reduction in the cumulative eroded volume is 23, 36.5, and 75%, while the reduction in peak discharge is 7, 14, and 17.35%, respectively. The experimental results are utilized to evaluate the performance of the Flow 3D software in replicating the erosion process. The numerical model successfully simulates the erosion process of non-cohesive earth dams.

    Keywords

    Earth dam, Eroded volume, Flow 3D model, Non-cohesive soil, Overtopping failure, Tailwater depth

    Notation

    d50

    Mean partical diameterWc

    Optimum water contentZo

    Dam height (cm)do

    Tailwater depth (cm)Zeroded

    Eroded height of the dam measured at distance of 0.7 m from the dam heel (cm)t

    Total time of failure (sec)t1

    Time of crest width erosion (sec)Zcrest

    The crest height (cm)Vtotal

    Total volume of the dam (m3)Veroded

    Cumulative eroded volume (m3)RMSE

    The statistical variable root- mean- square errord

    Degree of agreement indexyu.s.

    The upstream water depth (cm)yd.s

    The downstream water depth (cm)H

    Water surface elevation over sharp crested weir (cm)Q

    Outflow discharge (liter/sec)Qpeak

    Peak discharge (liter/sec)

    1. Introduction

    Earth dams are compacted structures composed of natural materials that are usually mined or quarried from local locations. The failures of the earth dams have proven to be deadly, destructive, and costly. According to People’s Daily, two earthen dams, Yong’an Dam and Xinfa Dam located in Hulun Buir City in North China’s Inner Mongolia failed on 2021, due to a surge in the water level of the Nuomin River caused by heavy rain. The dam breach affected 16,660 people, flooded 325,622 mu of farmland (21708.1 ha), and destroyed 22 bridges, 124 culverts, and 15.6 km of roadways. Also, the failure of south fork dam (earth and rock fill dam) near Johnstown on 1889 is considered the worst U.S dam disaster in terms of loss of life. The dam was overtopped and washed away due to unexpected heavy rains, releasing 20 million tons of water which destroyed Johnstown and resulted in 2209 deaths, [1], [2]. Piping or shear sliding, failure due to natural factors, and failure due to overtopping are all possible causes of earth dam failure. However, overtopping failure is the most frequent cause of dam failure. According to The International Committee on Large Dams (ICOLD, 1995), and [3], more than one-third of the total known dam failures were caused by dam overtopping.

    Overtopping occurs as the result of insufficient flood design or freeboard in some cases. Extreme rainstorms can cause floods which can overtop the dam and cause it to fail. The size and geometry of the reservoir or the dam (side slopes, top width, height, etc.), the homogeneity of the material used in the construction of the dam, overtopping depth, and the presence or absence of tailwater are all elements that influence this type of failure which will be illustrated in the following literature. Overtopping failures of earth dams may be divided into several failure mechanisms based on the material composition and the inner structure of the dam. For cohesive earth dams because of low permeability, no seepage exists on the slopes. Erosion often begins at the earth dam toe during turbulent erosion and moves upstream, undercutting the slope, causing the removal of large chunks of materials. While for non-cohesive earth dams the downstream face of the dam flattens progressively and is often said to rotate around a point near the downstream toe [4], [5], [6] In the last few decades, the study of failures due to overtopping has gained popularity among researchers. The overtopping failure, in fact, has been widely investigated in coastal and river hydraulics and morpho dynamic. In addition, several laboratory experimental studies have been conducted in this field in order to better understand different involved factors. Also, many numerical types of research have been conducted to investigate the process of overtopping failure as well as the elements that influence this type of failure.

    Tabrizi et al. [5] conducted a series of embankment overtopping tests to find the effect of compaction on the failure of a homogenous sand embankment. A plane breach process occurred across the flume width due to the narrow flume width. They measured the downstream hydrographs and embankment surface profile for every case. They concluded that the peak discharge decreased with a high compaction level, while the time to peak increased. Kansoh et al. [6] studied experimentally the failure of compacted homogeneous non-cohesive earthen embankment due to overtopping. They investigated the influence of different shape parameters including the downstream slope, the crest width, and the height of the embankment on the erosion process. The erosion process was initiated by carving a pilot channel into the embankment crest. They evaluated the time of embankment failure for different shape parameters. They concluded that the failure time increases with increasing the downstream slope and the crest width. Zhu et al. [7] investigated experimentally the breaching of five embankments, one constructed with pure sand, and four with different sand-silt–clay mixtures. The erosion pattern was similar across the flume width. They stated that for cohesive soil mixtures the head cut erosion was the most important factor that affected the breach growth, while for non-cohesive soil the breach erosion was affected by shear erosion.

    Amaral et al. [8] studied experimentally the failure by overtopping for two embankments built from silt sand material. They studied the effect of the degree of compaction of the embankment and the geometry of the pilot channel carved at the centre of the dam crest. They studied two shapes of pilot channel a rectangular shape and triangular shape. They stated that the breach development is influenced by a higher degree of compaction, however, the pilot channel geometry did not influence the breach’s final form. Bereta et al. [9] studied experimentally the breach formation of five dam models, three of them were homogenous clay soil while two were sandy-clay mixtures. The erosion process was initiated by cutting a pilot channel at the centre of the dam crest. They observed the initiation of erosion, flow shear erosion, sidewall bottom erosion, and distinguished the soil mechanical slope mass failure from the head cut vertically and laterally during these tests. Verma et al. [10] investigated experimentally a two-dimensional erosion phenomenon due to overtopping by using a wooden fuse plug model and five different soils. They concluded that the erosion process was affected mostly by cohesiveness and degree of compaction. For cohesive soils, a head cut erosion was observed, while for non-cohesive soils surface erosion occurred gradually. Also, the dimensions of fuse plug, type of fill material, reservoir capacity, and inflow were found to affect the behaviour of the overall breaching process.

    Wu and Qin [11] studied the effect of adding coarse grains to the downstream face of a non-cohesive dam as a result of tailings deposition. The process of overtopping during tailings dam failures is analyzed and its effect on delaying the dam-break process and disaster mitigation are investigated. They found that the tested protective measures decreased the breach area, the maximum breaching flow discharge and flow velocity, and the downstream inundated area. Khankandi et al. [12] studied experimentally the effect of reservoir geometry on dam break flow in case of dry and wet bed conditions. They considered four different reservoir shapes, a long reservoir, a wide, a trapezoidal shaped and one with a 90◦ bend all with identical water volume and horizontal bed. The dam break is simulated by the sudden gate removal using a pneumatic jack. They measured the variation of water level over time with ultrasonic sensors and flow velocity component with an acoustic Doppler velocimeter. Also, the experimental results of water level variation are compared with Ritters solution (1892) [13]. They stated that for dry bed condition the long and 90 bend reservoirs results are close to the analytical solution by ritter also in these two shapes a 1D flow is noticed. However, for wide and trapezoidal reservoirs a 2D effect is significant due to flow contraction at channel entrance.

    Rifai et al. [14] conducted a series of experiments to investigate the effect of tailwater depth on the outflow discharge and breach geometry during non-cohesive homogenous fluvial dikes overtopping failure. They cut an initial notch in the crest at 0.8 m from the upstream end of the dike to initiate overtopping. They compared their results to previous experiments under different main channel inflow discharges combined with a free floodplain. They divided the dike breaching process into three stages: gradual start of overtopping flow resulting in slow initiation of dike erosion, deepening and widening breach due to large flow depth and velocity, finally the flow depth starts stabilizing at its minimal level with or without sustained breach expansion. They stated that breach discharge has lower values than in free floodplain tests. Jiang [15] studied the effect of bed slope on breach parameters and peak discharge in non-cohesive embankment failure. An initial triangular breach with a depth and width of 4 cm was pre-set on one side of the dam. He stated that peak discharge increases with the increase of bed slope and then decreases.

    Ozmen-cagatay et al. [16] studied experimentally flood wave propagation resulted from a sudden dam break event. For dam-break modelling, they used a mechanism that permitted the rapid removal of a vertical plate with a thickness of 4 mm and made of rigid plastic. They conducted three tests, one with dry bed condition and two tests with tailwater depths equal 0.025 m and 0.1 m respectively. They recorded the free surface profile during initial stages of dam break by using digital image processing. Finally, they compared the experimental results with the with a commercially available VOF-based CFD program solving the Reynolds-averaged Navier –Stokes equations (RANS) with the k– Ɛ turbulence model and the shallow water equations (SWEs). They concluded that Wave breaking was delayed with increasing the tailwater depth to initial reservoir depth ratio. They also stated that the SWE approach is sufficient more to represent dam break flows for wet bed condition. Evangelista [17] investigated experimentally and numerically using a depth-integrated two-phase model, the erosion of sand dike caused by the impact of a dam break wave. The dam break is simulated by a sudden opening of an upstream reservoir gate resulting in the overtopping of a downstream trapezoidal sand dike. The evolution of the water wave caused from the gate opening and dike erosion process are recorded by using a computer-controlled camera. The experimental results demonstrated that the progression of the wave front and dike erosion have a considerable influence on each other during the process. In addition, the dike constructed from fine sands was more resistant to erosion than the one built with coarse sand. They also stated that the numerical model can is capable of accurately predicting wave front position and dike erosion. Also, Di Cristo et al. [18] studied the effect of dam break wave propagation on a sand embankment both experimentally and numerically using a two-phase shallow-water model. The evolution of free surface and of the embankment bottom are recorded and used in numerical model assessment. They stated that the model allows reasonable simulation of the experimental trends of the free surface elevation regardeless of the geofailure operator.

    Lots of numerical models have been developed over the past few years to simulate the dam break flooding problem. A one-dimensional model, such as Hec-Ras, DAMBRK and MIKE 11, ect. A two-dimensional model such as iRIC Nay2DH is used in earth embankment breach simulation. Other researchers studied the failure process numerically using (3D) computational fluid dynamics (CFD) models, such as FLOW-3D, and FLUENT. Goharnejad et al. [19] determined the outflow hydrograph which results from the embankment dam break due to overtopping. Hu et al. [20] performed a comparison between Flow-3D and MIKE3 FM numerical models in simulating a dam break event under dry and wet bed conditions with different tailwater depths. Kaurav et al. [21] simulated a planar dam breach process due to overtopping. They conducted a sensitivity analysis to find the effect of dam material, dam height, downstream slope, crest width, and inlet discharge on the erosion process and peak discharge through breach. They concluded that downstream slope has a significant influence on breaching process. Yusof et al. [22] studied the effect of embankment sediment sizes and inflow rates on breaching geometric and hydrodynamic parameters. They stated that the peak outflow hydrograph increases with increasing sediment size and inflow rates while time of failure decreases.

    In the present work, the effect of tailwater depth on earth dam failure during overtopping is studied experimentally. The relation between the eroded volume of the dam and the tailwater depth is presented. Also, the percentage of reduction in peak discharge due to tailwater existence is calculated. An assessment of Flow 3D software performance in simulating the erosion process during earth dam failure is introduced. The statistical variable root- mean- square error, RMSE, and the agreement degree index, d, are used in model assessment.

    2. Material and methods

    The tests are conducted in a straight rectangular flume in the laboratory of Irrigation Engineering and Hydraulics Department, Faculty of Engineering, Alexandria University, Egypt. The flume dimensions are 10 m long, 0.86 m wide, and 0.5 m deep. The front part of the flume is connected to a storage basin 1 m long by 0.86 m wide. The storage basin is connected to a collecting tank for water recirculation during the experiments as shown in Fig. 1, Fig. 2. A sharp-crested weir is placed at a distance of 4 m downstream the constructed dam to keep a constant tailwater depth in each experiment and to measure the outflow discharge.

    To measure the eroded volume with time a rods technique is used. This technique consists of two parallel wooden plates with 10 cm distance in between and five rows of stainless-steel rods passing vertically through the wooden plates at a spacing of 20 cm distributed across flume width. Each row consists of four rods with 15 cm spacing between them. Also, a graph board is provided to measure the drop in each rod with time as shown in Fig. 3, Fig. 4. After dam construction the rods are carefully rested on the dam, with the first line of rods resting in the middle of the dam crest and then a constant distance of 15 cm between rods lines is maintained.

    A soil sample is taken and tested in the laboratory of the soil mechanics to find the soil geotechnical parameters. The soil particle size distribution is also determined by sieve analysis as shown in Fig. 5. The soil mean diameter d50,equals 0.38 mm and internal friction angle equals 32.6°.

    2.1. Experimental procedures

    To investigate the effect of the tailwater depth (do), the tailwater depth is changed four times 5, 15, 20, and 25 cm on the sand dam model. The dam profile is 35 cm height, with crest width = 15 cm, the dam base width is 155 cm, and the upstream and downstream slopes are 2:1 as shown in Fig. 6. The dam dimensions are set as the flume permitted to allow observation of the dam erosion process under the available flume dimensions and conditions. All of the conducted experiments have the same dimensions and configurations.

    The optimum water content, Wc, from the standard proctor test is found to be 8 % and the maximum dry unit weight is 19.42 kN/m3. The soil and water are mixed thoroughly to ensure consistency and then placed on three horizontal layers. Each layer is compacted according to ASTM standard with 25 blows by using a rammer (27 cm × 20.5 cm) weighing 4 kg. Special attention is paid to the compaction of the soil to guarantee the repeatability of the tests.

    After placing and compacting the three layers, the dam slopes are trimmed carefully to form the trapezoidal shape of the dam. A small triangular pilot channel with 1 cm height and 1:1 side slopes is cut into the dam crest to initiate the erosion process. The position of triangular pilot channel is presented in Fig. 1. Three digital video cameras with a resolution of 1920 × 1080 pixels and a frame rate of 60 fps are placed in three different locations. One camera on one side of the flume to record the progress of the dam profile during erosion. Another to track the water level over the sharp-crested rectangular weir placed at the downstream end of the flume. And the third camera is placed above the flume at the downstream side of the dam and in front of the rods to record the drop of the tip of the rods with time as shown previously in Fig. 1.

    Before starting the experiment, the water is pumped into the storage basin by using pump with capacity 360 m3/hr, and then into the upstream section of the flume. The upstream boundary is an inflow condition. The flow discharge provided to the storage basin is kept at a constant rate of 6 L/sec for all experiments, while the downstream boundary is an outflow boundary condition.

    Also, the required tailwater depth for each experiment is filled to the desired depth. A dye container valve is opened to color the water upstream of the dam to make it easy to distinguish the dam profile from the water profile. A wooden board is placed just upstream of the dam to prevent water from overtopping the dam until the water level rises to a certain level above the dam crest and then the wooden board is removed slowly to start the experiment.

    2.2. Repeatability

    To verify the accuracy of the results, each experiment is repeated two times under the same conditions. Fig. 7 shows the relative eroded crest height, Zeroded / Zo, with time for 5 cm tailwater depth. From the Figure, it can be noticed that results for all runs are consistent, and accuracy is achieved.

    3. Numerical model

    The commercially available numerical model, Flow 3D is used to simulate the dam failure due to overtopping for the cases of 15 cm, 20 cm and 25 cm tailwater depths. For numerical model calibration, experimental results for dam surface evolution are used. The numerical model is calibrated for selection of the optimal turbulence model (RNG, K-e, and k-w) and sediment scour equations (Van Rin, Meyer- peter and Muller, and Nielsen) that produce the best results. In this, the flow field is solved by the RNG turbulence model, and the van Rijn equation is used for the sediment scour model. A geometry file is imported before applying the mesh.

    A Mesh sensitivity is analyzed and checked for various cell sizes, and it is found that decreasing the cell size significantly increases the simulation time with insignificant differences in the result. It is noticed that the most important factor influencing cell size selection is the value of the dam’s upstream and downstream slopes. For example, the slopes in the dam model are 2:1, thus the cell size ratio in X and Z directions should be 2:1 as well. The cell size in a mesh block is set to be 0.02 m, 0.025 m, and 0.01 m in X, Y and Z directions respectively.

    In the numerical computations, the boundary conditions employed are the walls for sidewalls and the channel bottom. The pressure boundary condition is applied at the top, at the air–water interface, to account for atmospheric pressure on the free surface. The upstream boundary is volume flow rate while the downstream boundary is outflow discharge.

    The initial condition is a fluid region, which is used to define fluid areas both upstream and downstream of the dam. To assess the model accuracy, the statistical variable root- mean- square error, RMSE, and the agreement degree index, d, are calculated as(1)RMSE=1N∑i=1N(Pi-Mi)2(2)d=1-∑Mi-Pi2∑Mi-M¯+Pi-P¯2

    where N is the number of samples, Pi and Mi are the models and experimental values, P and M are the means of the model and experimental values. The best fit between the experimental and model results would have an RMSE = 0 and degree of agreement, d = 1.

    4. Results of experimental work

    The results of the total time of failure, t (defined as the time from when the water begins to overtop the dam crest until the erosion reaches a steady state, when no erosion occurs), time of crest width erosion t1, cumulative eroded volume Veroded, and peak discharge Qpeak for each experiment are listed in Table 1. The case of 5 cm tailwater depth is considered as a reference case in this work.

    Table 1. Results of experimental work.

    Tailwater depth, do (cm)Total time of failure, t (sec)Time of crest width erosion, t1 (sec)cumulative eroded volume, Veroded (m3)Peak discharge, Qpeak (liter/sec)
    5255220.2113.12
    15165300.1612.19
    20140340.1311.29
    25110390.0510.84

    5. Discussion

    5.1. Side erosion

    The evolution of the bathymetry of the erosion line recorded by the video camera1. The videos are split into frames (60 frames/sec) by the Free Video to JPG Converter v.5.063 build and then converted into an excel spreadsheet using MATLAB code as shown in Fig. 8.

    Fig. 9 shows a sample of numerical model output. Fig. 10, Fig. 11, Fig. 12 show a dam profile development for different time steps from both experimental and numerical model, for tailwater depths equal 15 cm, 20 cm and 25 cm. Also, the values of RMSE and d for each figure are presented. The comparison shows that the Flow 3D software can simulate the erosion process of non-cohesive earth dam during overtopping with an RMSE value equals 0.023, 0.0218, and 0.0167 and degree of agreement, d, equals 0.95, 0.968, and 0.988 for relative tailwater depths, do/(do)ref, = 3, 4 and 5, respectively. The low values of RMSE and high values of d show that the Flow 3D can effectively simulate the erosion process. From Fig. 10, Fig. 11, Fig. 12, it can be noticed that the model is not capable of reproducing the head cut, while it can simulate well the degradation of the crest height with a minor difference from experimental work. The reason of this could be due to inability of simulation of all physical conditions which exists in the experimental work, such as channel friction and the grain size distribution of the dam soil which is surely has a great effect on the erosion process and breach development. In the experimental work the grain size distribution is shown in Fig. 5, while the numerical model considers that the soil is uniform and exactly 50 % of the dam particles diameter are equal to the d50 value. Another reason is that the model is not considering the increased resistance of the dam due to the apparent cohesion which happens due to dam saturation [23].

    It is clear from both the experimental and numerical results that for a 5 cm tailwater depth, do/(do)ref = 1.0, erosion begins near the dam toe and continues upward on the downstream slope until it reaches the crest. After eroding the crest width, the crest is lowered, resulting in increased flow rates and the speeding up of the erosion process. While for relative tailwater depths, do/(do)ref = 3, 4, and 5 erosion starts at the point of intersection between the downstream slope and tailwater. The existence of tailwater works as an energy dissipater for the falling water which reduces the erosion process and prevents the dam from failure as shown in Fig. 13. It is found that the time of the failure decreases with increasing the tailwater depth because most of the dam height is being submerged with water which decreases the erosion process. The reduction in time of failure from the referenced case is found to be 35.3, 45, and 57 % for relative tailwater depth, do /(do)ref equals 3, 4, and 5, respectively.

    The relation between the relative eroded crest height, Zeroded /Zo, with time is drawn as shown in Fig. 14. It is found that the relative eroded crest height decreases with increasing tailwater depth by 10, 41, and 77.6 % for relative tailwater depth, do /(do)ref equals 3, 4, and 5, respectively. The time required for the erosion of the crest width, t1, is calculated for each experiment. The relation between relative tailwater depth and relative time of crest width erosion is shown in Fig. 15. It is found that the time of crest width erosion increases linearly with increasing, do /Zo. The percent of increase is 36.4, 54.5 and 77.3 % for relative tailwater depth, do /(do)ref = 3, 4 and 5, respectively.

    Crest height, Zcrest is calculated from the experimental results and the Flow 3D results for relative tailwater depths, do/(do)ref, = 3, 4, and 5. A relation between relative crest height, Zcrest/Zo with time from experimental and numerical results is presented in Fig. 16. From Fig. 16, it is seen that there is a good consistency between the results of numerical model and the experimental results in the case of tracking the erosion of the crest height with time.

    5.2. Upstream and downstream water depths

    It is noticed that at the beginning of the erosion process, both upstream and downstream water depths increase linearly with time as long as erosion of the crest height did not take place. However, when the crest height starts to lower the upstream water depth decreases with time while the downstream water depth increases. At the end of the experiment, the two depths are nearly equal. A relation between relative downstream and upstream water depths with time is drawn for each experiment as shown in Fig. 17.

    5.3. Eroded volume

    A MATLAB code is used to calculate the cumulative eroded volume every time interval for each experiment. The total volume of the dam, Vtotal is 0.256 m3. The cumulative eroded volume, Veroded is 0.21, 0.16, 0.13, and 0.05 m3 for tailwater depths, do = 5, 15, 20, and 25 cm, respectively. Fig. 18 presents the relation between cumulative eroded volume, Veroded and time. From Fig. 18, it is observed that the cumulative eroded volume decreases with increasing the tailwater depth. The reduction in cumulative eroded volume is 23, 36.5, and 75 % for relative tailwater depth, do /(do)ref = 3, 4, and 5, respectively. The relative remained volume of the dam equals 0.18, 0.375, 0.492, and 0.8 for tailwater depths = 5, 15, 20, and 25 cm, respectively. Fig. 19 shows a relation between relative tailwater depth and relative cumulative eroded volume from experimental results. From that figure, it is noticed that the eroded volume decreases exponentially with increasing relative tailwater depth.

    5.4. The outflow discharge

    The inflow discharge provided to the storage tank is maintained constant for all experiments. The water surface elevation, H, over the sharp-crested weir placed at the downstream side is recorded by the video camera 2. For each experiment, the outflow discharge is then calculated by using the sharp-crested rectangular weir equation every 10 sec.

    The outflow discharge is found to increase rapidly until it reaches its peak then it decreases until it is constant. For high values of tailwater depths, the peak discharge becomes less than that in the case of small tailwater depth as shown in Fig. 20 which agrees well with the results of Rifai et al. [14] The reduction in peak discharge is 7, 14, and 17.35 % for relative tailwater depth, do /(do)ref = 3, 4, and 5, respectively.

    The scenario presented in this article in which the tailwater depth rises due to unexpected heavy rainfall, is investigated to find the effect of rising tailwater depth on earth dam failure. The results revealed that rising tailwater depth positively affects the process of dam failure in terms of preventing the dam from complete failure and reducing the outflow discharge.

    6. Conclusions

    The effect of tailwater depth on earth dam failure due to overtopping is investigated experimentally in this work. The study focuses on the effect of tailwater depth on side erosion, upstream and downstream water depths, eroded volume, outflow hydrograph, and duration of the failure process. The Flow 3D numerical software is used to simulate the dam failure, and a comparison is made between the experimental and numerical results to find the ability of this software to simulate the erosion process. The following are the results of the investigation:

    The existence of tailwater with high depths prevents the dam from completely collapsing thereby turning it into a broad crested weir. The failure time decreases with increasing the tailwater depth and the reduction from the reference case is found to be 35.3, 45, and 57 % for relative tailwater depth, do /(do)ref = 3, 4, and 5, respectively. The difference between the upstream and downstream water depths decreases with time till it became almost negligible at the end of the experiment. The reduction in cumulative eroded volume is 23, 36.5, and 75 % for relative tailwater depth, do /(do)ref = 3, 4, and 5, respectively. The peak discharge decreases by 7, 14, and 17.35 % for relative tailwater depth, do /(do)ref = 3, 4, and 5, respectively. The relative eroded crest height decreases linearly with increasing the tailwater depth by 10, 41, and 77.6 % for relative tailwater depth, do /(do)ref = 3, 4, and 5, respectively. The numerical model can reproduce the erosion process with a minor deviation from the experimental results, particularly in terms of tracking the degradation of the crest height with time.

    Declaration of Competing Interest

    The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

    Reference

    [1]

    D. McCullough

    The Johnstown Flood

    Simon and Schuster, NY (1968)

    Google Scholar[2]Rose AT. The influence of dam failures on dam safety laws in Pennsylvania. Association of State Dam Safety Officials Annual Conference 2013, Dam Safety 2013. 2013;1:738–56.

    Google Scholar[3]

    M. Foster, R. Fell, M. Spannagle

    The statistics of embankment dam failures and accidents

    Can Geotech J, 37 (5) (2000), pp. 1000-1024, 10.1139/t00-030 View PDF

    View Record in ScopusGoogle Scholar[4]Pickert, G., Jirka, G., Bieberstein, A., Brauns, J. Soil/water interaction during the breaching process of overtopped embankments. In: Greco, M., Carravetta, A., Morte, R.D. (Eds.), Proceedings of the Conference River-Flow 2004, Balkema.

    Google Scholar[5]

    A. Asghari Tabrizi, E. Elalfy, M. Elkholy, M.H. Chaudhry, J. Imran

    Effects of compaction on embankment breach due to overtopping

    J Hydraul Res, 55 (2) (2017), pp. 236-247, 10.1080/00221686.2016.1238014 View PDF

    View Record in ScopusGoogle Scholar[6]

    R.M. Kansoh, M. Elkholy, G. Abo-Zaid

    Effect of Shape Parameters on Failure of Earthen Embankment due to Overtopping

    KSCE J Civ Eng, 24 (5) (2020), pp. 1476-1485, 10.1007/s12205-020-1107-x View PDF

    View Record in ScopusGoogle Scholar[7]

    YongHui Zhu, P.J. Visser, J.K. Vrijling, GuangQian Wang

    Experimental investigation on breaching of embankments

    Experimental investigation on breaching of embankments, 54 (1) (2011), pp. 148-155 View PDF

    CrossRefView Record in ScopusGoogle Scholar[8]Amaral S, Jónatas R, Bento AM, Palma J, Viseu T, Cardoso R, et al. Failure by overtopping of earth dams. Quantification of the discharge hydrograph. Proceedings of the 3rd IAHR Europe Congress: 14-15 April 2014, Portugal. 2014;(1):182–93.

    Google Scholar[9]

    G. Bereta, P. Hui, H. Kai, L. Guang, P. Kefan, Y.Z. Zhao

    Experimental study of cohesive embankment dam breach formation due to overtopping

    Periodica Polytechnica Civil Engineering, 64 (1) (2020), pp. 198-211, 10.3311/PPci.14565 View PDF

    View Record in ScopusGoogle Scholar[10]

    D.K. Verma, B. Setia, V.K. Arora

    Experimental study of breaching of an earthen dam using a fuse plug model

    Int J Eng Trans A, 30 (4) (2017), pp. 479-485, 10.5829/idosi.ije.2017.30.04a.04 View PDF

    View Record in ScopusGoogle Scholar[11]Wu T, Qin J. Experimental Study of a Tailings Impoundment Dam Failure Due to Overtopping. Mine Water and the Environment [Internet]. 2018;37(2):272–80. Available from: doi: 10.1007/s10230-018-0529-x.

    Google Scholar[12]

    A. Feizi Khankandi, A. Tahershamsi, S. Soares-Frazo

    Experimental investigation of reservoir geometry effect on dam-break flow

    J Hydraul Res, 50 (4) (2012), pp. 376-387 View PDF

    CrossRefView Record in ScopusGoogle Scholar[13]

    A. Ritter

    Die Fortpflanzung der Wasserwellen (The propagation of water waves)

    Zeitschrift Verein Deutscher Ingenieure, 36 (33) (1892), pp. 947-954

    [in German]

    View Record in ScopusGoogle Scholar[14]

    I. Rifai, K. El Kadi Abderrezzak, S. Erpicum, P. Archambeau, D. Violeau, M. Pirotton, et al.

    Floodplain Backwater Effect on Overtopping Induced Fluvial Dike Failure

    Water Resour Res, 54 (11) (2018), pp. 9060-9073 View PDF

    This article is free to access.

    CrossRefView Record in ScopusGoogle Scholar[15]

    X. Jiang

    Laboratory Experiments on Breaching Characteristics of Natural Dams on Sloping Beds

    Advances in Civil Engineering, 2019 (2019), pp. 1-14

    View Record in ScopusGoogle Scholar[16]

    H. Ozmen-Cagatay, S. Kocaman

    Dam-break flows during initial stage using SWE and RANS approaches

    J Hydraul Res, 48 (5) (2010), pp. 603-611 View PDF

    CrossRefView Record in ScopusGoogle Scholar[17]

    S. Evangelista

    Experiments and numerical simulations of dike erosion due to a wave impact

    Water (Switzerland), 7 (10) (2015), pp. 5831-5848 View PDF

    CrossRefView Record in ScopusGoogle Scholar[18]

    C. Di Cristo, S. Evangelista, M. Greco, M. Iervolino, A. Leopardi, A. Vacca

    Dam-break waves over an erodible embankment: experiments and simulations

    J Hydraul Res, 56 (2) (2018), pp. 196-210 View PDF

    CrossRefView Record in ScopusGoogle Scholar[19]Goharnejad H, Sm M, Zn M, Sadeghi L, Abadi K. Numerical Modeling and Evaluation of Embankment Dam Break Phenomenon (Case Study : Taleghan Dam) ISSN : 2319-9873. 2016;5(3):104–11.

    Google Scholar[20]Hu H, Zhang J, Li T. Dam-Break Flows : Comparison between Flow-3D , MIKE 3 FM , and Analytical Solutions with Experimental Data. 2018;1–24. doi: 10.3390/app8122456.

    Google Scholar[21]

    R. Kaurav, P.K. Mohapatra, D. Ph

    Studying the Peak Discharge through a Planar Dam Breach, 145 (6) (2019), pp. 1-8 View PDF

    CrossRef[22]

    Z.M. Yusof, Z.A.L. Shirling, A.K.A. Wahab, Z. Ismail, S. Amerudin

    A hydrodynamic model of an embankment breaching due to overtopping flow using FLOW-3D

    IOP Conference Series: Earth and Environmental Science, 920 (1) (2021)

    Google Scholar[23]

    G. Pickert, V. Weitbrecht, A. Bieberstein

    Breaching of overtopped river embankments controlled by apparent cohesion

    J Hydraul Res, 49 (2) (Apr. 2011), pp. 143-156, 10.1080/00221686.2011.552468 View PDF

    View Record in ScopusGoogle Scholar

    Cited by (0)

    My name is Shaimaa Ibrahim Mohamed Aman and I am a teaching assistant in Irrigation and Hydraulics department, Faculty of Engineering, Alexandria University. I graduated from the Faculty of Engineering, Alexandria University in 2013. I had my MSc in Irrigation and Hydraulic Engineering in 2017. My research interests lie in the area of earth dam Failures.

    Peer review under responsibility of Ain Shams University.

    © 2022 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Ain Shams University.

    PDF 원문보기
    Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

    AZ91 합금 주물 내 연행 결함에 대한 캐리어 가스의 영향

    TianLiabJ.M.T.DaviesaXiangzhenZhuc
    aUniversity of Birmingham, Birmingham B15 2TT, United Kingdom
    bGrainger and Worrall Ltd, Bridgnorth WV15 5HP, United Kingdom
    cBrunel Centre for Advanced Solidification Technology, Brunel University London, Kingston Ln, London, Uxbridge UB8 3PH, United Kingdom

    Abstract

    An entrainment defect (also known as a double oxide film defect or bifilm) acts a void containing an entrapped gas when submerged into a light-alloy melt, thus reducing the quality and reproducibility of the final castings. Previous publications, carried out with Al-alloy castings, reported that this trapped gas could be subsequently consumed by the reaction with the surrounding melt, thus reducing the void volume and negative effect of entrainment defects. Compared with Al-alloys, the entrapped gas within Mg-alloy might be more efficiently consumed due to the relatively high reactivity of magnesium. However, research into the entrainment defects within Mg alloys has been significantly limited. In the present work, AZ91 alloy castings were produced under different carrier gas atmospheres (i.e., SF6/CO2, SF6/air). The evolution processes of the entrainment defects contained in AZ91 alloy were suggested according to the microstructure inspections and thermodynamic calculations. The defects formed in the different atmospheres have a similar sandwich-like structure, but their oxide films contained different combinations of compounds. The use of carrier gases, which were associated with different entrained-gas consumption rates, affected the reproducibility of AZ91 castings.

    연행 결함(이중 산화막 결함 또는 이중막이라고도 함)은 경합금 용융물에 잠길 때 갇힌 가스를 포함하는 공극으로 작용하여 최종 주물의 품질과 재현성을 저하시킵니다. Al-합금 주물을 사용하여 수행된 이전 간행물에서는 이 갇힌 가스가 주변 용융물과의 반응에 의해 후속적으로 소모되어 공극 부피와 연행 결함의 부정적인 영향을 줄일 수 있다고 보고했습니다. Al-합금에 비해 마그네슘의 상대적으로 높은 반응성으로 인해 Mg-합금 내에 포집된 가스가 더 효율적으로 소모될 수 있습니다. 그러나 Mg 합금 내 연행 결함에 대한 연구는 상당히 제한적이었습니다. 현재 작업에서 AZ91 합금 주물은 다양한 캐리어 가스 분위기(즉, SF6/CO2, SF6/공기)에서 생산되었습니다. AZ91 합금에 포함된 연행 결함의 진화 과정은 미세 조직 검사 및 열역학 계산에 따라 제안되었습니다. 서로 다른 분위기에서 형성된 결함은 유사한 샌드위치 구조를 갖지만 산화막에는 서로 다른 화합물 조합이 포함되어 있습니다. 다른 동반 가스 소비율과 관련된 운반 가스의 사용은 AZ91 주물의 재현성에 영향을 미쳤습니다.

    Keywords

    Magnesium alloy, Casting, Oxide film, Bifilm, Entrainment defect, Reproducibility

    1. Introduction

    As the lightest structural metal available on Earth, magnesium became one of the most attractive light metals over the last few decades. The magnesium industry has consequently experienced a rapid development in the last 20 years [1,2], indicating a large growth in demand for Mg alloys all over the world. Nowadays, the use of Mg alloys can be found in the fields of automobiles, aerospace, electronics and etc.[3,4]. It has been predicted that the global consumption of Mg metals will further increase in the future, especially in the automotive industry, as the energy efficiency requirement of both traditional and electric vehicles further push manufactures lightweight their design [3,5,6].

    The sustained growth in demand for Mg alloys motivated a wide interest in the improvement of the quality and mechanical properties of Mg-alloy castings. During a Mg-alloy casting process, surface turbulence of the melt can lead to the entrapment of a doubled-over surface film containing a small quantity of the surrounding atmosphere, thus forming an entrainment defect (also known as a double oxide film defect or bifilm) [7], [8], [9], [10]. The random size, quantity, orientation, and placement of entrainment defects are widely accepted to be significant factors linked to the variation of casting properties [7]. In addition, Peng et al. [11] found that entrained oxides films in AZ91 alloy melt acted as filters to Al8Mn5 particles, trapping them as they settle. Mackie et al. [12] further suggested that entrained oxide films can act to trawl the intermetallic particles, causing them to cluster and form extremely large defects. The clustering of intermetallic compounds made the entrainment defects more detrimental for the casting properties.

    Most of the previous studies regarding entrainment defects were carried out on Al-alloys [7,[13], [14], [15], [16], [17], [18], and a few potential methods have been suggested for diminishing their negative effect on the quality of Al-alloy castings. Nyahumwa et al.,[16] shows that the void volume within entrainment defects could be reduced by a hot isostatic pressing (HIP) process. Campbell [7] suggested the entrained gas within the defects could be consumed due to reaction with the surrounding melt, which was further verified by Raiszedeh and Griffiths [19].The effect of the entrained gas consumption on the mechanical properties of Al-alloy castings has been investigated by [8,9], suggesting that the consumption of the entrained gas promoted the improvement of the casting reproducibility.

    Compared with the investigation concerning the defects within Al-alloys, research into the entrainment defects within Mg-alloys has been significantly limited. The existence of entrainment defects has been demonstrated in Mg-alloy castings [20,21], but their behaviour, evolution, as well as entrained gas consumption are still not clear.

    In a Mg-alloy casting process, the melt is usually protected by a cover gas to avoid magnesium ignition. The cavities of sand or investment moulds are accordingly required to be flushed with the cover gas prior to the melt pouring [22]. Therefore, the entrained gas within Mg-alloy castings should contain the cover gas used in the casting process, rather than air only, which may complicate the structure and evolution of the corresponding entrainment defects.

    SF6 is a typical cover gas widely used for Mg-alloy casting processes [23], [24], [25]. Although this cover gas has been restricted to use in European Mg-alloy foundries, a commercial report has pointed out that this cover is still popular in global Mg-alloy industry, especially in the countries which dominated the global Mg-alloy production, such as China, Brazil, India, etc. [26]. In addition, a survey in academic publications also showed that this cover gas was widely used in recent Mg-alloy studies [27]. The protective mechanism of SF6 cover gas (i.e., the reaction between liquid Mg-alloy and SF6 cover gas) has been investigated by several previous researchers, but the formation process of the surface oxide film is still not clearly understood, and even some published results are conflicting with each other. In early 1970s, Fruehling [28] found that the surface film formed under SF6 was MgO mainly with traces of fluorides, and suggested that SF6 was absorbed in the Mg-alloy surface film. Couling [29] further noticed that the absorbed SF6 reacted with the Mg-alloy melt to form MgF2. In last 20 years, different structures of the Mg-alloy surface films have been reported, as detailed below.(1)

    Single-layered film. Cashion [30,31] used X-ray Photoelectron Spectroscopy (XPS) and Auger Spectroscopy (AES) to identify the surface film as MgO and MgF2. He also found that composition of the film was constant throughout the thickness and the whole experimental holding time. The film observed by Cashion had a single-layered structure created from a holding time from 10 min to 100 min.(2)

    Double-layered film. Aarstad et. al [32] reported a doubled-layered surface oxide film in 2003. They observed several well-distributed MgF2 particles attached to the preliminary MgO film and grew until they covered 25–50% of the total surface area. The inward diffusion of F through the outer MgO film was the driving force for the evolution process. This double-layered structure was also supported by Xiong’s group [25,33] and Shih et al. [34].(3)

    Triple-layered film. The triple-layered film and its evolution process were reported in 2002 by Pettersen [35]. Pettersen found that the initial surface film was a MgO phase and then gradually evolved to the stable MgF2 phase by the inward diffusion of F. In the final stage, the film has a triple-layered structure with a thin O-rich interlayer between the thick top and bottom MgF2 layers.(4)

    Oxide film consisted of discrete particles. Wang et al [36] stirred the Mg-alloy surface film into the melt under a SF6 cover gas, and then inspect the entrained surface film after the solidification. They found that the entrained surface films were not continues as the protective surface films reported by other researchers but composed of discrete particles. The young oxide film was composed of MgO nano-sized oxide particles, while the old oxide films consist of coarse particles (about 1  µm in average size) on one side that contained fluorides and nitrides.

    The oxide films of a Mg-alloy melt surface or an entrained gas are both formed due to the reaction between liquid Mg-alloy and the cover gas, thus the above-mentioned research regarding the Mg-alloy surface film gives valuable insights into the evolution of entrainment defects. The protective mechanism of SF6 cover gas (i.e., formation of a Mg-alloy surface film) therefore indicated a potential complicated evolution process of the corresponding entrainment defects.

    However, it should be noted that the formation of a surface film on a Mg-alloy melt is in a different situation to the consumption of an entrained gas that is submerged into the melt. For example, a sufficient amount of cover gas was supported during the surface film formation in the studies previously mentioned, which suppressed the depletion of the cover gas. In contrast, the amount of entrained gas within a Mg-alloy melt is finite, and the entrained gas may become fully depleted. Mirak [37] introduced 3.5%SF6/air bubbles into a pure Mg-alloy melt solidifying in a specially designed permanent mould. It was found that the gas bubbles were entirely consumed, and the corresponding oxide film was a mixture of MgO and MgF2. However, the nucleation sites (such as the MgF2 spots observed by Aarstad [32] and Xiong [25,33]) were not observed. Mirak also speculated that the MgF2 formed prior to MgO in the oxide film based on the composition analysis, which was opposite to the surface film formation process reported in previous literatures (i.e., MgO formed prior to MgF2). Mirak’s work indicated that the oxide-film formation of an entrained gas may be quite different from that of surface films, but he did not reveal the structure and evolution of the oxide films.

    In addition, the use of carrier gas in the cover gases also influenced the reaction between the cover gas and the liquid Mg-alloy. SF6/air required a higher content of SF6 than did a SF6/CO2 carrier gas [38], to avoid the ignition of molten magnesium, revealing different gas-consumption rates. Liang et.al [39] suggested that carbon was formed in the surface film when CO2 was used as a carrier gas, which was different from the films formed in SF6/air. An investigation into Mg combustion [40] reported a detection of Mg2C3 in the Mg-alloy sample after burning in CO2, which not only supported Liang’s results, but also indicated a potential formation of Mg carbides in double oxide film defects.

    The work reported here is an investigation into the behaviour and evolution of entrainment defects formed in AZ91 Mg-alloy castings, protected by different cover gases (i.e., SF6/air and SF6/CO2). These carrier gases have different protectability for liquid Mg alloy, which may be therefore associated with different consumption rates and evolution processes of the corresponding entrained gases. The effect of the entrained-gas consumption on the reproducibility of AZ91 castings was also studied.

    2. Experiment

    2.1. Melting and casting

    Three kilograms AZ91 alloy was melted in a mild steel crucible at 700 ± 5 °C. The composition of the AZ91 alloy has been shown in Table 1. Prior to heating, all oxide scale on the ingot surface was removed by machining. The cover gases used were 0.5%SF6/air or 0.5%SF6/CO2 (vol.%) at a flow rate of 6 L/min for different castings. The melt was degassed by argon with a flow rate of 0.3 L/min for 15 min [41,42], and then poured into sand moulds. Prior to pouring, the sand mould cavity was flushed with the cover gas for 20 min [22]. The residual melt (around 1 kg) was solidified in the crucible.

    Table 1. Composition (wt.%) of the AZ91 alloy used in this study.

    AlZnMnSiFeNiMg
    9.40.610.150.020.0050.0017Residual

    Fig. 1(a) shows the dimensions of the casting with runners. A top-filling system was deliberately used to generate entrainment defects in the final castings. Green and Campbell [7,43] suggested that a top-filling system caused more entrainment events (i.e., bifilms) during a casting process, compared with a bottom-filling system. A melt flow simulation (Flow-3D software) of this mould, using Reilly’s model [44] regarding the entrainment events, also predicted that a large amount of bifilms would be contained in the final casting (denoted by the black particles in Fig. 1b).

    Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

    Shrinkage defects also affect the mechanical properties and reproducibility of castings. Since this study focused on the effect of bifilms on the casting quality, the mould has been deliberately designed to avoid generating shrinkage defects. A solidification simulation using ProCAST software showed that no shrinkage defect would be contained in the final casting, as shown in Fig. 1c. The casting soundness has also been confirmed using a real time X-ray prior to the test bar machining.

    The sand moulds were made from resin-bonded silica sand, containing 1wt. % PEPSET 5230 resin and 1wt. % PEPSET 5112 catalyst. The sand also contained 2 wt.% Na2SiF6 to act as an inhibitor [45]. The pouring temperature was 700 ± 5 °C. After the solidification, a section of the runner bars was sent to the Sci-Lab Analytical Ltd for a H-content analysis (LECO analysis), and all the H-content measurements were carried out on the 5th day after the casting process. Each of the castings was machined into 40 test bars for a tensile strength test, using a Zwick 1484 tensile test machine with a clip extensometer. The fracture surfaces of the broken test bars were examined using Scanning Electron Microscope (SEM, Philips JEOL7000) with an accelerating voltage of 5–15 kV. The fractured test bars, residual Mg-alloy solidified in the crucible, and the casting runners were then sectioned, polished and also inspected using the same SEM. The cross-section of the oxide film found on the test-bar fracture surface was exposed by the Focused Ion Beam milling technique (FIB), using a CFEI Quanta 3D FEG FIB-SEM. The oxide film required to be analysed was coated with a platinum layer. Then, a gallium ion beam, accelerated to 30 kV, milled the material substrate surrounding the platinum coated area to expose the cross section of the oxide film. EDS analysis of the oxide film’s cross section was carried out using the FIB equipment at accelerating voltage of 30 kV.

    2.2. Oxidation cell

    As previously mentioned, several past researchers investigated the protective film formed on a Mg-alloy melt surface [38,39,[46], [47], [48], [49], [50], [51], [52]. During these experiments, the amount of cover gas used was sufficient, thus suppressing the depletion of fluorides in the cover gas. The experiment described in this section used a sealed oxidation cell, which limited the supply of cover gas, to study the evolution of the oxide films of entrainment defects. The cover gas contained in the oxidation cell was regarded as large-size “entrained bubble”.

    As shown in Fig. 2, the main body of the oxidation cell was a closed-end mild steel tube which had an inner length of 400 mm, and an inner diameter of 32 mm. A water-cooled copper tube was wrapped around the upper section of the cell. When the tube was heated, the cooling system created a temperature difference between the upper and lower sections, causing the interior gas to convect within the tube. The temperature was monitored by a type-K thermocouple located at the top of the crucible. Nie et al. [53] suggested that the SF6 cover gas would react with the steel wall of the holding furnace when they investigated the surface film of a Mg-alloy melt. To avoid this reaction, the interior surface of the steel oxidation cell (shown in Fig. 2) and the upper half section of the thermocouple were coated with boron nitride (the Mg-alloy was not in contact with boron nitride).

    Fig. 2. Schematic of the oxidation cell used to study the evolution of the oxide films of the entrainment defects (unit mm).

    During the experiment, a block of solid AZ91 alloy was placed in a magnesia crucible located at the bottom of the oxidation cell. The cell was heated to 100 °C in an electric resistance furnace under a gas flow rate of 1 L/min. The cell was held at this temperature for 20 min, to replace the original trapped atmosphere (i.e. air). Then, the oxidation cell was further heated to 700 °C, melting the AZ91 sample. The gas inlet and exit valves were then closed, creating a sealed environment for oxidation under a limited supply of cover gas. The oxidation cell was then held at 700 ± 10 °C for periods of time from 5 min to 30 min in 5-min intervals. At the end of each holding time, the cell was quenched in water. After cooling to room temperature, the oxidised sample was sectioned, polished, and subsequently examined by SEM.

    3. Results

    3.1. Structure and composition of the entrainment defects formed in SF6/air

    The structure and composition of the entrainment defect formed in the AZ91 castings under a cover gas of 0.5%SF6/air was observed by SEM and EDS. The results indicate that there exist two types of entrainment defects which are sketched in Fig. 3: (1) Type A defect whose oxide film has a traditional single-layered structure and (2) Type B defect, whose oxide film has two layers. The details of these defects were introduced in the following. Here it should be noticed that, as the entrainment defects are also known as biofilms or double oxide film, the oxide films of Type B defect were referred to as “multi-layered oxide film” or “multi-layered structure” in the present work to avoid a confusing description such as “the double-layered oxide film of a double oxide film defect”.

    Fig. 3. Schematic of the different types of entrainment defects found in AZ91 castings. (a) Type A defect with a single-layered oxide film and (b) Type B defect with two-layered oxide film.

    Fig. 4(a-b) shows a Type A defect having a compact single-layered oxide film with about 0.4 µm thickness. Oxygen, fluorine, magnesium and aluminium were detected in this film (Fig. 4c). It is speculated that oxide film is the mixture of fluoride and oxide of magnesium and aluminium. The detection of fluorine revealed that an entrained cover gas was contained in the formation of this defect. That is to say that the pores shown in Fig. 4(a) were not shrinkage defects or hydrogen porosity, but entrainment defects. The detection of aluminium was different with Xiong and Wang’s previous study [47,48], which showed that no aluminium was contained in their surface film of an AZ91 melt protected by a SF6 cover gas. Sulphur could not be clearly recognized in the element map, but there was a S-peak in the corresponding ESD spectrum.

    Fig. 4. (a) A Type A entrainment defect formed in SF6/air and having a single-layered oxide film, (b) the oxide film of this defect, (c) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area highlighted in (b).

    Fig. 5(a-b) shows a Type B entrainment defect having a multi-layered oxide film. The compact outer layers of the oxide films were enriched with fluorine and oxygen (Fig. 5c), while their relatively porous inner layers were only enriched with oxygen (i.e., poor in fluorine) and partly grew together, thus forming a sandwich-like structure. Therefore, it is speculated that the outer layer is the mixture of fluoride and oxide, while the inner layer is mainly oxide. Sulphur could only be recognized in the EDX spectrum and could not be clearly identified in the element map, which might be due to the small S-content in the cover gas (i.e., 0.5% volume content of SF6 in the cover gas). In this oxide film, aluminium was contained in the outer layer of this oxide film but could not be clearly detected in the inner layer. Moreover, the distribution of Al seems to be uneven. It can be found that, in the right side of the defect, aluminium exists in the film but its concentration can not be identified to be higher than the matrix. However, there is a small area with much higher aluminium concentration in the left side of the defect. Such an uneven distribution of aluminium was also observed in other defects (shown in the following), and it is the result of the formation of some oxide particles in or under the film.

    Fig. 5. (a) A Type B entrainment defect formed in SF6/air and having a multi-layered oxide film, (b) the oxide films of this defect have grown together, (c) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area shown in (b).

    Figs. 4 and 5 show cross sectional observations of the entrainment defects formed in the AZ91 alloy sample cast under a cover gas of SF6/air. It is not sufficient to characterize the entrainment defects only by the figures observed from the two-dimensional section. To have a further understanding, the surface of the entrainment defects (i.e. the oxide film) was further studied by observing the fracture surface of the test bars.

    Fig. 6(a) shows fracture surfaces of an AZ91 alloy tensile test bar produced in SF6/air. Symmetrical dark regions can be seen on both sides of the fracture surfaces. Fig. 6(b) shows boundaries between the dark and bright regions. The bright region consisted of jagged and broken features, while the surface of the dark region was relatively smooth and flat. In addition, the EDS results (Fig. 6c-d and Table 2) show that fluorine, oxygen, sulphur, and nitrogen were only detected in the dark regions, indicating that the dark regions were surface protective films entrained into the melt. Therefore, it could be suggested that the dark regions were an entrainment defect with consideration of their symmetrical nature. Similar defects on fracture surfaces of Al-alloy castings have been previously reported [7]. Nitrides were only found in the oxide films on the test-bar fracture surfaces but never detected in the cross-sectional samples shown in Figs. 4 and 5. An underlying reason is that the nitrides contained in these samples may have hydrolysed during the sample polishing process [54].

    Fig. 6. (a) A pair of the fracture surfaces of a AZ91 alloy tensile test bar produced under a cover gas of SF6/air. The dimension of the fracture surface is 5 mm × 6 mm, (b) a section of the boundary between the dark and bright regions shown in (a), (c-d) EDS spectrum of the (c) bright regions and (d) dark regions, (e) schematic of an entrainment defect contained in a test bar.

    Table 2. EDS results (wt.%) corresponding to the regions shown in Fig. 6 (cover gas: SF6/air).

    Empty CellCOMgFAlZnSN
    Dark region in Fig. 6(b)3.481.3279.130.4713.630.570.080.73
    Bright region in Fig. 6(b)3.58–84.48–11.250.68––

    In conjunction with the cross-sectional observation of the defects shown in Figs. 4 and 5, the structure of an entrainment defect contained in a tensile test bar was sketched as shown in Fig. 6(e). The defect contained an entrained gas enclosed by its oxide film, creating a void section inside the test bar. When the tensile force applied on the defect during the fracture process, the crack was initiated at the void section and propagated along the entrainment defect, since cracks would be propagated along the weakest path [55]. Therefore, when the test bar was finally fractured, the oxide films of entrainment defect appeared on both fracture surfaces of the test bar, as shown in Fig. 6(a).

    3.2. Structure and composition of the entrainment defects formed in SF6/CO2

    Similar to the entrainment defect formed in SF6/air, the defects formed under a cover gas of 0.5%SF6/CO2 also had two types of oxide films (i.e., single-layered and multi-layered types). Fig. 7(a) shows an example of the entrainment defects containing a multi-layered oxide film. A magnified observation to the defect (Fig. 7b) shows that the inner layers of the oxide films had grown together, presenting a sandwich-like structure, which was similar to the defects formed in an atmosphere of SF6/air (Fig. 5b). An EDS spectrum (Fig. 7c) revealed that the joint area (inner layer) of this sandwich-like structure mainly contained magnesium oxides. Peaks of fluorine, sulphur, and aluminium were recognized in this EDS spectrum, but their amount was relatively small. In contrast, the outer layers of the oxide films were compact and composed of a mixture of fluorides and oxides (Fig. 7d-e).

    Fig. 7. (a) An example of entrainment defects formed in SF6/CO2 and having a multi-layered oxide film, (b) magnified observation of the defect, showing the inner layer of the oxide films has grown together, (c) EDS spectrum of the point denoted in (b), (d) outer layer of the oxide film, (e) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area shown in (d).

    Fig. 8(a) shows an entrainment defect on the fracture surfaces of an AZ91 alloy tensile test bar, which was produced in an atmosphere of 0.5%SF6/CO2. The corresponding EDS results (Table 3) showed that oxide film contained fluorides and oxides. Sulphur and nitrogen were not detected. Besides, a magnified observation (Fig. 8b) indicated spots on the oxide film surface. The diameter of the spots ranged from hundreds of nanometres to a few micron meters.

    Fig. 8. (a) A pair of the fracture surfaces of a AZ91 alloy tensile test bar, produced in an atmosphere of SF6/CO2. The dimension of the fracture surface is 5 mm × 6 mm, (b) surface appearance of the oxide films on the fracture surfaces, showing spots on the film surface.

    To further reveal the structure and composition of the oxide film clearly, the cross-section of the oxide film on a test-bar fracture surface was onsite exposed using the FIB technique (Fig. 9). As shown in Fig. 9a, a continuous oxide film was found between the platinum coating layer and the Mg-Al alloy substrate. Fig. 9 (b-c) shows a magnified observation to oxide films, indicating a multi-layered structure (denoted by the red box in Fig. 9c). The bottom layer was enriched with fluorine and oxygen and should be the mixture of fluoride and oxide, which was similar to the “outer layer” shown in Figs. 5 and 7, while the only-oxygen-enriched top layer was similar to the “inner layer” shown in Figs. 5 and 7.

    Fig. 9. (a) A cross-sectional observation of the oxide film on the fracture surface of the AZ91 casting produced in SF6/CO2, exposed by FIB, (b) a magnified observation of area highlighted in (a), and (c) SEM-EDS elements map of the area shown in (b), obtained by CFEI Quanta 3D FEG FIB-SEM.

    Except the continuous film, some individual particles were also observed in or below the continuous film, as shown in Fig. 9. An Al-enriched particle was detected in the left side of the oxide film shown in Fig. 9b and might be speculated to be spinel Mg2AlO4 because it also contains abundant magnesium and oxygen elements. The existing of such Mg2AlO4 particles is responsible for the high concentration of aluminium in small areas of the observed film and the uneven distribution of aluminium, as shown in Fig. 5(c). Here it should be emphasized that, although the other part of the bottom layer of the continuous oxide film contains less aluminium than this Al-enriched particle, the Fig. 9c indicated that the amount of aluminium in this bottom layer was still non-negligible, especially when comparing with the outer layer of the film. Below the right side of the oxide film shown in Fig. 9b, a particle was detected and speculated to be MgO because it is rich in Mg and O. According to Wang’s result [56], lots of discrete MgO particles can be formed on the surface of the Mg melt by the oxidation of Mg melt and Mg vapor. The MgO particles observed in our present work may be formed due to the same reasons. While, due to the differences in experimental conditions, less Mg melt can be vapored or react with O2, thus only a few of MgO particles formed in our work. An enrichment of carbon was also found in the film, revealing that CO2 was able to react with the melt, thus forming carbon or carbides. This carbon concentration was consistent with the relatively high carbon content of the oxide film shown in Table 3 (i.e., the dark region). In the area next to the oxide film.

    Table 3. EDS results (wt.%) corresponding to the regions shown in Fig. 8 (cover gas: SF6/ CO2).

    Empty CellCOMgFAlZnSN
    Dark region in Fig. 8(a)7.253.6469.823.827.030.86––
    Bright region in Fig. 8(a)2.100.4482.83–13.261.36––

    This cross-sectional observation of the oxide film on a test bar fracture surface (Fig. 9) further verified the schematic of the entrainment defect shown in Fig. 6(e). The entrainment defects formed in different atmospheres of SF6/CO2 and SF6/air had similar structures, but their compositions were different.

    3.3. Evolution of the oxide films in the oxidation cell

    The results in Section 3.1 and 3.2 have shown the structures and compositions of entrainment defects formed in AZ91 castings under cover gases of SF6/air and SF6/CO2. Different stages of the oxidation reaction may lead to the different structures and compositions of entrainment defects. Although Campbell has conjectured that an entrained gas may react with the surrounding melt, it is rarely reported that the reaction occurring between the Mg-alloy melt and entrapped cover gas. Previous researchers normally focus on the reaction between a Mg-alloy melt and the cover gas in an open environment [38,39,[46], [47], [48], [49], [50], [51], [52], which was different from the situation of a cover gas trapped into the melt. To further understand the formation of the entrainment defect in an AZ91 alloy, the evolution process of oxide films of the entrainment defect was further studied using an oxidation cell.

    Fig. 10 (a and d) shows a surface film held for 5 min in the oxidation cell, protected by 0.5%SF6/air. There was only one single layer consisting of fluoride and oxide (MgF2 and MgO). In this surface film. Sulphur was detected in the EDS spectrum, but its amount was too small to be recognized in the element map. The structure and composition of this oxide film was similar to the single-layered films of entrainment defects shown in Fig. 4.

    Fig. 10. Oxide films formed in the oxidation cell under a cover gas of 0.5%SF6/air and held at 700 °C for (a) 5 min; (b) 10 min; (c) 30 min, and (d-f) the SEM-EDS element maps (using Philips JEOL7000) corresponding to the oxide film shown in (a-c) respectively, (d) 5 min; (e) 10 min; (f) 30 min. The red points in (c and f) are the location references, denoting the boundary of the F-enriched layer in different element maps.

    After a holding time of 10 min, a thin (O, S)-enriched top layer (around 700 nm) appeared upon the preliminary F-enriched film, forming a multi-layered structure, as shown in Fig. 10(b and e). The thickness of the (O, S)-enriched top layer increased with increased holding time. As shown in Fig. 10(c and f), the oxide film held for 30 min also had a multi-layered structure, but the thickness of its (O, S)-enriched top layer (around 2.5 µm) was higher than the that of the 10-min oxide film. The multi-layered oxide films shown in Fig. 10(b-c) presented a similar appearance to the films of the sandwich-like defect shown in Fig. 5.

    The different structures of the oxide films shown in Fig. 10 indicated that fluorides in the cover gas would be preferentially consumed due to the reaction with the AZ91 alloy melt. After the depletion of fluorides, the residual cover gas reacted further with the liquid AZ91 alloy, forming the top (O, S)-enriched layer in the oxide film. Therefore, the different structures and compositions of entrainment defects shown in Figs. 4 and 5 may be due to an ongoing oxidation reaction between melt and entrapped cover gas.

    This multi-layered structure has not been reported in previous publications concerning the protective surface film formed on a Mg-alloy melt [38,[46], [47], [48], [49], [50], [51]. This may be due to the fact that previous researchers carried out their experiments with an un-limited amount of cover gas, creating a situation where the fluorides in the cover gas were not able to become depleted. Therefore, the oxide film of an entrainment defect had behaviour traits similar to the oxide films shown in Fig. 10, but different from the oxide films formed on the Mg-alloy melt surface reported in [38,[46], [47], [48], [49], [50], [51].

    Similar with the oxide films held in SF6/air, the oxide films formed in SF6/CO2 also had different structures with different holding times in the oxidation cell. Fig. 11(a) shows an oxide film, held on an AZ91 melt surface under a cover gas of 0.5%SF6/CO2 for 5 min. This film had a single-layered structure consisting of MgF2. The existence of MgO could not be confirmed in this film. After the holding time of 30 min, the film had a multi-layered structure; the inner layer was of a compact and uniform appearance and composed of MgF2, while the outer layer is the mixture of MgF2 and MgO. Sulphur was not detected in this film, which was different from the surface film formed in 0.5%SF6/air. Therefore, fluorides in the cover gas of 0.5%SF6/CO2 were also preferentially consumed at an early stage of the film growth process. Compared with the film formed in SF6/air, the MgO in film formed in SF6/CO2 appeared later and sulphide did not appear within 30 min. It may mean that the formation and evolution of film in SF6/air is faster than SF6/CO2. CO2 may have subsequently reacted with the melt to form MgO, while sulphur-containing compounds accumulated in the cover gas and reacted to form sulphide in very late stage (may after 30 min in oxidation cell).

    Fig. 11. Oxide films formed in the oxidation cell under a cover gas of 0.5%SF6/CO2, and their SEM-EDS element maps (using Philips JEOL7000). They were held at 700 °C for (a) 5 min; (b) 30 min. The red points in (b) are the location references, denoting the boundary between the top and bottom layers in the oxide film.

    4. Discussion

    4.1. Evolution of entrainment defects formed in SF6/air

    HSC software from Outokumpu HSC Chemistry for Windows (http://www.hsc-chemistry.net/) was used to carry out thermodynamic calculations needed to explore the reactions which might occur between the trapped gases and liquid AZ91 alloy. The solutions to the calculations suggest which products are most likely to form in the reaction process between a small amount of cover gas (i.e., the amount within a trapped bubble) and the AZ91-alloy melt.

    In the trials, the pressure was set to 1 atm, and the temperature set to 700 °C. The amount of the cover gas was assumed to be 7 × 10−7 kg, with a volume of approximately 0.57 cm3 (3.14 × 10−8 kmol) for 0.5%SF6/air, and 0.35 cm3 (3.12 × 10−8 kmol) for 0.5%SF6/CO2. The amount of the AZ91 alloy melt in contact with the trapped gas was assumed to be sufficient to complete all reactions. The decomposition products of SF6 were SF5, SF4, SF3, SF2, F2, S(g), S2(g) and F(g) [57], [58], [59], [60].

    Fig. 12 shows the equilibrium diagram of the thermodynamic calculation of the reaction between the AZ91 alloy and 0.5%SF6/air. In the diagram, the reactants and products with less than 10−15 kmol have not been shown, as this was 5 orders of magnitude less than the amount of SF6 present (≈ 1.57 × 10−10 kmol) and therefore would not affect the observed process in a practical way.

    Fig. 12. An equilibrium diagram for the reaction between 7e-7 kg 0.5%SF6/air and a sufficient amount of AZ91 alloy. The X axis is the amount of AZ91 alloy melt having reacted with the entrained gas, and the vertical Y-axis is the amount of the reactants and products.

    This reaction process could be divided into 3 stages.

    Stage 1: The formation of fluorides. the AZ91 melt preferentially reacted with SF6 and its decomposition products, producing MgF2, AlF3, and ZnF2. However, the amount of ZnF2 may have been too small to be detected practically (1.25 × 10−12 kmol of ZnF2 compared with 3 × 10−10 kmol of MgF2), which may be the reason why Zn was not detected in any the oxide films shown in Sections 3.1–3.3. Meanwhile, sulphur accumulated in the residual gas as SO2.

    Stage 2: The formation of oxides. After the liquid AZ91 alloy had depleted all the available fluorides in the entrapped gas, the amount of AlF3 and ZnF2 quickly reduced due to a reaction with Mg. O2(g) and SO2 reacted with the AZ91 melt, forming MgO, Al2O3, MgAl2O4, ZnO, ZnSO4 and MgSO4. However, the amount of ZnO and ZnSO4 would have been too small to be found practically by EDS (e.g. 9.5 × 10−12 kmol of ZnO,1.38 × 10−14 kmol of ZnSO4, in contrast to 4.68 × 10−10 kmol of MgF2, when the amount of AZ91 on the X-axis is 2.5 × 10−9 kmol). In the experimental cases, the concentration of F in the cover gas is very low, whole the concentration f O is much higher. Therefore, the stage 1 and 2, i.e, the formation of fluoride and oxide may happen simultaneously at the beginning of the reaction, resulting in the formation of a singer-layered mixture of fluoride and oxide, as shown in Figs. 4 and 10(a). While an inner layer consisted of oxides but fluorides could form after the complete depletion of F element in the cover gas.

    Stages 1- 2 theoretically verified the formation process of the multi-layered structure shown in Fig. 10.

    The amount of MgAl2O4 and Al2O3 in the oxide film was of a sufficient amount to be detected, which was consistent with the oxide films shown in Fig. 4. However, the existence of aluminium could not be recognized in the oxide films grown in the oxidation cell, as shown in Fig. 10. This absence of Al may be due to the following reactions between the surface film and AZ91 alloy melt:(1)

    Al2O3 + 3Mg + = 3MgO + 2Al, △G(700 °C) = -119.82 kJ/mol(2)

    Mg + MgAl2O4 = MgO + Al, △G(700 °C) =-106.34 kJ/molwhich could not be simulated by the HSC software since the thermodynamic calculation was carried out under an assumption that the reactants were in full contact with each other. However, in a practical process, the AZ91 melt and the cover gas would not be able to be in contact with each other completely, due to the existence of the protective surface film.

    Stage 3: The formation of Sulphide and nitride. After a holding time of 30 min, the gas-phase fluorides and oxides in the oxidation cell had become depleted, allowing the melt reaction with the residual gas, forming an additional sulphur-enriched layer upon the initial F-enriched or (F, O)-enriched surface film, thus resulting in the observed multi-layered structure shown in Fig. 10 (b and c). Besides, nitrogen reacted with the AZ91 melt until all reactions were completed. The oxide film shown in Fig. 6 may correspond to this reaction stage due to its nitride content. However, the results shows that the nitrides were not detected in the polished samples shown in Figs. 4 and 5, but only found on the test bar fracture surfaces. The nitrides may have hydrolysed during the sample preparation process, as follows [54]:(3)

    Mg3N2 + 6H2O =3Mg(OH)2 + 2NH3↑(4)

    AlN+ 3H2O =Al(OH)3 + NH3↑

    In addition, Schmidt et al. [61] found that Mg3N2 and AlN could react to form ternary nitrides (Mg3AlnNn+2, n= 1, 2, 3…). HSC software did not contain the database of ternary nitrides, and it could not be added into the calculation. The oxide films in this stage may also contain ternary nitrides.

    4.2. Evolution of entrainment defects formed in SF6/CO2

    Fig. 13 shows the results of the thermodynamic calculation between AZ91 alloy and 0.5%SF6/CO2. This reaction processes can also be divided into three stages.

    Fig. 13. An equilibrium diagram for the reaction between 7e-7 kg 0.5%SF6/CO2 and a sufficient amount of AZ91 alloy. The X axis denotes the amount of Mg alloy melt having reacted with the entrained gas, and the vertical Y-axis denotes the amounts of the reactants and products.

    Stage 1: The formation of fluorides. SF6 and its decomposition products were consumed by the AZ91 melt, forming MgF2, AlF3, and ZnF2. As in the reaction of AZ91 in 0.5%SF6/air, the amount of ZnF2 was too small to be detected practically (1.51 × 10−13 kmol of ZnF2 compared with 2.67 × 10−10 kmol of MgF2). Sulphur accumulated in the residual trapped gas as S2(g) and a portion of the S2(g) reacted with CO2, to form SO2 and CO. The products in this reaction stage were consistent with the film shown in Fig. 11(a), which had a single layer structure that contained fluorides only.

    Stage 2: The formation of oxides. AlF3 and ZnF2 reacted with the Mg in the AZ91 melt, forming MgF2, Al and Zn. The SO2 began to be consumed, producing oxides in the surface film and S2(g) in the cover gas. Meanwhile, the CO2 directly reacted with the AZ91 melt, forming CO, MgO, ZnO, and Al2O3. The oxide films shown in Figs. 9 and 11(b) may correspond to this reaction stage due to their oxygen-enriched layer and multi-layered structure.

    The CO in the cover gas could further react with the AZ91 melt, producing C. This carbon may further react with Mg to form Mg carbides, when the temperature reduced (during solidification period) [62]. This may be the reason for the high carbon content in the oxide film shown in Figs. 8–9. Liang et al. [39] also reported carbon-detection in an AZ91 alloy surface film protected by SO2/CO2. The produced Al2O3 may be further combined with MgO, forming MgAl2O4 [63]. As discussed in Section 4.1, the alumina and spinel can react with Mg, causing an absence of aluminium in the surface films, as shown in Fig. 11.

    Stage 3: The formation of Sulphide. the AZ91 melt began to consume S2(g) in the residual entrapped gas, forming ZnS and MgS. These reactions did not occur until the last stage of the reaction process, which could be the reason why the S-content in the defect shown Fig. 7(c) was small.

    In summary, thermodynamic calculations indicate that the AZ91 melt will react with the cover gas to form fluorides firstly, then oxides and sulphides in the last. The oxide film in the different reaction stages would have different structures and compositions.

    4.3. Effect of the carrier gases on consumption of the entrained gas and the reproducibility of AZ91 castings

    The evolution processes of entrainment defects, formed in SF6/air and SF6/CO2, have been suggested in Sections 4.1 and 4.2. The theoretical calculations were verified with respect to the corresponding oxide films found in practical samples. The atmosphere within an entrainment defect could be efficiently consumed due to the reaction with liquid Mg-alloy, in a scenario dissimilar to the Al-alloy system (i.e., nitrogen in an entrained air bubble would not efficiently react with Al-alloy melt [64,65], however, nitrogen would be more readily consumed in liquid Mg alloys, commonly referred to as “nitrogen burning” [66]).

    The reaction between the entrained gas and the surrounding liquid Mg-alloy converted the entrained gas into solid compounds (e.g. MgO) within the oxide film, thus reducing the void volume of the entrainment defect and hence probably causing a collapse of the defect (e.g., if an entrained gas of air was depleted by the surrounding liquid Mg-alloy, under an assumption that the melt temperature is 700 °C and the depth of liquid Mg-alloy is 10 cm, the total volume of the final solid products would be 0.044% of the initial volume taken by the entrapped air).

    The relationship between the void volume reduction of entrainment defects and the corresponding casting properties has been widely studied in Al-alloy castings. Nyahumwa and Campbell [16] reported that the Hot Isostatic Pressing (HIP) process caused the entrainment defects in Al-alloy castings to collapse and their oxide surfaces forced into contact. The fatigue lives of their castings were improved after HIP. Nyahumwa and Campbell [16] also suggested a potential bonding of the double oxide films that were in contact with each other, but there was no direct evidence to support this. This binding phenomenon was further investigated by Aryafar et.al.[8], who re-melted two Al-alloy bars with oxide skins in a steel tube and then carried out a tensile strength test on the solidified sample. They found that the oxide skins of the Al-alloy bars strongly bonded with each other and became even stronger with an extension of the melt holding time, indicating a potential “healing” phenomenon due to the consumption of the entrained gas within the double oxide film structure. In addition, Raidszadeh and Griffiths [9,19] successfully reduced the negative effect of entrainment defects on the reproducibility of Al-alloy castings, by extending the melt holding time before solidification, which allowed the entrained gas to have a longer time to react with the surrounding melt.

    With consideration of the previous work mentioned, the consumption of the entrained gas in Mg-alloy castings may diminish the negative effect of entrainment defects in the following two ways.

    (1) Bonding phenomenon of the double oxide films. The sandwich-like structure shown in Fig. 5 and 7 indicated a potential bonding of the double oxide film structure. However, more evidence is required to quantify the increase in strength due to the bonding of the oxide films.

    (2) Void volume reduction of entrainment defects. The positive effect of void-volume reduction on the quality of castings has been widely demonstrated by the HIP process [67]. As the evolution processes discussed in Section 4.1–4.2, the oxide films of entrainment defects can grow together due to an ongoing reaction between the entrained gas and surrounding AZ91 alloy melt. The volume of the final solid products was significant small compared with the entrained gas (i.e., 0.044% as previously mentioned).

    Therefore, the consumption rate of the entrained gas (i.e., the growth rate of oxide films) may be a critical parameter for improving the quality of AZ91 alloy castings. The oxide film growth rate in the oxidization cell was accordingly further investigated.

    Fig. 14 shows a comparison of the surface film growth rates in different cover gases (i.e., 0.5%SF6/air and 0.5%SF6/CO2). 15 random points on each sample were selected for film thickness measurements. The 95% confidence interval (95%CI) was computed under an assumption that the variation of the film thickness followed a Gaussian distribution. It can be seen that all the surface films formed in 0.5%SF6/air grew faster than those formed in 0.5%SF6/CO2. The different growth rates suggested that the entrained-gas consumption rate of 0.5%SF6/air was higher than that of 0.5%SF6/CO2, which was more beneficial for the consumption of the entrained gas.

    Fig. 14. A comparison of the AZ91 alloy oxide film growth rates in 0.5%SF6/air and 0.5%SF6/CO2

    It should be noted that, in the oxidation cell, the contact area of liquid AZ91 alloy and cover gas (i.e. the size of the crucible) was relatively small with consideration of the large volume of melt and gas. Consequently, the holding time for the oxide film growth within the oxidation cell was comparatively long (i.e., 5–30 min). However, the entrainment defects contained in a real casting are comparatively very small (i.e., a few microns size as shown in Figs. 3–6, and [7]), and the entrained gas is fully enclosed by the surrounding melt, creating a relatively large contact area. Hence the reaction time for cover gas and the AZ91 alloy melt may be comparatively short. In addition, the solidification time of real Mg-alloy sand castings can be a few minutes (e.g. Guo [68] reported that a Mg-alloy sand casting with 60 mm diameter required 4 min to be solidified). Therefore, it can be expected that an entrained gas trapped during an Mg-alloy melt pouring process will be readily consumed by the surrounding melt, especially for sand castings and large-size castings, where solidification times are long.

    Therefore, the different cover gases (0.5%SF6/air and 0.5%SF6/CO2) associated with different consumption rates of the entrained gases may affect the reproducibility of the final castings. To verify this assumption, the AZ91 castings produced in 0.5%SF6/air and 0.5%SF6/CO2 were machined into test bars for mechanical evaluation. A Weibull analysis was carried out using both linear least square (LLS) method and non-linear least square (non-LLS) method [69].

    Fig. 15(a-b) shows a traditional 2-p linearized Weibull plot of the UTS and elongation of the AZ91 alloy castings, obtained by the LLS method. The estimator used is P= (i-0.5)/N, which was suggested to cause the lowest bias among all the popular estimators [69,70]. The casting produced in SF6/air has an UTS Weibull moduli of 16.9, and an elongation Weibull moduli of 5.0. In contrast, the UTS and elongation Weibull modulus of the casting produced in SF6/CO2 are 7.7 and 2.7 respectively, suggesting that the reproducibility of the casting protected by SF6/CO2 were much lower than that produced in SF6/air.

    Fig. 15. The Weibull modulus of AZ91 castings produced in different atmospheres, estimated by (a-b) the linear least square method, (c-d) the non-linear least square method, where SSR is the sum of residual squares.

    In addition, the author’s previous publication [69] demonstrated a shortcoming of the linearized Weibull plots, which may cause a higher bias and incorrect R2 interruption of the Weibull estimation. A Non-LLS Weibull estimation was therefore carried out, as shown in Fig. 15 (c-d). The UTS Weibull modulus of the SF6/air casting was 20.8, while the casting produced under SF6/CO2 had a lower UTS Weibull modulus of 11.4, showing a clear difference in their reproducibility. In addition, the SF6/air elongation (El%) dataset also had a Weibull modulus (shape = 5.8) higher than the elongation dataset of SF6/CO2 (shape = 3.1). Therefore, both the LLS and Non-LLS estimations suggested that the SF6/air casting has a higher reproducibility than the SF6/CO2 casting. It supports the method that the use of air instead of CO2 contributes to a quicker consumption of the entrained gas, which may reduce the void volume within the defects. Therefore, the use of 0.5%SF6/air instead of 0.5%SF6/CO2 (which increased the consumption rate of the entrained gas) improved the reproducibility of the AZ91 castings.

    However, it should be noted that not all the Mg-alloy foundries followed the casting process used in present work. The Mg-alloy melt in present work was degassed, thus reducing the effect of hydrogen on the consumption of the entrained gas (i.e., hydrogen could diffuse into the entrained gas, potentially suppressing the depletion of the entrained gas [7,71,72]). In contrast, in Mg-alloy foundries, the Mg-alloy melt is not normally degassed, since it was widely believed that there is not a ‘gas problem’ when casting magnesium and hence no significant change in tensile properties [73]. Although studies have shown the negative effect of hydrogen on the mechanical properties of Mg-alloy castings [41,42,73], a degassing process is still not very popular in Mg-alloy foundries.

    Moreover, in present work, the sand mould cavity was flushed with the SF6 cover gas prior to pouring [22]. However, not all the Mg-alloy foundries flushed the mould cavity in this way. For example, the Stone Foundry Ltd (UK) used sulphur powder instead of the cover-gas flushing. The entrained gas within their castings may be SO2/air, rather than the protective gas.

    Therefore, although the results in present work have shown that using air instead of CO2 improved the reproducibility of the final casting, it still requires further investigations to confirm the effect of carrier gases with respect to different industrial Mg-alloy casting processes.

    7. Conclusion

    Entrainment defects formed in an AZ91 alloy were observed. Their oxide films had two types of structure: single-layered and multi-layered. The multi-layered oxide film can grow together forming a sandwich-like structure in the final casting.2.

    Both the experimental results and the theoretical thermodynamic calculations demonstrated that fluorides in the trapped gas were depleted prior to the consumption of sulphur. A three-stage evolution process of the double oxide film defects has been suggested. The oxide films contained different combinations of compounds, depending on the evolution stage. The defects formed in SF6/air had a similar structure to those formed in SF6/CO2, but the compositions of their oxide films were different. The oxide-film formation and evolution process of the entrainment defects were different from that of the Mg-alloy surface films previous reported (i.e., MgO formed prior to MgF2).3.

    The growth rate of the oxide film was demonstrated to be greater under SF6/air than SF6/CO2, contributing to a quicker consumption of the damaging entrapped gas. The reproducibility of an AZ91 alloy casting improved when using SF6/air instead of SF6/CO2.

    Acknowledgements

    The authors acknowledge funding from the EPSRC LiME grant EP/H026177/1, and the help from Dr W.D. Griffiths and Mr. Adrian Carden (University of Birmingham). The casting work was carried out in University of Birmingham.

    Reference

    [1]

    M.K. McNutt, SALAZAR K.

    Magnesium, Compounds & Metal, U.S. Geological Survey and U.S. Department of the Interior

    Reston, Virginia (2013)

    Google Scholar[2]

    Magnesium

    Compounds & Metal, U.S. Geological Survey and U.S. Department of the Interior

    (1996)

    Google Scholar[3]

    I. Ostrovsky, Y. Henn

    ASTEC’07 International Conference-New Challenges in Aeronautics, Moscow (2007), pp. 1-5

    Aug 19-22

    View Record in ScopusGoogle Scholar[4]

    Y. Wan, B. Tang, Y. Gao, L. Tang, G. Sha, B. Zhang, N. Liang, C. Liu, S. Jiang, Z. Chen, X. Guo, Y. Zhao

    Acta Mater., 200 (2020), pp. 274-286

    ArticleDownload PDFView Record in Scopus[5]

    J.T.J. Burd, E.A. Moore, H. Ezzat, R. Kirchain, R. Roth

    Appl. Energy, 283 (2021), Article 116269

    ArticleDownload PDFView Record in Scopus[6]

    A.M. Lewis, J.C. Kelly, G.A. Keoleian

    Appl. Energy, 126 (2014), pp. 13-20

    ArticleDownload PDFView Record in Scopus[7]

    J. Campbell

    Castings

    Butterworth-Heinemann, Oxford (2004)

    Google Scholar[8]

    M. Aryafar, R. Raiszadeh, A. Shalbafzadeh

    J. Mater. Sci., 45 (2010), pp. 3041-3051 View PDF

    CrossRefView Record in Scopus[9]

    R. Raiszadeh, W.D. Griffiths

    Metall. Mater. Trans. B-Process Metall. Mater. Process. Sci., 42 (2011), pp. 133-143 View PDF

    CrossRefView Record in Scopus[10]

    R. Raiszadeh, W.D. Griffiths

    J. Alloy. Compd., 491 (2010), pp. 575-580

    ArticleDownload PDFView Record in Scopus[11]

    L. Peng, G. Zeng, T.C. Su, H. Yasuda, K. Nogita, C.M. Gourlay

    JOM, 71 (2019), pp. 2235-2244 View PDF

    CrossRefView Record in Scopus[12]

    S. Ganguly, A.K. Mondal, S. Sarkar, A. Basu, S. Kumar, C. Blawert

    Corros. Sci., 166 (2020)[13]

    G.E. Bozchaloei, N. Varahram, P. Davami, S.K. Kim

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 548 (2012), pp. 99-105

    View Record in Scopus[14]

    S. Fox, J. Campbell

    Scr. Mater., 43 (2000), pp. 881-886

    ArticleDownload PDFView Record in Scopus[15]

    M. Cox, R.A. Harding, J. Campbell

    Mater. Sci. Technol., 19 (2003), pp. 613-625

    View Record in Scopus[16]

    C. Nyahumwa, N.R. Green, J. Campbell

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 32 (2001), pp. 349-358

    View Record in Scopus[17]

    A. Ardekhani, R. Raiszadeh

    J. Mater. Eng. Perform., 21 (2012), pp. 1352-1362 View PDF

    CrossRefView Record in Scopus[18]

    X. Dai, X. Yang, J. Campbell, J. Wood

    Mater. Sci. Technol., 20 (2004), pp. 505-513

    View Record in Scopus[19]

    E.M. Elgallad, M.F. Ibrahim, H.W. Doty, F.H. Samuel

    Philos. Mag., 98 (2018), pp. 1337-1359 View PDF

    CrossRefView Record in Scopus[20]

    W.D. Griffiths, N.W. Lai

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 38A (2007), pp. 190-196 View PDF

    CrossRefView Record in Scopus[21]

    A.R. Mirak, M. Divandari, S.M.A. Boutorabi, J. Campbell

    Int. J. Cast Met. Res., 20 (2007), pp. 215-220 View PDF

    CrossRefView Record in Scopus[22]

    C. Cingi

    Laboratory of Foundry Engineering

    Helsinki University of Technology, Espoo, Finland (2006)

    Google Scholar[23]

    Y. Jia, J. Hou, H. Wang, Q. Le, Q. Lan, X. Chen, L. Bao

    J. Mater. Process. Technol., 278 (2020), Article 116542

    ArticleDownload PDFView Record in Scopus[24]

    S. Ouyang, G. Yang, H. Qin, S. Luo, L. Xiao, W. Jie

    Mater. Sci. Eng. A, 780 (2020), Article 139138

    ArticleDownload PDFView Record in Scopus[25]

    S.-m. Xiong, X.-F. Wang

    Trans. Nonferrous Met. Soc. China, 20 (2010), pp. 1228-1234

    ArticleDownload PDFView Record in Scopus[26]

    G.V. Research

    Grand View Research

    (2018)

    USA

    Google Scholar[27]

    T. Li, J. Davies

    Metall. Mater. Trans. A, 51 (2020), pp. 5389-5400 View PDF

    CrossRefView Record in Scopus[28]J.F. Fruehling, The University of Michigan, 1970.

    Google Scholar[29]

    S. Couling

    36th Annual World Conference on Magnesium, Norway (1979), pp. 54-57

    View Record in ScopusGoogle Scholar[30]

    S. Cashion, N. Ricketts, P. Hayes

    J. Light Met., 2 (2002), pp. 43-47

    ArticleDownload PDFView Record in Scopus[31]

    S. Cashion, N. Ricketts, P. Hayes

    J. Light Met., 2 (2002), pp. 37-42

    ArticleDownload PDFView Record in Scopus[32]

    K. Aarstad, G. Tranell, G. Pettersen, T.A. Engh

    Various Techniques to Study the Surface of Magnesium Protected by SF6

    TMS (2003)

    Google Scholar[33]

    S.-M. Xiong, X.-L. Liu

    Metall. Mater. Trans. A, 38 (2007), pp. 428-434 View PDF

    CrossRefView Record in Scopus[34]

    T.-S. Shih, J.-B. Liu, P.-S. Wei

    Mater. Chem. Phys., 104 (2007), pp. 497-504

    ArticleDownload PDFView Record in Scopus[35]

    G. Pettersen, E. Øvrelid, G. Tranell, J. Fenstad, H. Gjestland

    Mater. Sci. Eng. A, 332 (2002), pp. 285-294

    ArticleDownload PDFView Record in Scopus[36]

    H. Bo, L.B. Liu, Z.P. Jin

    J. Alloy. Compd., 490 (2010), pp. 318-325

    ArticleDownload PDFView Record in Scopus[37]

    A. Mirak, C. Davidson, J. Taylor

    Corros. Sci., 52 (2010), pp. 1992-2000

    ArticleDownload PDFView Record in Scopus[38]

    B.D. Lee, U.H. Beak, K.W. Lee, G.S. Han, J.W. Han

    Mater. Trans., 54 (2013), pp. 66-73 View PDF

    View Record in Scopus[39]

    W.Z. Liang, Q. Gao, F. Chen, H.H. Liu, Z.H. Zhao

    China Foundry, 9 (2012), pp. 226-230 View PDF

    CrossRef[40]

    U.I. Gol’dshleger, E.Y. Shafirovich

    Combust. Explos. Shock Waves, 35 (1999), pp. 637-644[41]

    A. Elsayed, S.L. Sin, E. Vandersluis, J. Hill, S. Ahmad, C. Ravindran, S. Amer Foundry

    Trans. Am. Foundry Soc., 120 (2012), pp. 423-429[42]

    E. Zhang, G.J. Wang, Z.C. Hu

    Mater. Sci. Technol., 26 (2010), pp. 1253-1258

    View Record in Scopus[43]

    N.R. Green, J. Campbell

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 173 (1993), pp. 261-266

    ArticleDownload PDFView Record in Scopus[44]

    C Reilly, MR Jolly, NR Green

    Proceedings of MCWASP XII – 12th Modelling of Casting, Welding and Advanced Solidifcation Processes, Vancouver, Canada (2009)

    Google Scholar[45]H.E. Friedrich, B.L. Mordike, Springer, Germany, 2006.

    Google Scholar[46]

    C. Zheng, B.R. Qin, X.B. Lou

    Proceedings of the 2010 International Conference on Mechanical, Industrial, and Manufacturing Technologies, ASME (2010), pp. 383-388

    Mimt 2010 View PDF

    CrossRefView Record in ScopusGoogle Scholar[47]

    S.M. Xiong, X.F. Wang

    Trans. Nonferrous Met. Soc. China, 20 (2010), pp. 1228-1234

    ArticleDownload PDFView Record in Scopus[48]

    S.M. Xiong, X.L. Liu

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 38A (2007), pp. 428-434 View PDF

    CrossRefView Record in Scopus[49]

    T.S. Shih, J.B. Liu, P.S. Wei

    Mater. Chem. Phys., 104 (2007), pp. 497-504

    ArticleDownload PDFView Record in Scopus[50]

    K. Aarstad, G. Tranell, G. Pettersen, T.A. Engh

    Magn. Technol. (2003), pp. 5-10[51]

    G. Pettersen, E. Ovrelid, G. Tranell, J. Fenstad, H. Gjestland

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 332 (2002), pp. 285-294

    ArticleDownload PDFView Record in Scopus[52]

    X.F. Wang, S.M. Xiong

    Corros. Sci., 66 (2013), pp. 300-307

    ArticleDownload PDFView Record in Scopus[53]

    S.H. Nie, S.M. Xiong, B.C. Liu

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 422 (2006), pp. 346-351

    ArticleDownload PDFView Record in Scopus[54]

    C. Bauer, A. Mogessie, U. Galovsky

    Zeitschrift Fur Metallkunde, 97 (2006), pp. 164-168 View PDF

    CrossRef[55]

    Q.G. Wang, D. Apelian, D.A. Lados

    J. Light Met., 1 (2001), pp. 73-84

    ArticleDownload PDFView Record in Scopus[56]

    S. Wang, Y. Wang, Q. Ramasse, Z. Fan

    Metall. Mater. Trans. A, 51 (2020), pp. 2957-2974[57]

    S. Hayashi, W. Minami, T. Oguchi, H.J. Kim

    Kag. Kog. Ronbunshu, 35 (2009), pp. 411-415 View PDF

    CrossRefView Record in Scopus[58]

    K. Aarstad

    Norwegian University of Science and Technology

    (2004)

    Google Scholar[59]

    R.L. Wilkins

    J. Chem. Phys., 51 (1969), p. 853

    -&

    View Record in Scopus[60]

    O. Kubaschewski, K. Hesselemam

    Thermo-Chemical Properties of Inorganic Substances

    Springer-Verlag, Belin (1991)

    Google Scholar[61]

    R. Schmidt, M. Strobele, K. Eichele, H.J. Meyer

    Eur. J. Inorg. Chem. (2017), pp. 2727-2735 View PDF

    CrossRefView Record in Scopus[62]

    B. Hu, Y. Du, H. Xu, W. Sun, W.W. Zhang, D. Zhao

    J. Min. Metall. Sect. B-Metall., 46 (2010), pp. 97-103

    View Record in Scopus[63]

    O. Salas, H. Ni, V. Jayaram, K.C. Vlach, C.G. Levi, R. Mehrabian

    J. Mater. Res., 6 (1991), pp. 1964-1981

    View Record in Scopus[64]

    S.S.S. Kumari, U.T.S. Pillai, B.C. Pai

    J. Alloy. Compd., 509 (2011), pp. 2503-2509

    ArticleDownload PDFView Record in Scopus[65]

    H. Scholz, P. Greil

    J. Mater. Sci., 26 (1991), pp. 669-677

    View Record in Scopus[66]

    P. Biedenkopf, A. Karger, M. Laukotter, W. Schneider

    Magn. Technol., 2005 (2005), pp. 39-42

    View Record in Scopus[67]

    H.V. Atkinson, S. Davies

    Metall. Mater. Trans. A, 31 (2000), pp. 2981-3000 View PDF

    CrossRefView Record in Scopus[68]

    E.J. Guo, L. Wang, Y.C. Feng, L.P. Wang, Y.H. Chen

    J. Therm. Anal. Calorim., 135 (2019), pp. 2001-2008 View PDF

    CrossRefView Record in Scopus[69]

    T. Li, W.D. Griffiths, J. Chen

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 48A (2017), pp. 5516-5528 View PDF

    CrossRefView Record in Scopus[70]

    M. Tiryakioglu, D. Hudak

    J. Mater. Sci., 42 (2007), pp. 10173-10179 View PDF

    CrossRefView Record in Scopus[71]

    Y. Yue, W.D. Griffiths, J.L. Fife, N.R. Green

    Proceedings of the 1st International Conference on 3d Materials Science (2012), pp. 131-136 View PDF

    CrossRefView Record in ScopusGoogle Scholar[72]

    R. Raiszadeh, W.D. Griffiths

    Metall. Mater. Trans. B-Process Metall. Mater. Process. Sci., 37 (2006), pp. 865-871

    View Record in Scopus[73]

    Z.C. Hu, E.L. Zhang, S.Y. Zeng

    Mater. Sci. Technol., 24 (2008), pp. 1304-1308 View PDF

    CrossRefView Record in Scopus

    원문보기
    Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.

    알루미늄 합금 겹침 용접 중 용접 형성, 용융 흐름 및 입자 구조에 대한 사인파 발진 레이저 빔의 영향

    린 첸 가오 양 미시 옹 장 춘밍 왕
    Lin Chen , Gaoyang Mi , Xiong Zhang , Chunming Wang *
    중국 우한시 화중과학기술대학 재료공학부, 430074

    Effects of sinusoidal oscillating laser beam on weld formation, melt flow and grain structure during aluminum alloys lap welding

    Abstract

    A numerical model of 1.5 mm 6061/5182 aluminum alloys thin sheets lap joints under laser sinusoidal oscillation (sine) welding and laser welding (SLW) weld was developed to simulate temperature distribution and melt flow. Unlike the common energy distribution of SLW, the sinusoidal oscillation of laser beam greatly homogenized the energy distribution and reduced the energy peak. The energy peaks were located at both sides of the sine weld, resulting in the tooth-shaped sectional formation. This paper illustrated the effect of the temperature gradient (G) and solidification rate (R) on the solidification microstructure by simulation. Results indicated that the center of the sine weld had a wider area with low G/R, promoting the formation of a wider equiaxed grain zone, and the columnar grains were slenderer because of greater GR. The porosity-free and non-penetration welds were obtained by the laser sinusoidal oscillation. The reasons were that the molten pool volume was enlarged, the volume proportion of keyhole was reduced and the turbulence in the molten pool was gentled, which was observed by the high-speed imaging and simulation results of melt flow. The tensile test of both welds showed a tensile fracture form along the fusion line, and the tensile strength of sine weld was significantly better than that of the SLW weld. This was because that the wider equiaxed grain area reduced the tendency of cracks and the finer grain size close to the fracture location. Defect-free and excellent welds are of great significance to the new energy vehicles industry.

    온도 분포 및 용융 흐름을 시뮬레이션하기 위해 레이저 사인파 진동 (사인) 용접 및 레이저 용접 (SLW) 용접에서 1.5mm 6061/5182 알루미늄 합금 박판 랩 조인트 의 수치 모델이 개발되었습니다. SLW의 일반적인 에너지 분포와 달리 레이저 빔의 사인파 진동은 에너지 분포를 크게 균질화하고 에너지 피크를 줄였습니다. 에너지 피크는 사인 용접의 양쪽에 위치하여 톱니 모양의 단면이 형성되었습니다. 이 논문은 온도 구배(G)와 응고 속도 의 영향을 설명했습니다.(R) 시뮬레이션에 의한 응고 미세 구조. 결과는 사인 용접의 중심이 낮은 G/R로 더 넓은 영역을 가짐으로써 더 넓은 등축 결정립 영역의 형성을 촉진하고 더 큰 GR로 인해 주상 결정립 이 더 가늘다는 것을 나타냅니다. 다공성 및 비관통 용접은 레이저 사인파 진동에 의해 얻어졌습니다. 그 이유는 용융 풀의 부피가 확대되고 열쇠 구멍의 부피 비율이 감소하며 용융 풀의 난류가 완만해졌기 때문이며, 이는 용융 흐름의 고속 이미징 및 시뮬레이션 결과에서 관찰되었습니다. 두 용접부 의 인장시험 은 융착선을 따라 인장파괴형태를인장강도사인 용접의 경우 SLW 용접보다 훨씬 우수했습니다. 이는 등축 결정립 영역이 넓을수록 균열 경향이 감소하고 파단 위치에 근접한 입자 크기가 미세 하기 때문입니다. 결함이 없고 우수한 용접은 신에너지 자동차 산업에 매우 중요합니다.

    Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.
    Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.
    Fig. 2. Finite element mesh.
    Fig. 2. Finite element mesh.
    Fig. 3. Weld morphologies of cross-section and upper surface for the two welds: (a) sine pattern weld; (b) SLW weld.
    Fig. 3. Weld morphologies of cross-section and upper surface for the two welds: (a) sine pattern weld; (b) SLW weld.
    Fig. 4. Calculation of laser energy distribution: (a)-(c) sine pattern weld; (d)-(f) SLW weld.
    Fig. 4. Calculation of laser energy distribution: (a)-(c) sine pattern weld; (d)-(f) SLW weld.
    Fig. 5. The partially melted region of zone A.
    Fig. 5. The partially melted region of zone A.
    Fig. 6. The simulated profiles of melted region for the two welds: (a) SLW weld; (b) sine pattern weld.
    Fig. 6. The simulated profiles of melted region for the two welds: (a) SLW weld; (b) sine pattern weld.
    Fig. 7. The temperature field simulation results of cross section for sine pattern weld.
    Fig. 7. The temperature field simulation results of cross section for sine pattern weld.
    Fig. 8. Dynamic behavior of the molten pool at the same time interval of 0.004 s within one oscillating period: (a) SLW weld; (b) sine pattern weld.
    Fig. 8. Dynamic behavior of the molten pool at the same time interval of 0.004 s within one oscillating period: (a) SLW weld; (b) sine pattern weld.
    Fig. 9. The temperature field and flow field of the molten pool for the SLW weld: (a)~(f) t = 80 ms~100 ms.
    Fig. 9. The temperature field and flow field of the molten pool for the SLW weld: (a)~(f) t = 80 ms~100 ms.
    Fig. 10. The temperature field and flow field of the molten pool for the sine pattern weld: (a)~(f) t = 151 ms~171 ms.
    Fig. 10. The temperature field and flow field of the molten pool for the sine pattern weld: (a)~(f) t = 151 ms~171 ms.
    Fig. 11. The evolution of the molten pool volume and keyhole depth within one period.
    Fig. 11. The evolution of the molten pool volume and keyhole depth within one period.
    Fig. 12. The X-ray inspection results for the two welds: (a) SLW weld, (b) sine pattern weld.
    Fig. 12. The X-ray inspection results for the two welds: (a) SLW weld, (b) sine pattern weld.
    Fig. 13. Comparison of the solidification parameters for sine and SLW patterns: (a) the temperature field simulated results of the molten pool upper surfaces; (b) temperature gradient G and solidification rate R along the molten pool boundary isotherm from weld centerline to the fusion boundary; (c) G/R; (d) GR.
    Fig. 13. Comparison of the solidification parameters for sine and SLW patterns: (a) the temperature field simulated results of the molten pool upper surfaces; (b) temperature gradient G and solidification rate R along the molten pool boundary isotherm from weld centerline to the fusion boundary; (c) G/R; (d) GR.
    Fig. 14. The EBSD results of equiaxed grain zone in the weld center of: (a) sine pattern weld; (b) SLW weld; (c) grain size.
    Fig. 14. The EBSD results of equiaxed grain zone in the weld center of: (a) sine pattern weld; (b) SLW weld; (c) grain size.
    Fig. 15. (a) EBSD results of horizontal sections of SLW weld and sine pattern weld; (b) The columnar crystal widths of SLW weld and sine pattern weld.
    Fig. 15. (a) EBSD results of horizontal sections of SLW weld and sine pattern weld; (b) The columnar crystal widths of SLW weld and sine pattern weld.
    Fig. 16. (a) The tensile test results of the two welds; (b) Fracture location of SLW weld; (b) Fracture location of sine pattern weld.
    Fig. 16. (a) The tensile test results of the two welds; (b) Fracture location of SLW weld; (b) Fracture location of sine pattern weld.

    Keywords

    Laser welding, Sinusoidal oscillating, Energy distribution, Numerical simulation, Molten pool flow, Grain structure

    References

    Chen, X., 2014. Study on laser-MAG Hybrid Weaving Welding Charateristics. Master
    thesis. Harbin Institute of Technology, China.
    Chen, G., Wang, B., Mao, S., Zhong, P., He, J., 2019. Research on the “∞”-shaped laser
    scanning welding process for aluminum alloy. Opt. Laser Technol. 115, 32–41.
    Cho, W.-I., Na, S.-J., Cho, M.-H., Lee, J.-S., 2010. Numerical study of alloying element
    distribution in CO2 laser–GMA hybrid welding. Comput. Mater. Sci. 49, 792–800.
    Cho, W.-I., Na, S.-J., Thomy, C., Vollertsen, F., 2012. Numerical simulation of molten
    pool dynamics in high power disk laser welding. J. Mater. Process. Technol. 212,
    262–275.
    Das, A., Butterworth, I., Masters, I., Williams, D., 2018. Microstructure and mechanical
    properties of gap-bridged remote laser welded (RLW) automotive grade AA 5182
    joints. Mater. Charact. 145, 697–712.
    Fetzer, F., Sommer, M., Weber, R., Weberpals, J.-P., Graf, T., 2018. Reduction of pores by
    means of laser beam oscillation during remote welding of AlMgSi. Opt. Lasers Eng.
    108, 68–77.
    Geng, S., Jiang, P., Shao, X., Guo, L., Gao, X., 2020. Heat transfer and fluid flow and their
    effects on the solidification microstructure in full-penetration laser welding of
    aluminum sheet. J. Mater. Sci. Technol. 46, 50–63.
    Hagenlocher, C., Sommer, M., Fetzer, F., Weber, R., Graf, T., 2018a. Optimization of the
    solidification conditions by means of beam oscillation during laser beam welding of
    aluminum. Mater. Des. 160, 1178–1185.
    Hagenlocher, C., Weller, D., Weber, R., Graf, T., 2018b. Reduction of the hot cracking
    susceptibility of laser beam welds in AlMgSi alloys by increasing the number of grain
    boundaries. Sci. Technol. Weld. Join. 24, 313–319.
    Hagenlocher, C., Fetzer, F., Weller, D., Weber, R., Graf, T., 2019. Explicit analytical
    expressions for the influence of welding parameters on the grain structure of laser
    beam welds in aluminium alloys. Mater. Des. 174, 107791.
    Han, X., Tang, X., Wang, T., Shao, C., Lu, F., Cui, H., 2018. Role of ambient pressure in
    keyhole dynamics based on beam transmission path method for laser welding on Al
    alloy. Int. J. Adv. Manuf. Technol. 99, 1639–1651.
    Hao, K., Li, G., Gao, M., Zeng, X., 2015. Weld formation mechanism of fiber laser
    oscillating welding of austenitic stainless steel. J. Mater. Process. Technol. 225,
    77–83.
    Hirt, C.W., Nichols, B.D., 1981. Volume of fluid (VOF) method for the dynamics of free
    boundaries. J. Comput. Phys. 39, 201–225.
    Jiang, Z., Chen, X., Li, H., Lei, Z., Chen, Y., Wu, S., Wang, Y., 2020. Grain refinement and
    laser energy distribution during laser oscillating welding of Invar alloy. Mater. Des.
    186, 108195.
    Kaplan, A., 1994. A model of deep penetration laser welding based on calculation of the
    keyhole profile. J. Phys. D Appl. Phys. 27, 1805–1814.
    Kou, S., 2002. Welding Metallurgy, 2nd ed. Wiley-Interscience, New Jersey, USA.
    Kuryntsev, S.V., Gilmutdinov, A.K., 2015. The effect of laser beam wobbling mode in
    welding process for structural steels. Int. J. Adv. Manuf. Technol. 81, 1683–1691.
    Li, P., Nie, F., Dong, H., Li, S., Yang, G., Zhang, H., 2018. Pulse MIG welding of 6061-T6/
    A356-T6 aluminum alloy dissimilar T-joint. J. Mater. Eng. Perform. 27, 4760–4769.
    Liu, T., Mu, Z., Hu, R., Pang, S., 2019. Sinusoidal oscillating laser welding of 7075
    aluminum alloy: hydrodynamics, porosity formation and optimization. Int. J. Heat
    Mass Transf. 140, 346–358.
    Seto, N., Katayama, S., Matsunawa, A., 2000. High-speed simultaneous observation of
    plasma and keyhole behavior during high power CO2 laser welding: effect of
    shielding gas on porosity formation. J. Laser Appl. 12, 245–250.
    Tang, Z., Vollertsen, F., 2014. Influence of grain refinement on hot cracking in laser
    welding of aluminum. Weld. World 58, 355–366.
    Wang, L., Gao, M., Zhang, C., Zeng, X., 2016. Effect of beam oscillating pattern on weld
    characterization of laser welding of AA6061-T6 aluminum alloy. Mater. Des. 108,
    707–717.
    Wang, L., Gao, M., Zeng, X., 2018. Experiment and prediction of weld morphology for
    laser oscillating welding of AA6061 aluminium alloy. Sci. Technol. Weld. Join. 24,
    334–341.
    Yamazaki, Y., Abe, Y., Hioki, Y., Nakatani, M., Kitagawa, A., Nakata, K., 2016.
    Fundamental study of narrow-gap welding with oscillation laser beam. Weld. Int. 30,
    699–707.
    Yuan, Z., Tu, Y., Yuan, T., Zhang, Y., Huang, Y., 2021. Size effects on mechanical
    properties of pure industrial aluminum sheet for micro/meso scale plastic
    deformation: experiment and modeling. J. Alloys. Compd. 859, 157752.
    Zou, J., 2016. Characteristics of laser scanning welding process for 5A06 aluminum alloy
    thick plate with narrow gap. Materials Processing Engineering. Harbin Welding
    Institute, China. Master thesis.

    PDF VIEW
    Fig. 8. Variation of water surface profile (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.

    Numerical study of the dam-break waves and Favre waves down sloped wet rigid-bed at laboratory scale

    WenjunLiuaBoWangaYakunGuobaState Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, Chengdu 610065, ChinabFaculty of Engineering & Informatics, University of Bradford, BD7 1DP, UK

    Highlights

    경사진 습윤층에서 댐파괴유동과 FFavre 파를 수치적으로 조사하였다.
    수직 대 수평 속도의 비율이 먼저 정량화됩니다.
    유동 상태는 유상 경사가 큰 후기 단계에서 크게 변경됩니다.
    Favre 파도는 수직 속도와 수직 가속도에 큰 영향을 미칩니다.
    베드 전단응력의 변화는 베드 기울기와 꼬리물의 영향을 받습니다.

    Abstract

    The bed slope and the tailwater depth are two important ones among the factors that affect the propagation of the dam-break flood and Favre waves. Most previous studies have only focused on the macroscopic characteristics of the dam-break flows or Favre waves under the condition of horizontal bed, rather than the internal movement characteristics in sloped channel. The present study applies two numerical models, namely, large eddy simulation (LES) and shallow water equations (SWEs) models embedded in the CFD software package FLOW-3D to analyze the internal movement characteristics of the dam-break flows and Favre waves, such as water level, the velocity distribution, the fluid particles acceleration and the bed shear stress, under the different bed slopes and water depth ratios. The results under the conditions considered in this study show that there is a flow state transition in the flow evolution for the steep bed slope even in water depth ratio α = 0.1 (α is the ratio of the tailwater depth to the reservoir water depth). The flow state transition shows that the wavefront changes from a breaking state to undular. Such flow transition is not observed for the horizontal slope and mild bed slope. The existence of the Favre waves leads to a significant increase of the vertical velocity and the vertical acceleration. In this situation, the SWEs model has poor prediction. Analysis reveals that the variation of the maximum bed shear stress is affected by both the bed slope and tailwater depth. Under the same bed slope (e.g., S0 = 0.02), the maximum bed shear stress position develops downstream of the dam when α = 0.1, while it develops towards the end of the reservoir when α = 0.7. For the same water depth ratio (e.g., α = 0.7), the maximum bed shear stress position always locates within the reservoir at S0 = 0.02, while it appears in the downstream of the dam for S0 = 0 and 0.003 after the flow evolves for a while. The comparison between the numerical simulation and experimental measurements shows that the LES model can predict the internal movement characteristics with satisfactory accuracy. This study improves the understanding of the effect of both the bed slope and the tailwater depth on the internal movement characteristics of the dam-break flows and Favre waves, which also provides a valuable reference for determining the flood embankment height and designing the channel bed anti-scouring facility.

    Fig. 1. Sketch of related variables involved in shallow water model.
    Fig. 1. Sketch of related variables involved in shallow water model.
    Fig. 2. Flume model in numerical simulation.
    Fig. 2. Flume model in numerical simulation.
    Fig. 3. Grid sensitivity analysis (a) water surface profile; (b) velocity profile.
    Fig. 3. Grid sensitivity analysis (a) water surface profile; (b) velocity profile.
    Fig. 4. Sketch of experimental set-up for validating the velocity profile.
    Fig. 4. Sketch of experimental set-up for validating the velocity profile.
    Fig. 5. Sketch of experimental set-up for validating the bed shear stress.
    Fig. 5. Sketch of experimental set-up for validating the bed shear stress.
    Fig. 6. Model validation results (a) variation of the velocity profile; (b) error value of the velocity profile; (c) variation of the bed shear stress; (d) error value of the bed shear stress.
    Fig. 6. Model validation results (a) variation of the velocity profile; (b) error value of the velocity profile; (c) variation of the bed shear stress; (d) error value of the bed shear stress.
    Fig. 7. Schematic diagram of regional division.
    Fig. 7. Schematic diagram of regional division.
    Fig. 8. Variation of water surface profile (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 8. Variation of water surface profile (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 8. (continued).
    Fig. 8. (continued).
    Fig. 8. (continued).
    Fig. 8. (continued).
    Fig. 8. (continued).
    Fig. 8. (continued).
    Fig. 9. Froude number for α = 0.1 (a) variation with time; (b) variation with wavefront position.
    Fig. 9. Froude number for α = 0.1 (a) variation with time; (b) variation with wavefront position.
    Fig. 10. Characteristics of velocity distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 10. Characteristics of velocity distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 11. Average proportion of the vertical velocity (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 11. Average proportion of the vertical velocity (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 12. Bed shear stress distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 12. Bed shear stress distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 12. (continued).
    Fig. 12. (continued).
    Fig. 13. Variation of the maximum bed shear stress position with time (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 13. Variation of the maximum bed shear stress position with time (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 14. Time when the maximum bed shear stress appears at different positions (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 14. Time when the maximum bed shear stress appears at different positions (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 15. Movement characteristics of the fluid particles (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 15. Movement characteristics of the fluid particles (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
    Fig. 15. (continued).
    Fig. 15. (continued).

    Keywords

    Dam-break flow, Bed slope, Wet bed, Velocity profile, Bed shear stress, Large eddy simulation

    References

    Barnes, M.P., Baldock, T.E. 2006. Bed shear stress measurements in dam break and swash
    flows. Proceedings of International Conference on Civil and Environmental
    Engineering. Hiroshima University, Japan, 28–29 September.
    Biscarini, C., Francesco, S.D., Manciola, P., 2010. CFD modelling approach for dam break
    flow studies. Hydrol. Earth Syst. Sc. 14, 705–718. https://doi.org/10.5194/hess-14-
    705-2010.
    Fig. 15. (continued).
    W. Liu et al.
    Journal of Hydrology 602 (2021) 126752
    19
    Bristeau, M.-O., Goutal, N., Sainte-Marie, J., 2011. Numerical simulations of a nonhydrostatic shallow water model. Comput. Fluids. 47 (1), 51–64. https://doi.org/
    10.1016/j.compfluid.2011.02.013.
    Bung, D.B., Hildebrandt, A., Oertel, M., Schlenkhoff, A., Schlurmann, T. 2008. Bore
    propagation over a submerged horizontal plate by physical and numerical
    simulation. Proc. 31st Intl.Conf. Coastal Eng., Hamburg, Germany, 3542–3553.
    Cantero-Chinchilla, F.N., Castro-Orgaz, O., Dey, S., Ayuso, J.L., 2016. Nonhydrostatic
    dam break flows. I: physical equations and numerical schemes. J. Hydraul. Eng. 142
    (12), 04016068. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001205.
    Castro-Orgaz, O., Chanson, H., 2020. Undular and broken surges in dam-break flows: A
    review of wave breaking strategies in a boussinesq-type framework. Environ. Fluid
    Mech. 154 https://doi.org/10.1007/s10652-020-09749-3.
    Chang, T.-J., Kao, H.-M., Chang, K.-H., Hsu, M.-H., 2011. Numerical simulation of
    shallow-water dam break flows in open channels using smoothed particle
    hydrodynamics. J. Hydrol. 408 (1-2), 78–90. https://doi.org/10.1016/j.
    jhydrol.2011.07.023.
    Chen, H., Xu, W., Deng, J., Xue, Y., Li, J., 2009. Experimental investigation of pressure
    load exerted on a downstream dam by dam-break flow. J. Hydraul. Eng. 140,
    199–207. https://doi.org/10.1061/(ASCE)HY.1943-7900.0000743.
    Favre H. 1935. Etude th´eorique et exp´erimentale des ondes de translation dans les
    canaux d´ecouverts. Dunod, Paris. (in French).
    Flow Science Inc. 2016. Flow-3D User’s Manuals. Santa Fe NM.
    Fraccarollo, L., Toro, E.F., 1995. Experimental and numerical assessment of the shallow
    water model for two-dimensional dam-break type problems. J. Hydraul. Res. 33 (6),
    843–864. https://doi.org/10.1080/00221689509498555.
    Guo, Y., Wu, X., Pan, C., Zhang, J., 2012. Numerical simulation of the tidal flow and
    suspended sediment transport in the qiantang estuary. J Waterw. Port Coastal. 138
    (3), 192–202. https://doi.org/10.1061/(ASCE)WW.1943-5460.0000118.
    Guo, Y., Zhang, Z., Shi, B., 2014. Numerical simulation of gravity current descending a
    slope into a linearly stratified environment. J. Hydraulic Eng. 140 (12), 04014061.
    https://doi.org/10.1061/(ASCE)HY.1943-7900.0000936.
    Khosronejad, A., Kang, S., Flora, K., 2019. Fully coupled free-surface flow and sediment
    transport modelling of flash floods in a desert stream in the mojave desert, california.
    Hydrol. Process 33 (21), 2772–2791. https://doi.org/10.1002/hyp.v33.2110.1002/
    hyp.13527.
    Khosronejad, A., Arabi, M.G., Angelidis, D., Bagherizadeh, E., Flora, K., Farhadzadeh, A.,
    2020a. A comparative study of rigid-lid and level-set methods for LES of openchannel flows: morphodynamics. Environ. Fluid Mech. 20 (1), 145–164. https://doi.
    org/10.1007/s10652-019-09703-y.
    Khosronejad, A., Flora, K., Zhang, Z.X., Kang, S., 2020b. Large-eddy simulation of flash
    flood propagation and sediment transport in a dry-bed desert stream. Int. J.
    Sediment Res. 35 (6), 576–586. https://doi.org/10.1016/j.ijsrc.2020.02.002.
    Khoshkonesh, A., Nsom, B., Gohari, S., Banejad, H., 2019. A comprehensive study of dam
    break over the dry and wet beds. Ocean Eng. 188, 106279.1–106279.18. https://doi.
    org/10.1016/j.oceaneng.2019.106279.
    Kocaman, S., Ozmen-Cagatay, H., 2012. The effect of lateral channel contraction on dam
    break flows: laboratory experiment. J. Hydrol. 432–433, 145–153. https://doi.org/
    10.1016/j.jhydrol.2012.02.035.
    Kocaman, S., Ozmen-Cagatay, H., 2015. Investigation of dam-break induced shock waves
    impact on a vertical wall. J. Hydrol. 525, 1–12. https://doi.org/10.1016/j.
    jhydrol.2015.03.040.
    LaRocque, L.A., Imran, J., Chaudhry, M.H., 2013a. Experimental and numerical
    investigations of two-dimensional dam-break flows. J. Hydraul. Eng. 139 (6),
    569–579. https://doi.org/10.1061/(ASCE)HY.1943-7900.0000705.
    Larocque, L.A., Imran, J., Chaudhry, M.H., 2013b. 3D numerical simulation of partial
    breach dam-break flow using the LES and k-ε turbulence models. J. Hydraul. Res. 51,
    145–157. https://doi.org/10.1080/00221686.2012.734862.
    Lauber, G., Hager, W.H., 1998a. Experiments to dam break wave: Horizontal channel.
    J. Hydraul. Res. 36 (3), 291–307. https://doi.org/10.1080/00221689809498620.
    Lauber, G., Hager, W.H., 1998b. Experiments to dam break wave: Sloping channel.
    J. Hydraul. Res. 36 (5), 761–773. https://doi.org/10.1080/00221689809498601.
    Leal, J.G., Ferreira, R.M., Cardoso, A.H., 2006. Dam-break wave-front celerity.
    J. Hydraul. Eng. 132 (1), 69–76. https://doi.org/10.1061/(ASCE)0733-9429(2006)
    132:1(69).
    Liu, W., Wang, B., Guo, Y., Zhang, J., Chen, Y., 2020. Experimental investigation on the
    effects of bed slope and tailwater on dam-break flows. J. Hydrol. 590, 125256.
    https://doi.org/10.1016/j.jhydrol.2020.125256.
    Marche, C., Beauchemin P. EL Kayloubi, A. 1995. Etude num´erique et exp´erimentale des
    ondes secondaires de Favre cons´ecutives a la rupture d’un harrage. Can. J. Civil Eng.
    22, 793–801, (in French). https://doi.org/10.1139/l95-089.
    Marra, D., Earl, T., Ancey, C. 2011. Experimental investigations of dam break flows down
    an inclined channel. Proceedings of the 34th World Congress of the International
    Association for Hydro-Environment Research and Engineering: 33rd Hydrology and
    Water Resources Symposium and 10th Conference on Hydraulics in Water
    Engineering, Brisbane, Australia.
    Marsooli, R., Wu, W., 2014. 3-D finite-volume model of dam-break flow over uneven
    beds based on vof method. Adv. Water Resour. 70, 104–117. https://doi.org/
    10.1016/j.advwatres.2014.04.020.
    Miller, S., Chaudhry, M.H., 1989. Dam-break flows in curved channel. J. Hydraul. Eng.
    115 (11), 1465–1478. https://doi.org/10.1061/(ASCE)0733-9429(1989)115:11
    (1465).
    Mohapatra, P.K., Chaudhry, M.H., 2004. Numerical solution of Boussinesq equations to
    simulate dam-break flows. J. Hydraul. Eng. 130 (2), 156–159. https://doi.org/
    10.1061/(ASCE)0733-9429(2004)130:2(156).
    Oertel, M., Bung, D.B., 2012. Initial stage of two-dimensional dam-break waves:
    laboratory versus VOF. J. Hydraul. Res. 50 (1), 89–97. https://doi.org/10.1080/
    00221686.2011.639981.
    Ozmen-Cagatay, H., Kocaman, S., 2012. Investigation of dam-break flow over abruptly
    contracting channel with trapezoidal-shaped lateral obstacles. J. Fluids Eng. 134,
    081204 https://doi.org/10.1115/1.4007154.
    Ozmen-Cagatay, H., Kocaman, S., Guzel, H., 2014. Investigation of dam-break flood
    waves in a dry channel with a hump. J. Hydro-environ. Res. 8 (3), 304–315. https://
    doi.org/10.1016/j.jher.2014.01.005.
    Park, I.R., Kim, K.S., Kim, J., Van, S.H., 2012. Numerical investigation of the effects of
    turbulence intensity on dam-break flows. Ocean Eng. 42, 176–187. https://doi.org/
    10.1016/j.oceaneng.2012.01.005.
    Peregrine, D.H., 1966. Calculations of the development of an undular bore. J. Fluid
    Mech. 25 (2), 321–330. https://doi.org/10.1017/S0022112066001678.
    Savic, L.j., Holly, F.M., 1993. Dam break flood waves computed by modified Godunov
    method. J. Hydraul. Res. 31 (2), 187–204. https://doi.org/10.1080/
    00221689309498844.
    Shigematsu, T., Liu, P., Oda, K., 2004. Numerical modeling of the initial stages of dambreak waves. J. Hydraul. Res. 42 (2), 183–195. https://doi.org/10.1080/
    00221686.2004.9628303.
    Smagorinsky, J., 1963. General circulation experiments with the primitive equations.
    Part I: the basic experiment. Mon. Weather Rev. 91, 99–164. https://doi.org/
    10.1126/science.27.693.594.
    Soares-Frazao, S., Zech, Y., 2002. Undular bores and secondary waves – Experiments and
    hybrid finite-volume modeling. J. Hydraul. Res. 40, 33–43. https://doi.org/
    10.1080/00221680209499871.
    Stansby, P.K., Chegini, A., Barnes, T.C.D., 1998. The initial stages of dam-break flow.
    J. Fluid Mech. 370, 203–220. https://doi.org/10.1017/022112098001918.
    Treske, A., 1994. Undular bores (favre-waves) in open channels – experimental studies.
    J. Hydraul. Res. 32 (3), 355–370. https://doi.org/10.1080/00221689409498738.
    Wang, B., Chen, Y., Wu, C., Dong, J., Ma, X., Song, J., 2016. A semi-analytical approach
    for predicting peak discharge of floods caused by embankment dam failures. Hydrol.
    Process 30 (20), 3682–3691. https://doi.org/10.1002/hyp.v30.2010.1002/
    hyp.10896.
    Wang, B., Chen, Y., Wu, C., Peng, Y., Ma, X., Song, J., 2017. Analytical solution of dambreak flood wave propagation in a dry sloped channel with an irregular-shaped
    cross-section. J. Hydro-environ. Res. 14, 93–104. https://doi.org/10.1016/j.
    jher.2016.11.003.
    Wang, B., Chen, Y., Wu, C., Peng, Y., Song, J., Liu, W., Liu, X., 2018. Empirical and semianalytical models for predicting peak outflows caused by embankment dam failures.
    J. Hydrol. 562, 692–702. https://doi.org/10.1016/j.jhydrol.2018.05.049.
    Wang, B., Zhang, J., Chen, Y., Peng, Y., Liu, X., Liu, W., 2019. Comparison of measured
    dam-break flood waves in triangular and rectangular channels. J. Hydrol. 575,
    690–703. https://doi.org/10.1016/j.jhydrol.2019.05.081.
    Wang, B., Liu, W., Zhang, J., Chen, Y., Wu, C., Peng, Y., Wu, Z., Liu, X., Yang, S., 2020a.
    Enhancement of semi-theoretical models for predicting peak discharges in breached
    embankment dams. Environ. Fluid Mech. 20 (4), 885–904. https://doi.org/10.1007/
    s10652-019-09730-9.
    Wang, B., Chen, Y., Peng, Y., Zhang, J., Guo, Y., 2020b. Analytical solution of shallow
    water equations for ideal dam-break flood along a wet bed slope. J. Hydraul. Eng.
    146 (2), 06019020. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001683.
    Wang, B., Liu, W., Wang, W., Zhang, J., Chen, Y., Peng, Y., Liu, X., Yang, S., 2020c.
    Experimental and numerical investigations of similarity for dam-break flows on wet
    bed. J. Hydrol. 583, 124598. https://doi.org/10.1016/j.jhydrol.2020.124598.
    Wang, B., Liu, X., Zhang, J., Guo, Y., Chen, Y., Peng, Y., Liu, W., Yang, S., Zhang, F.,
    2020d. Analytical and experimental investigations of dam-break flows in triangular
    channels with wet-bed conditions. J. Hydraul. Eng. 146 (10), 04020070. https://doi.
    org/10.1061/(ASCE)HY.1943-7900.0001808.
    Wu, W., Wang, S., 2007. One-dimensional modeling of dam-break flow over movable
    beds. J. Hydraul. Eng. 133 (1), 48–58. https://doi.org/10.1061/(ASCE)0733-9429
    (2007)133:1(48).
    Xia, J., Lin, B., Falconer, R.A., Wang, G., 2010. Modelling dam-break flows over mobile
    beds using a 2d coupled approach. Adv. Water Resour. 33 (2), 171–183. https://doi.
    org/10.1016/j.advwatres.2009.11.004.
    Yang, S., Yang, W., Qin, S., Li, Q., Yang, B., 2018a. Numerical study on characteristics of
    dam-break wave. Ocean Eng. 159, 358–371. https://doi.org/10.1016/j.
    oceaneng.2018.04.011.
    Yang, S., Yang, W., Qin, S., Li, Q., 2018b. Comparative study on calculation methods of
    dam-break wave. J. Hydraul. Res. 57 (5), 702–714. https://doi.org/10.1080/
    00221686.2018.1494057.

    Figure 9. Turbulent kinetic energy (TKE) contour map on different sections.

    Numerical Simulation Research on the Diversion
    Characteristics of a Trapezoidal Channel

    Yong Cheng, Yude Song, Chunye Liu, Wene Wang * and Xiaotao Hu
    Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas, Ministry of Education, Northwest A&F University, Yangling 712100, China

    • Correspondence: wangwene@nwsuaf.edu.cn

    Abstract

    개방 채널 분기점은 관개 지역에서 가장 일반적인 물 전환 구조입니다. 관개용수 운반에서는 물 운반 효율과 침전이 주요 관심사입니다. 따라서 이 연구는 관개 지역의 물 공급에 대한 개방 채널 분기점의 영향을 분석합니다.

    여기에서 FLOW-3D 소프트웨어를 사용하고 15 세트의 작업 조건을 포함하는 수치 시뮬레이션을 통해 개방 채널 분기점에서의 3차원 유동을 연구했습니다. 개수로 분기점 부근의 재순환 구역 및 유동 구조의 수리학적 특성을 분석하였다.

    그런 다음 사다리꼴 채널에서 표면 및 바닥층의 흐름 전환 폭에 대한 방정식을 얻었습니다. 수심에 따른 흐름 전환 폭은 사다리꼴 채널과 직사각형 채널에서 다른 것으로 나타났습니다. 결과는 또한 개방 수로 분기점이 주 수로의 유속에 상당한 영향을 미친다는 것을 보여줍니다.

    개방 채널 분기점의 재순환 영역에서의 유속은 작았지만 맥동 속도와 난류 운동 에너지는 컸다. 이 지역에서 소산되는 에너지는 상대적으로 커서 수로 물 전달에 도움이 되지 않았습니다.

    이 연구는 관개구역의 수로 최적화 및 운영 관리에 대한 참고 자료를 제공합니다.

    Open-channel bifurcations are the most common water diversion structures in irrigation districts. In irrigation water conveyance, water transport efficiency and sedimentation are primary concerns. This study accordingly analyzes the influence of open-channel bifurcations on water delivery in irrigation areas. Herein, the three-dimensional flow at an open-channel bifurcation was studied via numerical simulations using FLOW-3D software and including 15 sets of working conditions. The hydraulic characteristics of the recirculation zone and flow structures in the vicinity of the open-channel bifurcation were analyzed. Equations for the flow diversion width of the surface and bottom layers in the trapezoidal channel were then obtained. The flow diversion widths along the water depth were found to differ between trapezoidal and rectangular channels. The results also show that open-channel bifurcations considerably influence the flow velocity in the main channel. The flow velocity in the recirculation zone of open-channel bifurcations was small, but the pulsation velocity and the turbulent kinetic energy were large. The energy dissipated in this area was relatively large, which was not conducive to channel water delivery. This study provides a reference for channel optimization and operation management in irrigation districts.

    Keywords

    trapezoidal open channel; numerical simulation; the recirculation zone; flow diversion
    width; turbulence kinetic energy

    Figure 1. Experimental plan and section measurement layout. Note: Red points in the figure represent the measurement point arrangement, and Roman numerals represent measurement section numbers.
    Figure 1. Experimental plan and section measurement layout. Note: Red points in the figure represent the measurement point arrangement, and Roman numerals represent measurement section numbers.
    Figure 5. Froude number (Fr) contour map at different water depths. Note: Q1 = 40 L/s; b = 30 cm. X* and Y* are obtained by dimensionless processing of X-axis and Y-axis coordinates. (a) depth of water below the sill height; (b) depth of water above the sill height.
    Figure 5. Froude number (Fr) contour map at different water depths. Note: Q1 = 40 L/s; b = 30 cm. X* and Y* are obtained by dimensionless processing of X-axis and Y-axis coordinates. (a) depth of water below the sill height; (b) depth of water above the sill height.
    PDF 원문보기
    Figure 1 Mitochondrial Weir Dam

    The Three-dimensional Simulation of Granular
    Mixtures Weir

    Shen Zhen-dong*1, 2, Zhang Yang1, 2
    1Zhejiang Guangchuan Engineering Consultation Co., Ltd., Hangzhou, 310020,
    Zhejiang, China
    2Zhejiang Institute of Hydraulics &Estuary, Hangzhou 310020, Zhejiang, China
    E-mail: zdshen1991@126.com

    Abstract

    최근 몇 년 동안 생태학적 수자원 보존 공학의 발전으로 많은 새로운 댐 디자인이 등장했습니다. 본 논문에서는 체계적인 소면보 연구와 조사를 바탕으로 새로운 종류의 입상 혼합물 위어를 제시하였습니다.

    입상보의 수치해석은 Flow-3D를 이용하여 수행하였으며, 그 결과를 물리적 모델 실험결과와 비교하였습니다. 유속, 유속 분포 및 둑의 파손에 대한 수치 시뮬레이션 결과는 실험 결과와 잘 일치하며, 이는 3차원 수학적 모델이 물리적 모델 실험과 결합되어 모든 입상 혼합물 둑을 시뮬레이션할 수 있음을 나타냅니다.

    이 방법을 이용하여 특성 및 수리학적 매개변수를 분석하면 생태보의 후속 연구를 위한 기술적 지원을 제공할 수 있습니다.

    In recent years, with the development of ecological water conservancy engineering,
    many new weir designs have also emerged. This paper has put forward a new kind of granular
    mixtures weir based on the systematic carding weir researches, combined with investigation. The
    numerical simulation of granular weir is carried out by using Flow-3D,and the results are
    compared with the physical model experiment results. The numerical simulation results of the
    flow velocity, flow distribution and the failure of the weir are in good agreement with the
    experimental results, which indicates that the 3-D mathematical model can be combined with
    physical model experiments to simulate the granular mixtures weir in all directions. Using this
    method to analysis the characteristics and hydraulic parameters can provide technical support
    for the follow-up research of ecological weir.

    Figure 1 Mitochondrial Weir Dam
    Figure 1 Mitochondrial Weir Dam
    Table 1 Numerical simulation programme table
    Table 1 Numerical simulation programme table
    Figure 4 Final Damage of Weir in Different Projects
    Figure 4 Final Damage of Weir in Different Projects

    References

    [1] Ma Y.Y, Yan Y, Wang S.Y, Jin D, Gong Y.x, Lu Q, Wang Y.T, Yue F.J. (2012) Study on
    Distribution Characteristics and Historical Value of Ancient Weirs in Zhejiang Province .
    Zhejiang Hydrotechnics, 04:47-50.
    [2] Jin H.J. (2016) Design of Weir Dam in Flood Control Engineering. A Brief Discussion Science
    and Technology Economic Guide 9.
    [3] Chang Q. (2017) Experimental Study on Flow Characteristics of Tooth Weir and Z Weir.
    Shandong Agricultural University.
    [4] Wu G.J, Liu X.P, Fang S.S, Sun W.H, Hou B. (2011) Hydraulic Characteristics of Low Practical
    Weir and Its Influence on Engineering; Journal of Yangtze River Scientific Research Institute,
    28(09):21-24.
    [5] Jiang D, Li G.D, Li S.S. (2019) Experimental study on discharge characteristics of different
    upstream-downstream overhang ratios of piano key weir; Water Resources and Hydropower
    Engineering, 50(07):124-130.
    [6] Liu X.P, Hu S.L, Ren Q.M, Zhao J. (2015) Study on impact from sedimentation of low-head
    broken line practical weir. Water Resources and Hydropower Engineering, (03):136-140.
    [7] GUAN D,MELVILLE B,FRIEDRICH H. (2014) Flow patterns and turbulence structures in a
    scour hole downstream of a submerged weir. Journal of Hydraulic Engineering, 140(1)：68-
    76.
    [8] Lu WANG. GUAN D.W, Yan Y.X, Zheng J.H, Bruce MELVILLE, Lu W. (2017) Research
    Progress on scour at weir-like structures. Advances Water Science , 28(02):311-318.
    [9] Zhang C, Sun S.K. (2017) Study and improvement on hydraulic characteristics of turning-section
    pools with various angles for vertical slot fish way. Water Resources and Hydropower
    Engineering, 48(11):20-25.
    [10] Bian Y.H. (2015)Study on Several Hydraulic Problems of Vertical Slot Fishways. China Institute
    of Water Resources and Hydropower Research.
    [11] Zhang D.R. The Influence of Water-related Engineering on Flood-control in Mountainous
    Watershed on Mike21FM. China Institute of Water Resources and Hydropower Research.
    [12] Chen D.H, Chen Z. (2005) Three dimensional simulation of flow over weirs. Engineering Journal
    of Wuhan University, (05):56-58+64.
    [13] MOHAMMADPOUR R,GHANI A A, AZAMATHULLA H M. (2013) Numerical modeling of
    3-d flow on porous broad crested weirs. Applied Mathematical Modelling, 37(22)：9324-9337.

    PDF 원문보기
    Flow Field in a Sloped Channel with Damaged and Undamaged Piers: Numerical and Experimental Studies

    Flow Field in a Sloped Channel with Damaged and Undamaged Piers: Numerical and Experimental Studies

    Ehsan Oveici, Omid Tayari & Navid Jalalkamali
    KSCE Journal of Civil Engineering volume 25, pages4240–4251 (2021)Cite this article

    Abstract

    본 논문은 경사가 완만한 수로에서 손상되거나 손상되지 않은 교각 주변의 유동 패턴을 분석했습니다. 실험은 길이가 12m이고 기울기가 0.008인 직선 수로에서 수행되었습니다. Acoustic Doppler Velocimeter(ADV)를 이용하여 3차원 유속 데이터를 수집하였고, 그 결과를 PIV(Particle Image Velocimetry) 데이터와 분석하여 비교하였습니다.

    다중 블록 옵션이 있는 취수구의 퇴적물 시뮬레이션(SSIIM)은 이 연구에서 흐름의 수치 시뮬레이션을 위해 통합되었습니다. 일반적으로 비교에서 얻은 결과는 수치 데이터와 실험 데이터 간의 적절한 일치를 나타냅니다. 결과는 모든 경우에 수로 입구에서 2m 거리에서 기복적 수압 점프가 발생했음을 보여주었습니다.

    경사진 수로의 최대 베드 전단응력은 2개의 손상 및 손상되지 않은 교각을 설치하기 위한 수평 수로의 12배였습니다. 이와 같은 경사수로 교각의 위치에 따라 상류측 수위는 수평수로의 유사한 조건에 비해 72.5% 감소한 반면, 이 감소량은 경사면에서 다른 경우에 비해 8.3% 감소하였다. 채널 또한 두 교각이 있는 경우 최대 Froude 수는 수평 수로의 5.7배였습니다.

    This paper analyzed the flow pattern around damaged and undamaged bridge piers in a channel with a mild slope. The experiments were carried out on a straight channel with a length of 12 meters and a slope of 0.008. Acoustic Doppler velocimeter (ADV) was employed to collect three-dimensional flow velocity data, and the results were analyzed and compared with particle image velocimetry (PIV) data. Sediment Simulation in Intakes with Multiblock option (SSIIM) was incorporated for the numerical simulation of the flow in this study. Generally, the results obtained from the comparisons referred to the appropriate agreement between the numerical and the experimental data. The results showed that an undular hydraulic jump occurred at a distance of two meters from the channel entrance in every case; the maximum bed shear stress in the sloped channel was 12 times that in a horizontal channel for installing two damaged and undamaged piers. With this position of the piers in the sloped channel, the upstream water level underwent a 72.5% reduction compared to similar conditions in a horizontal channel, while the amount of this water level decrease was equal to 8.3% compared to the other cases in a sloped channel. In addition, with the presence of both piers, the maximum Froude number was 5.7 times that in a horizontal channel.

    This is a preview of subscription content, access via your institution.

    References

    Download references

    하류하천의 영향 최소화를 위한 보조 여수로 최적 활용방안 검토

    The Optimal Operation on Auxiliary Spillway to Minimize the Flood Damage in Downstream River with Various Outflow Conditions

    하류하천의 영향 최소화를 위한 보조 여수로 최적 활용방안 검토

    Hyung Ju Yoo1, Sung Sik Joo2, Beom Jae Kwon3, Seung Oh Lee4*

    유 형주1, 주 성식2, 권 범재3, 이 승오4*

    1Ph.D Student, Dept. of Civil & Environmental Engineering, Hongik University
    2Director, Water Resources & Environment Department, HECOREA
    3Director, Water Resources Department, ISAN
    4Professor, Dept. of Civil & Environmental Engineering, Hongik University

    1홍익대학교 건설환경공학과 박사과정
    2㈜헥코리아 수자원환경사업부 이사
    3㈜이산 수자원부 이사
    4홍익대학교 건설환경공학과 교수

    ABSTRACT

    최근 기후변화로 인해 강우강도 및 빈도의 증가에 따른 집중호우의 영향 및 기존 여수로의 노후화에 대비하여 홍수 시 하류 하천의 영향을 최소화할 수 있는 보조 여수로 활용방안 구축이 필요한 실정이다. 이를 위해, 수리모형 실험 및 수치모형 실험을 통하여 보조 여수로 운영에 따른 흐름특성 변화 검토에 관한 연구가 많이 진행되어 왔다. 그러나 대부분의 연구는 여수로에서의 흐름특성 및 기능성에 대한 검토를 수행하였을 뿐 보조 여수로의 활용방안에 따른 하류하천 영향 검토 및 호안 안정성 검토에 관한 연구는 미비한 실정이다. 이에 본 연구에서는 기존 여수로 및 보조 여수로 방류 조건에 따른 하류영향 분석 및 호안 안정성 측면에서 최적 방류 시나리오 검토를 3차원 수치모형인 FLOW-3D를 사용하여 검토하였다. 또한 FLOW-3D 수치모의 수행을 통한 유속, 수위 결과와 소류력 산정 결과를 호안 설계허용 기준과 비교하였다. 수문 완전 개도 조건으로 가정하고 계획홍수량 유입 시 다양한 보조 여수로 활용방안에 대하여 수치모의를 수행한 결과, 보조 여수로 단독 운영 시 기존 여수로 단독운영에 비하여 최대유속 및 최대 수위의 감소효과를 확인하였다. 다만 계획홍수량의 45% 이하 방류 조건에서 대안부의 호안 안정성을 확보하였고 해당 방류량 초과 경우에는 처오름 현상이 발생하여 월류에 대한 위험성 증가를 확인하였다. 따라서 기존 여수로와의 동시 운영 방안 도출이 중요하다고 판단하였다. 여수로의 배분 비율 및 총 허용 방류량에 대하여 검토한 결과 보조 여수로의 방류량이 기존 여수로의 방류량보다 큰 경우 하류하천의 흐름이 중심으로 집중되어 대안부의 유속 저감 및 수위 감소를 확인하였고, 계획 홍수량의 77% 이하의 조건에서 호안의 허용 유속 및 허용 소류력 조건을 만족하였다. 이를 통하여 본 연구에서 제안한 보조 여수로 활용방안으로는 기존 여수로와 동시 운영 시 총 방류량에 대하여 보조 여수로의 배분량이 기존 여수로의 배분량보다 크게 설정하는 것이 하류하천의 영향을 최소화 할 수 있는 것으로 나타났다. 그러나 본 연구는 여수로 방류에 따른 대안부에서의 영향에 대해서만 검토하였고 수문 전면 개도 조건에서 검토하였다는 한계점은 분명히 있다. 이에 향후에는 다양한 수문 개도 조건 및 방류 시나리오를 적용 및 검토한다면 보다 효율적이고, 효과적인 보조 여수로 활용방안을 도출이 가능할 것으로 기대 된다.

    키워드 : 보조 여수로, FLOW-3D, 수치모의, 호안 안정성, 소류력

    1. 서 론

    최근 기후변화로 인한 집중호우의 영향으로 홍수 시 댐으로 유입되는 홍수량이 설계 홍수량보다 증가하여 댐 안정성 확보가 필요한 실정이다(Office for Government Policy Coordination, 2003). MOLIT & K-water(2004)에서는 기존댐의 수문학적 안정성 검토를 수행하였으며 이상홍수 발생 시 24개 댐에서 월류 등으로 인한 붕괴위험으로 댐 하류지역의 극심한 피해를 예상하여 보조여수로 신설 및 기존여수로 확장 등 치수능력 증대 기본계획을 수립하였고 이를 통하여 극한홍수 발생 시 홍수량 배제능력을 증대하여 기존댐의 안전성 확보 및 하류지역의 피해를 방지하고자 하였다. 여기서 보조 여수로는 기존 여수로와 동시 또는 별도 운영하는 여수로로써 비상상황 시 방류 기능을 포함하고 있고(K-water, 2021), 최근에는 기존 여수로의 노후화에 따라 보조여수로의 활용방안에 대한 관심이 증가하고 있다. 따라서 본 연구에서는 3차원 수치해석을 수행하여 기존 및 보조 여수로의 방류량 조합에 따른 하류 영향을 분석하고 하류 호안 안정성 측면에서 최적 방류 시나리오를 검토하고자 한다.

    기존의 댐 여수로 검토에 관한 연구는 주로 수리실험을 통하여 방류조건 별 흐름특성을 검토하였으나 최근에는 수치모형 실험결과가 수리모형실험과 비교하여 근사한 것을 확인하는 등 점차 수치모형실험을 수리모형실험의 대안으로 활용하고 있다(Jeon et al., 2006; Kim, 2007; Kim et al., 2008). 국내의 경우, Jeon et al.(2006)은 수리모형 실험과 수치모의를 이용하여 임하댐 바상여수로의 기본설계안을 도출하였고, Kim et al.(2008)은 가능최대홍수량 유입 시 비상여수로 방류에 따른 수리학적 안정성과 기능성을 3차원 수치모형인 FLOW-3D를 활용하여 검토하였다. 또한 Kim and Kim(2013)은 충주댐의 홍수조절 효과 검토 및 방류량 변화에 따른 상·하류의 수위 변화를 수치모형을 통하여 검토하였다. 국외의 경우 Zeng et al.(2017)은 3차원 수치모형인 Fluent를 활용한 여수로 방류에 따른 흐름특성 결과와 측정결과를 비교하여 수치모형 결과의 신뢰성을 검토하였다. Li et al.(2011)은 가능 최대 홍수량(Probable Maximum Flood, PMF)조건에서 기존 여수로와 신규 보조 여수로 유입부 주변의 흐름특성에 대하여 3차원 수치모형 Fluent를 활용하여 검토하였고, Lee et al.(2019)는 서로 근접해있는 기존 여수로와 보조여수로 동시 운영 시 방류능 검토를 수리모형 실험 및 수치모형 실험(FLOW-3D)을 통하여 수행하였으며 기존 여수로와 보조 여수로를 동시운영하게 되면 배수로 간섭으로 인하여 총 방류량이 7.6%까지 감소되어 댐의 방류능력이 감소하였음을 확인하였다.

    그러나 대부분의 여수로 검토에 대한 연구는 여수로 내에서의 흐름특성 및 기능성에 대한 검토를 수행하였고. 이에 기존 여수로와 보조 여수로 방류운영에 따른 하류하천의 흐름특성 변화 및 호안 안정성 평가에 관한 추가적인 검토가 필요한 실정이다. 따라서 본 연구에서는 기존 여수로 및 보조 여수로 방류 조건에 따른 하류하천의 흐름특성 및 호안 안정성분석을 3차원 수치모형인 FLOW-3D를 이용하여 검토하였다. 또한 다양한 방류 배분 비율 및 허용 방류량 조건 변화에 따른 하류하천의 흐름특성 및 소류력 분석결과를 호안 설계 허용유속 및 허용 소류력 기준과 비교하여 하류하천의 영향을 최소화 할 수 있는 최적의 보조 여수로 활용방안을 도출하고자 한다.

    2. 본 론

    2.1 이론적 배경

    2.1.1 3차원 수치모형의 기본이론

    FLOW-3D는 미국 Flow Science, Inc에서 개발한 범용 유체역학 프로그램(CFD, Computational Fluid Dynamics)으로 자유 수면을 갖는 흐름모의에 사용되는 3차원 수치해석 모형이다. 난류모형을 통해 난류 해석이 가능하고, 댐 방류에 따른 하류 하천의 흐름 해석에도 많이 사용되어 왔다(Flow Science, 2011). 본 연구에서는 FLOW-3D(version 12.0)을 이용하여 홍수 시 기존 여수로의 노후화에 대비하여 보조 여수로의 활용방안에 대한 검토를 하류하천의 호안 안정성 측면에서 검토하였다.

    2.1.2 유동해석의 지배방정식

    1) 연속 방정식(Continuity Equation)

    FLOW-3D는 비압축성 유체에 대하여 연속방정식을 사용하며, 밀도는 상수항으로 적용된다. 연속 방정식은 Eqs. (1), (2)와 같다.

    (1)

    ∇·v=0

    (2)

    ∂∂x(uAx)+∂∂y(vAy)+∂∂z(wAz)=RSORρ

    여기서, ρ는 유체 밀도(kg/m3), u, v, w는 x, y, z방향의 유속(m/s), Ax, Ay, Az는 각 방향의 요소면적(m2), RSOR는 질량 생성/소멸(mass source/sink)항을 의미한다.

    2) 운동량 방정식(Momentum Equation)

    각 방향 속도성분 u, v, w에 대한 운동방정식은 Navier-Stokes 방정식으로 다음 Eqs. (3), (4), (5)와 같다.

    (3)

    ∂u∂t+1VF(uAx∂u∂x+vAy∂v∂y+wAz∂w∂z)=-1ρ∂p∂x+Gx+fx-bx-RSORρVFu

    (4)

    ∂v∂t+1VF(uAx∂u∂x+vAy∂v∂y+wAz∂w∂z)=-1ρ∂p∂y+Gy+fy-by-RSORρVFv

    (5)

    ∂w∂t+1VF(uAx∂u∂x+vAy∂v∂y+wAz∂w∂z)=-1ρ∂p∂z+Gz+fz-bz-RSORρVFw

    여기서, Gx, Gy, Gz는 체적력에 의한 가속항, fx, fy, fz는 점성에 의한 가속항, bx, by, bz는 다공성 매체에서의 흐름손실을 의미한다.

    2.1.3 소류력 산정

    호안설계 시 제방사면 호안의 안정성 확보를 위해서는 하천의 흐름에 의하여 호안에 작용하는 소류력에 저항할 수 있는 재료 및 공법 선택이 필요하다. 국내의 경우 하천공사설계실무요령(MOLIT, 2016)에서 계획홍수량 유하 시 소류력 산정 방법을 제시하고 있다. 소류력은 하천의 평균유속을 이용하여 산정할 수 있으며, 소류력 산정식은 Eqs. (6), (7)과 같다.

    1) Schoklitsch 공식

    Schoklitsch(1934)는 Chezy 유속계수를 적용하여 소류력을 산정하였다.

    (6)

    τ=γRI=γC2V2

    여기서, τ는 소류력(N/m2), R은 동수반경(m), γ는 물의 단위중량(10.0 kN/m3), I는 에너지경사, C는 Chezy 유속계수, V는 평균유속(m/s)을 의미한다.

    2) Manning 조도계수를 고려한 공식

    Chezy 유속계수를 대신하여 Manning의 조도계수를 고려하여 소류력을 산정할 수 있다.

    (7)

    τ=γn2V2R1/3

    여기서, τ는 소류력(N/m2), R은 동수반경(m), γ는 물의 단위중량(10.0 kN/m3), n은 Manning의 조도계수, V는 평균유속(m/s)을 의미한다.

    FLOW-3D 수치모의 수행을 통하여 하천의 바닥 유속을 도출할 수 있으며, 본 연구에서는 Maning 조도계수롤 고려하여 소류력을 산정하고자 한다. 소류력을 산정하기 위해서 여수로 방류에 따른 대안부의 바닥유속 변화를 검토하여 최대 유속 값을 이용하였다. 최종적으로 산정한 소류력과 호안의 재료 및 공법에 따른 허용 소류력과 비교하여 제방사면 호안의 안정성 검토를 수행하게 된다.

    2.2 하천호안 설계기준

    하천 호안은 계획홍수위 이하의 유수작용에 대하여 안정성이 확보되도록 계획하여야 하며, 호안의 설계 시에는 사용재료의 확보용이성, 시공상의 용이성, 세굴에 대한 굴요성(flexibility) 등을 고려하여 호안의 형태, 시공방법 등을 결정한다(MOLIT, 2019). 국내의 경우, 하천공사설계실무요령(MOLIT, 2016)에서는 다양한 호안공법에 대하여 비탈경사에 따라 설계 유속을 비교하거나, 허용 소류력을 비교함으로써 호안의 안정성을 평가한다. 호안에 대한 국외의 설계기준으로 미국의 경우, ASTM(미국재료시험학회)에서 호안블록 및 식생매트 시험방법을 제시하였고 제품별로 ASTM 시험에 의한 허용유속 및 허용 소류력을 제시하였다. 일본의 경우, 호안 블록에 대한 축소실험을 통하여 항력을 측정하고 이를 통해서 호안 블록에 대한 항력계수를 제시하고 있다. 설계 시에는 항력계수에 의한 블록의 안정성을 평가하고 있으나, 최근에는 세굴의 영향을 고려할 수 있는 호안 안정성 평가의 필요성을 제기하고 있다(MOLIT, 2019). 관련된 국내·외의 하천호안 설계기준은 Table 1에 정리하여 제시하였고, 본 연구에서 하천 호안 안정성 평가 시 하천공사설계실무요령(MOLIT, 2016)과 ASTM 시험에서 제시한 허용소류력 및 허용유속 기준을 비교하여 각각 0.28 kN/m2, 5.0 m/s 미만일 경우 호안 안정성을 확보하였다고 판단하였다.

    Table 1.

    Standard of Permissible Velocity and Shear on Revetment

    Country (Reference)MaterialPermissible velocity (Vp, m/s)Permissible Shear (τp, kN/m2)
    KoreaRiver Construction Design Practice Guidelines
    (MOLIT, 2016)    		Vegetated5.00.50
    Stone5.00.80
    USAASTM D’6460Vegetated6.10.81
    Unvegetated5.00.28
    JAPANDynamic Design Method of Revetment–5.0–

    2.3. 보조여수로 운영에 따른 하류하천 영향 분석

    2.3.1 모형의 구축 및 경계조건

    본 연구에서는 기존 여수로의 노후화에 대비하여 홍수 시 보조여수로의 활용방안에 따른 하류하천의 흐름특성 및 호안안정성 평가를 수행하기 위해 FLOW-3D 모형을 이용하였다. 기존 여수로 및 보조 여수로는 치수능력 증대사업(MOLIT & K-water, 2004)을 통하여 완공된 ○○댐의 제원을 이용하여 구축하였다. ○○댐은 설계빈도(100년) 및 200년빈도 까지는 계획홍수위 이내로 기존 여수로를 통하여 운영이 가능하나 그 이상 홍수조절은 보조여수로를 통하여 조절해야 하며, 또한 2011년 기존 여수로 정밀안전진단 결과 사면의 표층 유실 및 옹벽 밀림현상 등이 확인되어 노후화에 따른 보수·보강이 필요한 상태이다. 이에 보조여수로의 활용방안 검토가 필요한 것으로 판단하여 본 연구의 대상댐으로 선정하였다. 하류 하천의 흐름특성을 예측하기 위하여 격자간격을 0.99 ~ 8.16 m의 크기로 하여 총 격자수는 49,102,500개로 구성하였으며, 여수로 방류에 따른 하류하천의 흐름해석을 위한 경계조건으로 상류는 유입유량(inflow), 바닥은 벽면(wall), 하류는 수위(water surface elevation)조건으로 적용하도록 하였다(Table 2, Fig. 1 참조). FLOW-3D 난류모형에는 혼합길이 모형, 난류에너지 모형, k-ϵ모형, RNG(Renormalized Group Theory) k-ϵ모형, LES 모형 등이 있으며, 본 연구에서는 여수로 방류에 따른 복잡한 난류 흐름 및 높은 전단흐름을 정확하게 모의(Flow Science, 2011)할 수 있는 RNG k-ϵ모형을 사용하였고, 하류하천 호안의 안정성 측면에서 보조여수로의 활용방안을 검토하기 위하여 방류시나리오는 Table 3에 제시된 것 같이 설정하였다. Case 1 및 Case 2를 통하여 계획홍수량에 대하여 기존 여수로와 보조 여수로의 단독 운영이 하류하천에 미치는 영향을 확인하였고 보조 여수로의 방류량 조절을 통하여 호안 안정성 측면에서 보조 여수로 방류능 검토를 수행하였다(Case 3 ~ Case 6). 또한 기존 여수로와 보조 여수로의 방류량 배분에 따른 하류하천의 영향 검토(Case 7 ~ Case 10) 및 방류 배분에 따른 허용 방류량을 호안 안정성 측면에서 검토를 수행하였다(Case 11 ~ Case 14).

    수문은 완전개도 조건으로 가정하였으며 하류하천의 계획홍수량에 대한 기존 여수로와 보조여수로의 배분량을 조절하여 모의를 수행하였다. 여수로는 콘크리트의 조도계수 값(Chow, 1959)을 채택하였고, 댐 하류하천의 조도계수는 하천기본계획(Busan Construction and Management Administration, 2009) 제시된 조도계수 값을 채택하였으며 FLOW-3D의 적용을 위하여 Manning-Strickler 공식(Vanoni, 2006)을 이용하여 조도계수를 조고값으로 변환하여 사용하였다. Manning-Strickler 공식은 Eq. (8)과 같으며, FLOW-3D에 적용한 조도계수 및 조고는 Table 4와 같다.

    (8)

    n=ks1/68.1g1/2

    여기서, kS는 조고 (m), n은 Manning의 조도계수, g는 중력가속도(m/s2)를 의미한다.

    시간에 따라 동일한 유량이 일정하게 유입되도록 모의를 수행하였으며, 시간간격(Time Step)은 0.0001초로 설정(CFL number < 1.0) 하였다. 또한 여수로 수문을 통한 유량의 변동 값이 1.0%이내일 경우는 연속방정식을 만족하고 있다고 가정하였다. 이는, 유량의 변동 값이 1.0%이내일 경우 유속의 변동 값 역시 1.0%이내이며, 수치모의 결과 1.0%의 유속변동은 호안의 유속설계기준에 크게 영향을 미치지 않는다고 판단하였다. 그 결과 모든 수치모의 Case에서 2400초 이내에 결과 값이 수렴하는 것을 확인하였다.

    Table 2.

    Mesh sizes and numerical conditions

    MeshNumbers49,102,500 EA
    Increment (m)DirectionExisting SpillwayAuxiliary Spillway
    ∆X0.99 ~ 4.301.00 ~ 4.30
    ∆Y0.99 ~ 8.161.00 ~ 5.90
    ∆Z0.50 ~ 1.220.50 ~ 2.00
    Boundary ConditionsXmin / YmaxInflow / Water Surface Elevation
    Xmax, Ymin, Zmin / ZmaxWall / Symmetry
    Turbulence ModelRNG model
    Table 3.

    Case of numerical simulation (Qp : Design flood discharge)

    CaseExisting Spillway (Qe, m3/s)Auxiliary Spillway (Qa, m3/s)Remarks
    1Qp0Reference case
    20Qp
    300.58QpReview of discharge capacity on
    auxiliary spillway
    400.48Qp
    500.45Qp
    600.32Qp
    70.50Qp0.50QpDetermination of optimal division
    ratio on Spillways
    80.61Qp0.39Qp
    90.39Qp0.61Qp
    100.42Qp0.58Qp
    110.32Qp0.45QpDetermination of permissible
    division on Spillways
    120.35Qp0.48Qp
    130.38Qp0.53Qp
    140.41Qp0.56Qp
    Table 4.

    Roughness coefficient and roughness height

    CriteriaRoughness coefficient (n)Roughness height (ks, m)
    Structure (Concrete)0.0140.00061
    River0.0330.10496
    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F1.jpg
    Fig. 1

    Layout of spillway and river in this study

    2.3.2 보조 여수로의 방류능 검토

    본 연구에서는 기존 여수로와 보조 여수로의 방류량 배분에 따른 하류하천 대안부의 유속분포 및 수위분포를 검토하기 위해 수치모의 Case 별 다음과 같이 관심구역을 설정하였다(Fig. 2 참조). 관심구역(대안부)의 길이(L)는 총 1.3 km로 10 m 등 간격으로 나누어 검토하였으며, Section 1(0 < X/L < 0.27)은 기존 여수로 방류에 따른 영향이 지배적인 구간, Section 2(0.27 < X/L < 1.00)는 보조 여수로 방류에 따른 영향이 지배적인 구간으로 각 구간에서의 수위, 유속, 수심결과를 확인하였다. 기존 여수로의 노후화에 따른 보조 여수로의 방류능 검토를 위하여 Case 1 – Case 6까지의 결과를 비교하였다.

    보조 여수로의 단독 운영 시 기존 여수로 운영 시 보다 하류하천의 대안부의 최대 유속(Vmax)은 약 3% 감소하였으며, 이는 보조 여수로의 하천 유입각이 기존 여수로 보다 7°작으며 유입하천의 폭이 증가하여 유속이 감소한 것으로 판단된다. 대안부의 최대 유속 발생위치는 하류 쪽으로 이동하였으며 교량으로 인한 단면의 축소로 최대유속이 발생하는 것으로 판단된다. 또한 보조 여수로의 배분량(Qa)이 증가함에 따라 하류하천 대안부의 최대 유속이 증가하였다. 하천호안 설계기준에서 제시하고 있는 허용유속(Vp)과 비교한 결과, 계획홍수량(Qp)의 45% 이하(Case 5 & 6)를 보조 여수로에서 방류하게 되면 허용 유속(5.0 m/s)조건을 만족하여 호안안정성을 확보하였다(Fig. 3 참조). 허용유속 외에도 대안부에서의 소류력을 산정하여 하천호안 설계기준에서 제시한 허용 소류력(τp)과 비교한 결과, 유속과 동일하게 보조 여수로의 방류량이 계획홍수량의 45% 이하일 경우 허용소류력(0.28 kN/m2) 조건을 만족하였다(Fig. 4 참조). 각 Case 별 호안설계조건과 비교한 결과는 Table 5에 제시하였다.

    하류하천의 수위도 기존 여수로 운영 시 보다 보조 여수로 단독 운영 시 최대 수위(ηmax)가 약 2% 감소하는 효과를 보였으며 최대 수위 발생위치는 수충부로 여수로 방류시 처오름에 의한 수위 상승으로 판단된다. 기존 여수로의 단독운영(Case 1)의 수위(ηref)를 기준으로 보조 여수로의 방류량이 증가함에 따라 수위는 증가하였으나 계획홍수량의 58%까지 방류할 경우 월류에 대한 안정성(ηmax/ηref<0.97(=기설제방고))은 확보되었다(Fig. 5 참조). 그러나 계획홍수량 조건에서는 월류에 대한 위험성이 존재하기 때문에 기존여수로와 보조여수로의 적절한 방류량 배분 조합을 도출하는 것이 중요하다고 판단되어 진다.

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F2.jpg
    Fig. 2

    Region of interest in this study

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F3.jpg
    Fig. 3

    Maximum velocity and location of Vmax according to Qa

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F4.jpg
    Fig. 4

    Maximum shear according to Qa

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F5.jpg
    Fig. 5

    Maximum water surface elevation and location of ηmax according to Qa

    Table 5.

    Numerical results for each cases (Case 1 ~ Case 6)

    CaseMaximum Velocity
    (Vmax, m/s)    		Maximum Shear
    (τmax, kN/m2)    		Evaluation
    in terms of Vp    		Evaluation
    in terms of τp
    1
    (Qa = 0)    		9.150.54No GoodNo Good
    2
    (Qa = Qp)    		8.870.56No GoodNo Good
    3
    (Qa = 0.58Qp)    		6.530.40No GoodNo Good
    4
    (Qa = 0.48Qp)    		6.220.36No GoodNo Good
    5
    (Qa = 0.45Qp)    		4.220.12AccpetAccpet
    6
    (Qa = 0.32Qp)    		4.040.14AccpetAccpet

    2.3.3 기존 여수로와 보조 여수로 방류량 배분 검토

    기존 여수로 및 보조 여수로 단독운영에 따른 하류하천 및 호안의 안정성 평가를 수행한 결과 계획홍수량 방류 시 하류하천 대안부에서 호안 설계 조건(허용유속 및 허용 소류력)을 초과하였으며, 처오름에 의한 수위 상승으로 월류에 대한 위험성 증가를 확인하였다. 따라서 계획 홍수량 조건에서 기존 여수로와 보조 여수로의 방류량 배분을 통하여 호안 안정성을 확보하고 하류하천에 방류로 인한 피해를 최소화할 수 있는 배분조합(Case 7 ~ Case 10)을 검토하였다. Case 7은 기존 여수로와 보조여수로의 배분 비율을 균등하게 적용한 경우이고, Case 8은 기존 여수로의 배분량이 보조 여수로에 비하여 많은 경우, Case 9는 보조 여수로의 배분량이 기존 여수로에 비하여 많은 경우를 의미한다. 최대유속을 비교한 결과 보조 여수로의 배분 비율이 큰 경우 기존 여수로의 배분량에 의하여 흐름이 하천 중심에 집중되어 대안부의 유속을 저감하는 효과를 확인하였다. 보조여수로의 방류량 배분 비율이 증가할수록 기존 여수로 대안부 측(0.00<X/L<0.27, Section 1) 유속 분포는 감소하였으나, 신규여수로 대안부 측(0.27<X/L<1.00, Section 2) 유속은 증가하는 것을 확인하였다(Fig. 6 참조). 그러나 유속 저감 효과에도 대안부 전구간에서 설계 허용유속 조건을 초과하여 제방의 안정성을 확보하지는 못하였다. 소류력 산정 결과 유속과 동일하게 보조 여수로의 방류량이 기존 여수로의 방류량 보다 크면 감소하는 것을 확인하였고 일부 구간에서는 허용 소류력 조건을 만족하는 것을 확인하였다(Fig. 7 참조).

    따라서 유속 저감효과가 있는 배분 비율 조건(Qa>Qe)에서 Section 2에 유속 저감에 영향을 미치는 기존 여수로 방류량 배분 비율을 증가시켜 추가 검토(Case 10)를 수행하였다. 단독운영과 비교 시 하류하천에 유입되는 유량은 증가하였음에도 불구하고 기존 여수로 방류량에 의해 흐름이 하천 중심으로 집중되는 현상에 따라 대안부의 유속은 단독 운영에 비하여 감소하는 것을 확인하였고(Fig. 8 참조), 호안 설계 허용유속 및 허용 소류력 조건을 만족하는 구간이 발생하여 호안 안정성도 확보한 것으로 판단되었다. 최종적으로 각 Case 별 수위 결과의 경우 여수로 동시 운영을 수행하게 되면 대안부 전 구간에서 월류에 대한 안정성(ηmax/ηref<0.97(=기설제방고))은 확보하였다(Fig. 9 참조). 각 Case 별 대안부에서 최대 유속결과 및 산정한 소류력은 Table 6에 제시하였다.

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F6.jpg
    Fig. 6

    Maximum velocity on section 1 & 2 according to Qa

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F7.jpg
    Fig. 7

    Maximum shear on section 1 & 2 according to Qa

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F8.jpg
    Fig. 8

    Velocity results of FLOW-3D (a: auxiliary spillway operation only , b : simultaneous operation of spillways)

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F9.jpg
    Fig. 9

    Maximum water surface elevation on section 1 & 2 according to Qa

    Table 6.

    Numerical results for each cases (Case 7 ~ Case 10)

    Case (Qe &amp; Qa)Maximum Velocity (Vmax, m/s)Maximum Shear
    (τmax, kN/m2)    		Evaluation in terms of VpEvaluation in terms of τp
    Section 1Section 2Section 1Section 2Section 1Section 2Section 1Section 2
    7
    Qe : 0.50QpQa : 0.50Qp    		8.106.230.640.30No GoodNo GoodNo GoodNo Good
    8
    Qe : 0.61QpQa : 0.39Qp    		8.886.410.610.34No GoodNo GoodNo GoodNo Good
    9
    Qe : 0.39QpQa : 0.61Qp    		6.227.330.240.35No GoodNo GoodAcceptNo Good
    10
    Qe : 0.42QpQa : 0.58Qp    		6.394.790.300.19No GoodAcceptNo GoodAccept

    2.3.4 방류량 배분 비율의 허용 방류량 검토

    계획 홍수량 방류 시 기존 여수로와 보조 여수로의 배분 비율 검토 결과 Case 10(Qe = 0.42Qp, Qa = 0.58Qp)에서 방류에 따른 하류 하천의 피해를 최소화시킬 수 있는 것을 확인하였다. 그러나 대안부 전 구간에 대하여 호안 설계조건을 만족하지 못하였다. 따라서 기존 여수로와 보조 여수로의 방류 배분 비율을 고정시킨 후 총 방류량을 조절하여 허용 방류량을 검토하였다(Case 11 ~ Case 14).

    호안 안정성 측면에서 검토한 결과 계획홍수량 대비 총 방류량이 감소하면 최대 유속 및 최대 소류력이 감소하고 최종적으로 계획 홍수량의 77%를 방류할 경우 하류하천의 대안부에서 호안 설계조건을 모두 만족하는 것을 확인하였다(Fig. 10, Fig. 11 참조). 각 Case 별 대안부에서 최대 유속결과 및 산정한 소류력은 Table 7에 제시하였다. 또한 Case 별 수위 검토 결과 처오름으로 인한 대안부 전 구간에서 월류에 대한 안정성(ηmax/ηref<0.97(=기설제방고))은 확보하였다(Fig. 12 참조).

    Table 7.

    Numerical results for each cases (Case 11 ~ Case 14)

    Case (Qe &amp; Qa)Maximum Velocity
    (Vmax, m/s)    		Maximum Shear
    (τmax, kN/m2)    		Evaluation in terms of VpEvaluation in terms of τp
    Section 1Section 2Section 1Section 2Section 1Section 2Section 1Section 2
    11
    Qe : 0.32QpQa : 0.45Qp    		3.634.530.090.26AcceptAcceptAcceptAccept
    12
    Qe : 0.35QpQa : 0.48Qp    		5.745.180.230.22No GoodNo GoodAcceptAccept
    13
    Qe : 0.38QpQa : 0.53Qp    		6.704.210.280.11No GoodAcceptAcceptAccept
    14
    Qe : 0.41QpQa : 0.56Qp    		6.545.240.280.24No GoodNo GoodAcceptAccept
    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F10.jpg
    Fig. 10

    Maximum velocity on section 1 & 2 according to total outflow

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F11.jpg
    Fig. 11

    Maximum shear on section 1 & 2 according to total outflow

    /media/sites/ksds/2021-014-02/N0240140207/images/ksds_14_02_07_F12.jpg
    Fig. 12

    Maximum water surface elevation on section 1 & 2 according to total outflow

    3. 결 론

    본 연구에서는 홍수 시 기존 여수로의 노후화로 인한 보조 여수로의 활용방안에 대하여 하류하천의 호안 안정성 측면에서 검토하였다. 여수로 방류로 인한 하류하천의 흐름특성을 검토하기 위하여 3차원 수치모형인 FLOW-3D를 활용하였고, 여수로 지형은 치수능력 증대사업을 통하여 완공된 ○○댐의 제원을 이용하였다. 하류하천 조도 계수 및 여수로 방류량은 하천기본계획을 참고하여 적용하였다. 최종적으로 여수로 방류로 인한 하류하천의 피해를 최소화 시킬 수 있는 적절한 보조 여수로의 활용방안을 도출하기 위하여 보조 여수로 단독 운영과 기존 여수로와의 동시 운영에 따른 하류 하천의 흐름특성 및 소류력의 변화를 검토하였다.

    수문은 완전 개도 상태에서 방류한다는 가정으로 계획 홍수량 조건에서 보조 여수로 단독 운영 시 하류하천 대안부의 유속 및 수위를 검토한 결과 기존 여수로 단독운영에 비하여 최대 유속 및 최대 수위가 감소하는 것을 확인할 수 있었으며, 이는 보조 여수로 단독 운영 시 하류하천으로 유입각도가 작아지고, 유입되는 하천의 폭이 증가되기 때문이다. 그러나 계획 홍수량 조건에서 하천호안 설계기준에서 제시한 허용 유속(5.0 m/s)과 허용 소류력(0.28 kN/m2)과 비교하였을 때 호안 안정성을 확보하지 못하였으며, 계획홍수량의 45% 이하 방류 시에 대안부의 호안 안정성을 확보하였다. 수위의 경우 여수로 방류에 따른 대안부에서 처오름 현상이 발생하여 월류에 대한 위험성을 확인하였고 이를 통하여 기존 여수로와의 동시 운영 방안을 도출하는 것이 중요하다고 판단된다. 따라서 기존 여수로와의 동시 운영 측면에서 기존 여수로와 보조 여수로의 배분 비율 및 총 방류량을 변화시켜가며 하류 하천의 흐름특성 및 소류력의 변화를 검토하였다. 배분 비율의 경우 기존 여수로와 보조 여수로의 균등 배분(Case 7) 및 편중 배분(Case 8 & Case 9)을 검토하여 보조 여수로의 방류량이 기존 여수로의 방류량보다 큰 경우 하류하천의 중심부로 집중되어 대안부의 최대유속, 최대소류력 및 최대수위가 감소하는 것을 확인하였다. 이를 근거로 기존 여수로의 방류 비율을 증가(Qe=0.42Qp, Qa=0.58Qp)시켜 검토한 결과 대안부 일부 구간에서 허용 유속 및 허용소류력 조건을 만족하는 것을 확인하였다. 이를 통하여 기존 여수로와 보조 여수로의 동시 운영을 통하여 적절한 방류량 배분 비율을 도출하는 것이 방류로 인한 하류하천의 피해를 저감하는데 효과적인 것으로 판단된다. 그러나 설계홍수량 방류 시 전 구간에서 허용 유속 및 소류력 조건을 만족하지 못하였다. 최종적으로 전체 방류량에서 기존 여수로의 방류 비율을 42%, 보조 여수로의 방류 비율을 58%로 설정하여 허용방류량을 검토한 결과, 계획홍수량의 77%이하로 방류 시 대안부의 최대유속은 기존여수로 방류의 지배영향구간(section 1)에서 3.63 m/s, 기존 여수로와 보조 여수로 방류의 영향구간(section 2)에서 4.53 m/s로 허용유속 조건을 만족하였고, 산정한 소류력도 각각 0.09 kN/m2 및 0.26 kN/m2로 허용 소류력 조건을 만족하여 대안부 호안의 안정성을 확보하였다고 판단된다.

    본 연구 결과는 기후변화 및 기존여수로의 노후화로 인하여 홍수 시 기존여수로의 단독운영으로 하류하천의 피해가 발생할 수 있는 현시점에서 치수증대 사업으로 완공된 보조 여수로의 활용방안에 대한 기초자료로 활용될 수 있고, 향후 계획 홍수량 유입 시 최적의 배분 비율 및 허용 방류량 도출에 이용할 수 있다. 다만 본 연구는 여수로 방류에 따른 제방에 작용하는 수충력은 검토하지 못하고, 허용 유속 및 허용소류력은 제방과 유수의 방향이 일정한 구간에 대하여 검토하였다. 또한 여수로 방류에 따른 대안부에서의 영향에 대해서만 검토하였고 수문 전면 개도 조건에서 검토하였다는 한계점은 분명히 있다. 이에 향후에는 다양한 수문 개도 조건 및 방류 시나리오를 적용 및 검토하여 보다 효율적이고, 효과적인 보조 여수로 활용방안을 도출하고자 한다.

    Acknowledgements

    본 결과물은 K-water에서 수행한 기존 및 신규 여수로 효율적 연계운영 방안 마련(2021-WR-GP-76-149)의 지원을 받아 연구되었습니다.

    References

    1 Busan Construction and Management Administration (2009). Nakdonggang River Master Plan. Busan: BCMA.

    2 Chow, V. T. (1959). Open-channel Hydraulics. McGraw-Hill. New York.

    3 Flow Science (2011). Flow3D User Manual. Santa Fe: NM.

    4 Jeon, T. M., Kim, H. I., Park, H. S., and Baek, U. I. (2006). Design of Emergency Spillway Using Hydraulic and Numerical Model-ImHa Multipurpose Dam. Proceedings of the Korea Water Resources Association Conference. 1726-1731.

    5 Kim, D. G., Park, S. J., Lee, Y. S., and Hwang, J. H. (2008). Spillway Design by Using Numerical Model Experiment – Case Study of AnDong Multipurpose Dam. Proceedings of the Korea Water Resources Association Conference. 1604-1608.

    6 Kim, J. S. (2007). Comparison of Hydraulic Experiment and Numerical Model on Spillway. Water for Future. 40(4): 74-81.

    7 Kim, S. H. and Kim, J. S. (2013). Effect of Chungju Dam Operation for Flood Control in the Upper Han River. Journal of the Korean Society of Civil Engineers. 33(2): 537-548. 10.12652/Ksce.2013.33.2.537

    8 K-water (2021). Regulations of Dam Management. Daejeon: K-water.

    9 K-water and MOLIT (2004). Report on the Establishment of Basic Plan for the Increasing Flood Capacity and Review of Hydrological Stability of Dams. Sejong: K-water and MOLIT.

    10 Lee, J. H., Julien, P. Y., and Thornton, C. I. (2019). Interference of Dual Spillways Operations. Journal of Hydraulic Engineering. 145(5): 1-13. 10.1061/(ASCE)HY.1943-7900.0001593

    11 Li, S., Cain, S., Wosnik, M., Miller, C., Kocahan, H., and Wyckoff, R. (2011). Numerical Modeling of Probable Maximum Flood Flowing through a System of Spillways. Journal of Hydraulic Engineering. 137(1): 66-74. 10.1061/(ASCE)HY.1943-7900.0000279

    12 MOLIT (2016). Practice Guidelines of River Construction Design. Sejong: MOLIT.

    13 MOLIT (2019). Standards of River Design. Sejong: MOLIT.

    14 Prime Minister’s Secretariat (2003). White Book on Flood Damage Prevention Measures. Sejong: PMS.

    15 Schoklitsch, A. (1934). Der Geschiebetrieb und Die Geschiebefracht. Wasserkraft Wasserwirtschaft. 4: 1-7.

    16 Vanoni, V. A. (Ed.). (2006). Sedimentation Engineering. American Society of Civil Engineers. Virginia: ASCE. 10.1061/9780784408230

    17 Zeng, J., Zhang, L., Ansar, M., Damisse, E., and González-Castro, J. A. (2017). Applications of Computational Fluid Dynamics to Flow Ratings at Prototype Spillways and Weirs. I: Data Generation and Validation. Journal of Irrigation and Drainage Engineering. 143(1): 1-13. 10.1061/(ASCE)IR.1943-4774.0001112

    Korean References Translated from the English

    1 건설교통부·한국수자원공사 (2004). 댐의 수문학적 안정성 검토 및 치수능력증대방안 기본계획 수립 보고서. 세종: 국토교통부.

    2 국무총리실 수해방지대책단 (2003). 수해방지대책 백서. 세종: 국무총리실.

    3 국토교통부 (2016). 하천공사 설계실무요령. 세종: 국토교통부.

    4 국토교통부 (2019). 하천설계기준해설. 세종: 국토교통부.

    5 김대근, 박선중, 이영식, 황종훈 (2008). 수치모형실험을 이용한 여수로 설계 – 안동다목적댐. 한국수자원학회 학술발표회. 1604-1608.

    6 김상호, 김지성 (2013). 충주댐 방류에 따른 댐 상하류 홍수위 영향 분석. 대한토목학회논문집. 33(2): 537-548. 10.12652/Ksce.2013.33.2.537

    7 김주성 (2007). 댐 여수로부 수리 및 수치모형실험 비교 고찰. Water for Future. 40(4): 74-81.

    8 부산국토관리청 (2009). 낙동강수계 하천기본계획(변경). 부산: 부산국토관리청.

    9 전태명, 김형일, 박형섭, 백운일 (2006). 수리모형실험과 수치모의를 이용한 비상여수로 설계-임하댐. 한국수자원학회 학술발표회. 1726-1731.

    10 한국수자원공사 (2021). 댐관리 규정. 대전: 한국수자원공사.

    원문보기
    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process

    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process

    반고체 레오 다이 캐스팅 공정으로 제작된 알루미늄 합금 브래킷의 수치 시뮬레이션 및 생산 실험 검증을 기반으로 한 게이팅 시스템 설계

    International Journal of Metalcasting volume 16, pages878–893 (2022)Cite this article

    Abstract

    In this study a gating system including sprue, runner and overflows for semi-solid rheocasting of aluminum alloy was designed by means of numerical simulations with a commercial software. The effects of pouring temperature, mold temperature and injection speed on the filling process performance of semi-solid die casting were studied. Based on orthogonal test analysis, the optimal die casting process parameters were selected, which were metal pouring temperature 590 °C, mold temperature 260 °C and injection velocity 0.5 m/s. Semi-solid slurry preparation process of Swirled Enthalpy Equilibration Device (SEED) was used for die casting production experiment. Aluminum alloy semi-solid bracket components were successfully produced with the key die casting process parameters selected, which was consistent with the simulation result. The design of semi-solid gating system was further verified by observing and analyzing the microstructure of different zones of the casting. The characteristic parameters, particle size and shape factor of microstructure of the produced semi-solid casting showed that the semi-solid aluminum alloy components are of good quality.

    이 연구에서 알루미늄 합금의 반고체 레오캐스팅을 위한 스프루, 러너 및 오버플로를 포함하는 게이팅 시스템은 상용 소프트웨어를 사용한 수치 시뮬레이션을 통해 설계되었습니다. 주입 온도, 금형 온도 및 사출 속도가 반고체 다이캐스팅의 충전 공정 성능에 미치는 영향을 연구했습니다. 직교 테스트 분석을 기반으로 금속 주입 온도 590°C, 금형 온도 260°C 및 사출 속도 0.5m/s인 최적의 다이 캐스팅 공정 매개변수가 선택되었습니다. Swirled Enthalpy Equilibration Device(SEED)의 반고체 슬러리 제조 공정을 다이캐스팅 생산 실험에 사용하였다. 알루미늄 합금 반고체 브래킷 구성 요소는 시뮬레이션 결과와 일치하는 주요 다이 캐스팅 공정 매개변수를 선택하여 성공적으로 생산되었습니다. 반고체 게이팅 시스템의 설계는 주조의 다른 영역의 미세 구조를 관찰하고 분석하여 추가로 검증되었습니다. 생산된 반고체 주조물의 특성 매개변수, 입자 크기 및 미세 구조의 형상 계수는 반고체 알루미늄 합금 부품의 품질이 양호함을 보여주었습니다.

    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process
    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process

    References

    1. G. Li, H. Lu, X. Hu et al., Current progress in rheoforming of wrought aluminum alloys: a review. Met. Open Access Metall. J. 10(2), 238 (2020)CAS Google Scholar 
    2. G. Eisaabadi, A. Nouri, Effect of Sr on the microstructure of electromagnetically stirred semi-solid hypoeutectic Al–Si alloys. Int. J. Metalcast. 12, 292–297 (2018). https://doi.org/10.1007/s40962-017-0161-8CAS Article Google Scholar 
    3. C. Xghab, D. Qza, E. Spma et al., Blistering in semi-solid die casting of aluminium alloys and its avoidance. Acta Mater. 124, 446–455 (2017)Article Google Scholar 
    4. M. Modigell, J. Koke, Rheological modelling on semi-solid metal alloys and simulation of thixocasting processes. J. Mater. Process. Technol. 111(1–3), 53–58 (2001)CAS Article Google Scholar 
    5. A. Pola, M. Tocci, P. Kapranos, Microstructure and properties of semi-solid aluminum alloys: a literature review. Met. Open Access Metall. J. 8(3), 181 (2018)Google Scholar 
    6. M.C. Flemings, Behavior of metal alloys in the semisolid state. Metall. Trans. B 22, 269–293 (1991). https://doi.org/10.1007/BF02651227Article Google Scholar 
    7. Q. Zhu, Semi-solid moulding: competition to cast and machine from forging in making automotive complex components. Trans. Nonferrous Met. Soc. China 20, 1042–1047 (2010)Article Google Scholar 
    8. K. Prapasajchavet, Y. Harada, S. Kumai, Microstructure analysis of Al–5.5 at.%Mg alloy semi-solid slurry by Weck’s reagent. Int. J. Metalcast. 11(1), 123 (2017). https://doi.org/10.1007/s40962-016-0084-9Article Google Scholar 
    9. P. Das, S.K. Samanta, S. Tiwari, P. Dutta, Die filling behaviour of semi solid A356 Al alloy slurry during rheo pressure die casting. Trans. Indian Inst. Met. 68(6), 1215–1220 (2015). https://doi.org/10.1007/s12666-015-0706-6CAS Article Google Scholar 
    10. B. Zhou, S. Lu, K. Xu et al., Microstructure and simulation of semisolid aluminum alloy castings in the process of stirring integrated transfer-heat (SIT) with water cooling. Int. J. Metalcast. 14(2), 396–408 (2019). https://doi.org/10.1007/s40962-019-00357-6CAS Article Google Scholar 
    11. S. Ji, Z. Fan, Solidification behavior of Sn–15 wt Pct Pb alloy under a high shear rate and high intensity of turbulence during semisolid processing. Metall. Mater. Trans. A. 33(11), 3511–3520 (2002). https://doi.org/10.1007/s11661-002-0338-4Article Google Scholar 
    12. P. Kapranos, P.J. Ward, H.V. Atkinson, D.H. Kirkwood, Near net shaping by semi-solid metal processing. Mater. Des. 21, 387–394 (2000). https://doi.org/10.1016/S0261-3069(99)00077-1Article Google Scholar 
    13. H.V. Atkinson, Alloys for semi-solid processing. Solid State Phenom. 192–193, 16–27 (2013)Google Scholar 
    14. L. Rogal, Critical assessment: opportunities in developing semi-solid processing: aluminium, magnesium, and high-temperature alloys. Mater. Sci. Technol. Mst A Publ. Inst. Met. 33, 759–764 (2017)CAS Article Google Scholar 
    15. H. Guo, Rheo-diecasting process for semi-solid aluminum alloys. J. Wuhan Univ. Technol. Mater. Sci. Ed. 22(004), 590–595 (2007)CAS Article Google Scholar 
    16. T. Chucheep, J. Wannasin, R. Canyook, T. Rattanochaikul, S. Janudom, S. Wisutmethangoon, M.C. Flemings, Characterization of flow behavior of semi-solid slurries with low solid fractions. Metall. Mater. Trans. A 44(10), 4754–4763 (2013)CAS Article Google Scholar 
    17. M. Li, Y.D. Li, W.L. Yang et al., Effects of forming processes on microstructures and mechanical properties of A356 aluminum alloy prepared by self-inoculation method. Mater. Res. 22(3) (2019)
    18. P. Côté, M.E. Larouche, X.G. Chen et al., New developments with the SEED technology. Solid State Phenom. 192(3), 373–378 (2012)Article Google Scholar 
    19. I. Dumanić, S. Jozić, D. Bajić et al., Optimization of semi-solid high-pressure die casting process by computer simulation, Taguchi method and grey relational analysis. Inter Metalcast. 15, 108–118 (2021). https://doi.org/10.1007/s40962-020-00422-5Article Google Scholar 
    20. Y. Bai et al., Numerical simulation on the rheo-diecasting of the semi-solid A356 aluminum alloy. Int. J. Miner. Metall. Mater. 16, 422 (2009). https://doi.org/10.1016/S1674-4799(09)60074-1CAS Article Google Scholar 
    21. B.C. Bhunia, Studies on die filling of A356 Al alloy and development of a steering knuckle component using rheo pressure die casting system. J. Mater. Process. Technol. 271, 293–311 (2019). https://doi.org/10.1016/j.jmatprotec.2019.04.014CAS Article Google Scholar 
    22. A. Guo, J. Zhao, C. Xu et al., Effects of pouring temperature and electromagnetic stirring on porosity and mechanical properties of A357 aluminum alloy rheo-diecasting. J. Mater. Eng. Perform. (2018). https://doi.org/10.1007/s11665-018-3310-1Article Google Scholar 
    23. C.G. Kang, S.M. Lee, B.M. Kim, A study of die design of semi-solid die casting according to gate shape and solid fraction. J. Mater. Process. Technol. 204(1–3), 8–21 (2008)CAS Article Google Scholar 
    24. Z. Liu, W. Mao, T. Wan et al., Study on semi-solid A380 aluminum alloy slurry prepared by water-cooling serpentine channel and its rheo-diecasting. Met. Mater. Int. (2020). https://doi.org/10.1007/s12540-020-00672-2Article Google Scholar 
    25. Z.Y. Liu, W.M. Mao, W.P. Wang et al., Investigation of rheo-diecasting mold filling of semi-solid A380 aluminum alloy slurry. Int. J. Miner. Metall. Mater. 24(006), 691–700 (2017)CAS Article Google Scholar 
    26. M. Arif, M.Z. Omar, N. Muhamad et al., Microstructural evolution of solid-solution-treated Zn–22Al in the semisolid state. J. Mater. Sci. Technol. 29(008), 765–774 (2013)CAS Article Google Scholar 

    Keywords

    • semi-solid rheo-die casting
    • gating system
    • process parameters
    • numerical simulation
    • microstructure
    원문보기
    Fig. 1- Schematic of the general pattern of flow and aeration process in the aerators

    2상 유동 해석을 통한 슈트 폭기 시스템 효율에 대한 램프 각도의 영향 조사

    Investigation of the Effect of Ramp Angle on Chute Aeration System Efficiency by Two-Phase Flow Analysis

    Authors

    1 Associate Professor, Civil Engineering Department, Jundi-Shapur University of Technology, Dezful, Iran

    2 Instructor in Civil Engineering Department Jundi-Shapur University of Technology, Dezful,Iran.

     10.22055/JISE.2021.37743.1980

    Abstract

    Flow aeration in chute spillway is one of the most effective and economic ways to prevent cavitation damage. Surface damage is significantly reduced when very small values of air are scattered in a water prism. A structure known as an aerator may be used for this purpose. Besides, ramp angle is one of the factors influencing aerator efficiency. In this research, the value of air entraining the flow through the Jarreh Dam’s spillway at the ramp angles of 6, 8 and 10 degrees, as three different scenarios, was simulated using the Flow-3D software. In order to validate the results of the inlet air into the flowing fluid at a ramp angle of 6 degrees, the observational results of the dam spillway physical model from the laboratory of TAMAB Company in Iran were used. According to the results, raising the ramp angle increases the inlet air to the water jet nappe, and a ten-degree ramp angle provides the best aeration efficiency. The Flow-3D model can also simulate the two-phase water-air flow on spillways, according to the results.

    슈트 여수로의 흐름 폭기는 캐비테이션 손상을 방지하는 가장 효과적이고 경제적인 방법 중 하나입니다. 수중 프리즘에 아주 작은 양의 공기가 흩어지면 표면 손상이 크게 줄어듭니다. 이를 위해 폭기 장치로 알려진 구조를 사용할 수 있습니다. 또한, 램프 각도는 폭기 효율에 영향을 미치는 요인 중 하나입니다. 이 연구에서는 FLOW-3D 소프트웨어를 사용하여 3가지 다른 시나리오인 6, 8 및 10도의 램프 각도에서 Jarreh 댐의 방수로를 통해 흐름을 동반하는 공기의 값을 시뮬레이션했습니다. 6도의 경사각에서 유동 유체로 유입되는 공기의 결과를 검증하기 위해이란 TAMAB Company의 실험실에서 댐 방수로 물리적 모델의 관찰 결과를 사용했습니다. 결과에 따르면 램프 각도를 높이면 워터제트 기저귀로 유입되는 공기가 증가하고 10도 램프 각도는 최고의 폭기 효율을 제공합니다. Flow-3D 모델은 결과에 따라 여수로의 2단계 물-공기 흐름을 시뮬레이션할 수도 있습니다.

    Keywords

    Fig. 1- Schematic of the general pattern of flow and aeration process in the aerators
    Fig. 1- Schematic of the general pattern of flow and aeration process in the aerators
    (a) The full-scale map of the Jarreh spillway’s plan and profile.
    (a) The full-scale map of the Jarreh spillway’s plan and profile.
    Fig. 2- Experimental setup (Shamloo et al., 2012)
    Fig. 2- Experimental setup (Shamloo et al., 2012)

    References

    1- Baharvand, S., & Lashkar-Ara, B. (2021). Hydraulic design criteria of the modified meander C-type
    fishway using the combined experimental and CFD models. Ecological Engineering, 164.
    https://doi.org/10.1016/j.ecoleng.2021.106207
    2- Bayon, A., Toro, J. P., Bombardelli, F. A., Matos, J., & López-Jiménez, P. A. (2018). Influence of VOF
    technique, turbulence model and discretization scheme on the numerical simulation of the non-aerated,
    skimming flow in stepped spillways. Journal of Hydro-Environment Research, 19, 137–149.
    https://doi.org/10.1016/j.jher.2017.10.002
    3- Brethour, J. M., & Hirt, C. W. (2009). Drift Model for Two-Component Flows. Flow Science, Inc., FSI09-TN83Rev, 1–7.
    4- Chanson, H. (1989). Study of air entrainment and aeration devices. Journal of Hydraulic Research, 27(3),
    301–319. https://doi.org/10.1080/00221688909499166
    5- Dong, Z., Wang, J., Vetsch, D. F., Boes, R. M., & Tan, G. (2019). Numerical simulation of air-water twophase flow on stepped spillways behind X-shaped flaring gate piers under very high unit discharge. Water
    (Switzerland), 11(10). https://doi.org/10.3390/w11101956
    6- Flow-3D, V. 11. 2. (2017). User Manual. Flow Science Inc.: Santa Fe, NM, USA;
    7- Hirt, C. W. (2003). Modeling Turbulent Entrainment of Air at a Free Surface. Flow Science, Inc., FSI-03-
    TN6, 1–9.
    8- Hirt, C. W. (2016). Dynamic Droplet Sizes for Drift Fluxes. Flow Science, Inc., 1–10.
    9- Hirt, C. W., & Nichols, B. D. (1981). Volume of fluid (VOF) method for the dynamics of free boundaries.
    Journal of Computational Physics, 39(1), 201–225. https://doi.org/10.1016/0021-9991(81)90145-5
    10- Kherbache, K., Chesneau, X., Zeghmati, B., Abide, S., & Benmamar, S. (2017). The effects of step
    inclination and air injection on the water flow in a stepped spillway: A numerical study. Journal of
    Hydrodynamics, 29(2), 322–331. https://doi.org/10.1016/S1001-6058(16)60742-4
    11- Kramer, M., & Chanson, H. (2019). Optical flow estimations in aerated spillway flows: Filtering and
    discussion on sampling parameters. Experimental Thermal and Fluid Science, 103, 318–328.
    https://doi.org/10.1016/j.expthermflusci.2018.12.002
    12- Mahmoudian, Z., Baharvand, S., & Lashkarara, B. (2019). Investigating the Flow Pattern in Baffle
    Fishway Denil Type. Irrigation Sciences and Engineering (JISE), 42(3), 179–196.
    13- Meireles, I. C., Bombardelli, F. A., & Matos, J. (2014). Air entrainment onset in skimming flows on
    steep stepped spillways: An analysis. Journal of Hydraulic Research, 52(3).
    https://doi.org/10.1080/00221686.2013.878401
    14- Parsaie, A., & Haghiabi, A. H. (2019). Inception point of flow aeration on quarter-circular crested stepped
    spillway. Flow Measurement and Instrumentation, 69.
    https://doi.org/10.1016/j.flowmeasinst.2019.101618
    15- Richardson, J. F., & Zaki W N. (1979). Sedimentation and Fluidisation. Part 1. Trans. Inst. Chem. Eng,
    32, 35–53.
    16- Shamloo, H., Hoseini Ghafari, S., & Kavianpour, M. (2012). Experimental study on the effects of inlet
    flows on aeration in chute spillway (Case study: Jare Dam, Iran). 10th International Congress on
    Advances in Civil Engineering, Middle East Technical University, Ankara, Turkey.
    17- Wang, S. Y., Hou, D. M., & Wang, C. H. (2012). Aerator of stepped chute in Murum Hydropower
    Station. Procedia Engineering, 28, 803–807. https://doi.org/10.1016/j.proeng.2012.01.813.
    18- Wei, W., Deng, J., & Zhang, F. (2016). Development of self-aeration process for supercritical chute
    flows. International Journal of Multiphase Flow, 79, 172–180.
    https://doi.org/10.1016/j.ijmultiphaseflow.2015.11.003
    19- Wu, J., QIAN, S., & MA, F. (2016). A new design of ski-jump-step spillway. Journal of Hydrodynamics,
    05, 914–917.
    20- Xu, Y., Wang, W., Yong, H., & Zhao, W. (2012). Investigation on the cavity backwater of the jet flow from the chute aerators. Procedia Engineering, 31, 51–56. https://doi.org/10.1016/j.proeng.2012.01.989
    21- Yakhot, V., & Orszag, S. A. (1986). Renormalization group analysis of turbulence. I. Basic theory.
    Journal of Scientific Computing, 1(1), 3–51. https://doi.org/10.1007/BF01061452
    22- Yang, J., Teng, P., & Lin, C. (2019). Air-vent layouts and water-air flow behaviors of a wide spillway
    aerator. Theoretical and Applied Mechanics Letters, 9(2), 130–143.
    https://doi.org/10.1016/j.taml.2019.02.009
    23- Zhang, G., & Chanson, H. (2016). Interaction between free-surface aeration and total pressure on a
    stepped chute. Experimental Thermal and Fluid Science, 74, 368–381.
    https://doi.org/10.1016/j.expthermflusci.2015.12.011

    PDF 원문 보기
    Fig 2(b) Observed velocity field for aspect ratio 0.25(Sukhodolov 2002)

    고정 베드의 불침투성 토양에서 흐름 패턴의 수치 시뮬레이션

    NUMERICAL SIMULATION OF FLOW PATTERN IN SERIES OF IMPERMEABLE GROYNES IN FIXED BED

    Kafle, Mukesh Raj1
    1Asst. Professor, Department of Civil Engineering, Institute of Engineering, Pulchowk Campus, Nepal
    Email: mkafle@pcampus.edu.np

    Abstract

    This paper presents a numerical simulation of recirculating flow patterns in groyne fields. Moreover, it entails the concept determination of proper spacing of vertical unsubmerged and impermeable groynesin seriesto control the bank erosion. Flow pattern between the groynes varies along their space. The flow in groyne field may significantly affect the flow change, bed change, bank erosion and condition of habitat. In this regard, an assessment of flow along the space of groynes will yield important data needed to diversify the object of groyne installation. So, knowledge about determination of the proper spacing of groynes in groyne field is important. Space of vertical groynes was set from 1.5 to 10 times the length of groynes. The velocity field between groynes was simulated by using Computational Fluid Dynamics (CFD) model Nays 2D. Simulated velocity field was compared with existing experimentaldata for the same parameter, which agreed satisfactorily. Based on simulated results,the optimal spacing of vertical groynes to control the bank erosion was recommended.

    이 논문은 groyne 필드에서 재순환 흐름 패턴의 수치 시뮬레이션을 제공합니다. 더욱이, 그것은 제방 침식을 제어하기 위해 수직 비침수 및 불침투성 그로이네신 시리즈의 적절한 간격의 개념 결정을 수반합니다. groynes 사이의 흐름 패턴은 공간에 따라 다릅니다. groyne field의 흐름은 흐름 변화, 하상 변화, 제방 침식 및 서식지 상태에 중대한 영향을 미칠 수 있습니다. 이와 관련하여, groyne 공간을 따른 흐름의 평가는 groyne 설치 대상을 다양화하는 데 필요한 중요한 데이터를 산출할 것입니다. 따라서, groyne field에서 groyne의 적절한 간격 결정에 대한 지식이 중요합니다. 수직 여백의 간격은 여아 길이의 1.5배에서 10배 사이로 설정하였다. groyne 사이의 속도장은 CFD(Computational Fluid Dynamics) 모델 Nays 2D를 사용하여 시뮬레이션되었습니다. 시뮬레이션된 속도장은 동일한 매개변수에 대해 기존 실험 데이터와 비교되었으며 만족스럽게 일치했습니다. 모의 결과를 바탕으로 제방 침식을 억제하기 위한 최적의 수직 제방 간격을 제안하였다.

    1. Introduction
      Spur dikes or groynes are used to protect river banks from erosion and also keep the channel
      navigable.Depending upon the flow characteristics, spur-dikes may be classified as submerged and unsubmerged. Also, based on the permeability, spur dikes are further classified as permeable and
      impermeable. Herein, un-submerged !impermeable spur dikes are dealt. These structures are built from the river bank into the stream flow and usually built in group. Construction of groyne against the flow causes significant changes in flow pattern in channel. Those changes may result in scour phenomenon around groynes which may lead structure instability and changes in river morphology. Moreover, in series of groynes, spacing of groynes leads different types of recirculating flow patterns.Therefore, investigating the characteristics of flow pattern around groynes have been a great interest in river engineering. Numerous researchers like Sukhodolov et al. (2002), Hao Zhang et al.(2009), Beheshti (2010), Duan (2009), Naji(2010), Karami(2011) made a variety of experiments in order to determine the flow pattern around groynes. Most of these researchers studied effect of single groyne, while using series of groynes is more effective in protection of rivers. Besides experimental studies, variety of CFD models have been developed for computing flow pattern around hydraulic structures; like Fluent, Flow 3D, Nays 2D, Nays CUBE and SSIIM. In this study, Nays 2D numerical modelling has been used to investigate flow and recirculating pattern around a series of groynes and streamlines including components of velocities.
    1. Flow pattern in groyne fields
      Under conditions where the groynes are not submerged, the groyne fields are not really part of the wetted cross section of a river. Because of that, the flow pattern in the groyne-field is not directly the result of the discharge in the main channel. Reducing the main stream velocity has no effect on the flow pattern itself, whereas lowering the water level does (Uijttewaal et al.2001). Moreover, the flow pattern inside a groyne field may change with the change of its geometry, location along the river (inner curve, outer curve, or straight part), and/ or the groynes orientation( Przedwojski et al.1995). However, there is an indirect effect of the discharge on the flow pattern in the groyne field. Because of the flow that is diverted from the main channel into the groyne fields, water flows into the groyne field with low velocity through the downstream half of the interfacial section between the groyne field and the main channel. This water flows back to the main channel through a small width of, just downstream the upstream groyne of the groyne field ( Termes et al.1991). Flow separates on a groyne head and forms a secondary flow represented by a large scale vortex with a vertical axis of rotation called primary gyre. Deflection of the flow inside the groyne field by banks and upstream groynes leads to the development of a secondary gyre with an opposite direction of rotation to the primary gyre. Location, mutual interactions, and energy exchange between gyres are the factors that create a specific recirculation pattern, and, consequently assuming correspondence with sedimentation processes, they define deposition patterns.
    2. Model Formulation
      The CFD model selected for this study is the publically available software NAYS 2D (iRIC 2.0), which is an analytical solver for calculation of unsteady two-dimensional plane flow and riverbed deformation using boundary-fitted coordinates within general curvilinear coordinates. A numerical channel of length 8.0m and width 0.9m was created with grid size of 0.01m im stream wise and 0.03m in cross stream directions. Groynes or spur dikes of length 0.15 and width 0.01m were chosen in series. Groyne field with various aspect ratio (b/x) 0.7, 0.25, 0.17, 0.125 and 0.10, where b=length of spur dike, x=spacing of two dikes. Discharge of 0.0175 m3 /s was applied. For boundary conditions, water surface at downstream and velocity at upstream were considered as uniform flow. Relaxation coefficient for water surface calculation was considered as 0.8. For the finite-difference method, the CIP method was applied to the advection terms in equations of motion. For the turbulent field calculation, Constant eddy viscosity, Zero-equation model and k-G models were applied and compared. The model!s accuracy in predicting the velocity magnitudes is evaluated using statistical parameters- mean absolute error (MAE), mean square error(MSE), and root mean square error (RMSE). The comparison of results shows the importance of selecting an appropriate turbulence model in simulating flow field around a spur dike. From the comparison, k-I model is found superior over zero energy model and eddy viscosity model. So, k-I model is chosen as appropriate turbulence closure model.
    3. Model!s Validation
      The capability of CFD model Nays 2D to simulate the velocity field and recirculation pattern in groyne field was compared with experimental data of laboratory experiments by Sukhodolov et al. (2002). The numerical simulation was validated for aspect ratio (R=b/x=0.7) and R=0.25. For aspect ratio R=0.7, one gyre system occupies the whole area of the groyne field. The areas with lower-than-average velocity values are clearly seen in the central part of the gyre and near its corners. Velocities increase towards the margins of the gyre. For aspect ratio R=0.25, two gyre velocity fields were observed in the groyne field. In the downstream part of the groyne field a large gyre, covering two-thirds of the area is clearly visible. The left part(upstream) contains second gyre rotating much more slowly and in the direction opposed to the primary gyre. The simulated and observed velocity field pattern and gyre found satisfactorily agreed. Now, after validation, the model was used for further analysis of velocity field for various aspect ratios.
    Fig 2(b) Observed velocity field for aspect ratio 0.25(Sukhodolov 2002)
    Fig 2(b) Observed velocity field for aspect ratio 0.25(Sukhodolov 2002)
    1. Results and Discussions
      The calibrated model was applied to five different cases of un-submerged and impermeable groyne fields with aspect ratios R=0.70,0.25,0.17,0.125 & 0.10 and flow pattern was numerically simulated. For aspect ratio R=0.7 i.e x/b=1.5, Fig 1(a) only one lateral primary gyre was formed inside the groyne field. The circulation pattern in this case is distinguished by the main flow that is deflected outside the groyne field. The developed primary gyre prevents the main flow from penetrating the groyne field. Therefore, this pattern is desirable for navigation purposes as a continuous deep channel is maintained along the face of the groyne field. Simulated velocity pattern satisfactorily agrees with the observed velocity field Fig 1(b) for the same aspect ratio by Sukhodolov (2002). The spacing of the groyne was further increased maintaining aspect ratio R= 0.25 i. e x/b=4 Fig 2(a) and flow pattern inside the groyne field was simulated. In this case, in the downstream part of the groyne field, a primary gyre occupying almost two-third area was formed. In addition, deflection of the flow inside the groyne field by banks and upstream groynes leads to the development of a secondary gyre with an opposite direction of rotation to the primary gyre covering almost one-third part of the groyne field. Likewise in the first case, the main current is maintained deflected outside the groyne field. Simulated velocity pattern satisfactorily agrees with the observed velocity field Fig 2(b) for the same aspect ratio by Sukhodolov (2002). The spacing of the groyne was further increased maintaining aspect ratio R=0.17 i.e x/b=6. In this case the flow pattern was similar to the aspect ratio R=0.25. The spacing of the groynes was further increased maintaining aspect ratio R=0.125 i. e x/b=8. In this case, similar to the previous scenarios two longitudinal gyres but with different positions are formed. The main current is directed in to the groyne field (Fig 3) creating a much more stronger eddy near the upstream groyne and greater turbulence along the upstream face and at the groyne lower head. As the spacing between groynes increased maintaining aspect ratio R=0.10 i. e x/b=10 (Fig 4), still primary and secondary gyres are generated. The formed gyres deflect the main flow thus preventing to enter in to the groyne field in upstream part. However, in the downstream of the primary gyre and just upstream of the second groyne, the flow attacks the bank directly. The resultant velocity profiles at the deflected region y/b=3 were plotted and how the spacing of second groyne affect the result was analyzed. Spacing of groynes makes little change in upstream resultant velocity. However, in the deflected region, its effect is significant. Higher value of spacing of groyne leads higher average deviation in resultant velocity. For aspect ratio R=0.7, the average deviation estimated as 0.02%. In the case of aspect ratio R=0.25, this value was reached to 1.57%. Further increment of spacing i. e decreasing the aspect ratio R=0.17, average deviation was found 3.82%. For the aspect ratio R=0.125, that value was estimated as 4.16%.
    3. Conclusions
      Geometry of the groyne fields; width and length of the groyne field mainly cause the specific flow patterns including number and shape of eddies or gyres. Eddies developed inside the groyne field deflects the main flow preventing it entering into the dead zone. An aspect ratio close to unity gives rise to a single eddy. A smaller aspect ratio (higher spacing between groynes) gives room to two stationary eddies, a large one called primary eddy, in the downstream part of the groyne field, and a smaller secondary eddy emerges near the upstream groyne. The extreme long groyne field -case of length to width ratio of larger thaneight shows penetration of main flow into the groyne field. The two eddies remain in a relatively stable position, while the main flow zone starts to penetrate into groyne field further downstream. In all cases, there is an eddy detaches from the upstream groyne tip that travels along the main channel groyne field interface and eventually merges with the primary eddy. The simulated results indicate that the spacing of groynes or spur dikes from the controlling of bank erosion point of view should be limited within six times the length of groyne.
    Fig 3 Computed velocity field for aspect ratio 0.125
    Fig 3 Computed velocity field for aspect ratio 0.125
    Fig 4 Computed velocity field for aspect ratio 0.10
    Fig 4 Computed velocity field for aspect ratio 0.10
    Fig 5 Resultant velocity profiles at y/b=3
    Fig 5 Resultant velocity profiles at y/b=3
    Fig 5 Resultant velocity profiles at y/b=3
    Fig 5 Resultant velocity profiles at y/b=3

    References

    1. Holtz, K.P  Numerical simulation of recirculating flow at groynes.Å Computer Methods in Water Resources, Vol 2, No 2 (1991).
    2. Hossein, Bassar; Abdollah, Ardeshir; Hojat, Karami.  Numerical simulation of flow pattern around spur dikes series in rigid bed.Å 9th international congress on civil engineering, May 8- 10,2012, Isfahan University of Technology (IUT) , Isfahan, Iran (2012).
    3. Kang, J.G; Yeo, H.K; Kim,S.J An experimental study on a characteristics of flow around groyne area by install conditions.Å www.SciRP.org/journal/eng(2012).
    4. Shimizu,Y; Nelson,JIntroduction of Nays solver in iRIC.Åwww.i-ric.org(2012).
    5. Sukhodolov, A. Uijttewaal, W. S. J., and Engelhardt, C. On the correspondence between morphological and hydro dynamical patterns of groyne fields.Å Earth Surf. Processes Landforms, 27(3) (2002).
    6. Uijttewall, W.S.J; Lehman,D; VanMazijk,A.  Exchange process between a river and its groyne fields-model experiments.Å Journal of Hydraulic Engineering, ASCE, 127(11) (2001).
    7. Uijttewall, W.S.J Groyne field velocity patterns determined with particle tracking
      velocimetryÅ.28th IAHR congress, Graz, Austria (1999).
    8. Yossef, Mohamed  Flow details near groynes: Experimental investigations.Å Journal of Hydraulic Engineering, ASCE, 137 (2011).
    Figura 4.5 – Superfície de contorno de velocidades a 3,9 m do piso após 20 s de acionamento da motosserra.

    전기톱 절단 시험실에서 배기 시스템의 CFD 시뮬레이션

    CFD Simulation of an exhaust system in chainsaw cutting test room

    Área de Concentração: Energia e Fenômenos de Transporte
    Orientador: Prof. Diogo Elias da Vinha Andrade
    Comissão de Avaliação:
    Profa
    . Letícia Jenisch Rodrigues
    Prof. Francis Henrique Ramos França
    Prof. Paulo Smith Schneider

    Abstract

    The objective of the present work is to improve an exhaust system for a chain saw
    cutting test room through a fluid dynamic computational simulation (CFD). The purpose of the
    designed system is to remove combustion gases, such as carbon monoxide (CO), which is
    extremely toxic, colourless and inodorous. The current system consists of a set of exhaust fans,
    a hood and an insufflation set. From experimental tests, the input data of the simulation were
    collected to define the variables and boundary conditions such as volumetric flow of CO, its
    temperature and density and the supply of fresh air in the room. The necessary means of
    instrumentation are presented so that it is possible to obtain the correlation with the results of
    the simulation and, once validated, a study of mesh refinement was carried out. With this, the
    possible solutions to the problem are evaluated through a case study involving the geometry of
    the hood and the exhaust and insufflation systems. By changing the hood geometry, the most
    satisfactory result was obtained for the problem, as it was shown to be able to remove all CO
    from the room, respecting the proposed operational limits.

    현재 연구의 목적은 유체 역학 계산 시뮬레이션(CFD)을 통해 체인 톱 절단 시험실의 배기 시스템을 개선하는 것입니다. 설계된 시스템의 목적은 매우 유독하고 무색이며 냄새가 나는 일산화탄소(CO)와 같은 연소 가스를 제거하는 것입니다. 현재 시스템은 배기 팬 세트, 후드 및 흡입 세트로 구성됩니다. 실험 테스트에서 시뮬레이션의 입력 데이터는 CO의 체적 유량, 온도 및 밀도, 실내의 신선한 공기 공급과 같은 변수 및 경계 조건을 정의하기 위해 수집되었습니다. 시뮬레이션 결과와의 상관관계를 얻을 수 있도록 필요한 계측 수단을 제시하고 검증 후 메쉬 미세화 연구를 수행했습니다. 이를 통해 후드의 기하학적 구조와 배기 및 흡입 시스템과 관련된 사례 연구를 통해 문제에 대한 가능한 솔루션을 평가합니다. 후드 형상을 변경함으로써 제안된 작동 한계를 준수하면서 실내에서 모든 CO를 제거할 수 있는 것으로 나타났기 때문에 문제에 대해 가장 만족스러운 결과를 얻었습니다.

    Keywords

    carbon monoxide, exhaust system, CFD simulation.

    Figura 3.2 – Geometria simplificada da sala de testes da primeira versão.
    Figura 3.2 – Geometria simplificada da sala de testes da primeira versão.
    Figura 3.4 – Velocidade nos sensores de velocidade para verificar independência de malha para cada refino após 20 s do acionamento da motosserra.
    Figura 3.4 – Velocidade nos sensores de velocidade para verificar independência de malha para cada refino após 20 s do acionamento da motosserra.
    Figura 3.4 – Velocidade nos sensores de velocidade para verificar independência de malha para cada refino após 20 s do acionamento da motosserra.
    Figura 3.4 – Velocidade nos sensores de velocidade para verificar independência de malha para cada refino após 20 s do acionamento da motosserra.
    Figura 3.5 – Vista em detalhe da coifa e os elementos que a compõe.
    Figura 3.5 – Vista em detalhe da coifa e os elementos que a compõe.
    Figura 3.6 – Geometrias das versões simuladas do Teste de Casos.
    Figura 3.6 – Geometrias das versões simuladas do Teste de Casos.
    Figura 4.1 – Concentração de CO medida pelos sensores da simulação.
    Figura 4.1 – Concentração de CO medida pelos sensores da simulação.
    Figura 4.2 – Plano indicando os três cortes realizados na simulação para as superfícies de contorno sendo (1) a altura do escape da máquina, (2) a altura dos detectores de CO e (3) a altura dos exaustores
    Figura 4.2 – Plano indicando os três cortes realizados na simulação para as superfícies de contorno sendo (1) a altura do escape da máquina, (2) a altura dos detectores de CO e (3) a altura dos exaustores
    Figura 4.3 – Superfície de contorno de velocidades a 1,3 m do piso após 20 s de acionamento da motosserra.
    Figura 4.3 – Superfície de contorno de velocidades a 1,3 m do piso após 20 s de acionamento da motosserra.
    Figura 4.4 – Superfície de contorno de velocidades a 1,5 m do piso após 20 s de acionamento da motosserra.
    Figura 4.4 – Superfície de contorno de velocidades a 1,5 m do piso após 20 s de acionamento da motosserra.
    Figura 4.5 – Superfície de contorno de velocidades a 3,9 m do piso após 20 s de acionamento da motosserra.
    Figura 4.5 – Superfície de contorno de velocidades a 3,9 m do piso após 20 s de acionamento da motosserra.
    Figura 4.6 – Superfície de contorno de massas específicas a 1,3 m do piso após 20 s de acionamento da motosserra.
    Figura 4.6 – Superfície de contorno de massas específicas a 1,3 m do piso após 20 s de acionamento da motosserra.
    Figura 4.7 – Superfície de contorno de massas específicas a 1,5 m do piso após 20 s de acionamento da motosserra.
    Figura 4.7 – Superfície de contorno de massas específicas a 1,5 m do piso após 20 s de acionamento da motosserra.
    Figura 4.8 – Superfície de contorno de massas específicas a 3,9 m do piso após 20 s de acionamento da motosserra.
    Figura 4.8 – Superfície de contorno de massas específicas a 3,9 m do piso após 20 s de acionamento da motosserra.
    Figura 4.9 – Volume total de monóxido ao longo do tempo na sala.
    Figura 4.9 – Volume total de monóxido ao longo do tempo na sala.
    Figura 4.9 – Vazão volumétrica de CO ao longo do tempo através da superfície de controle. As linhas contínuas representam curvas de ajuste aos dados simulados.
    Figura 4.9 – Vazão volumétrica de CO ao longo do tempo através da superfície de controle. As linhas contínuas representam curvas de ajuste aos dados simulados.

    REFERENCIAS

    CROWL, Daniel A.; LOUVAR, Joseph F. Chemical process safety: fundamentals with applications. Second Edition, Pearson Education, 2001. BURNETT, J.; CHAN, M. Y. Criteria for air quality in enclosed car parks. Em: Proceedings of the Institution of Civil Engineers-Transport. Thomas Telford-ICE Virtual Library, 1997. Disponível em: < http://www.icevirtuallibrary.com/doi/10.1680/itran.1997.29379> SITTISAK, P.; CHARINPANITKUL T.; CHALERMSINSUWAN, B. Enhancement of carbon monoxide removal in an underground car park using ventilation system with single and twin jet fans. Em: Tunnelling and Underground Space Technology. Volume 97, 2020. VERSTEEG, H.K.; MALALASEKERA, W. Computational Fluid Dynamics: The Finite Volume Method. Second Edition, Pearson Education, 2007. BULIŃSKA, A.; POPIOŁEK, Z.; BULIŃSKI, Z.; Experimentally validated CFD analysis on sampling region determination of average indoor carbon dioxide concentration in occupied space. Em: Building and Environment. Volume 72, 2014. KARIMI, H.; RIAZI, B.; MOHHAMMADI, M. Application of Computational Fluid Dynamics in the Simulation of Carbon Monoxide Distribution, a Case Study: Sayad Underground Tunnel in Tehran. Disponível em: YAKHOT, V.; ORSZAG, S. Renormalization group analysis of turbulence. I. Basic theory. Journal of scientific computing, v. 1, n. 1, p. 3-51, 1986. VAN HOOFF, T.; BLOCKEN, B. CFD evaluation of natural ventilation of indoor environments by the concentration decay method: CO2 gas dispersion from a semi-enclosed stadium. Building and Environment, v. 61, p. 1-17, 2013. Disponível em: < https://www.sciencedirect.com/science/article/pii/S0360132312003216> YANG, L., YE, M., HE, B. CFD simulation research on residential indoor air quality. Em Science of The Total Environment. Volume 472, 2014. Disponível em: < https://www.sciencedirect.com/science/article/pii/S0048969713014228> Flow-3D. Flow-3D User’s Guide. Versão 12, 2020. LAUNDER, B. E. e SPALDING, D. B. The numerical computation of turbulent flows. Em Computer Methods in Applied Mechanics and Engineering, vol. 3, 1974. pp. 269-289 MALISKA, Clovis R. Transferência de Calor e Mecânica dos Fluidos Computacional: fundamentos e coordenadas generalizadas. Segunda Edição. Rio de Janeiro, LTC, 2004. ROACHE, P. J. Perspective: A Method for Uniform Reporting of Grid Refinement Studies, Journal of Fluids Engineering, Vol. 116, 1994; 405-413.

    원문 보기
    Figure 16: Velocity Vectors of Flow at Ghulmet

    댐 붕괴 홍수파 및 범람 매핑 시뮬레이션: A
    아타바드 호수 사례 연구

    Simulation of Dam-Break Flood Wave and Inundation Mapping: A
    Case study of Attabad Lake

    Wasim Karam1, Fayaz A. Khan2, Muhammad Alam3, Sajjad Ali4
    1Lab. Engineer, Department of Civil Engineering, University of Engineering and Technology Mardan, Pakistan,
    wasim10karam@gmail.com
    2Assistant Professor, National Institute of Urban Infrastructure Planning, University of Engineering and Technology Peshawar,
    Pakistan, fayazuet@yahoo.com
    3,4Assistant Professor, Department of Civil Engineering, University of Engineering and Technology Mardan, Pakistan,
    emalam82@gmail.com, sajjadali@uetmardan.edu.pk

    ABSTRACT

    산사태 또는 제방 댐의 파손 연구는 구성이 불확실하고 자연적이며 재해에 대해 적절하게 설계되지 않았기 때문에 다른 자연적 사건에 대한 대응 지식이 부족하기 때문에 더 중요합니다. 이 논문은 댐 ​​파괴의 수력학적 모델링의 다양한 방법을 개선하는 것을 목표로 합니다.

    현재 이 연구에서 Attabad 호수의 댐 붕괴는 전산 유체 역학 기술을 사용하여 시뮬레이션됩니다. 수치 모델(FLOW-3D)은 Reynolds 평균 Navier-Stoke 방정식을 완전히 3D로 풀어서 다양한 단면에서의 피크 유량 깊이, 피크 속도, 피크 방전, 피크 깊이까지의 시간 및 피크 방전까지의 시간을 예측하기 위해 개발되었습니다.

    표준 RNG 난류 모델을 사용하여 난류를 시뮬레이션한 다음 마을의 흐름에 대한 홍수 범람 지도와 속도 벡터를 그립니다. 결과는 Hunza 강의 수로를 통해 모델링된 홍수파의 대부분이 Hunza 강의 범람원에 포함되지만 Hunza 강의 범람원 내부에 위치한 Miaun 및 chalat와 같은 일부 마을의 경우 더 높은 위험에 있음을 보여줍니다.

    그러나 이들 마을의 예상 홍수 도달 시간은 각각 31분과 44분으로 인구를 안전한 지역으로 대피시키기에 충분한 시간인 반면, 알리 아바드에 인접한 하산 아바드와 같은 일부 마을의 경우 침수 위험이 더 높은 반면 마을의 예상 홍수 도착 시간은 12분으로 인구 대피에 충분하지 않으므로 홍수 억제를 위한 추가 홍수 보호 구조가 필요합니다.

    최고속도의 추정치는 하천평야의 더 높은 전단응력, 심한 침식의 위험, 농경지 피해, 주거지 및 형태학적 변화가 예상됨을 의미한다. 댐 파손 분석(예: 최고 깊이, 최고 속도, 홍수 도달 시간 및 홍수 범람 지도)은 향후 위험 분석 및 홍수 관리의 지침으로만 사용해야 합니다.

    Figure 2: Case Study Location on Map of Pakistan
    Figure 2: Case Study Location on Map of Pakistan
    Figure 3: Lake Condition 3 months after Landslide
    Figure 3: Lake Condition 3 months after Landslide
    Figure 5: 3D Model from the Merged DEM
    Figure 5: 3D Model from the Merged DEM
    Figure 7: Free Surface Elevation relative to local origin
    Figure 7: Free Surface Elevation relative to local origin
    Figure 8: Model of lake referenced over Google Earth Image
    Figure 8: Model of lake referenced over Google Earth Image
    Figure 9: Meshing in the 3D Terrain Model
    Figure 9: Meshing in the 3D Terrain Model
    Figure 10: Flow Depth Hydrographs of the downstream villages (A) Karim Abad (B) Ghulmet (C) Thol (D) Chalat (E) Nomal
    Figure 10: Flow Depth Hydrographs of the downstream villages (A) Karim Abad (B) Ghulmet (C) Thol (D) Chalat (E) Nomal
    Figure 11: Flow Hydrograph at Karim Abad and Nomal Bridge
    Figure 11: Flow Hydrograph at Karim Abad and Nomal Bridge
    Figure 12: Flood Inundation Map of Karim Abad
    Figure 12: Flood Inundation Map of Karim Abad
    Figure 13: Flood Inundation Map of Ghulmet
    Figure 13: Flood Inundation Map of Ghulmet
    Figure 14: Flood Inundation Map of Chalat
    Figure 14: Flood Inundation Map of Chalat
    Figure 15: Velocity Vectors of flow at Karim Abad
    Figure 15: Velocity Vectors of flow at Karim Abad
    Figure 16: Velocity Vectors of Flow at Ghulmet
    Figure 16: Velocity Vectors of Flow at Ghulmet
    Figure 17: Velocity Vectors of Flow at Chalat
    Figure 17: Velocity Vectors of Flow at Chalat

    REFERENCES

    [1]. Zhang, L. & Peng, M. & Chang, D.S. & Xu, Y. (2015).
    Dam Failure Mechanisms and Risk Assessment, First
    Ed. John Wiley and Sons, Singapore 473 pp.
    10.1002/9781118558522.
    [2]. T. L. Wahl, “Dam Breach Modeling – an Overview of
    Analysis Methods,” 2nd Jt. Fed. Interagency Conf. Las
    Vegas, NV, pp. 1–12, 2010.
    [3]. Khosravi K. “Dam Break Analysis and Flood
    Inundation Mapping : The Case Study of Sefid-Rud
    Dam,” no. August 2019. DOI:
    10.1016/B978-0-12-815998-9.00031-2
    [4]. Robb, D. M., & Vasquez, J. A. (2015). Numerical
    simulation of dam-break flows using depth-averaged
    hydrodynamic and three-dimensional CFD models.
    22nd Canadian Hydrotechnical Conference, (June).
    [5]. Mohammad Rostami, M. S. (2015). Human Life Saving
    by Simulation of Dam Break using Flow-3D. Trend in
    Life Sciences, 4(3), 308–316
    [6]. Gharbi, M., Soualmia, A., Dartus, D., & Masbernat, L.
    (2016). Comparison of 1D and 2D hydraulic models
    for floods simulation on the Medjerda River in
    Tunisia. Journal of Materials and Environmental
    Science, 7(8), 3017–3026. https://doi.org/10.1080/153
    [7]. Andrei, A., Robert, B., & Erika, B. (2017). Numerical
    Limitations of 1D Hydraulic Models Using MIKE11
    or HEC-RAS software – A case study of Baraolt
    River, Romania. IOP Conference Series: Materials
    Science and Engineering, 245(7).
    https://doi.org/10.1088/1757-899X/245/7/072010
    [8]. Henderson, F.M. (1966). Open Channel Flow. MacMillan
    Company, New York, USA, P. No 304-313
    [9]. Betsholtz, A., & Nordlöf, B. (2017). Potentials and
    limitations of 1D, 2D and coupled 1D-2D flood
    modeling in HEC-RAS. Lund University, 128.
    https://doi.org/10.1016/S0300-9440(03)00139-5
    [10].Ozmen-Cagatay, H., & Kocaman, S. (2011). Dam-break
    flow in the presence of obstacle: Experiment and CFD
    simulation. Engineering Applications of Computational
    Fluid Mechanics, 5(4), 541–552.
    https://doi.org/10.1080/19942060.2011.11015393
    [11].Toombes, L., & Chanson, H. (2011). Numerical
    Limitations of Hydraulic Models. 10th Hydraulics
    Conference, (July), 2322–2329.
    https://doi.org/10.1016/j.jalz.2016.06.1613
    [12].Zarein, M. (2015). Modeling Dam-Break Flows Using
    a 3d Mike 3 Flow Model, (January).
    [13].George, A. C., & Nair, B. T. (2015). Dam Break
    Analysis Using BOSS DAMBRK. Aquatic Procedia,
    4(Icwrcoe), 853–860.
    https://doi.org/10.1016/j.aqpro.2015.02.10
    [14].S. Roga and K. M. Pandey, “Computational Analysis of
    Supersonic Flow Regime Using Ramp Injector with
    Standard K- ω Turbulence Model” .World Academy of
    research in Science and Engineering, vol. 2, no. 1, pp.
    31–40, 2013.http:// doi.org/10.1.1.348.5862.

    원문 PDF 보기
    Fig. 5. The predicted shapes of initial breach (a) Rectangular (b) V-notch. Fig. 6. Dam breaching stages.

    Investigating the peak outflow through a spatial embankment dam breach

    공간적 제방댐 붕괴를 통한 최대 유출량 조사

    Mahmoud T.GhonimMagdy H.MowafyMohamed N.SalemAshrafJatwaryFaculty of Engineering, Zagazig University, Zagazig 44519, Egypt

    Abstract

    Investigating the breach outflow hydrograph is an essential task to conduct mitigation plans and flood warnings. In the present study, the spatial dam breach is simulated by using a three-dimensional computational fluid dynamics model, FLOW-3D. The model parameters were adjusted by making a comparison with a previous experimental model. The different parameters (initial breach shape, dimensions, location, and dam slopes) are studied to investigate their effects on dam breaching. The results indicate that these parameters have a significant impact. The maximum erosion rate and peak outflow for the rectangular shape are higher than those for the V-notch by 8.85% and 5%, respectively. Increasing breach width or decreasing depth by 5% leads to increasing maximum erosion rate by 11% and 15%, respectively. Increasing the downstream slope angle by 4° leads to an increase in both peak outflow and maximum erosion rate by 2.0% and 6.0%, respectively.

    유출 유출 수문곡선을 조사하는 것은 완화 계획 및 홍수 경보를 수행하는 데 필수적인 작업입니다. 본 연구에서는 3차원 전산유체역학 모델인 FLOW-3D를 사용하여 공간 댐 붕괴를 시뮬레이션합니다. 이전 실험 모델과 비교하여 모델 매개변수를 조정했습니다.

    다양한 매개변수(초기 붕괴 형태, 치수, 위치 및 댐 경사)가 댐 붕괴에 미치는 영향을 조사하기 위해 연구됩니다. 결과는 이러한 매개변수가 상당한 영향을 미친다는 것을 나타냅니다. 직사각형 형태의 최대 침식율과 최대 유출량은 V-notch보다 각각 8.85%, 5% 높게 나타났습니다.

    위반 폭을 늘리거나 깊이를 5% 줄이면 최대 침식률이 각각 11% 및 15% 증가합니다. 하류 경사각을 4° 증가시키면 최대 유출량과 최대 침식률이 각각 2.0% 및 6.0% 증가합니다.

    Keywords

    Spatial dam breach; FLOW-3D; Overtopping erosion; Computational fluid dynamics (CFD)

    1. Introduction

    There are many purposes for dam construction, such as protection from flood disasters, water storage, and power generation. Embankment failures may have a catastrophic impact on lives and infrastructure in the downstream regions. One of the most common causes of embankment dam failure is overtopping. Once the overtopping of the dam begins, the breach formation will start in the dam body then end with the dam failure. This failure occurs within a very short time, which threatens to be very dangerous. Therefore, understanding and modeling the embankment breaching processes is essential for conducting mitigation plans, flood warnings, and forecasting flood damage.

    The analysis of the dam breaching process is implemented by different techniques: comparative methods, empirical models with dimensional and dimensionless solutions, physical-based models, and parametric models. These models were described in detail [1]. Parametric modeling is commonly used to simulate breach growth as a time-dependent linear process and calculate outflow discharge from the breach using hydraulics principles [2]. Alhasan et al. [3] presented a simple one-dimensional mathematical model and a computer code to simulate the dam breaching process. These models were validated by small dams breaching during the floods in 2002 in the Czech Republic. Fread [4] developed an erosion model (BREACH) based on hydraulics principles, sediment transport, and soil mechanics to estimate breach size, time of formation, and outflow discharge. Říha et al. [5] investigated the dam break process for a cascade of small dams using a simple parametric model for piping and overtopping erosion, as well as a 2D shallow-water flow model for the flood in downstream areas. Goodarzi et al. [6] implemented mathematical and statistical methods to assess the effect of inflows and wind speeds on the dam’s overtopping failure.

    Dam breaching studies can be divided into two main modes of erosion. The first mode is called “planar dam breach” where the flow overtops the whole dam width. While the second mode is called “spatial dam breach” where the flow overtops through the initial pilot channel (i.e., a channel created in the dam body). Therefore, the erosion will be in both vertical and horizontal directions [7].

    The erosion process through the embankment dams occurs due to the shear stress applied by water flows. The dam breaching evolution can be divided into three stages [8], [9], but Y. Yang et al. [10] divided the breach development into five stages: Stage I, the seepage erosion; Stage II, the initial breach formation; Stage III, the head erosion; Stage IV, the breach expansion; and Stage V, the re-equilibrium of the river channel through the breach. Many experimental tests have been carried out on non-cohesive embankment dams with an initial breach to examine the effect of upstream inflow discharges on the longitudinal profile evolution and the time to inflection point [11].

    Zhang et al. [12] studied the effect of changing downstream slope angle, sediment grain size, and dam crest length on erosion rates. They noticed that increasing dam crest length and decreasing downstream slope angle lead to decreasing sediment transport rate. While the increase in sediment grain size leads to an increased sediment transport rate at the initial stages. Höeg et al. [13] presented a series of field tests to investigate the stability of embankment dams made of various materials. Overtopping and piping were among the failure tests carried out for the dams composed of homogeneous rock-fill, clay, or gravel with a height of up to 6.0 m. Hakimzadeh et al. [14] constructed 40 homogeneous cohesive and non-cohesive embankment dams to study the effect of changing sediment diameter and dam height on the breaching process. They also used genetic programming (GP) to estimate the breach outflow. Refaiy et al. [15] studied different scenarios for the downstream drain geometry, such as length, height, and angle, to minimize the effect of piping phenomena and therefore increase dam safety.

    Zhu et al. [16] examined the effect of headcut erosion on dam breach growth, especially in the case of cohesive dams. They found that the breach growth in non-cohesive embankments is slower than cohesive embankments due to the little effect of headcut. Schmocker and Hager [7] proposed a relationship for estimating peak outflow from the dam breach process.(1)QpQin-1=1.7exp-20hc23d5013H0

    where: Qp = peak outflow discharge.

    Qin = inflow discharge.

    hc = critical flow depth.

    d50 = mean sediment diameter.

    Ho = initial dam height.

    Yu et al. [17] carried out an experimental study for homogeneous non-cohesive embankment dams in a 180° bending rectangular flume to determine the effect of overtopping flows on breaching formation. They found that the main factors influencing breach formation are water level, river discharge, and embankment material diameter.

    Wu et al. [18] carried out a series of experiments to investigate the effect of breaching geometry on both non-cohesive and cohesive embankment dams in a U-bend flume due to overtopping flows. In the case of non-cohesive embankments, the non-symmetrical lateral expansion was noticed during the breach formation. This expansion was described by a coefficient ranging from 2.7 to 3.3.

    The numerical models of the dam breach can be categorized according to different parameters, such as flow dimensions (1D, 2D, or 3D), flow governing equations, and solution methods. The 1D models are mainly used to predict the outflow hydrograph from the dam breach. Saberi et al. [19] applied the 1D Saint-Venant equation, which is solved by the finite difference method to investigate the outflow hydrograph during dam overtopping failure. Because of the ability to study dam profile evolution and breach formation, 2D models are more applicable than 1D models. Guan et al. [20] and Wu et al. [21] employed both 2D shallow water equations (SWEs) and sediment erosion equations, which are solved by the finite volume method to study the effect of the dam’s geometry parameters on outflow hydrograph and dam profile evolution. Wang et al. [22] also proposed a second-order hybrid-type of total variation diminishing (TVD) finite-difference to estimate the breach outflow by solving the 2D (SWEs). The accuracy of (SWEs) for both vertical flow contraction and surface roughness has been assessed [23]. They noted that the accuracy of (SWEs) is acceptable for milder slopes, but in the case of steeper slopes, modelers should be more careful. Generally, the accuracy of 2D models is still low, especially with velocity distribution over the flow depth, lateral momentum exchange, density-driven flows, and bottom friction [24]. Therefore, 3D models are preferred. Larocque et al. [25] and Yang et al. [26] started to use three-dimensional (3D) models that depend on the Reynolds-averaged Navier-Stokes (RANS) equations.

    Previous experimental studies concluded that there is no clear relationship between the peak outflow from the dam breach and the initial breach characteristics. Some of these studies depend on the sharp-crested weir fixed at the end of the flume to determine the peak outflow from the breach, which leads to a decrease in the accuracy of outflow calculations at the microscale. The main goals of this study are to carry out a numerical simulation for a spatial dam breach due to overtopping flows by using (FLOW-3D) software to find an empirical equation for the peak outflow discharge from the breach and determine the worst-case that leads to accelerating the dam breaching process.

    2. Numerical simulation

    The current study for spatial dam breach is simulated by using (FLOW-3D) software [27], which is a powerful computational fluid dynamics (CFD) program.

    2.1. Geometric presentations

    A stereolithographic (STL) file is prepared for each change in the initial breach geometry and dimensions. The CAD program is useful for creating solid objects and converting them to STL format, as shown in Fig. 1.

    2.2. Governing equations

    The governing equations for water flow are three-dimensional Reynolds Averaged Navier-Stokes equations (RANS).

    The continuity equation:(2)∂ui∂xi=0

    The momentum equation:(3)∂ui∂t+1VFuj∂ui∂xj=1ρ∂∂xj-pδij+ν∂ui∂xj+∂uj∂xi-ρu`iu`j¯

    where u is time-averaged velocity,ν is kinematic viscosity, VF is fractional volume open to flow, p is averaged pressure and -u`iu`j¯ are components of Reynold’s stress. The Volume of Fluid (VOF) technique is used to simulate the free surface profile. Hirt et al. [28] presented the VOF algorithm, which employs the function (F) to express the occupancy of each grid cell with fluid. The value of (F) varies from zero to unity. Zero value refers to no fluid in the grid cell, while the unity value refers to the grid cell being fully occupied with fluid. The free surface is formed in the grid cells having (F) values between zero and unity.(4)∂F∂t+1VF∂∂xFAxu+∂∂yFAyv+∂∂zFAzw=0

    where (u, v, w) are the velocity components in (x, y, z) coordinates, respectively, and (Ax, Ay, Az) are the area fractions.

    2.3. Boundary and initial conditions

    To improve the accuracy of the results, the boundary conditions should be carefully determined. In this study, two mesh blocks are used to minimize the time consumed in the simulation. The boundary conditions for mesh block 1 are as follows: The inlet and sides boundaries are defined as a wall boundary condition (wall boundary condition is usually used for bound fluid by solid regions. In the case of viscous flows, no-slip means that the tangential velocity is equal to the wall velocity and the normal velocity is zero), the outlet is defined as a symmetry boundary condition (symmetry boundary condition is usually used to reduce computational effort during CFD simulation. This condition allows the flow to be transferred from one mesh block to another. No inputs are required for this boundary condition except that its location should be defined accurately), the bottom boundary is defined as a uniform flow rate boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure. The boundary conditions for mesh block 2 are as follows: The inlet is defined as a symmetry boundary condition, the outlet is defined as a free flow boundary condition, the bottom and sides boundaries are defined as a wall boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure as shown in Fig. 2. The initial conditions required to be set for the fluid (i.e., water) inside of the domain include configuration, temperature, velocities, and pressure distribution. The configuration of water depends on the dimensions and shape of the dam reservoir. While the other conditions have been assigned as follows: temperature is normal water temperature (25 °c) and pressure distribution is hydrostatic with no initial velocity.

    2.4. Numerical method

    FLOW-3D uses the finite volume method (FVM) to solve the governing equation (Reynolds-averaged Navier-Stokes) over the computational domain. A finite-volume method is an Eulerian approach for representing and evaluating partial differential equations in algebraic equations form [29]. At discrete points on the mesh geometry, values are determined. Finite volume expresses a small volume surrounding each node point on a mesh. In this method, the divergence theorem is used to convert volume integrals with a divergence term to surface integrals. After that, these terms are evaluated as fluxes at each finite volume’s surfaces.

    2.5. Turbulent models

    Turbulence is the chaotic, unstable motion of fluids that occurs when there are insufficient stabilizing viscous forces. In FLOW-3D, there are six turbulence models available: the Prandtl mixing length model, the one-equation turbulent energy model, the two-equation (k – ε) model, the Renormalization-Group (RNG) model, the two-equation (k – ω) models, and a large eddy simulation (LES) model. For simulating flow motion, the RNG model is adopted to simulate the motion behavior better than the k – ε and k – ω.

    models [30]. The RNG model consists of two main equations for the turbulent kinetic energy KT and its dissipation.εT(5)∂kT∂t+1VFuAx∂kT∂x+vAy∂kT∂y+wAz∂kT∂z=PT+GT+DiffKT-εT(6)∂εT∂t+1VFuAx∂εT∂x+vAy∂εT∂y+wAz∂εT∂z=C1.εTKTPT+c3.GT+Diffε-c2εT2kT

    where KT is the turbulent kinetic energy, PT is the turbulent kinetic energy production, GT is the buoyancy turbulence energy, εT is the turbulent energy dissipation rate, DiffKT and Diffε are terms of diffusion, c1, c2 and c3 are dimensionless parameters, in which c1 and c3 have a constant value of 1.42 and 0.2, respectively, c2 is computed from the turbulent kinetic energy (KT) and turbulent production (PT) terms.

    2.6. Sediment scour model

    The sediment scour model available in FLOW-3D can calculate all the sediment transport processes including Entrainment transport, Bedload transport, Suspended transport, and Deposition. The erosion process starts once the water flows remove the grains from the packed bed and carry them into suspension. It happens when the applied shear stress by water flows exceeds critical shear stress. This process is represented by entrainment transport in the numerical model. After entrained, the grains carried by water flow are represented by suspended load transport. After that, some suspended grains resort to settling because of the combined effect of gravity, buoyancy, and friction. This process is described through a deposition. Finally, the grains sliding motions are represented by bedload transport in the model. For the entrainment process, the shear stress applied by the fluid motion on the packed bed surface is calculated using the standard wall function as shown in Eq.7.(7)ks,i=Cs,i∗d50

    where ks,i is the Nikuradse roughness and Cs,i is a user-defined coefficient. The critical bed shear stress is defined by a dimensionless parameter called the critical shields number as expressed in Eq.8.(8)θcr,i=τcr,i‖g‖diρi-ρf

    where θcr,i is the critical shields number, τcr,i is the critical bed shear stress, g is the absolute value of gravity acceleration, di is the diameter of the sediment grain, ρi is the density of the sediment species (i) and ρf is the density of the fluid. The value of the critical shields number is determined according to the Soulsby-Whitehouse equation.(9)θcr,i=0.31+1.2d∗,i+0.0551-exp-0.02d∗,i

    where d∗,i is the dimensionless diameter of the sediment, given by Eq.10.(10)d∗,i=diρfρi-ρf‖g‖μf213

    where μf is the fluid dynamic viscosity. For the sloping bed interface, the value of the critical shields number is modified according to Eq.11.(11)θ`cr,i=θcr,icosψsinβ+cos2βtan2φi-sin2ψsin2βtanφi

    where θ`cr,i is the modified critical shields number, φi is the angle of repose for the sediment, β is the angle of bed slope and ψ is the angle between the flow and the upslope direction. The effects of the rolling, hopping, and sliding motions of grains along the packed bed surface are taken by the bedload transport process. The volumetric bedload transport rate (qb,i) per width of the bed is expressed in Eq.12.(12)qb,i=Φi‖g‖ρi-ρfρfdi312

    where Φi is the dimensionless bedload transport rate is calculated by using Meyer Peter and Müller equation.(13)Φi=βMPM,iθi-θ`cr,i1.5cb,i

    where βMPM,i is the Meyer Peter and Müller user-defined coefficient and cb,i is the volume fraction of species i in the bed material. The suspended load transport is calculated as shown in Eq.14.(14)∂Cs,i∂t+∇∙Cs,ius,i=∇∙∇DCs,i

    where Cs,i is the suspended sediment mass concentration, D is the diffusivity, and us,i is the grain velocity of species i. Entrainment and deposition are two opposing processes that take place at the same time. The lifting and settling velocities for both entrainment and deposition processes are calculated according to Eq.15 and Eq.16, respectively.(15)ulifting,i=αid∗,i0.3θi-θ`cr,igdiρiρf-1(16)usettling,i=υfdi10.362+1.049d∗,i3-10.36

    where αi is the entrainment coefficient of species i and υf is the kinematic viscosity of the fluid.

    2.7. Grid type

    Using simple rectangular orthogonal elements in planes and hexahedral in volumes in the (FLOW-3D) program makes the mesh generation process easier, decreases the required memory, and improves numerical accuracy. Two mesh blocks were used in a joined form with a size ratio of 2:1. The first mesh block is coarser, which contains the reservoir water, and the second mesh block is finer, which contains the dam. For achieving accuracy and efficiency in results, the mesh size is determined by using a grid convergence test. The optimum uniform cell size for the first mesh block is 0.012 m and for the second mesh block is 0.006 m.

    2.8. Time step

    The maximum time step size is determined by using a Courant number, which controls the distance that the flow will travel during the simulation time step. In this study, the Courant number was taken equal to 0.25 to prevent the flow from traveling through more than one cell in the time step. Based on the Courant number, a maximum time step value of 0.00075 s was determined.

    2.9. Numerical model validation

    The numerical model accuracy was achieved by comparing the numerical model results with previous experimental results. The experimental study of Schmocker and Hager [7] was based on 31 tests with changes in six parameters (d50, Ho, Bo, Lk, XD, and Qin). All experimental tests were conducted in a straight open glass-sided flume. The horizontal flume has a rectangular cross-section with a width of 0.4 m and a height of 0.7 m. The flume was provided with a flow straightener and an intake with a length of 0.66 m. All tested dams were inserted at various distances (XD) from the intake. Test No.1 from this experimental program was chosen to validate the numerical model. The different parameters used in test No.1 are as follows:

    (1) uniform sediment with a mean diameter (d50 = 0.31 mm), (2) Ho = 0.2 m, (3) Bo = 0.2 m, (4) Lk = 0.1 m,

    (5) XD = 1.0 m, (6) Qin = 6.0 lit/s, (7) Su and Sd = 2:1, (8) mass density (ρs = 2650 kg/m3) (9) Homogenous and non-cohesive embankment dam. As shown in Fig. 2, the simulation is contained within a rectangular grid with dimensions: 3.56 m in the x-direction (where 0.66 m is used as inlet, 0.9 m as dam base width, and 1.0 m as outlet), in y-direction 0.2 m (dam length), and in the z-direction 0.3 m, which represents the dam height (0.2 m) with a free distance (0.1 m) above the dam. There are two main reasons that this experimental program is preferred for the validation process. The first reason is that this program deals with homogenous, non-cohesive soil, which is available in FLOW-3D. The second reason is that this program deals with small-scale models which saves time for numerical simulation. Finally, some important assumptions were considered during the validation process. The flow is assumed to be incompressible, viscous, turbulent, and three-dimensional.

    By comparing dam profiles at different time instants for the experimental test with the current numerical model, it appears that the numerical model gives good agreement as shown in Fig. 3 and Fig. 4, with an average error percentage of 9% between the experimental results and the numerical model.

    3. Analysis and discussions

    The current model is used to study the effects of different parameters such as (initial breach shapes, dimensions, locations, upstream and downstream dam slopes) on the peak outflow discharge, QP, time of peak outflow, tP, and rate of erosion, E.

    This study consists of a group of scenarios. The first scenario is changing the shapes of the initial breach according to Singh [1], the most predicted shapes are rectangular and V-notch as shown in Fig. 5. The second scenario is changing the initial breach dimensions (i.e., width and depth). While the third scenario is changing the location of the initial breach. Eventually, the last scenario is changing the upstream and downstream dam slopes.

    All scenarios of this study were carried out under the same conditions such as inflow discharge value (Qin=1.0lit/s), dimensions of the tested dam, where dam height (Ho=0.20m), crest width.

    (Lk=0.1m), dam length (Bo=0.20m), and homogenous & non-cohesive soil with a mean diameter (d50=0.31mm).

    3.1. Dam breaching process evolution

    The dam breaching process is a very complex process due to the quick changes in hydrodynamic conditions during dam failure. The dam breaching process starts once water flows reach the downstream face of the dam. During the initial stage of dam breaching, the erosion process is relatively quiet due to low velocities of flow. As water flows continuously, erosion rates increase, especially in two main zones: the crest and the downstream face. As soon as the dam crest is totally eroded, the water levels in the dam reservoir decrease rapidly, accompanied by excessive erosion in the dam body. The erosion process continues until the water levels in the dam reservoir equal the remaining height of the dam.

    According to Zhou et al. [11], the breaching process consists of three main stages. The first stage starts with beginning overtopping flow, then ends when the erosion point directed upstream and reached the inflection point at the inflection time (ti). The second stage starts from the end of the stage1 until the occurrence of peak outflow discharge at the peak outflow time (tP). The third stage starts from the end of the stage2 until the value of outflow discharge becomes the same as the value of inflow discharge at the final time (tf). The outflow discharge from the dam breach increases rapidly during stage1 and stage2 because of the large dam storage capacity (i.e., the dam reservoir is totally full of water) and excessive erosion. While at stage3, the outflow values start to decrease slowly because most of the dam’s storage capacity was run out. The end of stage3 indicates that the dam storage capacity was totally run out, so the outflow equalized with the inflow discharge as shown in Fig. 6 and Fig. 7.

    3.2. The effect of initial breach shape

    To identify the effect of the initial breach shape on the evolution of the dam breaching process. Three tests were carried out with different cross-section areas for each shape. The initial breach is created at the center of the dam crest. Each test had an ID to make the process of arranging data easier. The rectangular shape had an ID (Rec5h & 5b), which means that its depth and width are equal to 5% of the dam height, and the V-notch shape had an ID (V-noch5h & 1:1) which means that its depth is equal to 5% of the dam height and its side slope is equal to 1:1. The comparison between rectangular and V-notch shapes is done by calculating the ratio between maximum dam height at different times (ZMax) to the initial dam height (Ho), rate of erosion, and hydrograph of outflow discharge for each test. The rectangular shape achieves maximum erosion rate and minimum inflection time, in addition to a rapid decrease in the dam reservoir levels. Therefore, the dam breaching is faster in the case of a rectangular shape than in a V-notch shape, which has the same cross-section area as shown in Fig. 8.

    Also, by comparing the hydrograph for each test, the peak outflow discharge value in the case of a rectangular shape is higher than the V-notch shape by 5% and the time of peak outflow for the rectangular shape is shorter than the V-notch shape by 9% as shown in Fig. 9.

    3.3. The effect of initial breach dimensions

    The results of the comparison between the different initial breach shapes indicate that the worst initial breach shape is rectangular, so the second scenario from this study concentrated on studying the effect of a change in the initial rectangular breach dimensions. Groups of tests were carried out with different depths and widths for the rectangular initial breach. The first group had a depth of 5% from the dam height and with three different widths of 5,10, and 15% from the dam height, the second group had a depth of 10% with three different widths of 5,10, and 15%, the third group had a depth of 15% with three different widths of 5,10, and 15% and the final group had a width of 15% with three different heights of 5, 10, and 15% for a rectangular breach shape. The comparison was made as in the previous section to determine the worst case that leads to the quick dam failure as shown in Fig. 10.

    The results show that the (Rec 5 h&15b) test achieves a maximum erosion rate for a shorter period of time and a minimum ratio for (Zmax / Ho) as shown in Fig. 10, which leads to accelerating the dam failure process. The dam breaching process is faster with the minimum initial breach depth and maximum initial breach width. In the case of a minimum initial breach depth, the retained head of water in the dam reservoir is high and the crest width at the bottom of the initial breach (L`K) is small, so the erosion point reaches the inflection point rapidly. While in the case of the maximum initial breach width, the erosion perimeter is large.

    3.4. The effect of initial breach location

    The results of the comparison between the different initial rectangular breach dimensions indicate that the worst initial breach dimension is (Rec 5 h&15b), so the third scenario from this study concentrated on studying the effect of a change in the initial breach location. Three locations were checked to determine the worst case for the dam failure process. The first location is at the center of the dam crest, which was named “Center”, the second location is at mid-distance between the dam center and dam edge, which was named “Mid”, and the third location is at the dam edge, which was named “Edge” as shown in Fig. 11. According to this scenario, the results indicate that the time of peak outflow discharge (tP) is the same in the three cases, but the maximum value of the peak outflow discharge occurs at the center location. The difference in the peak outflow values between the three cases is relatively small as shown in Fig. 12.

    The rates of erosion were also studied for the three cases. The results show that the maximum erosion rate occurs at the center location as shown in Fig. 13. By making a comparison between the three cases for the dam storage volume. The results show that the center location had the minimum values for the dam storage volume, which means that a large amount of water has passed to the downstream area as shown in Fig. 14. According to these results, the center location leads to increased erosion rate and accelerated dam failure process compared with the two other cases. Because the erosion occurs on both sides, but in the case of edge location, the erosion occurs on one side.

    3.5. The effect of upstream and downstream dam slopes

    The results of the comparison between the different initial rectangular breach locations indicate that the worst initial breach location is the center location, so the fourth scenario from this study concentrated on studying the effect of a change in the upstream (Su) and downstream (Sd) dam slopes. Three slopes were checked individually for both upstream and downstream slopes to determine the worst case for the dam failure process. The first slope value is (2H:1V), the second slope value is (2.5H:1V), and the third slope value is (3H:1V). According to this scenario, the results show that the decreasing downstream slope angle leads to increasing time of peak outflow discharge (tP) and decreasing value of peak outflow discharge. The difference in the peak outflow values between the three cases for the downstream slope is 2%, as shown in Fig. 15, but changing the upstream slope has a negligible impact on the peak outflow discharge and its time as shown in Fig. 16.

    The rates of erosion were also studied in the three cases for both upstream and downstream slopes. The results show that the maximum erosion rate increases by 6.0% with an increasing downstream slope angle by 4°, as shown in Fig. 17. The results also indicate that the erosion rates aren’t affected by increasing or decreasing the upstream slope angle, as shown in Fig. 18. According to these results, increasing the downstream slope angle leads to increased erosion rate and accelerated dam failure process compared with the upstream slope angle. Because of increasing shear stress applied by water flows in case of increasing downstream slope.

    According to all previous scenarios, the dimensionless peak outflow discharge QPQin is presented for a fixed dam height (Ho) and inflow discharge (Qin). Fig. 19 illustrates the relationship between QP∗=QPQin and.

    Lr=ho2/3∗bo2/3Ho. The deduced relationship achieves R2=0.96.(17)QP∗=2.2807exp-2.804∗Lr

    4. Conclusions

    A spatial dam breaching process was simulated by using FLOW-3D Software. The validation process was performed by making a comparison between the simulated results of dam profiles and the dam profiles obtained by Schmocker and Hager [7] in their experimental study. And also, the peak outflow value recorded an error percentage of 12% between the numerical model and the experimental study. This model was used to study the effect of initial breach shape, dimensions, location, and dam slopes on peak outflow discharge, time of peak outflow, and the erosion process. By using the parameters obtained from the validation process, the results of this study can be summarized in eight points as follows.1.

    The rectangular initial breach shape leads to an accelerating dam failure process compared with the V-notch.2.

    The value of peak outflow discharge in the case of a rectangular initial breach is higher than the V-notch shape by 5%.3.

    The time of peak outflow discharge for a rectangular initial breach is shorter than the V-notch shape by 9%.4.

    The minimum depth and maximum width for the initial breach achieve maximum erosion rates (increasing breach width, b0, or decreasing breach depth, h0, by 5% from the dam height leads to an increase in the maximum rate of erosion by 11% and 15%, respectively), so the dam failure is rapid.5.

    The center location of the initial breach leads to an accelerating dam failure compared with the edge location.6.

    The initial breach location has a negligible effect on the peak outflow discharge value and its time.7.

    Increasing the downstream slope angle by 4° leads to an increase in both peak outflow discharge and maximum rate of erosion by 2.0% and 6.0%, respectively.8.

    The upstream slope has a negligible effect on the dam breaching process.

    References

    [1]V. SinghDam breach modeling technologySpringer Science & Business Media (1996)Google Scholar[2]Wahl TL. Prediction of embankment dam breach parameters: a literature review and needs assessment. 1998.Google Scholar[3]Z. Alhasan, J. Jandora, J. ŘíhaStudy of dam-break due to overtopping of four small dams in the Czech RepublicActa Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 63 (3) (2015), pp. 717-729 View PDFCrossRefView Record in ScopusGoogle Scholar[4]D. FreadBREACH, an erosion model for earthen dam failures: Hydrologic Research LaboratoryNOAA, National Weather Service (1988)Google Scholar[5]J. Říha, S. Kotaška, L. PetrulaDam Break Modeling in a Cascade of Small Earthen Dams: Case Study of the Čižina River in the Czech RepublicWater, 12 (8) (2020), p. 2309, 10.3390/w12082309 View PDFView Record in ScopusGoogle Scholar[6]E. Goodarzi, L. Teang Shui, M. ZiaeiDam overtopping risk using probabilistic concepts–Case study: The Meijaran DamIran Ain Shams Eng J, 4 (2) (2013), pp. 185-197ArticleDownload PDFView Record in ScopusGoogle Scholar[7]L. Schmocker, W.H. HagerPlane dike-breach due to overtopping: effects of sediment, dike height and dischargeJ Hydraul Res, 50 (6) (2012), pp. 576-586 View PDFCrossRefView Record in ScopusGoogle Scholar[8]J.S. Walder, R.M. Iverson, J.W. Godt, M. Logan, S.A. SolovitzControls on the breach geometry and flood hydrograph during overtopping of noncohesive earthen damsWater Resour Res, 51 (8) (2015), pp. 6701-6724View Record in ScopusGoogle Scholar[9]H. Wei, M. Yu, D. Wang, Y. LiOvertopping breaching of river levees constructed with cohesive sedimentsNat Hazards Earth Syst Sci, 16 (7) (2016), pp. 1541-1551 View PDFCrossRefView Record in ScopusGoogle Scholar[10]Y. Yang, S.-Y. Cao, K.-J. Yang, W.-P. LiYang K-j, Li W-p. Experimental study of breach process of landslide dams by overtopping and its initiation mechanismsJ Hydrodynamics, 27 (6) (2015), pp. 872-883ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[11]G.G.D. Zhou, M. Zhou, M.S. Shrestha, D. Song, C.E. Choi, K.F.E. Cui, et al.Experimental investigation on the longitudinal evolution of landslide dam breaching and outburst floodsGeomorphology, 334 (2019), pp. 29-43ArticleDownload PDFView Record in ScopusGoogle Scholar[12]J. Zhang, Z.-x. Guo, S.-y. CaoYang F-g. Experimental study on scour and erosion of blocked damWater Sci Eng, 5 (2012), pp. 219-229ArticleDownload PDFView Record in ScopusGoogle Scholar[13]K. Höeg, A. Løvoll, K. VaskinnStability and breaching of embankment dams: Field tests on 6 m high damsInt J Hydropower Dams, 11 (2004), pp. 88-92View Record in ScopusGoogle Scholar[14]H. Hakimzadeh, V. Nourani, A.B. AminiGenetic programming simulation of dam breach hydrograph and peak outflow dischargeJ Hydrol Eng, 19 (4) (2014), pp. 757-768View Record in ScopusGoogle Scholar[15]A.R. Refaiy, N.M. AboulAtta, N.Y. Saad, D.A. El-MollaModeling the effect of downstream drain geometry on seepage through earth damsAin Shams Eng J, 12 (3) (2021), pp. 2511-2531ArticleDownload PDFView Record in ScopusGoogle Scholar[16]Y. Zhu, P.J. Visser, J.K. Vrijling, G. WangExperimental investigation on breaching of embankmentsScience China Technological Sci, 54 (1) (2011), pp. 148-155 View PDFCrossRefView Record in ScopusGoogle Scholar[17]M.-H. Yu, H.-Y. Wei, Y.-J. Liang, Y. ZhaoInvestigation of non-cohesive levee breach by overtopping flowJ Hydrodyn, 25 (4) (2013), pp. 572-579ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[18]S. Wu, M. Yu, H. Wei, Y. Liang, J. ZengNon-symmetrical levee breaching processes in a channel bend due to overtoppingInt J Sedim Res, 33 (2) (2018), pp. 208-215ArticleDownload PDFView Record in ScopusGoogle Scholar[19]O. Saberi, G. ZenzNumerical investigation on 1D and 2D embankment dams failure due to overtopping flowInt J Hydraulic Engineering, 5 (2016), pp. 9-18View Record in ScopusGoogle Scholar[20]M. Guan, N.G. Wright, P.A. Sleigh2D Process-Based Morphodynamic Model for Flooding by Noncohesive Dyke BreachJ Hydraul Eng, 140 (7) (2014), p. 04014022, 10.1061/(ASCE)HY.1943-7900.0000861 View PDFView Record in ScopusGoogle Scholar[21]W. Wu, R. Marsooli, Z. HeDepth-Averaged Two-Dimensional Model of Unsteady Flow and Sediment Transport due to Noncohesive Embankment Break/BreachingJ Hydraul Eng, 138 (6) (2012), pp. 503-516View Record in ScopusGoogle Scholar[22]Z. Wang, D.S. BowlesThree-dimensional non-cohesive earthen dam breach model. Part 1: Theory and methodologyAdv Water Resour, 29 (10) (2006), pp. 1528-1545ArticleDownload PDFView Record in ScopusGoogle Scholar[23]Říha J, Duchan D, Zachoval Z, Erpicum S, Archambeau P, Pirotton M, et al. Performance of a shallow-water model for simulating flow over trapezoidal broad-crested weirs. J Hydrology Hydromechanics. 2019;67:322-8.Google Scholar[24]C.B. VreugdenhilNumerical methods for shallow-water flowSpringer Science & Business Media (1994)Google Scholar[25]L.A. Larocque, J. Imran, M.H. Chaudhry3D numerical simulation of partial breach dam-break flow using the LES and k–∊ turbulence modelsJ Hydraul Res, 51 (2) (2013), pp. 145-157 View PDFCrossRefView Record in ScopusGoogle Scholar[26]C. Yang, B. Lin, C. Jiang, Y. LiuPredicting near-field dam-break flow and impact force using a 3D modelJ Hydraul Res, 48 (6) (2010), pp. 784-792 View PDFCrossRefView Record in ScopusGoogle Scholar[27]FLOW-3D. Version 11.1.1 Flow Science, Inc., Santa Fe, NM. https://wwwflow3dcom.Google Scholar[28]C.W. Hirt, B.D. NicholsVolume of fluid (VOF) method for the dynamics of free boundariesJ Comput Phys, 39 (1) (1981), pp. 201-225ArticleDownload PDFGoogle Scholar[29]S.V. PatankarNumerical heat transfer and fluid flow, Hemisphere PublCorp, New York, 58 (1980), p. 288View Record in ScopusGoogle Scholar[30]M. Alemi, R. MaiaNumerical simulation of the flow and local scour process around single and complex bridge piersInt J Civil Eng, 16 (5) (2018), pp. 475-487 View PDFCrossRefView Record in ScopusGoogle Scholar

    PDF 전체 원문 보기