Citizens’ daily needs such as; transportation, communication, clean water and sewage are supplied with infrastructure systems. Horizontal and vertical expansion in the cities due to the increase in population leads to serious demand for infrastructural improvements. The infrastructure systems in developing cities are required to be designed in a satisfactory capacity to supply the increasing demand for residential and industrial constructions. The districts having insufficient infrastructure systems inevitably confront heavy traffic, flood, air pollution problems, and also having difficulties with the inadequacy of parking area, clear and potable water, communication. The problems may cause social and health problems over time. At this point, it is wished to emphasize that the primary factor of citycivilization development depends on infrastructural systems and it is meaningful to name the engineering field like civil engineering, literally leads civilization. Dropshafts, commonly used in the urban storm and sewage water systems produced generally circular are used for energy dissipation and flow direction control. Aeration is significant for the working principle of the flow in dropshaft and this study is made mainly for this two-phase (air-water) physics of dropshafts. Chanson showed that aeration and energy dissipation is directly linked to each other (2002), but the influencing factors and the action mechanisms of the factors on the phenomena are not discovered entirely. By the comprehension of the factors, more effective dropshafts will be able to design. This study aims to guide the more comprehensive investigation of design factors using Computational Fluid Dynamics-CFD programs. The reasons for the preference of the programs are the cost-effectiveness of material, workmanship and duration relative to hydraulic modelling. The competence of the inputs, outputs and solution system of the CFD code is validated by the comparison of previous hydraulic modelling results.

Keywords

CFD, Dropshaft, Sewer system, Storm Water System, Two-Phase Flow

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 S_{eq} around USAF. At last, a parametric study was carried out to study the effects of the Froude number F_{r} and Euler number E_{u} for the S_{eq.} 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 KC_{s,p} < 8. The predicting results of the revised stochastic model are the most favorable for n = 10 when KC_{rms,a} < 4. The higher F_{r} and E_{u} both lead to the more intensive horseshoe vortex and larger S_{eq}.

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].

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 < θ_{cr}) or live bed scour (θ > θ_{cr}). 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(R_{d}) (R_{d} is pile Reynolds), which means the turbulence intensity induced by the flow-structure interaction would decrease with R_{d} increases, but the effects of R_{d} 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, U_{wm} 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.

where, γ is safety factor, depending on design process, typically γ = 1.5, K_{wave} is correction factor considering wave action, K_{hw} is correction factor considering water depth.

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 U_{m} and peak wave period T_{P} to calculate KC. Khalfin [35] recommended the RMS wave height H_{rms} and peak wave period T_{P} 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 U_{m} and mean zero-crossing wave period T_{z}. 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 (F_{r}) and Euler number (E_{u}) 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:

where, V_{F} is the volume fraction; u, v, and w are the velocity components in x, y, z direction respectively with Cartesian coordinates; A_{i} is the area fraction; ρ_{f} is the fluid density, f_{i} is the viscous fluid acceleration, G_{i} 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].

where, k_{T} is specific kinetic energy involved with turbulent velocity, G_{T} is the turbulent energy generated by buoyancy; ε_{T} is the turbulent energy dissipating rate, P_{T} is the turbulent energy, Diff_{ε} and Diff_{kT} are diffusion terms associated with V_{F}, A_{i}; CDIS1, CDIS2 and CDIS3 are dimensionless parameters, and CDIS1, CDIS3 have default values of 1.42, 0.2 respectively. CDIS2 can be obtained from P_{T} and k_{T}.

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]:

where, α_{i} is the entrainment parameter, n_{s} is the outward point perpendicular to the seabed, d_{*} is the dimensionless diameter of sand particles, which was calculated by Equation (15), θ_{cr} is the critical Shields parameter, g is the gravity acceleration, d_{i} is the diameter of sand particles, ρ_{i} is the density of seabed species.

In Equation (14), the entrainment parameter α_{i} 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]. n_{s} is the outward pointing normal to the seabed interface, and n_{s} = (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, U_{f,m} is the maximum value of the near-bed friction velocity; d_{50} is the median diameter of sand particles. The detailed calculation procedure of θ was available in Soulsby [44].

The critical shields parameter θ_{cr} was obtained from the Equation (17) [44]

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]:

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]:

where, q_{b,i} is the bed load transport rate, which was obtained from Equation (20), δ_{i} is the bed load thickness, which was calculated by Equation (21), c_{b,i} is the volume fraction of sand i in the multiple species, f_{b} is the critical packing fraction of the seabed.

where, C_{s,i} is the suspended sand particles mass concentration of sand i in the multiple species, u_{s,i} is the sand particles velocity of sand i, D_{f} is the diffusivity.

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

where, u¯�¯ is the velocity of mixed fluid-particles, which can be calculated by the RANS equation with turbulence model, c_{s,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 d_{50} = 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.

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.

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.

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]:

where, α is wave energy scale parameter, which is calculated by Equation (26), ω is frequency, ω_{p} 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 ≤ ω_{p} and 0.09 for ω > ω_{p} 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 U_{r} were acquired form Equations (28) and (29) respectively

ε=2πgHsT2a�=2���s�a2(28)

Ur=Hsk2h3w�r=�s�2ℎw3(29)

where, H_{s} is significant wave height, T_{a} is average wave period, k is wave number, h_{w} is water depth. The Shield parameter θ satisfies θ_{>}θ_{cr} for all simulations in current study, indicating the live bed scour prevails.

Table 2. Numerical simulating cases.

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.

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.

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.

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.

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 T_{c} is time scale of scour process.

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.

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.

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.

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

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.

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 U_{wm} and wave period T. For random waves, the U_{wm} can be denoted by the root-mean-square (RMS) value of near-bed velocity amplitude U_{wm,rms} or the significant value of near-bed velocity amplitude U_{wm,s}. The U_{wm,rms} and U_{wm,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 T_{a}, peak wave period T_{p}, significant wave period T_{s}, the maximum wave period T_{m}, 1/10′th highest wave period T_{n = 1/10} and 1/5′th highest wave period T_{n = 1/5} for random waves, so the different combinations of U_{wm} and T will acquire different KC. The Table 3 and Table 4 list 12 types of KC, for example, the KC_{rms,s} was calculated by U_{wm,rms} and T_{s}. 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 KC_{rms,p}. It should be noted that the Equation (2) is only suitable for KC > 6 under regular waves or KC_{rms,p} > 6 under random waves.

Table 3.U_{wm,rms} and KC for case 1~9.

Table 4.U_{wm,s} and KC for case 1~9.

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 S_{eq} between the present study and Raaijmakers’s equation was conducted. The position where the scour depth S_{eq} was evaluated is the location of the maximum scour depth, and it was depicted in Figure 14. The Figure 15 displays the comparison of S_{eq} with different KC between the present study and Raaijmakers’s model.

Figure 14. Sketch of the position where the S_{eq} was evaluated.

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

As shown in Figure 15, there is an error in predicting S_{eq} between the present study and Raaijmakers’s model, and Raaijmakers’s model underestimates the results generally. Although the error exists, the varying trend of S_{eq} 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 KC_{s,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

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 (KC_{s,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 d_{50} = 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 KC_{s,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 KC_{s,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 KC_{s,p} > 8.

Figure 16. Comparison of S_{eq} between the simulating results and the predicting values by Equation (31).

Figure 17. Comparison of S_{eq}/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 KC_{rms,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 S_{eq} with KC_{rms,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.

Figure 18. Comparison of S_{eq} between the simulating results and the predicting values by Equation (8).

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

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 S_{eq} with KC_{rms,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.

Figure 20. Comparison of S_{eq} 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 (KC_{rms,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 d_{50} = 0.2 mm and the water depth was maintained at 0.5 m. The JONSWAP wave spectrum was used and the KC_{rms,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 KC_{rms,a} (KC_{rms,a} < 4), the error between the predicting values and experimental results increases with decreasing of n for KC_{rms,a} > 4. Therefore, the predicting results are the most favorable for n = 2 when KC_{rms,a} > 4.

Figure 21. Comparison of S_{eq} 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 F_{r} is the key parameter to influence the scale and intensity of horseshoe vortex. The F_{r} under waves can be calculated by the following formula [42]

Fr=UwgD−−−√�r=�w��(34)

where U_{w} 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 U_{wm,rms} is used for calculating U_{wm}.

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

Tavouktsoglou et al. [25] proposed the following formula between F_{r} and the vertical location of the stagnation y

yh∝Fer�ℎ∝�r�(36)

where e is constant.

The Figure 23 displays the relationship between S_{eq}/D and F_{r} 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 F_{r} 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 F_{r}, 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 F_{r} 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 F_{r} 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 S_{eq}/D and F_{r} as following

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

where A, B and C are constant.

Figure 23. The fitting curve between S_{eq}/D and F_{r}.

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 S_{eq}/D and F_{r} in present study is consistent with Equation (37) basically, meaning the Equation (37) is applicable to express the relationship of S_{eq}/D with F_{r} around USAF under random waves.

4.4.2. Influence of Euler Number

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

Eu=U2mgD�u=�m2��(38)

where U_{m} is depth-averaged flow velocity.

The Figure 25 displays the relationship between S_{eq}/D and E_{u} 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 S_{eq}/D and F_{r}, the equilibrium scour depth appears a logarithmic increase as E_{u} 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 E_{u}, which is benefit for the wave boundary layer separating from seabed, inducing the high-intensity horseshoe vortex, hence, causing intensive scour around USAF.

Figure 25. The fitting curve between S_{eq}/D and E_{u}.

Therefore, the variation of F_{r} and E_{u} 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 S_{eq}/D and E_{u} in present study is consistent with Equation (37) basically, meaning the Equation (37) is also applicable to express the relationship of S_{eq}/D with E_{u} around USAF under random waves. Additionally, according to the above description of F_{r}, it can be inferred that the higher F_{r} and E_{u} 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 F_{r} and Euler number E_{u} 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 KC_{s,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 KC_{s,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 KC_{rms,a} < 4. However, contrary to the case of low KC_{rms,a}, the predicting results are the most favorable for n = 2 when KC_{rms,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 S_{eq}/D with E_{u} or F_{r}, and it can be inferred that the higher F_{r} and E_{u} both lead to the more intensive horseshoe vortex and larger S_{eq.}

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

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]

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]

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]

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]

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]

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]

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]

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. Energies2019, 12, 1709. [Google Scholar] [CrossRef][Green Version]

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]

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]

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]

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]

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]

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]

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]

Sumer, B.M.; Fredsøe, J. Scour around Pile in Combined Waves and Current. J. Hydraul. Eng.2001, 127, 403–411. [Google Scholar] [CrossRef]

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]

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]

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]

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]

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]

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]

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]

Ettema, R.; Melville, B.; Barkdoll, B. Scale Effect in Pier-Scour Experiments. J. Hydraul. Eng.1998, 124, 639–642. [Google Scholar] [CrossRef]

Umeda, S. Scour Regime and Scour Depth around a Pile in Waves. J. Coast. Res. Spec. Issue2011, 64, 845–849. [Google Scholar]

Umeda, S. Scour process around monopiles during various phases of sea storms. J. Coast. Res.2013, 165, 1599–1604. [Google Scholar] [CrossRef]

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]

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]

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]

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]

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]

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]

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]

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]

Myrhaug, D.; Rue, H. Scour below pipelines and around vertical piles in random waves. Coast. Eng.2003, 48, 227–242. [Google Scholar] [CrossRef]

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]

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]

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]

Yakhot, V.; Orszag, S.A. Renormalization group analysis of turbulence. I. Basic theory. J. Sci. Comput.1986, 1, 3–51. [Google Scholar] [CrossRef]

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]

Mastbergen, D.R.; Berg, J.V.D. Breaching in fine sands and the generation of sustained turbidity currents in submarine canyons. Sedimentology2003, 50, 625–637. [Google Scholar] [CrossRef]

Soulsby, R. Dynamics of Marine Sands; Thomas Telford Ltd.: London, UK, 1998. [Google Scholar] [CrossRef]

Van Rijn, L.C. Sediment Transport, Part I: Bed Load Transport. J. Hydraul. Eng.1984, 110, 1431–1456. [Google Scholar] [CrossRef][Green Version]

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]

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]

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]

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]

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]

Flow3D User Manual, version 11.0.3; Flow Science, Inc.: Santa Fe, NM, USA, 2013.

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]

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]

Breusers, H.N.C.; Nicollet, G.; Shen, H. Local Scour Around Cylindrical Piers. J. Hydraul. Res.1977, 15, 211–252. [Google Scholar] [CrossRef]

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.

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.

For vortex settling basins (VSBs) installed with a deflector, perforation is an effective retrofit to reduce the self-weight of the deflector and sediment deposition on it. The current study investigated experimentally the performance of VSBs the deflector of which was perforated at different locations with various opening ratios. The results showed that perforating the outside overflow area of the deflector was the optimum for reducing sediment deposition. With an opening ratio of 8.67–13% in the outside overflow area of the deflector, the VSB exhibited similar sediment removal efficiency to the original design without any openings on the deflector. The current study provided the design optimization for deflector perforation in VSBs.

디플렉터와 함께 설치된 와류 침전 분지(VSB)의 경우 천공은 디플렉터의 자체 중량과 침전물 증착을 줄이기 위한 효과적인 개조입니다. 현재 연구는 다양한 개방 비율로 다른 위치에서 디플렉터가 천공된 VSB의 성능을 실험적으로 조사했습니다. 결과는 디플렉터의 외부 오버플로 영역을 천공하는 것이 침전물 퇴적을 줄이는 데 최적임을 보여주었습니다. 디플렉터의 외부 오버플로 영역에서 8.67-13%의 개구부로 VSB는 디플렉터에 개구부가 없는 원래 설계와 유사한 침전물 제거 효율을 나타냈습니다. 현재 연구는 VSB의 디플렉터 천공에 대한 설계 최적화를 제공했습니다.

이 논문은 비대칭 인보드, 비대칭 아웃보드 및 다양한 스태거/분리 위치에서의 대칭을 포함하는 세 가지 대안적인 측면 선체 형태를 가진 웨이브 피어싱 3동선의 저항 성능에 대한 실험적 조사 결과를 제시했습니다.

모델 테스트는 0.225에서 0.60까지의 Froude 수에서 삼동선 축소 모형을 사용하여 National Iranian Marine Laboratory(NIMALA) 예인 탱크에서 수행되었습니다.

결과는 측면 선체를 주 선체 트랜섬의 앞쪽으로 이동함으로써 삼동선의 총 저항 계수가 감소하는 것으로 나타났습니다.

또한 조사 결과, 측면 선체의 대칭 형태가 3개의 측면 선체 형태 중 전체 저항에 대한 성능이 가장 우수한 것으로 나타났습니다. 본 연구의 결과는 저항 관점에서 측면 선체 구성을 선택하는 데 유용합니다.

Keywords

Resistance performance

Wave-piercing trimaran

Seakeeping characteristics

Side hull symmetry

Model test

Experimental study

References

Ackers BB, Thad JM, Tredennick OW, Landen CH, Miller EJ, Sodowsky JP, Hadler JB (1997) An investigation of the resistance characteristics of powered trimaran side-hull configurations. SNAME Transactions 105:349–373Google Scholar

ASME (2005) Test uncertainty, The American society of mechanical engineers performance test code, American Society of Mechanical Engineers, No. PTC 19. 1–2005, New York

Chen Y, Yang L, Xie Y, Yu S (2016) The research on characteristic parameters and resistance chart of operation and maintenance trimaran in the sea. Polish Maritime Research 23(s1):20–24. https://doi.org/10.1515/pomr-2016-0041ArticleGoogle Scholar

Claire M, Andrea M (2014) Resistance analysis for a trimaran. International Journal of Mechanical, Aerospace, Industrial, Mechatronic and Manufacturing Engineering 8(1):7–15Google Scholar

Doctors L, Scrace R (2003) The optimization of trimaran side hull position for minimum resistance. Seventh International Conference on Fast Transportation (FAST 2003), Ischia, Italy, 1–12

Ghadimi P, Nazemian A, Ghadimi A (2019) Numerical scrutiny of the influence of side hulls arrangement on the motion of a Trimaran vessel in regular waves through CFD analysis. Journal of the Brazilian Society of Mechanical Sciences and Engineering 41(1):1–10. https://doi.org/10.1007/s40430-018-1505-xArticleGoogle Scholar

Hafez K, El-Kot A-R (2011) Comparative analysis of the separation variation influence on the hydrodynamic performance of a high speed trimaran. Journal of Marine Science and Application 10(4):377–393. https://doi.org/10.1007/s11804-011-1083-0ArticleGoogle Scholar

Hafez KA, El-Kot AA (2012) Comparative investigation of the stagger variation influence on the hydrodynamic interference of high speed trimaran. Alexandria Engineering Journal 51(3):153–169. https://doi.org/10.1016/j.aej.2012.02.002ArticleGoogle Scholar

Hashimoto H, Amano S, Umeda N, Matsuda A (2011) Influence of side-hull positions on dynamic behaviors of a trimaran running in following and stern quartering seas. Proceedings of the 21th International Conference on Offshore and Polar Engineering, 573–580

ITTC (2014) Testing and extrapolation methods in resistance towing tank tests, Recommended Procedures, 7.5–02–02–01

Iqbal M, Utama IKAP (2014) An investigation into the effect of water depth on the resistance components of trimaran configuration. Proceedings of the 9th International Conference on Marine Technology, Surabaya

Lewis EV (1988) Principles of Naval Architecture. The Society of Naval Architects and Marine Engineers III: 323–324

Luhulima RB, Utama I, Sulisetyono A (2016) Experimental investigation into the resistance components of displacement trimaran at various lateral spacing. International Journal of Engineering Research & Science (IJOER) 2:21–29Google Scholar

Luhulima RB (2017) An Investigation into the resistance of displacement trimaran: a comparative analysis between experimental and CFD approaches. International Journal of Mechanical Engineering (IJME) 6:9–18Google Scholar

Molland AF, Turnock SR, Hudson DA (2011) Ship resistance and propulsion: practical estimation of ship propulsive power. Cambridge University Press, 544.

Verna S, Khan K, Praveen PC (2012) Trimaran hull form optimization, using ship flow. International Journal of Innovative Research and Development 1(10):5–15

Yanuar Y, Gunawan G, Talahatu MA, Indrawati RT, Jamaluddin A (2013) Resistance analysis of unsymmetrical trimaran model with outboard side hulls configuration. Journal of Marine Science and Application 12(3):293–297ArticleGoogle Scholar

Yanuar Y, Gunawan G, Talahatu MA, Indrawati RT, Jamaluddin A (2015a) Resistance reduction on trimaran ship model by biopolymer of eel slime. Journal of Naval Architecture and Marine Engineering 12(2):95–102. https://doi.org/10.3329/jname.v12i2.19549ArticleGoogle Scholar

Yanuar Y, Gunawan G, Waskito KT, Jamaluddin A (2015b) Experimental study resistances of asymmetrical Pentamaran model with separation and staggered hull variation of inner side-hulls. International Journal of Fluid Mechanics Research 42(1):82–94. https://doi.org/10.1615/interjfluidmechres.v42.i1.60ArticleGoogle Scholar

Zhang L, Zhang JN, Shang YC (2019) A potential flow theory and boundary layer theory based hybrid method for waterjet propulsion. Journal of Marine Science and Engineering 7(4):113–132. https://doi.org/10.3390/jmse7040113ArticleGoogle Scholar

Thermo-fluid modeling of influence of attenuated laser beam intensity profile on melt pool behavior in laser-assisted powder-based direct energy deposition

Mohammad Sattari, Amin Ebrahimi, Martin Luckabauer, Gert-willem R.B.E. Römer

A numerical framework based on computational fluid dynamics (CFD), using the finite volume method (FVM) and volume of fluid (VOF) technique is presented to investigate the effect of the laser beam intensity profile on melt pool behavior in laser-assisted powder-based directed energy deposition (L-DED). L-DED is an additive manufacturing (AM) process that utilizes a laser beam to fuse metal powder particles. To assure high-fidelity modeling, it was found that it is crucial to accurately model the interaction between the powder stream and the laser beam in the gas region above the substrate. The proposed model considers various phenomena including laser energy attenuation and absorption, multiple reflections of the laser rays, powder particle stream, particle-fluid interaction, temperature-dependent properties, buoyancy effects, thermal expansion, solidification shrinkage and drag, and Marangoni flow. The latter is induced by temperature and element-dependent surface tension. The model is validated using experimental results and highlights the importance of considering laser energy attenuation. Furthermore, the study investigates how the laser beam intensity profile affects melt pool size and shape, influencing the solidification microstructure and mechanical properties of the deposited material. The proposed model has the potential to optimize the L-DED process for a variety of materials and provides insights into the capability of numerical modeling for additive manufacturing optimization.

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.

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/10.1029/98JC02622ArticleGoogle Scholar

Celliers L, Schleyer MH (2002) Coral bleaching on high-latitude marginal reefs at Sodwana Bay, South Africa. Mar Pollut Bull 44:1380–1387ArticleGoogle Scholar

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

Flow Science Inc (2018) FLOW-3D, Version 12.0 Users Manual. Santa Fe, NM, https://www.flow3d.com/

Flow Science Inc (2019) FLOW-3D, Version 12.0 [Computer Software]. Santa Fe, NM, https://www.flow3d.com/

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):2255–2279ArticleGoogle Scholar

Fringer OB, Gerritsen M, Street RL (2006) An unstructured-grid, finite-volume, nonhydrostatic, parallel coastal ocean simulator. Ocean Model 14(3):139–173ArticleGoogle Scholar

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/

Hocker LO, Hruska MA (2004) Interleaving synchronous data and asynchronous data in a single data storage file

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-yArticleGoogle Scholar

Morris T (2009) Physical oceanography of Sodwana Bay and its effect on larval transport and coral bleaching. PhD thesis, Cape Peninsula University of Technology

Pope SB (2001) Turbulent flows. Cambridge University Press, CambridgeGoogle Scholar

Porter SN (2009) Biogeography and potential factors regulating shallow subtidal reef communities in the Western Indian Ocean. PhD thesis, University of Cape Town

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/d11040057ArticleGoogle Scholar

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

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)

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

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.10365ArticleGoogle Scholar

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/MF02049ArticleGoogle Scholar

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

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):347–360ArticleGoogle Scholar

Smagorinsky J (1963) General circulation experiments with the primitive equations. Mon Weather Rev 91(3):99–164ArticleGoogle Scholar

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.2018.0448ArticleGoogle Scholar

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 405:113–130ArticleGoogle Scholar

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

��

��

��

Quantity

0.012

4.38

1.42

1.68

1.39

1.39

0.084

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 x_{k} 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 Cell

Algorithm 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 Cell

Algorithm 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

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.

[2]A. Morteza, M. SadipourReza Saadati Fard, Saman Taheri, and Amirhossein AhmadiA dagging-based deep learning framework for transmission line flexibility assessment, IET Renewable Power Generation (2022)Google Scholar

[3]A. Morteza, M. Ilbeigi, J. SchwedA blockchain information management framework for construction safety. Comput-ingCivil Engineering (2021, 2022.)Google Scholar

[4]Jochem Weber, Ronan Costello, and John Ringwood. Wec technology performance levels (tpls)-metric for successful development of economic wec technology. Proceedings EWTEC 2013, 2013.Google Scholar

[5]K. Rahgooy, A. Bahmanpour, M. Derakhshandi, A.a. Bagherzadeh-KhalkhaliDistribution of elastoplastic modulus of subgrade reaction for analysis of raft foundationsGeomechanics and Engineering, 28 (1) (2022), pp. 89-105View in ScopusGoogle Scholar

[7]M. Penalba, G. Giorgi, J.V. RingwoodMathematical modelling of wave energy converters: A review of nonlinear approachesRenewable and Sustainable Energy Reviews, 78 (2017), pp. 1188-1207View PDFView articleView in ScopusGoogle Scholar

[8]C. Windt, J. Davidson, J.V. RingwoodHigh-fidelity numerical modelling of ocean wave energy systems: A review of computational fluid dynamics-based numerical wave tanksRenewable and Sustainable Energy Reviews, 93 (2018), pp. 610-630View PDFView articleView in ScopusGoogle Scholar

[10]J.V. Ringwood, G. Bacelli, F. FuscoEnergymaximizing control of wave-energy converters: The development of control system technology to optimize their operationIEEE control systems magazine, 34 (5) (2014), pp. 30-55View article CrossRefView in ScopusGoogle Scholar

[11]N. Faedo, S. Olaya, J.V. RingwoodOptimal control, mpc and mpc-like algorithms for wave energy systems: An overviewIFAC Journal of Systems and Control, 1 (2017), pp. 37-56View PDFView articleView in ScopusGoogle Scholar

[12]L. Wang, J. Isberg, E. TedeschiReview of control strategies for wave energy conversion systems and their validation: the wave-to-wire approachRenewable and Sustainable Energy Reviews, 81 (2018), pp. 366-379View PDFView articleView in ScopusGoogle Scholar

[15]E. Amini, D. Golbaz, R. Asadi, M. Nasiri, O. Ceylan, M.M. Nezhad, et al.A comparative study of metaheuristic algorithms for wave energy converter power take-off optimisation: A case study for eastern australiaJournal of Marine Science and Engineering, 9(5):490 (2021)Google Scholar

[16]Arthur Pecher and Jens Peter KofoedHandbook of ocean wave energySpringer Nature (2017)Google Scholar

[17]G. Chang, C.A. Jones, J.D. Roberts, V.S. NearyA comprehensive evaluation of factors affecting the levelized cost of wave energy conversion projectsRenewable Energy, 127 (2018), pp. 344-354View PDFView articleView in ScopusGoogle Scholar

[18]E. Amini, H. Mehdipour, E. Faraggiana, D. Golbaz, S. Mozaffari, G. Bracco, et al.Optimization of hydraulic power take-off system settings for point absorber wave energy converterRenewable Energy, 194 (2022), pp. 938-954View PDFView articleView in ScopusGoogle Scholar

[19]A. Garcia-Teruel, D.I.M. ForehandA review of geometry optimisa-tion of wave energy convertersRenewable and Sustainable Energy Reviews, 139 (2021), Article 110593View PDFView articleView in ScopusGoogle Scholar

[20]M.M. Nezhad, A. Heydari, M. Neshat, F. Keynia, G. Piras, D.A. GarciaA mediterranean sea offshore wind classification using merra-2 and machine learning modelsRenewable Energy, 190 (2022), pp. 156-166Google Scholar

[21]I. López, J. Andreu, S. Ceballos, I.M.D. Alegría, I. KortabarriaReview of wave energy technologies and the necessary power-equipmentRenewable and sustainable energy reviews, 27 (2013), pp. 413-434View PDFView articleView in ScopusGoogle Scholar

[22]R. Ekström, B. Ekergård, M. LeijonElectrical damping of linear generators for wave energy converters—a reviewRenewable and Sustainable Energy Reviews, 42 (2015), pp. 116-128View PDFView articleGoogle Scholar

[23]Danial Golbaz, Rojin Asadi, Erfan Amini, Hossein Mehdipour, Mahdieh Nasiri, Meysam Majidi Nezhad, Seyed Taghi Omid Naeeni, and Mehdi Neshat. Ocean wave energy converters optimization: A comprehensive review on research directions. arXiv preprint arXiv:2105.07180, 2021.Google Scholar

[24]Michael Choiniere, Jacob Davis, Nhu Nguyen, Nathan Tom, Matthew Fowler, and Krish Thiagarajan Sharman. Hydrodynamics and load shedding behavior of a variable geometry oscillating surge wave energy converter (oswec). Available at SSRN 3900951, 2022.Google Scholar

[25]Alan Henry, Olivier Kimmoun, Jonathan Nicholson, Guillaume Dupont, Yanji Wei, andFrederic Dias. A two dimensional experimental investigation of slamming of an oscillating wave surge converter. In The Twenty-fourth International Ocean and Polar Engineering Conference. OnePetro, 2014.Google Scholar

[26]S. Doyle, G.A. AggidisDevelopment of multioscillating water columns as wave energy convertersRenewable and Sustainable Energy Reviews, 107 (2019), pp. 75-86View PDFView articleView in ScopusGoogle Scholar

[28]Matthew Folley, TJT Whittaker, and Alan Henry. The effect of water depth on the performance of a small surging wave energy converter. Ocean Engineering, 34(8-9):1265–1274, 2007.Google Scholar

[30]D. Sarkar, E. Renzi, F. DiasEffect of a straight coast on the hydrodynamics and performance of the oscillating wave surge converterOcean Engineering, 105 (2015), pp. 25-32View PDFView articleView in ScopusGoogle Scholar

[31]Adrian de Andres, Jéromine Maillet, Jørgen Hals Todalshaug, Patrik Möller, and Henry Jeffrey. On the optimum sizing of a real wec from a techno-economic perspective. In International Conference on Offshore Mechanics and Arctic Engineering, volume 49972, page V006T09A013. American Society of Mechanical Engineers, 2016.Google Scholar

[34]T. Whittaker, M. FolleyNearshore oscillating wave surge converters and the development of oysterPhilosophical Transactions Sciences of the Royal Society A: Mathematical, Physical and Engineering, 370 (1959) (2012), pp. 345-364View article CrossRefView in ScopusGoogle Scholar

[35]Louise O’Boyle, Kenneth Doherty, Jos van’t Hoff, and Jessica Skelton. The value of full scale prototype data-testing oyster 800 at emec, orkney. In Proceedings of the 11th European wave and tidal energy conference (_{EWTEC), Nantes, France}, pages 6–11, 2015.Google Scholar

[37]Ishmail B Celik, Urmila Ghia, Patrick J Roache, and Christopher J Freitas. Procedure for estimation and reporting of uncertainty due to discretization in cfd applications. Journal of fluids EngineeringTransactions of the ASME, 130(7), 2008.Google Scholar

[38]Pal Schmitt, K Doherty, Darragh Clabby, and T Whittaker. The opportunities and limitations of using cfd in the development of wave energyconverters. Marine&OffshoreRenewableEnergy, pages 89–97, 2012.Google Scholar

[39]M. Choiniere, J. Davis, N.u. Nguyen, N. Tom, M. Fowler, K. ThiagarajanHydrodynamics and load shedding behavior of a variable-geometry oscillating surge wave energy converter (oswec)Renewable Energy (2022)Google Scholar

[40]Yong Li and Mian Lin. Regular and irregular wave impacts on floating body. Ocean Engineering, 42:93–101, 2012. Pal Manuel Schmitt. Investigation of the near flow field of bottom hinged flap type wave energy converters. PhD thesis, Queen’s University Belfast, 2014.Google Scholar

[41]Alan John Henry. The hydrodynamics of small seabed mounted bottom hinged wave energy conerverters in shallow water. PhD thesis, Queen’s University Belfast, 2009.Google Scholar

[42]N. Ghorbani, A. Korzeniowski, et al.Adaptive risk hedging for call options under cox-ingersoll-ross interest ratesJournal of Mathematical Finance, 10 (04) (2020), p. 697 View PDF CrossRefView in ScopusGoogle Scholar

[44]M. Abdel-Basset, L. Abdel-Fatah, A.K. SangaiahChapter 10metaheuristic algorithms: a comprehensive reviewcomputational intelligence for multimedia big data on the cloud with engineering applications (2018)Google Scholar

[47]Mohammad Shehab, Laith Abualigah, Husam Al Hamad, Hamzeh Alabool, Mohammad Alshinwan, and Ahmad M Khasawneh. Moth– flame optimization algorithm: variants and applications. Neural Computing and Applications, 32(14):9859–9884, 2020.Google Scholar

[48]Betül Sultan Yıldız and Ali Rıza YıldızMoth-flame optimization algorithm to determine optimal machining parameters in manufacturing processesMaterials Testing, 59 (5) (2017), pp. 425-429Google Scholar

[49]M Tegmark. Barrow, jd davies, pc harper, cl, jr eds. Science and Ultimate Reality Cambridge University Press Cambridge, 2004.Google Scholar

[52]M.S. Morris, K.S. ThorneWormholes in spacetime and their use for interstellar travel: A tool for teaching general relativityAmerican Journal of Physics, 56 (5) (1988), pp. 395-412View article CrossRefView in ScopusGoogle Scholar

[53]S. Mirjalili, S.M. Mirjalili, A. HatamlouMulti-verse optimizer: a nature-inspired algorithm for global optimizationNeural Computing and Applications, 27 (2) (2016), pp. 495-513View article CrossRefView in ScopusGoogle Scholar

[55]Farhad Soleimanian Gharehchopogh and Hojjat GholizadehA comprehensive survey: Whale optimization algorithm and its applicationsSwarm and Evolutionary Computation, 48 (2019), pp. 1-24Google Scholar

[56]L. AbualigahMulti-verse optimizer algorithm: a comprehensive survey of its results, variants, and applicationsNeural Computing and Applications, 32 (16) (2020), pp. 12381-12401View article CrossRefView in ScopusGoogle Scholar

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.

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

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].

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

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

Figure 7. The relationship between the potential core length

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

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

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

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

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

Figure 13. Along-range distribution of the flow angle φ of the jet at different Reynolds numbers.

Figure 14. Velocity distribution along the jet axis at different Reynolds numbers.

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

References

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]

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]

Gao, Y.; Han, P.; Wang, F.; Cao, J.; Zhang, S. Study on the Characteristics of Water Jet Breaking Coal Rock in a Drilling Hole. Sustainability2022, 14, 8258. [Google Scholar] [CrossRef]

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]

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]

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]

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]

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]

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]

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]

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]

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]

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]

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]

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]

Seo, H.; Kim, K.C. Experimental study on flow and turbulence characteristics of bubbly jet with low void fraction. Int. J. Multiph. Flow2021, 142, 103738. [Google Scholar] [CrossRef]

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]

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]

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]

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]

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]

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]

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]

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]

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]

Lucas, C.; Rusche, H.; Schroeder, A.; Koehler, J. Numerical investigation of a two-phase CO_{2} ejector. Int. J. Refrigeration2014, 43, 154–166. [Google Scholar] [CrossRef]

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]

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]

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]

Huang, J.; Ni, F.; Gu, L. Numerical method of FLOW-3D for sediment erosion simulation. China Harb. Eng.2019, 39, 6–11. [Google Scholar]

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]

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]

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]

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]

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]

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]

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. Machines2023, 11, 68. [Google Scholar] [CrossRef]

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]

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]

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]

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. Fluids2012, 52, 1475–1489. [Google Scholar] [CrossRef]

Fairweather, M.; Hargrave, G. Experimental investigation of an axisymmetric, impinging turbulent jet. 1. Velocity field. Exp. Fluids2002, 33, 464–471. [Google Scholar] [CrossRef]

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]

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. Shock2021, 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.

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. Sustainability2023, 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

Dissertação de Mestrado Ciclo de Estudos Integrados Conducentes ao Grau de Mestre em Engenharia Mecânica Trabalho efectuado sob a orientação do Doutor Hélder de Jesus Fernades Puga Professor Doutor José Joaquim Carneiro Barbosa

ABSTRACT

논문의 일부로 튜터 선택 가능성과 해결해야 할 주제가 설정되는 매개변수를 염두에 두고 개발 주제 ‘Flow- 3D ®에 의한 저압 충전 시스템 최적화’가 선택되었습니다. 이를 위해서는 달성해야 할 목표와 이를 달성하기 위한 방법을 정의하는 것이 필요했습니다.

충전 시스템을 시뮬레이션하고 검증할 수 있는 광범위한 소프트웨어에도 불구하고 Flow-3D®는 시장에서 최고의 도구 중 하나로 표시되어 전체 충전 프로세스 및 행동 표현과 관련하여 탁월한 정확도로 시뮬레이션하는 능력을 입증했습니다.

이를 위해 관련 프로세스를 더 잘 이해하고 충진 시스템 시뮬레이션을 위한 탐색적 기반 역할을 하기 위해 이 도구를 탐색하는 것이 중요합니다. 지연 및 재료 낭비에 반영되는 실제적인 측면에서 충전 장치의 치수를 완벽하게 만드는 비용 및 시간 낭비. 이러한 방식으로 저압 주조 공정에서 충진 시스템을 설계하고 물리적 모델을 탐색하여 특성화하는 방법론을 검증하기 위한 것입니다.

이를 위해 다음 주요 단계를 고려하십시오.

시뮬레이션 소프트웨어 Flow 3D® 탐색; 충전 시스템 모델링; 모델의 매개변수를 탐색하여 모델링된 시스템의 시뮬레이션, 검증 및 최적화.

따라서 연구 중인 압력 곡선과 주조 분석에서 가장 관련성이 높은 정보의 최종 마이닝을 검증하기 위한 것입니다.

사용된 압력 곡선은 수집된 문헌과 이전에 수행된 실제 작업을 통해 얻었습니다. 결과를 통해 3단계 압력 곡선이 층류 충진 체계의 의도된 목적과 관련 속도가 0.5 𝑚/𝑠를 초과하지 않는다는 결론을 내릴 수 있었습니다.

충전 수준이 2인 압력 곡선은 0.5 𝑚/𝑠 이상의 속도로 영역을 채우는 더 난류 시스템을 갖습니다. 열전달 매개변수는 이전에 얻은 값이 주물에 대한 소산 거동을 확증하지 않았기 때문에 연구되었습니다.

이러한 방식으로 주조 공정에 더 부합하는 새로운 가치를 얻었습니다. 달성된 결과는 유사한 것으로 나타난 NovaFlow & Solid®에 의해 생성된 결과와 비교되어 시뮬레이션에서 설정된 매개변수를 검증했습니다. Flow 3D®는 주조 부품 시뮬레이션을 위한 강력한 도구로 입증되었습니다.

As part of the dissertation and bearing in mind the parameters in which the possibility of a choice of tutor and the subject to be addressed is established, the subject for development ’Optimization of filling systems for low pressure by Flow 3D ®’ was chosen. For this it was necessary to define the objectives to achieve and the methods to attain them. Despite the wide range of software able to simulate and validate filling systems, Flow 3D® has been shown as one of the best tools in the market, demonstrating its ability to simulate with distinctive accuracy with respect to the entire process of filling and the behavioral representation of the fluid obtained. To this end, it is important to explore this tool for a better understanding of the processes involved and to serve as an exploratory basis for the simulation of filling systems, simulation being one of the great strengths of the current industry due to the need to reduce costs and time waste, in practical terms, that lead to the perfecting of the dimensioning of filling devices, which are reflected in delays and wasted material. In this way it is intended to validate the methodology to design a filling system in lowpressure casting process, exploring their physical models and thus allowing for its characterization. For this, consider the following main phases: The exploration of the simulation software Flow 3D®; modeling of filling systems; simulation, validation and optimization of systems modeled by exploring the parameters of the models. Therefore, it is intended to validate the pressure curves under study and the eventual mining of the most relevant information in a casting analysis. The pressure curves that were used were obtained through the gathered literature and the practical work previously performed. Through the results it was possible to conclude that the pressure curve with 3 levels meets the intended purpose of a laminar filling regime and associated speeds never exceeding 0.5 𝑚/𝑠. The pressure curve with 2 filling levels has a more turbulent system, having filling areas with velocities above 0.5 𝑚/𝑠. The heat transfer parameter was studied due to the values previously obtained didn’t corroborate the behavior of dissipation regarding to the casting. In this way, new values, more in tune with the casting process, were obtained. The achieved results were compared with those generated by NovaFlow & Solid®, which were shown to be similar, validating the parameters established in the simulations. Flow 3D® was proven a powerful tool for the simulation of casting parts.

[1] E. Stanley and D. B. Sc, “Fluid Flow Aspects of Solidification Modelling : Simulation of Low Pressure Die Casting .” [2] Y. Sahin, “Computer aided foundry die-design,” Metallography, vol. 24, no. 8, pp. 671–679, 2003. [3] F. Bonollo, J. Urban, B. Bonatto, and M. Botter, “Gravity and low pressure die casting of aluminium alloys : a technical and economical benchmark,” La Metall. Ital., vol. 97, no. 6, pp. 23–32, 2005. [4] P. a and R. R, “Study of the effect of process parameters on the production of a nonsimmetric low pressure die casting part,” La Metall. Ital., pp. 57–63, 2009. [5] “Fundição em baixa pressão | Aluinfo.” [Online]. Available: http://www.aluinfo.com.br/novo/materiais/fundicao-em-baixa-pressao. [Accessed: 18- Sep-2015]. [6] “Low Pressure Sand Casting by Wolverine Bronze.” [Online]. Available: http://www.wolverinebronze.com/low-pressure-sand-casting.php. [Accessed: 18-Sep2015]. [7] A. Reikher, “Numerical Analysis of Die-Casting Process in Thin Cavities Using Lubrication Approximation,” no. December, 2012. [8] P. Fu, A. a. Luo, H. Jiang, L. Peng, Y. Yu, C. Zhai, and A. K. Sachdev, “Low-pressure die casting of magnesium alloy AM50: Response to process parameters,” J. Mater. Process. Technol., vol. 205, no. 1–3, pp. 224–234, 2008. [9] X. Li, Q. Hao, W. Jie, and Y. Zhou, “Development of pressure control system in counter gravity casting for large thin-walled A357 aluminum alloy components,” Trans. Nonferrous Met. Soc. China, vol. 18, no. 4, pp. 847–851, 2008. [10] J. a. Hines, “Determination of interfacial heat-transfer boundary conditions in an aluminum low-pressure permanent mold test casting,” Metall. Mater. Trans. B, vol. 35, no. 2, pp. 299–311, 2004. [11] A. Lima, A. Freitas, and P. Magalhães, “Processos de vazamento em moldações permanentes,” pp. 40–49, 2003. [12] Y. B. Choi, K. Matsugi, G. Sasaki, K. Arita, and O. Yanagisawa, “Analysis of Manufacturing Processes for Metal Fiber Reinforced Aluminum Alloy Composite Fabricated by Low-Pressure Casting,” Mater. Trans., vol. 47, no. 4, pp. 1227–1231, 68 2006. [13] G. Mi, X. Liu, K. Wang, and H. Fu, “Numerical simulation of low pressure die-casting aluminum wheel,” China Foundry, vol. 6, no. 1, pp. 48–52, 2009. [14] J. Kuo, F. Hsu, and W. Hwang, “ADVANCED Development of an interactive simulation system for the determination of the pressure ± time relationship during the ® lling in a low pressure casting process,” vol. 2, pp. 131–145, 2001. [15] S.-G. Liu, F.-Y. Cao, X.-Y. Zhao, Y.-D. Jia, Z.-L. Ning, and J.-F. Sun, “Characteristics of mold filling and entrainment of oxide film in low pressure casting of A356 alloy,” Mater. Sci. Eng. A, vol. 626, pp. 159–164, 2015. [16] “Casting Training Class – Lecture 10 – Solidification and Shrinkage-Casting.” FLOW3D®. [17] “UAB Casting Engineering Laboratory.” [Online]. Available: file:///C:/Users/Jos%C3%A9 Belo/Desktop/Artigo_Software/UAB Casting Engineering Laboratory.htm. [Accessed: 09-Nov-2015]. [18] A. Louvo, “Casting Simulation as a Tool in Concurrent Engineering,” pp. 1–12, 1997. [19] T. R. Vijayaram and P. Piccardo, “Computers in Foundries,” vol. 30, 2012. [20] M. Sadaiah, D. R. Yadav, P. V. Mohanram, and P. Radhakrishnan, “A generative computer-aided process planning system for prismatic components,” Int. J. Adv. Manuf. Technol., vol. 20, no. 10, pp. 709–719, 2002. [21] Ministry_of_Planning, “Digital Data,” vol. 67, pp. 1–6, 2004. [22] S. Shamasundar, D. Ramachandran, and N. S. Shrinivasan, “COMPUTER SIMULATION AND ANALYSIS OF INVESTMENTCASTING PROCESS.” [23] J. M. Siqueira and G. Motors, “Simulation applied to Aluminum High Pressure Die Casting,” pp. 1–5, 1998. [24] C. Fluid, COMPUTATIONAL FLUID DYNAMICS. Abdulnaser Sayma & Ventus Publishing ApS, 2009. [25] C. a. Felippa, “1 – Overview,” Adv. Finite Elem. Methods, pp. 1–9. [26] a. Meena and M. El Mansori, “Correlative thermal methodology for castability simulation of ductile iron in ADI production,” J. Mater. Process. Technol., vol. 212, no. 11, pp. 2484–2495, 2012. [27] T. R. Vijayaram, S. Sulaiman, a. M. S. Hamouda, and M. H. M. Ahmad, “Numerical simulation of casting solidification in permanent metallic molds,” J. Mater. Process. 69 Technol., vol. 178, pp. 29–33, 2006. [28] “General CFD FAQ — CFD-Wiki, the free CFD reference.” [Online]. Available: http://www.cfd-online.com/Wiki/General_CFD_FAQ. [Accessed: 10-Nov-2015]. [29] “FEM | FEA | CFD.” [Online]. Available: http://fem4analyze.blogspot.pt/. [Accessed: 09-Nov-2015]. [30] “Fundição; revista da Associação portuguesa de fundição,” Fundição, vol. N o 227. [31] “Casting Training Class – Lecture 1 – Introduction_to_FLOW-3D – Casting.” FLOW3D®. [32] F. Science, “FLOW-3D Cast Documentation,” no. 3.5, p. 80, 2012. [33] “Casting Training Class – Lecture 4 – Geometry Building – General.” FLOW-3D®. [34] F. Science, “FLOW-3D v11.0.3 User Manual,” pp. 1–132, 2015. [35] “Casting Training Class – Lecture 5 Meshing Concept – General.” FLOW-3D®. [36] “Casting Training Class – Lecture 6 – Boundary_Conditions – Casting.” FLOW-3D®. [37] “Casting Training Class – Lecture 9 – Physical Models-castings.” FLOW-3D®. [38] P. A. D. Jácome, M. C. Landim, A. Garcia, A. F. Furtado, and I. L. Ferreira, “The application of computational thermodynamics and a numerical model for the determination of surface tension and Gibbs–Thomson coefficient of aluminum based alloys,” Thermochim. Acta, vol. 523, no. 1–2, pp. 142–149, 2011. [39] J. P. Anson, R. A. L. Drew, and J. E. Gruzleski, “The surface tension of molten aluminum and Al-Si-Mg alloy under vacuum and hydrogen atmospheres,” Metall. Mater. Trans. B Process Metall. Mater. Process. Sci., vol. 30, no. 6, pp. XVI–1032, 1999.

Beom-Jin Kim ^{1}, Jae-Hong Hwang ^{2} and Byunghyun Kim ^{3},* ^{1} Advanced Structures and Seismic Safety Research Division, Korea Atomic Energy Research Institute, Daejeon 34057, Korea ^{2} Korea Water Resources Corporation (K-Water), Daejeon 34350, Korea ^{3 }Department of Civil Engineering, Kyungpook National University, Daegu 41566, Korea

Hydraulic structures installed in rivers inevitably create a water level difference between upstream and downstream regions. The potential energy due to this difference in water level is converted into kinetic energy, causing high-velocity flow and hydraulic jumps in the river. As a result, problems such as scouring and sloping downstream may occur around the hydraulic structures. In this study, a FLOW-3D model was constructed to perform a numerical analysis of the ChangnyeongHaman weir in the Republic of Korea. The constructed model was verified based on surface velocity measurements from a field gate operation experiment. In the simulation results, the flow discharge differed from the measured value by 9–15 m3/s, from which the accuracy was evaluated to be 82–87%. The flow velocity was evaluated with an accuracy of 92% from a difference of 0.01 to 0.16 m/s. Following this verification, a flow analysis of the hydraulic structures was performed according to boundary conditions and operation conditions for numerous scenarios. Since 2018, the ChangnyeongHaman weir gate has been fully opened due to the implementation of Korea’s eco-environmental policy; therefore, in this study, the actual gate operation history data prior to 2018 was applied and evaluated. The evaluation conditions were a 50% open gate condition and the flow discharge of two cases with a large difference in water level. As a result of the analysis, the actual operating conditions showed that the velocity and the Froude number were lower than the optimal conditions, confirming that the selected design was appropriate. It was also found that in the bed protection section, the average flow velocity was high when the water level difference was large, whereas the bottom velocity was high when the gate opening was large. Ultimately, through the reviewed status survey data in this study, the downstream flow characteristics of hydraulic structures along with adequacy verification techniques, optimal design techniques such as procedures for design, and important considerations were derived. Based on the current results, the constructed FLOW-3D-based model can be applied to creating or updating flow analysis guidelines for future repair and reinforcement measures as well as hydraulic structure design.

하천에 설치되는 수력구조물은 필연적으로 상류와 하류의 수위차를 발생시킨다. 이러한 수위차로 인한 위치에너지는 운동에너지로 변환되어 하천의 고속유동과 수압점프를 일으킨다. 그 결과 수력구조물 주변에서 하류의 세굴, 경사 등의 문제가 발생할 수 있다.

본 연구에서는 대한민국 창녕함안보의 수치해석을 위해 FLOW-3D 모델을 구축하였다. 구축된 모델은 현장 게이트 작동 실험에서 표면 속도 측정을 기반으로 검증되었습니다.

시뮬레이션 결과에서 유량은 측정값과 9~15 m3/s 차이가 나고 정확도는 82~87%로 평가되었다. 유속은 0.01~0.16m/s의 차이에서 92%의 정확도로 평가되었습니다.

검증 후 다양한 시나리오에 대한 경계조건 및 운전조건에 따른 수리구조물의 유동해석을 수행하였다. 2018년부터 창녕함안보 문은 한국의 친환경 정책 시행으로 전면 개방되었습니다.

따라서 본 연구에서는 2018년 이전의 실제 게이트 운영 이력 데이터를 적용하여 평가하였다. 평가조건은 50% open gate 조건과 수위차가 큰 2가지 경우의 유수방류로 하였다. 해석 결과 실제 운전조건은 속도와 Froude수가 최적조건보다 낮아 선정된 설계가 적합함을 확인하였다.

또한 베드보호구간에서는 수위차가 크면 평균유속이 높고, 수문개구가 크면 저저유속이 높은 것으로 나타났다. 최종적으로 본 연구에서 검토한 실태조사 자료를 통해 적정성 검증기법과 함께 수력구조물의 하류 유동특성, 설계절차 등 최적 설계기법 및 중요 고려사항을 도출하였다.

현재의 결과를 바탕으로 구축된 FLOW-3D 기반 모델은 수력구조 설계뿐만 아니라 향후 보수 및 보강 조치를 위한 유동해석 가이드라인 생성 또는 업데이트에 적용할 수 있습니다.

References

Wanoschek, R.; Hager, W.H. Hydraulic jump in trapezoidal channel. J. Hydraul. Res. 1989, 27, 429–446. [CrossRef]

Bohr, T.; Dimon, P.; Putkaradze, V. Shallow-water approach to the circular hydraulic jump. J. Fluid Mech. 1993, 254, 635–648. [CrossRef]

Chanson, H.; Brattberg, T. Experimental study of the air–water shear flow in a hydraulic jump. Int. J. Multiph. Flow 2000, 26, 583–607. [CrossRef]

Dhamotharan, S.; Gulliver, J.S.; Stefan, H.G. Unsteady one-dimensional settling of suspended sediment. Water Resour. Res. 1981, 17, 1125–1132. [CrossRef]

Ziegler, C.K.; Nisbet, B.S. Long-term simulation of fine-grained sediment transport in large reservoir. J. Hydraul. Eng. 1995, 121, 773–781. [CrossRef]

Olsen, N.R.B. Two-dimensional numerical modelling of flushing processes in water reservoirs. J. Hydraul. Res. 1999, 37, 3–16. [CrossRef]

Saad, N.Y.; Fattouh, E.M. Hydraulic characteristics of flow over weirs with circular openings. Ain Shams Eng. J. 2017, 8, 515–522. [CrossRef]

Bagheri, S.; Kabiri-Samani, A.R. Hydraulic Characteristics of flow over the streamlined weirs. Modares Civ. Eng. J. 2018, 17, 29–42.

Hussain, Z.; Khan, S.; Ullah, A.; Ayaz, M.; Ahmad, I.; Mashwani, W.K.; Chu, Y.-M. Extension of optimal homotopy asymptotic method with use of Daftardar–Jeffery polynomials to Hirota–Satsuma coupled system of Korteweg–de Vries equations. Open Phys. 2020, 18, 916–924. [CrossRef]

Arifeen, S.U.; Haq, S.; Ghafoor, A.; Ullah, A.; Kumam, P.; Chaipanya, P. Numerical solutions of higher order boundary value problems via wavelet approach. Adv. Differ. Equ. 2021, 2021, 347. [CrossRef]

Sharafati, A.; Haghbin, M.; Motta, D.; Yaseen, Z.M. The application of soft computing models and empirical formulations for hydraulic structure scouring depth simulation: A comprehensive review, assessment and possible future research direction. Arch. Comput. Methods Eng. 2021, 28, 423–447. [CrossRef]

Khan, S.; Selim, M.M.; Khan, A.; Ullah, A.; Abdeljawad, T.; Ayaz, M.; Mashwani, W.K. On the analysis of the non-Newtonian fluid flow past a stretching/shrinking permeable surface with heat and mass transfer. Coatings 2021, 11, 566. [CrossRef]

Khan, S.; Selim, M.M.; Gepreel, K.A.; Ullah, A.; Ayaz, M.; Mashwani, W.K.; Khan, E. An analytical investigation of the mixed convective Casson fluid flow past a yawed cylinder with heat transfer analysis. Open Phys. 2021, 19, 341–351. [CrossRef]

Ullah, A.; Selim, M.M.; Abdeljawad, T.; Ayaz, M.; Mlaiki, N.; Ghafoor, A. A Magnetite–Water-Based Nanofluid Three-Dimensional Thin Film Flow on an Inclined Rotating Surface with Non-Linear Thermal Radiations and Couple Stress Effects. Energies 2021, 14, 5531. [CrossRef]

Aamir, M.; Ahmad, Z.; Pandey, M.; Khan, M.A.; Aldrees, A.; Mohamed, A. The Effect of Rough Rigid Apron on Scour Downstream of Sluice Gates. Water 2022, 14, 2223. [CrossRef]

Gharebagh, B.A.; Bazargan, J.; Mohammadi, M. Experimental Investigation of Bed Scour Rate in Flood Conditions. Environ. Water Eng. 2022, in press. [CrossRef]

Laishram, K.; Devi, T.T.; Singh, N.B. Experimental Comparison of Hydraulic Jump Characteristics and Energy Dissipation Between Sluice Gate and Radial Gate. In Innovative Trends in Hydrological and Environmental Systems; Springer: Berlin/Heidelberg, Germany, 2022; pp. 207–218.

Varaki, M.E.; Sedaghati, M.; Sabet, B.S. Effect of apron length on local scour at the downstream of grade control structures with labyrinth planform. Arab. J. Geosci. 2022, 15, 1240. [CrossRef]

Rizk, D.; Ullah, A.; Elattar, S.; Alharbi, K.A.M.; Sohail, M.; Khan, R.; Khan, A.; Mlaiki, N. Impact of the KKL Correlation Model on the Activation of Thermal Energy for the Hybrid Nanofluid (GO+ ZnO+ Water) Flow through Permeable Vertically Rotating Surface. Energies 2022, 15, 2872. [CrossRef]

Kim, K.H.; Choi, G.W.; Jo, J.B. An Experimental Study on the Stream Flow by Discharge Ratio. Korea Water Resour. Assoc. Acad. Conf. 2005, 05b, 377–382.

Lee, D.S.; Yeo, H.G. An Experimental Study for Determination of the Material Diameter of Riprap Bed Protection Structure. Korea Water Resour. Assoc. Acad. Conf. 2005, 05b, 1036–1039.

Choi, G.W.; Byeon, S.J.; Kim, Y.G.; Cho, S.U. The Flow Characteristic Variation by Installing a Movable Weir having Water Drainage Equipment on the Bottom. J. Korean Soc. Hazard Mitig. 2008, 8, 117–122.

Jung, J.G. An Experimental Study for Estimation of Bed Protection Length. J. Korean Wetl. Soc. 2011, 13, 677–686.

Kim, S.H.; Kim, W.; Lee, E.R.; Choi, G.H. Analysis of Hydraulic Effects of Singok Submerged Weir in the Lower Han River. J. Korean Water Resour. Assoc. 2005, 38, 401–413. [CrossRef]

Kim, J.H.; Sim, M.P.; Choi, G.W.; Oh, J.M. Hydraulic Analysis of Air Entrainment by Weir Types. J. Korean Water Resour. Assoc. 2003, 36, 971–984. [CrossRef]

Jeong, S.; Yeo, C.G.; Yun, G.S.; Lee, S.O. Analysis of Characteristics for Bank Scour around Low Dam using 3D Numerical Simulation. Korean Soc. Hazard Mitig. Acad. Conf. 2011, 02a, 102.

Son, A.R.; Kim, B.H.; Moon, B.R.; Han, G.Y. An Analysis of Bed Change Characteristics by Bed Protection Work. J. Korean Soc. Civ. Eng. 2015, 35, 821–834.

French, R.H.; French, R.H. Open-Channel Hydraulics; McGraw-Hill: New York, NY, USA, 1985; ISBN 0070221340.

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.

Review of the governing equations in software Flow 3D

Continuity equation at three-dimensional Cartesian coordinates is given as Eq (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).

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.

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 cm^{3}.

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

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.

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.

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.

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.

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.

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.

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.

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.

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 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.

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 391–430ChapterGoogle Scholar

Chatila J, Tabbara M (2004) Computational modeling of flow over an ogee spillway. Comput Struct 82:1805–1812ArticleGoogle 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 11:127–132ArticleGoogle 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 9:161–169ArticleGoogle 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 16:265–272ArticleGoogle 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 22:680–686ArticleGoogle 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 487–498ChapterGoogle Scholar

Montagna F, Bellotti G, Di Risio M (2011) 3D numerical modeling of landslide-generated tsunamis around a conical island. Nat Hazards 58:591–608ArticleGoogle 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 29:973–985ArticleGoogle 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 1:1–6ArticleGoogle 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 21:415–422ArticleGoogle Scholar

Suprapto M (2013) Increase spillway capacity using Labyrinth Weir. Procedia Eng 54:440–446ArticleGoogle Scholar

Tabbara M, Chatila J, Awwad R (2005) Computational simulation of flow over stepped spillways. Comput Struct 83:2215–2224ArticleGoogle 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 15:561–565ArticleGoogle Scholar

Zhenwei MU, Zhiyan Z, Tao Z (2012) Numerical simulation of 3-D flow field of spillway based on VOF method. Procedia Eng 28:808–812ArticleGoogle Scholar

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

Numerical simulation of ship waves in the presence of a uniform current

CongfangAiYuxiangMaLeiSunGuohaiDongState Key Laboratory of Coastal and Offshore Engineering, Dalian University of Technology, Dalian, 116024, China

Highlights

• Ship waves in the presence of a uniform current are studied by a non-hydrostatic model.

• Effects of a following current on characteristic wave parameters are investigated.

• Effects of an opposing current on characteristic wave parameters are investigated.

• The response of the maximum water level elevation to the ship draft is discussed.

Abstract

이 논문은 균일한 해류가 존재할 때 선박파의 생성 및 전파를 시뮬레이션하기 위한 비정역학적 모델을 제시합니다. 선박 선체의 움직임을 표현하기 위해 움직이는 압력장 방법이 모델에 통합되었습니다.

뒤따르거나 반대 방향의 균일한 흐름이 있는 경우의 선박 파도의 수치 결과를 흐름이 없는 선박 파도의 수치 결과와 비교합니다. 추종 또는 반대 균일 전류가 존재할 때 계산된 첨단선 각도는 분석 솔루션과 잘 일치합니다. 추종 균일 전류와 반대 균일 전류가 특성파 매개변수에 미치는 영향을 제시하고 논의합니다.

선박 흘수에 대한 최대 수위 상승의 응답은 추종 또는 반대의 균일한 흐름이 있는 경우에도 표시되며 흐름이 없는 선박 파도의 응답과 비교됩니다. 선박 선체 측면의 최대 수위 상승은 Froude 수 Fr’=Us/gh의 특정 범위에 대해 다음과 같은 균일한 흐름의 존재에 의해 증가될 수 있음이 밝혀졌습니다.

여기서 Us는 선박 속도이고 h는 물입니다. 깊이. 균일한 해류를 무시하면 추종류나 반대류가 존재할 때 선박 흘수에 대한 최대 수위 상승의 응답이 과소평가될 수 있습니다.

본 연구는 선박파의 해석에 있어 균일한 해류의 영향을 고려해야 함을 시사합니다.

This paper presents a non-hydrostatic model to simulate the generation and propagation of ship waves in the presence of a uniform current. A moving pressure field method is incorporated into the model to represent the movement of a ship hull. Numerical results of ship waves in the presence of a following or an opposing uniform current are compared with those of ship waves without current. The calculated cusp-line angles in the presence of a following or opposing uniform current agree well with analytical solutions. The effects of a following uniform current and an opposing uniform current on the characteristic wave parameters are presented and discussed. The response of the maximum water level elevation to the ship draft is also presented in the presence of a following or an opposing uniform current and is compared with that for ship waves without current. It is found that the maximum water level elevation lateral to the ship hull can be increased by the presence of a following uniform current for a certain range of Froude numbers Fr′=Us/gh, where Us is the ship speed and h is the water depth. If the uniform current is neglected, the response of the maximum water level elevation to the ship draft in the presence of a following or an opposing current can be underestimated. The present study indicates that the effect of a uniform current should be considered in the analysis of ship waves.

Ship waves, Non-hydrostatic model, Following current, Opposing current, Wave parameters

1. Introduction

Similar to wind waves, ships sailing across the sea can also create free-surface undulations ranging from ripples to waves of large size (Grue, 2017, 2020). Ship waves can cause sediment suspension and engineering structures damage and even pose a threat to flora and fauna living near the embankments of waterways (Dempwolff et al., 2022). It is quite important to understand ship waves in various environments. The study of ship waves has been conducted over a century. A large amount of research (Almström et al., 2021; Bayraktar and Beji, 2013; David et al., 2017; Ertekin et al., 1986; Gourlay, 2001; Havelock, 1908; Lee and Lee, 2019; Samaras and Karambas, 2021; Shi et al., 2018) focused on the generation and propagation of ship waves without current. When a ship navigates in the sea or in a river where tidal flows or river flows always exist, the effect of currents should be taken into account. However, the effect of currents on the characteristic parameters of ship waves is still unclear, because very few publications have been presented on this topic.

Over the past two decades, many two-dimensional (2D) Boussinesq-type models (Bayraktar and Beji, 2013; Dam et al., 2008; David et al., 2017; Samaras and Karambas, 2021; Shi et al., 2018) were developed to examine ship waves. For example, Bayraktar and Beji (2013) solved Boussinesq equations with improved dispersion characteristics to simulate ship waves due to a moving pressure field. David et al. (2017) employed a Boussinesq-type model to investigate the effects of the pressure field and its propagation speed on characteristic wave parameters. All of these Boussinesq-type models aimed to simulate ship waves without current except for that of Dam et al. (2008), who investigated the effect of currents on the maximum wave height of ship waves in a narrow channel.

In addition to Boussinesq-type models, numerical models based on the Navier-Stokes equations (NSE) or Euler equations are also capable of resolving ship waves. Lee and Lee (2019, 2021) employed the FLOW-3D model to simulate ship waves without current and ship waves in the presence of a uniform current to confirm their equations for ship wave crests. FLOW-3D is a computational fluid dynamics (CFD) software based on the NSE, and the volume of fluid (VOF) method is used to capture the moving free surface. However, VOF-based NSE models are computationally expensive due to the treatment of the free surface. To efficiently track the free surface, non-hydrostatic models employ the so-called free surface equation and can be solved efficiently. One pioneering application for the simulation of ship waves by the non-hydrostatic model was initiated by Ma (2012) and named XBeach. Recently, Almström et al. (2021) validated XBeach with improved dispersive behavior by comparison with field measurements. XBeach employed in Almström et al. (2021) is a 2-layer non-hydrostatic model and is accurate up to Kh=4 for the linear dispersion relation (de Ridder et al., 2020), where K=2π/L is the wavenumber. L is the wavelength, and h is the still water depth. However, no applications of non-hydrostatic models on the simulation of ship waves in the presence of a uniform current have been published. For more advances in the numerical modelling of ship waves, the reader is referred to Dempwolff et al. (2022).

This paper investigates ship waves in the presence of a uniform current by using a non-hydrostatic model (Ai et al., 2019), in which a moving pressure field method is incorporated to represent the movement of a ship hull. The model solves the incompressible Euler equations by using a semi-implicit algorithm and is associated with iterating to solve the Poisson equation. The model with two, three and five layers is accurate up to Kh= 7, 15 and 40, respectively (Ai et al., 2019) in resolving the linear dispersion relation. To the best of our knowledge, ship waves in the presence of currents have been studied theoretically (Benjamin et al., 2017; Ellingsen, 2014; Li and Ellingsen, 2016; Li et al., 2019.) and numerically (Dam et al., 2008; Lee and Lee, 2019, 2021). However, no publications have presented the effects of a uniform current on characteristic wave parameters except for Dam et al. (2008), who investigated only the effect of currents on the maximum wave height in a narrow channel for the narrow relative Froude number Fr=(Us−Uc)/gh ranging from 0.47 to 0.76, where Us is the ship speed and Uc is the current velocity. To reveal the effect of currents on the characteristic parameters of ship waves, the main objectives of this paper are (1) to validate the capability of the proposed model to resolve ship waves in the presence of a uniform current, (2) to investigate the effects of a following or an opposing current on characteristic wave parameters including the maximum water level elevation and the leading wave period in the ship wave train, (3) to show the differences in characteristic wave parameters between ship waves in the presence of a uniform current and those without current when the same relative Froude number Fr is specified, and (4) to examine the response of the maximum water level elevation to the ship draft in the presence of a uniform current.

The remainder of this paper is organized as follows. The non-hydrostatic model for ship waves is described in Section 2. Section 3 presents numerical validations for ship waves. Numerical results and discussions about the effects of a uniform current on characteristic wave parameters are provided in Section 4, and a conclusion is presented in Section 5.

2. Non-hydrostatic model for ship waves

2.1. Governing equations

The 3D incompressible Euler equations are expressed in the following form:(1)∂u∂x+∂v∂y+∂w∂z=0(2)∂u∂t+∂u2∂x+∂uv∂y+∂uw∂z=−∂p∂x(3)∂v∂t+∂uv∂x+∂v2∂y+∂vw∂z=−∂p∂y(4)∂w∂t+∂uw∂x+∂vw∂y+∂w2∂z=−∂p∂z−gwhere t is the time; u(x,y,z,t), v(x,y,z,t) and w(x,y,z,t) are the velocity components in the horizontal x, y and vertical z directions, respectively; p(x,y,z,t) is the pressure divided by a constant reference density; and g is the gravitational acceleration.

The pressure p(x,y,z,t) can be expressed as(5)p=ps+g(η−z)+qwhere ps(x,y,t) is the pressure at the free surface, η(x,y,t) is the free surface elevation, and q(x,y,z,t) is the non-hydrostatic pressure.

η(x,y,t) is calculated by the following free-surface equation:(6)∂η∂t+∂∂x∫−hηudz+∂∂y∫−hηvdz=0where z=−h(x,y) is the bottom surface.

For −L/2≤x’≤L/2,−B/2≤y’≤B/2(7)ps(x,y,t)|t=0=pm[1−cL(x′/L)4][1−cB(y′/B)2]exp[−a(y′/B)2]where x′=x−x0 and y′=y−y0. (x0,y0) is the center of the pressure field, pm is the peak pressure defined at (x0,y0), and L and B are the lengthwise and breadthwise parameters, respectively. cL, cB and a are set to 16, 2 and 16, respectively.

2.2. Numerical algorithms

In this study, the generation of ship waves is incorporated into the semi-implicit non-hydrostatic model developed by Ai et al. (2019). The 3D grid system used in the model is built from horizontal rectangular grids by adding horizontal layers. The horizontal layers are distributed uniformly along the water depth, which means the layer thickness is defined by Δz=(η+h)/Nz, where Nz is the number of horizontal layers.

In the solution procedure, the first step is to generate ship waves by implementing Eq. (7) together with the prescribed ship track. In the second step, Eqs. (1), (2), (3), (4) are solved by the pressure correction method, which can be subdivided into three stages. The first stage is to compute intermediate velocities un+1/2, vn+1/2, and wn+1/2 by solving Eqs. (2), (3), (4), which contain the non-hydrostatic pressure at the preceding time level. In the second stage, the Poisson equation for the non-hydrostatic pressure correction term is solved on the graphics processing unit (GPU) in conjunction with the conjugate gradient method. The third stage is to compute the new velocities un+1, vn+1, and wn+1 by correcting the intermediate values after including the non-hydrostatic pressure correction term. In the discretization of Eqs. (2), (3), the gradient terms of the water surface ∂η/∂x and ∂η/∂y are discretized by means of the semi-implicit method (Vitousek and Fringer, 2013), in which the implicitness factor θ=0.5 is used. The model is second-order accurate in time for free-surface flows. More details about the model can be found in Ai et al. (2019).

3. Model validation

In this section, we validate the proposed model in resolving ship waves. The numerical experimental conditions are provided in Table 1 and Table 2. In Table 2, Case A with the current velocity of Uc = 0.0 m/s represents ship waves without current. Both Case B and Case C correspond to the cases in the presence of a following current, while Case D and Case E represent the cases in the presence of an opposing current. The current velocities are chosen based on the observed currents at 40.886° N, 121.812° E, which is in the Liaohe Estuary. The measured data were collected from 14:00 on September 18 (GMT + 08:00) to 19:00 on September 19 in 2021. The maximum flood velocity is 1.457 m/s, and the maximum ebb velocity is −1.478 m/s. The chosen current velocities are between the maximum flood velocity and the maximum ebb velocity.

Table 1. Summary of ship speeds.

Case

Water depth h (m)

Ship speed Us (m/s)

Froude number Fr′=Us/gh

1

6.0

4.57

0.6

2

6.0

5.35

0.7

3

6.0

6.15

0.8

4

6.0

6.90

0.9

5

6.0

7.093

0.925

6

6.0

7.28

0.95

7

6.0

7.476

0.975

8

6.0

7.86

1.025

9

6.0

8.06

1.05

10

6.0

8.243

1.075

11

6.0

8.45

1.1

12

6.0

9.20

1.2

13

6.0

9.97

1.3

14

6.0

10.75

1.4

15

6.0

11.50

1.5

16

6.0

12.30

1.6

17

6.0

13.05

1.7

18

6.0

13.80

1.8

19

6.0

14.60

1.9

20

6.0

15.35

2.0

Table 2. Summary of current velocities.

Case

A

B

C

D

E

Current velocity Uc (m/s)

0.0

0.5

1.0

−0.5

−1.0

Notably, the Froude number Fr′=Us/gh presented in Table 1 is defined by the ship speed Us only and is different from the relative Froude number Fr when a uniform current is presented. According to the theory of Lee and Lee (2021), with the same relative Froude number, the cusp-line angles in the presence of a following or an opposing uniform current are identical to those without current. As a result, for the test cases presented in Table 1, Table 2, all calculated cusp-line angles follow the analytical solution of Havelock (1908), when the relative Froude number Fr is introduced.

As shown in Fig. 1, the dimensions of the computational domain are −420≤x≤420 m and −200≤y≤200 m, which are similar to those of David et al. (2017). The ship track follows the x axis and ranges from −384 m to 384 m. The ship hull is represented by Eq. (7), in which the length L and the beam B are set to 14.0 m and 7.0 m, respectively, and the peak pressure value is pm= 5000 Pa. In the numerical simulations, grid convergence tests reveal that the horizontal grid spacing of Δx=Δy= 1.0 m and two horizontal layers are adequate. The numerical results with different numbers of horizontal layers are shown in the Appendix.

Fig. 2, Fig. 3 compare the calculated cusp-line angles θc with the analytical solutions of Havelock (1908) for ship waves in the presence of a following uniform current and an opposing uniform current, respectively. The calculated cusp-line angles without current are also depicted in Fig. 2, Fig. 3. All calculated cusp-line angles are in good agreement with the analytical solutions, except that the model tends to underpredict the cusp-line angle for 0.9<Fr<1.0. Notably, a similar underprediction of the cusp-line angle can also be found in David et al. (2017).

4. Results and discussions

This section presents the effects of a following current and opposing current on the maximum water level elevation and the leading wave period in the wave train based on the test cases presented in Table 1, Table 2. Moreover, the response of the maximum water level elevation to the ship draft in the presence of a uniform current is examined.

4.1. Effects of a following current on characteristic wave parameters

To present the effect of a following current on the maximum wave height, the variations of the maximum water level elevation ηmax with the Froude number Fr′ at gauge points G1 and G2 are depicted in Fig. 4. The positions of gauge points G1 and G2 are shown in Fig. 1. The maximum water level elevation is an analogue to the maximum wave height and is presented in this study, because maximum wave heights at different positions away from the ship track vary throughout the wave train (David et al., 2017). In general, the variations of ηmax with the Froude number Fr′ in the three cases show a similar behavior, in which with the increase in Fr′, ηmax increases and then decreases. The presence of the following currents decreases ηmax for Fr′≤0.8 and Fr′≥1.2. Specifically, the following currents have a significant effect on ηmax for Fr′≤0.8. Notably, ηmax can be increased by the presence of the following currents for 0.9≤Fr′≤1.1. Compared with Case A, at location G1 ηmax is amplified 1.25 times at Fr′=0.925 in Case B and 1.31 times at Fr′=1.025 in Case C. Similarly, at location G2 ηmax is amplified 1.15 times at Fr′=1.025 in Case B and 1.11 times at Fr′=1.075 in Case C. The fact that ηmax can be increased by the presence of a following current for 0.9≤Fr′≤1.1 implies that if a following uniform current is neglected, then ηmax may be underestimated.

To show the effect of a following current on the wave period, Fig. 5 depicts the variation of the leading wave period Tp in the wave train at gauge point G2 with the Froude number Fr′. Similar to David et al. (2017), Tp is defined by the wave period of the first wave with a leading trough in the wave train. The leading wave periods for Fr′= 0.6 and 0.7 were not given in Case B and Case C, because the leading wave heights for Fr′= 0.6 and 0.7 are too small to discern the leading wave periods. Compared with Case A, the presence of a following current leads to a larger Tp for 0.925≤Fr′≤1.1 and a smaller Tp for Fr′≥1.3. For Fr′= 0.8 and 0.9, Tp in Case B is larger than that in Case A and Tp in Case C is smaller than that in Case A. In all three cases, Tp decreases with increasing Fr′ for Fr′>1.0. However, this decreasing trend becomes very gentle after Fr′≥1.4. Notably, as shown in Fig. 5, Fr′=1.2 tends to be a transition point at which the following currents have a very limited effect on Tp. Moreover, before the transition point, Tp in Case B and Case C are larger than that in Case A (only for 0.925≤Fr′≤1.2), but after the transition point the reverse is true.

As mentioned previously, the cusp-line angles for ship waves in the presence of a following or an opposing current are identical to those for ship waves only with the same relative Froude number Fr. However, with the same Fr, the characteristic parameters of ship waves in the presence of a following or an opposing current are quite different from those of ship waves without current. Fig. 6 shows the variations of the maximum water level elevation ηmax with Fr at gauge points G1 and G2 for ship waves in the presence of a following uniform current. Overall, the relationship curves between ηmax and Fr in Case B and Case C are lower than those in Case A. It is inferred that with the same Fr, ηmax in the presence of a following current is smaller than that without current. Fig. 7 shows the variation of the leading wave period Tp in the wave train at gauge point G2 with Fr for ship waves in the presence of a following uniform current. The overall relationship curves between Tp and Fr in Case B and Case C are also lower than those in Case A for 0.9≤Fr≤2.0. It can be inferred that with the same Fr, Tp in the presence of a following current is smaller than that without current for Fr≥0.9.

To compare the numerical results between the case of ship waves only and the case of ship waves in the presence of a following current with the same Fr, Fig. 8 shows the wave patterns for Fr=1.2. To obtain the case of ship waves in the presence of a following current with Fr=1.2, the ship speed Us=9.7 m/s and the current velocity Uc=0.5 m/s are adopted. Fig. 8 indicates that both the calculated cusp-line angles for the case of Us=9.2 m/s and Uc=0.0 m/s and the case of Us=9.7 m/s and Uc=0.5 m/s are equal to 56.5°, which follows the theory of Lee and Lee (2021). Fig. 9 depicts the comparison of the time histories of the free surface elevation at gauge point G2 for Fr=1.2 between the case of ship waves only and the case of ship waves in the presence of a following current. The time when the ship wave just arrived at gauge point G2 is defined as t′=0. Both the maximum water level elevation and the leading wave period in the case of Us=9.2 m/s and Uc=0.0 m/s are larger than those in the case of Us=9.7 m/s and Uc=0.5 m/s, which is consistent with the inferences based on Fig. 6, Fig. 7.

Fig. 10 shows the response of the maximum water level elevation ηmax to the ship draft at gauge point G2 for Fr′= 1.2 in the presence of a following uniform current. pm ranges from 2500 Pa to 40,000 Pa with an interval of Δp= 2500 Pa pm0= 2500 Pa represents a reference case. ηmax0 denotes the maximum water level elevation corresponding to the case of pm0= 2500 Pa. The best-fit linear trend lines obtained by linear regression analysis for the three responses are also depicted in Fig. 10. In general, all responses of ηmax to the ship draft show a linear relationship. The coefficients of determination for the three linear trend lines are R2= 0.9901, 0.9941 and 0.9991 for Case A, Case B and Case C, respectively. R2 is used to measure how close the numerical results are to the linear trend lines. The closer R2 is to 1.0, the more linear the numerical results tend to be. As a result, the relationship curve between ηmax and the ship draft in the presence of a following uniform current tends to be more linear than that without current. Notably, with the increase in pmpm0, ηmax increases faster in Case B and Case C than Case A. This implies that neglecting the following currents can lead to the underestimation of the response of ηmax to the ship draft.

4.2. Effects of an opposing current on characteristic wave parameters

Fig. 11 shows the variations of the maximum water level elevation ηmax with the Froude number Fr′ at gauge points G1 and G2 for ship waves in the presence of an opposing uniform current. The presence of opposing uniform currents leads to a significant reduction in ηmax at the two gauge points for 0.6≤Fr′≤2.0. Especially for Fr′=0.6, the decrease in ηmax is up to 73.8% in Case D and 78.4% in Case E at location G1 and up to 93.8% in Case D and 95.3% in Case E at location G2 when compared with Case A. Fig. 12 shows the variations of the leading wave period Tp at gauge point G2 with the Froude number Fr′ for ship waves in the presence of an opposing uniform current. The leading wave periods for Fr′= 0.6 and 0.7 were also not provided in Case D and Case E due to the small leading wave heights. In general, Tp decreases with increasing Fr′ in Case D and Case E for 0.8≤Fr′≤2.0. Tp in Case D and Case E are larger than that in Case A for Fr′≥1.0.

Fig. 13 depicts the variations of the maximum water level elevation ηmax with the relative Froude number Fr at gauge points G1 and G2 for ship waves in the presence of an opposing uniform current. Similar to Case B and Case C shown in Fig. 6, the overall relationship curves between ηmax and Fr in Case D and Case E are lower than those in Case A. This implies that with the same Fr, ηmax in the presence of an opposing current is also smaller than that without current. Fig. 14 depicts the variations of the leading wave period Tp in the wave train at gauge point G2 with Fr for ship waves in the presence of an opposing uniform current. Similar to Case B and Case C shown in Fig. 7, the overall relationship curves between Tp and Fr in Case D and Case E are lower than those in Case A for 0.9≤Fr≤2.0. This also implies that with the same Fr, Tp in the presence of an opposing current is smaller than that without current.

Fig. 15 shows a comparison of the wave pattern for Fr=1.2 between the case of ship waves only and the case of ship waves in the presence of an opposing current. The case of the ship wave in the presence of an opposing current with Fr=1.2 is obtained by setting the ship speed Us=8.7 m/s and the current velocity Uc=−0.5 m/s. As expected (Lee and Lee, 2021), both calculated cusp-line angles are identical. Fig. 16 depicts the comparison of the time histories of the free surface elevation at gauge point G2 for Fr=1.2 between the case of ship waves only and the case of ship waves in the presence of an opposing current. The maximum water level elevation in the case of Us=9.2 m/s and Uc=0.0 m/s is larger than that in the case of Us=8.7 m/s and Uc=−0.5 m/s, while the reverse is true for the leading wave period. Fig. 16 is consistent with the inferences based on Fig. 13, Fig. 14.

Fig. 17 depicts the response of the maximum water level elevation ηmax to the ship draft at gauge point G2 for Fr′= 1.2 in the presence of an opposing uniform current. Similarly, the response of ηmax to the ship draft in the presence of an opposing uniform current shows a linear relationship. The coefficients of determination for the three linear trend lines are R2= 0.9901, 0.9955 and 0.9987 for Case A, Case D and Case E, respectively. This indicates that the relationship curve between ηmax and the ship draft in the presence of an opposing uniform current also tends to be more linear than that without current. In addition, ηmax increases faster with increasing pmpm0 in Case D and Case E than Case A, implying that the response of ηmax to the ship draft can also be underestimated by neglecting opposing currents.

5. Conclusions

A non-hydrostatic model incorporating a moving pressure field method was used to investigate characteristic wave parameters for ship waves in the presence of a uniform current. The calculated cusp-line angles for ship waves in the presence of a following or an opposing uniform current were in good agreement with analytical solutions, demonstrating that the proposed model can accurately resolve ship waves in the presence of a uniform current.

The model results showed that the presence of a following current can result in an increase in the maximum water level elevation ηmax for 0.9≤Fr′≤1.1, while the presence of an opposing current leads to a significant reduction in ηmax for 0.6≤Fr′≤2.0. The leading wave period Tp can be increased for 0.925≤Fr′≤1.2 and reduced for Fr′≥1.3 due to the presence of a following current. However, the presence of an opposing current leads to an increase in Tp for Fr′≥1.0.

Although with the same relative Froude number Fr, the cusp-line angles for ship waves in the presence of a following or an opposing current are identical to those for ship waves without current, the maximum water level elevation ηmax and leading wave period Tp in the presence of a following or an opposing current are quite different from those without current. The present model results imply that with the same Fr, ηmax in the presence of a following or an opposing current is smaller than that without current for Fr≥0.6, and Tp in the presence of a following or an opposing current is smaller than that without current for Fr≥0.9.

The response of ηmax to the ship draft in the presence of a following current or an opposing current is similar to that without current and shows a linear relationship. However, the presence of a following or an opposing uniform current results in more linear responses of ηmax to the ship draft. Moreover, more rapid responses of ηmax to the ship draft are obtained when a following current or an opposing current is presented. This implies that the response of ηmax to the ship draft in the presence of a following current or an opposing current can be underestimated if the uniform current is neglected.

The present results have implications for ships sailing across estuarine and coastal environments, where river flows or tidal flows are significant. In these environments, ship waves can be larger than expected and the response of the maximum water level elevation to the ship draft may be more remarkable. The effect of a uniform current should be considered in the analysis of ship waves.

The present study considered only slender-body type ships. For different hull shapes, the effects of a uniform current on characteristic wave parameters need to be further investigated. Moreover, the effects of an oblique uniform current on ship waves need to be examined in future work.

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.

Acknowledgments

This research is financially supported by the National Natural Science Foundation of China (Grant No. 52171248, 51720105010, 51979029), LiaoNing Revitalization Talents Program (Grant No. XLYC1807010) and the Fundamental Research Funds for the Central Universities (Grant No. DUT21LK01).

Appendix. Numerical results with different numbers of horizontal layers

Fig. 18 shows comparisons of the time histories of the free surface elevation at gauge point G1 for Case B and Fr′= 1.2 between the three sets of numerical results with different numbers of horizontal layers. The maximum water level elevations ηmax obtained by Nz= 3 and 4 are 0.24% and 0.35% larger than ηmax with Nz= 2, respectively. Correspondingly, the leading wave periods Tp obtained by Nz= 3 and 4 are 0.45% and 0.55% larger than Tp with Nz= 2, respectively. In general, the three sets of numerical results are very close. To reduce the computational cost, two horizontal layers Nz= 2 were chosen for this study.

워블 전략이 포함된 펄스 레이저 용접(PLW) 방법을 사용하여 알루미늄 및 구리 이종 랩 조인트의 제조를 위한 최적의 가공 매개변수에 대해 실험 및 수치 조사가 수행됩니다. 피크 레이저 출력과 접선 용접 속도의 대표적인 조합 43개를 선택하기 위해 원형 패킹 설계 알고리즘이 먼저 사용됩니다.

선택한 매개변수는 PLW 프로세스의 전산유체역학(CFD) 모델에 제공되어 용융 풀 형상(즉, 인터페이스 폭 및 침투 깊이) 및 구리 농도를 예측합니다. 시뮬레이션 결과는 설계 공간 내에서 PLW 매개변수의 모든 조합에 대한 용융 풀 형상 및 구리 농도를 예측하기 위해 3개의 대리 모델을 교육하는 데 사용됩니다.

마지막으로, 대체 모델을 사용하여 구성된 처리 맵은 용융 영역에 균열이나 기공이 없고 향상된 기계적 및 전기적 특성이 있는 이종 조인트를 생성하는 PLW 매개변수를 결정하기 위해 세 가지 품질 기준에 따라 필터링됩니다.

제안된 최적화 접근법의 타당성은 최적의 용접 매개변수를 사용하여 생성된 실험 샘플의 전단 강도, 금속간 화합물(IMC) 형성 및 전기 접촉 저항을 평가하여 입증됩니다.

결과는 최적의 매개변수가 1209N의 높은 전단 강도와 86µΩ의 낮은 전기 접촉 저항을 생성함을 확인합니다. 또한 용융 영역에는 균열 및 기공과 같은 결함이 없습니다.

An experimental and numerical investigation is performed into the optimal processing parameters for the fabrication of aluminum and copper dissimilar lap joints using a pulsed laser welding (PLW) method with a wobble strategy. A circle packing design algorithm is first employed to select 43 representative combinations of the peak laser power and tangential welding speed. The selected parameters are then supplied to a computational fluidic dynamics (CFD) model of the PLW process to predict the melt pool geometry (i.e., interface width and penetration depth) and copper concentration. The simulation results are used to train three surrogate models to predict the melt pool geometry and copper concentration for any combination of the PLW parameters within the design space. Finally, the processing maps constructed using the surrogate models are filtered in accordance with three quality criteria to determine the PLW parameters that produce dissimilar joints with no cracks or pores in the fusion zone and enhanced mechanical and electrical properties. The validity of the proposed optimization approach is demonstrated by evaluating the shear strength, intermetallic compound (IMC) formation, and electrical contact resistance of experimental samples produced using the optimal welding parameters. The results confirm that the optimal parameters yield a high shear strength of 1209 N and a low electrical contact resistance of 86 µΩ. Moreover, the fusion zone is free of defects, such as cracks and pores.

CrossRefView Record in ScopusGoogle Scholar[11]S. Smith, J. Blackburn, M. Gittos, P. de Bono, and P. Hilton, “Welding of dissimilar metallic materials using a scanned laser beam,” in International Congress on Applications of Lasers & Electro-Optics, 2013, vol. 2013, no. 1: Laser Institute of America, pp. 493-502.

Welding strategies for joining copper and aluminum by fast oscillating, high quality laser beam

High-Power Laser Materials Processing: Applications, Diagnostics, and Systems IX, vol. 11273, International Society for Optics and Photonics (2020), p. 112730C

Experimental investigation on the effect of spot diameter on continuous-wave laser welding of copper and aluminum thin sheets for battery manufacturing

Systematic approach for determining optimal processing parameters to produce parts with high density in selective laser melting process

Int. J. Adv. Manuf. Technol., 105 (10) (2019), pp. 4443-4460 View PDF

CrossRefView Record in ScopusGoogle Scholar[32]A. Ascari, A. Fortunato, E. Liverani, and A. Lutey, “Application of different pulsed laser sources to dissimilar welding of Cu and Al alloys,” in Proceedings of Lasers in Manufacturing Conference (LIM), 2019.

에피택셜 과 등축 응고 사이의 경쟁은 적층 제조에서 실행되는 레이저 용융 동안 CMSX-4 단결정 초합금에서 조사되었습니다. 단일 트랙 레이저 스캔은 레이저 출력과 스캐닝 속도의 여러 조합으로 방향성 응고된 CMSX-4 합금의 분말 없는 표면에서 수행되었습니다. EBSD(Electron Backscattered Diffraction) 매핑은 새로운 방향의 식별을 용이하게 합니다. 영역 분율 및 공간 분포와 함께 융합 영역 내에서 핵을 형성한 “스트레이 그레인”은 충실도가 높은 전산 유체 역학 시뮬레이션을 사용하여 용융 풀 내의 온도 및 유체 속도 필드를 모두 추정했습니다. 이 정보를 핵 생성 모델과 결합하여 용융 풀에서 핵 생성이 발생할 확률이 가장 높은 위치를 결정했습니다. 금속 적층 가공의 일반적인 경험에 따라 레이저 용융 트랙의 응고된 미세 구조는 에피택셜 입자 성장에 의해 지배됩니다. 더 높은 레이저 스캐닝 속도와 더 낮은 출력이 일반적으로 흩어진 입자 감소에 도움이 되지만,그럼에도 불구하고 길쭉한 용융 풀에서 흩어진 입자가 분명했습니다.

The competition between epitaxial vs. equiaxed solidification has been investigated in CMSX-4 single crystal superalloy during laser melting as practiced in additive manufacturing. Single-track laser scans were performed on a powder-free surface of directionally solidified CMSX-4 alloy with several combinations of laser power and scanning velocity. Electron backscattered diffraction (EBSD) mapping facilitated identification of new orientations, i.e., “stray grains” that nucleated within the fusion zone along with their area fraction and spatial distribution. Using high-fidelity computational fluid dynamics simulations, both the temperature and fluid velocity fields within the melt pool were estimated. This information was combined with a nucleation model to determine locations where nucleation has the highest probability to occur in melt pools. In conformance with general experience in metals additive manufacturing, the as-solidified microstructure of the laser-melted tracks is dominated by epitaxial grain growth; nevertheless, stray grains were evident in elongated melt pools. It was found that, though a higher laser scanning velocity and lower power are generally helpful in the reduction of stray grains, the combination of a stable keyhole and minimal fluid velocity further mitigates stray grains in laser single tracks.

Introduction

니켈 기반 초합금은 고온에서 긴 노출 시간 동안 높은 인장 강도, 낮은 산화 및 우수한 크리프 저항성을 포함하는 우수한 특성의 고유한 조합으로 인해 가스 터빈 엔진 응용 분야에서 광범위하게 사용됩니다. CMSX-4는 특히 장기 크리프 거동과 관련하여 초고강도의 2세대 레늄 함유 니켈 기반 단결정 초합금입니다. ^{[ 1 , 2 ]}입계의 존재가 크리프를 가속화한다는 인식은 가스 터빈 엔진의 고온 단계를 위한 단결정 블레이드를 개발하게 하여 작동 온도를 높이고 효율을 높이는 데 기여했습니다. 이러한 구성 요소는 사용 중 마모될 수 있습니다. 즉, 구성 요소의 무결성을 복원하고 단결정 미세 구조를 유지하는 수리 방법을 개발하기 위한 지속적인 작업이 있었습니다. ^{[ 3 , 4 , 5 ]}

적층 제조(AM)가 등장하기 전에는 다양한 용접 공정을 통해 단결정 초합금에 대한 수리 시도가 수행되었습니다. ^{균열 [ 6 , 7 ]} 및 흩어진 입자 ^{[ 8 , 9 ]} 와 같은 심각한 결함 이 이 수리 중에 자주 발생합니다. 일반적으로 “스트레이 그레인”이라고 하는 응고 중 모재의 방향과 다른 결정학적 방향을 가진 새로운 그레인의 형성은 니켈 기반 단결정 초합금의 수리 중 유해한 영향으로 인해 중요한 관심 대상입니다. ^{[ 3 , 10 ]}결과적으로 재료의 단결정 구조가 손실되고 원래 구성 요소에 비해 기계적 특성이 손상됩니다. 이러한 흩어진 입자는 특정 조건에서 에피택셜 성장을 대체하는 등축 응고의 시작에 해당합니다.

떠돌이 결정립 형성을 완화하기 위해 이전 작업은 용융 영역(FZ) 내에서 응고하는 동안 떠돌이 결정립 형성에 영향을 미치는 수지상 응고 거동 및 처리 조건을 이해하는 데 중점을 두었습니다. ^{[ 11 , 12 , 13 , 14 ]} 연구원들은 단결정 합금의 용접 중에 표류 결정립 형성에 대한 몇 가지 가능한 메커니즘을 제안했습니다. ^{[ 12 , 13 , 14 , 15 ]}응고 전단에 앞서 국부적인 구성 과냉각은 이질적인 핵 생성 및 등축 결정립의 성장을 유발할 수 있습니다. 또한 용융 풀에서 활발한 유체 흐름으로 인해 발생하는 덴드라이트 조각화는 용융 풀 경계 근처에서 새로운 결정립을 형성할 수도 있습니다. 두 메커니즘 모두에서, 표류 결정립 형성은 핵 생성 위치에 의존하며, 차이점은 수상 돌기 조각화는 수상 돌기 조각이 핵 생성 위치로 작용한다는 것을 의미하는 반면 다른 메커니즘은 재료, 예 를 들어 산화물 입자에서 발견되는 다른 유형의 핵 생성 위치를 사용한다는 것을 의미합니다. 잘 알려진 바와 같이, 많은 주물에 대한 반대 접근법은 TiB와 같은 핵제의 도입을 통해 등축 응고를 촉진하는 것입니다.22알루미늄 합금에서.

헌법적 과냉 메커니즘에서 Hunt ^{[ 11 ]} 는 정상 상태 조건에서 기둥에서 등축으로의 전이(CET)를 설명하는 모델을 개발했습니다. Gaumann과 Kurz는 Hunt의 모델을 수정하여 단결정이 응고되는 동안 떠돌이 결정립이 핵을 생성하고 성장할 수 있는 정도를 설명했습니다. ^{[ 12 , 14 ]} 이후 연구에서 Vitek은 Gaumann의 모델을 개선하고 출력 및 스캐닝 속도와 같은 용접 조건의 영향에 대한 보다 자세한 분석을 포함했습니다. Vitek은 또한 실험 및 모델링 기술을 통해 표류 입자 형성에 대한 기판 방향의 영향을 포함했습니다. ^{[ 3 , 10 ]}일반적으로 높은 용접 속도와 낮은 출력은 표류 입자의 양을 최소화하고 레이저 용접 공정 중 에피택셜 단결정 성장을 최대화하는 것으로 나타났습니다. ^{[ 3,10 ] 그러나 Vitek은 덴드라이트 조각화를 고려하지 않았으며 그의 연구는 불균질 핵형성이 레이저 용접된 CMSX} -4 단결정 합금에서 표류 결정립 형성을 이끄는 주요 메커니즘임을 나타냅니다. 현재 작업에서 Vitek의 수치적 방법이 채택되고 금속 AM의 급속한 특성의 더 높은 속도와 더 낮은 전력 특성으로 확장됩니다.

AM을 통한 금속 부품 제조 는 지난 10년 동안 급격한 인기 증가를 목격했습니다. ^{[ 16 ]} EBM(Electron Beam Melting)에 의한 CMSX-4의 제작 가능성은 자주 조사되었으나 ^{[ 17 , 18 , 19 , 20 , 21 ]} CMSX의 제조 및 수리에 대한 조사는 매우 제한적이었다. – 4개의 단결정 구성요소는 레이저 분말 베드 융합(LPBF)을 사용하며, AM의 인기 있는 하위 집합으로, 특히 표류 입자 형성을 완화하는 메커니즘과 관련이 있습니다. ^{[ 22 ]}이러한 조사 부족은 주로 이러한 합금 시스템과 관련된 처리 문제로 인해 발생합니다. ^{[ 2 , 19 , 22 , 23 , 24 ]} 공정 매개변수( 예: 열원 전력, 스캐닝 속도, 스폿 크기, 예열 온도 및 스캔 전략)의 엄격한 제어는 완전히 조밀한 부품을 만들고 유지 관리할 수 있도록 하는 데 필수적입니다. 단결정 미세구조. ^{[ 25 ]} EBM을 사용하여 단결정 합금의 균열 없는 수리가 현재 가능하지만 ^{[ 19 , 24 ]} 표류 입자를 생성하지 않는 수리는 쉽게 달성할 수 없습니다.^{[ 23 , 26 ]}

이 작업에서 LPBF를 대표하는 조건으로 레이저 용융을 사용하여 단결정 CMSX-4에서 표류 입자 완화를 조사했습니다. LPBF는 스캐닝 레이저 빔을 사용하여 금속 분말의 얇은 층을 기판에 녹이고 융합합니다. ^{층별} 증착에서 레이저 빔의 사용은 급격한 온도 구배, 빠른 가열/냉각 주기 및 격렬한 유체 흐름을 경험하는 용융 풀을 생성 ^{합니다 }^{. }이것은 일반적으로 부품에 결함을 일으킬 수 있는 매우 동적인 물리적 현상으로 이어집니다. ^{[ 28 , 29 , 30 ]} 레이저 유도 키홀의 동역학( 예:, 기화 유발 반동 압력으로 인한 위상 함몰) 및 열유체 흐름은 AM 공정에서 응고 결함과 강하게 결합되고 관련됩니다. ^{[ 31 , 32 , 33 , 34 ]} 기하 구조의 급격한 변화가 발생하기 쉬운 불안정한 키홀은 다공성, 볼링, 스패터 형성 및 흔하지 않은 미세 구조 상을 포함하는 유해한 물리적 결함을 유발할 수 있습니다. 그러나 키홀 진화와 유체 흐름은 자연적으로 다음을 통해 포착 ^{하기}^{ 어렵 }습니다 ^{.}전통적인 사후 특성화 기술. 고충실도 수치 모델링을 활용하기 위해 이 연구에서는 전산유체역학(CFD)을 적용하여 표면 아래의 레이저-물질 상호 작용을 명확히 했습니다. ^{[ 36 ]} 이것은 응고된 용융물 풀의 단면에 대한 오랫동안 확립된 사후 특성화와 비교하여 키홀 및 용융물 풀 유체 흐름 정량화를 실행합니다.

CMSX-4 구성 요소의 레이저 기반 AM 수리 및 제조를 위한 적절한 절차를 개발하기 위해 적절한 공정 창을 설정하고 응고 중 표류 입자 형성 경향에 대한 예측 기능을 개발하는 것부터 시작합니다. 다중 합금에 대한 단일 트랙 증착은 분말 층이 있거나 없는 AM 공정에서 용융 풀 형상 및 미세 구조의 정확한 분석을 제공하는 것으로 나타났습니다. ^{[ 37 , 38 , 39 ]}따라서 본 연구에서는 CMSX-4의 응고 거동을 알아보기 위해 분말을 사용하지 않는 단일 트랙 레이저 스캔 실험을 사용하였다. 이는 CMSX-4 단결정의 LPBF 제조를 위한 예비 실험 지침을 제공합니다. 또한 응고 모델링은 기존 용접에서 LPBF와 관련된 급속 용접으로 확장되어 표류 입자 감소를 위한 최적의 레이저 용융 조건을 식별했습니다. 가공 매개변수 최적화를 위한 추가 지침을 제공하기 위해 용융물 풀의 매우 동적인 유체 흐름을 모델링했습니다.

재료 및 방법

단일 트랙 실험

방전 가공(EDM)을 사용하여 CMSX-4 방향성 응고 단결정 잉곳으로부터 샘플을 제작했습니다. 샘플의 최종 기하학은 치수 20의 직육면체 형태였습니다.××20××6mm. 6개 중 하나⟨ 001 ⟩⟨001⟩잉곳의 결정학적 방향은 레이저 트랙이 이 바람직한 성장 방향을 따라 스캔되도록 절단 표면에 수직으로 위치했습니다. 단일 레이저 용융 트랙은 EOS M290 기계를 사용하여 분말이 없는 샘플 표면에 만들어졌습니다. 이 기계는 최대 출력 400W, 가우시안 빔 직경 100의 이터븀 파이버 레이저가 장착된 LPBF 시스템입니다. μμ초점에서 m. 실험 중에 직사각형 샘플을 LPBF 기계용 맞춤형 샘플 홀더의 포켓에 끼워 표면을 동일한 높이로 유지했습니다. 이 맞춤형 샘플 홀더에 대한 자세한 내용은 다른 곳에서 설명합니다. ^{실험} 은 아르곤 퍼지 분위기에서 수행되었으며 예열은 적용되지 ^{않았습니다 }^{. }단일 트랙 레이저 용융 실험은 다양한 레이저 출력(200~370W)과 스캔 속도(0.4~1.4m/s)에서 수행되었습니다.

성격 묘사

레이저 스캐닝 후, 레이저 빔 스캐닝 방향에 수직인 평면에서 FZ를 통해 다이아몬드 톱을 사용하여 샘플을 절단했습니다. 그 후, 샘플을 장착하고 220 그릿 SiC 페이퍼로 시작하여 콜로이드 실리카 현탁액 광택제로 마무리하여 자동 연마했습니다. 결정학적 특성화는 20kV의 가속 전압에서 TESCAN MIRA 3XMH 전계 방출 주사 전자 현미경(SEM)에서 수행되었습니다. EBSD 지도는0.4μm _0.4μ미디엄단계 크기. Bruker 시스템을 사용하여 EBSD 데이터를 정리하고 분석했습니다. EBSD 클린업은 그레인을 접촉시키기 위한 그레인 확장 루틴으로 시작한 다음 인덱스되지 않은 회절 패턴과 관련된 검은색 픽셀을 해결하기 위해 이웃 방향 클린업 루틴으로 이어졌습니다. 용융 풀 형태를 분석하기 위해 단면을 광학 현미경으로 분석했습니다. 광학 특성화의 대비를 향상시키기 위해 10g CuSO로 구성된 Marbles 시약의 변형으로 샘플을 에칭했습니다.44, 50mL HCl 및 70mL H22영형.

응고 모델링

구조적 과냉 기준에 기반한 응고 모델링을 수행하여 표유 입자의 성향 및 분포에 대한 가공 매개변수의 영향을 평가했습니다. 이 분석 모델링 접근 방식에 대한 자세한 내용은 이전 작업에서 제공됩니다. ^{[ 3 , 10 ] 참고문헌 }3 에 기술된 바와 같이 , 기본 재료의 결정학적 배향을 가진 용융 풀에서 총 표유 입자 면적 분율의 변화는 최소이므로 기본 재료 배향의 영향은 이 작업에서 고려되지 않았습니다. 우리의 LPBF 결과를 이전 작업과 비교하기 위해 Vitek의 작업에서 사용된 수학적으로 간단한 Rosenthal 방정식 ^{[ 3 ]}또한 레이저 매개변수의 함수로 용융 풀의 모양과 FZ의 열 조건을 계산하기 위한 기준으로 여기에서 채택되었습니다. Rosenthal 솔루션은 열이 일정한 재료 특성을 가진 반무한 판의 정상 상태 점원을 통해서만 전도를 통해 전달된다고 가정하며 일반적으로 다음과 같이 표현 됩니다 ^{[ 40 , 41 ]} .

티=티0+η피2 파이케이엑스2+와이2+지2———-√경험치[- 브이(엑스2+와이2+지2———-√− 엑스 )2α _] ,티=티0+η피2파이케이엑스2+와이2+지2경험치[-V(엑스2+와이2+지2-엑스)2α],(1)

여기서 T 는 온도,티0티0본 연구에서 313K( 즉 , EOS 기계 챔버 온도)로 설정된 주변 온도, P 는 레이저 빔 파워, V 는 레이저 빔 스캐닝 속도,ηη는 레이저 흡수율, k 는 열전도율,αα베이스 합금의 열확산율입니다. x , y , z 는 각각 레이저 스캐닝 방향, 가로 방향 및 세로 방향의 반대 방향과 정렬된 방향입니다 . 이 직교 좌표는 참조 3 의 그림 1에 있는 시스템을 따랐습니다 . CMSX-4에 대한 고상선 온도(1603K)와 액상선 온도(1669K)의 등온선 평균으로 응고 프런트( 즉 , 고체-액체 계면)를 정의했습니다. ^{[ 42 , 43 , 44 ]} 시뮬레이션에 사용된 열물리적 특성은 표 I 에 나열되어 있습니다.표 I CMSX-4의 응고 모델링에 사용된 열물리적 특성

어디θθ는 스캔 방향과 응고 전면의 법선 방향( 즉 , 최대 열 흐름 방향) 사이의 각도입니다. 이 연구의 용접 조건과 같은 제한된 성장에서 수지상 응고 전면은 고체-액체 등온선의 속도로 성장하도록 강제됩니다.V티V티. ^{[ 46 ]}

응고 전선이 진행되기 전에 새로 핵 생성된 입자의 국지적 비율ΦΦ, 액체 온도 구배 G 에 의해 결정 , 응고 선단 속도V티V티및 핵 밀도N0N0. 고정된 임계 과냉각에서 모든 입자가 핵형성된다고 가정함으로써△티N△티N, 등축 결정립의 반경은 결정립이 핵 생성을 시작하는 시점부터 주상 전선이 결정립에 도달하는 시간까지의 성장 속도를 통합하여 얻습니다. 과냉각으로 대체 시간d (ΔT_) / dt = – _V티G디(△티)/디티=-V티G, 열 구배 G 사이의 다음 관계 , 등축 입자의 국부적 부피 분율ΦΦ, 수상 돌기 팁 과냉각ΔT _△티, 핵 밀도N0N0, 재료 매개변수 n 및 핵생성 과냉각△티N△티N, Gäumann 외 여러분 에 의해 파생되었습니다 . ^{[ 12 , 14 ]} Hunt의 모델 ^{[ 11 ]} 의 수정에 기반함 :

계산을 단순화하기 위해 덴드라이트 팁 과냉각을 전적으로 구성 과냉각의 것으로 추정합니다.△티씨△티씨, 멱법칙 형식으로 근사화할 수 있습니다.△티씨= ( _V티)1 / 엔△티씨=(ㅏV티)1/N, 여기서 a 와 n 은 재료 종속 상수입니다. CMSX-4의 경우 이 값은a = 1.25 ×106ㅏ=1.25×106 s K ^{3.4m}− 1-1,엔 = 3.4N=3.4, 그리고N0= 2 ×1015N0=2×1015미디엄− 3,-삼,참고문헌 3 에 의해 보고된 바와 같이 .△티N△티N2.5K이며 보다 큰 냉각 속도에서 응고에 대해 무시할 수 있습니다.106106 K/s. 에 대한 표현ΦΦ위의 방정식을 재배열하여 해결됩니다.

As proposed by Hunt,^{[11]} a value of Φ≤0.66Φ≤0.66 pct represents fully columnar epitaxial growth condition, and, conversely, a value of Φ≥49Φ≥49 pct indicates that the initial single crystal microstructure is fully replaced by an equiaxed microstructure. To calculate the overall stray grain area fraction, we followed Vitek’s method by dividing the FZ into roughly 19 to 28 discrete parts (depending on the length of the melt pool) of equal length from the point of maximum width to the end of melt pool along the x direction. The values of G and vTvT were determined at the center on the melt pool boundary of each section and these values were used to represent the entire section. The area-weighted average of ΦΦ over these discrete sections along the length of melt pool is designated as Φ¯¯¯¯Φ¯, and is given by:

Φ¯¯¯¯=∑kAkΦk∑kAk,Φ¯=∑kAkΦk∑kAk,

(6)

where k is the index for each subsection, and AkAk and ΦkΦk are the areas and ΦΦ values for each subsection. The summation is taken over all the sections along the melt pool. Vitek’s improved model allows the calculation of stray grain area fraction by considering the melt pool geometry and variations of G and vTvT around the tail end of the pool.

수년에 걸쳐 용융 풀 현상 모델링의 정확도를 개선하기 위해 많은 고급 수치 방법이 개발되었습니다. 우리는 FLOW-3D와 함께 고충실도 CFD를 사용했습니다. FLOW-3D는 여러 물리 모델을 통합하는 상용 FVM(Finite Volume Method)입니다. ^{[ 47 , 48 ]} CFD는 유체 운동과 열 전달을 수치적으로 시뮬레이션하며 여기서 사용된 기본 물리 모델은 레이저 및 표면력 모델이었습니다. 레이저 모델에서는 레이 트레이싱 기법을 통해 다중 반사와 프레넬 흡수를 구현합니다. ^{[ 36 ]}먼저, 레이저 빔은 레이저 빔에 의해 조명되는 각 그리드 셀을 기준으로 여러 개의 광선으로 이산화됩니다. 그런 다음 각 입사 광선에 대해 입사 벡터가 입사 위치에서 금속 표면의 법선 벡터와 정렬될 때 에너지의 일부가 금속에 의해 흡수됩니다. 흡수율은 Fresnel 방정식을 사용하여 추정됩니다. ^{나머지}^{ 에너지는 반사광선 에 의해 유지되며 }^{,} 반사광선은 재료 표면에 부딪히면 새로운 입사광선으로 처리됩니다. 두 가지 주요 힘이 액체 금속 표면에 작용하여 자유 표면을 변형시킵니다. 금속의 증발에 의해 생성된 반동 압력은 증기 억제를 일으키는 주요 힘입니다. 본 연구에서 사용된 반동 압력 모델은피아르 자형= 특급 _{ B ( 1- _티V/ 티) }피아르 자형=ㅏ경험치{비(1-티V/티)}, 어디피아르 자형피아르 자형는 반동압력, A 와 B 는 재료의 물성에 관련된 계수로 각각 75와 15이다.티V티V는 포화 온도이고 T 는 키홀 벽의 온도입니다. 표면 흐름 및 키홀 형성의 다른 원동력은 표면 장력입니다. 표면 장력 계수는 Marangoni 흐름을 포함하기 위해 온도의 선형 함수로 추정되며,σ =1.79-9.90⋅10− 4( 티− 1654케이 )σ=1.79-9.90⋅10-4(티-1654년케이)엔엠− 1-1. ^{[ 49 ]} 계산 영역은 베어 플레이트의 절반입니다(2300 μμ미디엄××250 μμ미디엄××500 μμm) xz 평면 에 적용된 대칭 경계 조건 . 메쉬 크기는 8입니다. μμm이고 시간 단계는 0.15입니다. μμs는 계산 효율성과 정확성 간의 균형을 제공합니다.

결과 및 논의

용융 풀 형태

이 작업에 사용된 5개의 레이저 파워( P )와 6개의 스캐닝 속도( V )는 서로 다른 29개의 용융 풀을 생성했습니다.피- 브이피-V조합. P 와 V 값이 가장 높은 것은 그림 1 을 기준으로 과도한 볼링과 관련이 있기 때문에 본 연구에서는 분석하지 않았다 .

단일 트랙 용융 풀은 그림 1 과 같이 형상에 따라 네 가지 유형으로 분류할 수 있습니다 ^{[ 39 ]} : (1) 전도 모드(파란색 상자), (2) 키홀 모드(빨간색), (3) 전환 모드(마젠타), (4) 볼링 모드(녹색). 높은 레이저 출력과 낮은 스캐닝 속도의 일반적인 조합인 키홀 모드에서 용융물 풀은 일반적으로 너비/깊이( W / D ) 비율이 0.5보다 훨씬 큰 깊고 가느다란 모양을 나타냅니다 . 스캐닝 속도가 증가함에 따라 용융 풀이 얕아져 W / D 가 약 0.5인 반원형 전도 모드 용융 풀을 나타냅니다. W / D _전환 모드 용융 풀의 경우 1에서 0.5 사이입니다. 스캐닝 속도를 1200 및 1400mm/s로 더 높이면 충분히 큰 캡 높이와 볼링 모드 용융 풀의 특징인 과도한 언더컷이 발생할 수 있습니다.

힘과 속도의 함수로서의 용융 풀 깊이와 너비는 각각 그림 2 (a)와 (b)에 표시되어 있습니다. 용융 풀 폭은 기판 표면에서 측정되었습니다. 그림 2 (a)는 깊이가 레이저 출력과 매우 선형적인 관계를 따른다는 것을 보여줍니다. 속도가 증가함에 따라 깊이 대 파워 곡선의 기울기는 꾸준히 감소하지만 더 높은 속도 곡선에는 약간의 겹침이 있습니다. 이러한 예상치 못한 중첩은 종종 용융 풀 형태의 동적 변화를 유발하는 유체 흐름의 영향과 레이저 스캔당 하나의 이미지만 추출되었다는 사실 때문일 수 있습니다. 이러한 선형 동작은 그림 2 (b) 의 너비에 대해 명확하지 않습니다 . 그림 2(c)는 선형 에너지 밀도 P / V 의 함수로서 용융 깊이와 폭을 보여줍니다 . 선형 에너지 밀도는 퇴적물의 단위 길이당 에너지 투입량을 측정한 것입니다. ^{[ 50 ]} 용융 풀 깊이는 에너지 밀도에 따라 달라지며 너비는 더 많은 분산을 나타냅니다. 동일한 에너지 밀도가 준공 부품의 용융 풀, 미세 구조 또는 속성에서 반드시 동일한 유체 역학을 초래하지는 않는다는 점에 유의하는 것이 중요합니다. ^{[ 50 ]}

레이저 흡수율 평가

레이저 흡수율은 LPBF 조건에서 재료 및 가공 매개변수에 따라 크게 달라진다는 것은 잘 알려져 있습니다. ^{[ 31 , 51 , 52 ]} 적분구를 이용한 전통적인 흡수율의 직접 측정은 일반적으로 높은 비용과 구현의 어려움으로 인해 쉽게 접근할 수 없습니다. ^{[ 51 ]} 그 외 . ^{[ 39 ]} 전도 모드 용융 풀에 대한 Rosenthal 방정식을 기반으로 경험적 레이저 흡수율 모델을 개발했지만 기본 가정으로 인해 키홀 용융 풀에 대한 정확한 예측을 제공하지 못했습니다. ^{[ 40 ]} 최근 간외 . ^{[ 53 ]} Ti–6Al–4V에 대한 30개의 고충실도 다중 물리 시뮬레이션 사례를 사용하여 레이저 흡수에 대한 스케일링 법칙을 확인했습니다. 그러나 연구 중인 특정 재료에 대한 최소 흡수(평평한 용융 표면의 흡수율)에 대한 지식이 필요하며 이는 CMSX-4에 대해 알려지지 않았습니다. 다양한 키홀 모양의 용융 풀에 대한 레이저 흡수의 정확한 추정치를 얻기가 어렵기 때문에 상한 및 하한 흡수율로 분석 시뮬레이션을 실행하기로 결정했습니다. 깊은 키홀 모양의 용융 풀의 경우 대부분의 빛을 가두는 키홀 내 다중 반사로 인해 레이저 흡수율이 0.8만큼 높을 수 있습니다. 이것은 기하학적 현상이며 기본 재료에 민감하지 않습니다. ^{[ 51, 52 , 54 ]} 따라서 본 연구에서는 흡수율의 상한을 0.8로 설정하였다. 참고 문헌 51 에 나타낸 바와 같이 , 전도 용융 풀에 해당하는 최저 흡수율은 약 0.3이었으며, 이는 이 연구에서 합리적인 하한 값입니다. 따라서 레이저 흡수율이 스트레이 그레인 형성에 미치는 영향을 보여주기 위해 흡수율 값을 0.55 ± 0.25로 설정했습니다. Vitek의 작업에서는 1.0의 고정 흡수율 값이 사용되었습니다. ^{[ 3 ]}

퓨전 존 미세구조

그림 3 은 200~300W 및 600~300W 및 600~300W 범위의 레이저 출력 및 속도로 9가지 다른 처리 매개변수에 의해 생성된 CMSX-4 레이저 트랙의 yz 단면 에서 취한 EBSD 역극점도와 해당 역극점도를 보여 줍니다. 각각 1400mm/s. EBSD 맵에서 여러 기능을 쉽게 관찰할 수 있습니다. 스트레이 그레인은 EBSD 맵에서 그 방향에 해당하는 다른 RGB 색상으로 나타나고 그레인 경계를 묘사하기 위해 5도의 잘못된 방향이 사용되었습니다. 여기, 그림 3 에서 스트레이 그레인은 대부분 용융 풀의 상단 중심선에 집중되어 있으며, 이는 용접된 단결정 CMSX-4의 이전 보고서와 일치합니다. ^{[ 10 ]}역 극점도에서, 점 근처에 집중된 클러스터⟨ 001 ⟩⟨001⟩융합 경계에서 유사한 방향을 유지하는 단결정 기반 및 에피택셜로 응고된 덴드라이트를 나타냅니다. 그러나 흩어진 곡물은 식별할 수 있는 질감이 없는 흩어져 있는 점으로 나타납니다. 단결정 기본 재료의 결정학적 방향은 주로⟨ 001 ⟩⟨001⟩비록 샘플을 절단하는 동안 식별할 수 없는 기울기 각도로 인해 또는 단결정 성장 과정에서 약간의 잘못된 방향이 있었기 때문에 약간의 편차가 있지만. 용융 풀 내부의 응고된 수상 돌기의 기본 방향은 다시 한 번⟨ 001 ⟩⟨001⟩주상 결정립 구조와 유사한 에피택셜 성장의 결과. 그림 3 과 같이 용융 풀에서 수상돌기의 성장 방향은 하단의 수직 방향에서 상단의 수평 방향으로 변경되었습니다 . 이 전이는 주로 온도 구배 방향의 변화로 인한 것입니다. 두 번째 전환은 CET입니다. FZ의 상단 중심선 주변에서 다양한 방향의 흩어진 입자가 관찰되며, 여기서 안쪽으로 성장하는 수상돌기가 서로 충돌하여 용융 풀에서 응고되는 마지막 위치가 됩니다.

더 깊은 키홀 모양을 특징으로 하는 샘플에서 용융 풀의 경계 근처에 침전된 흩어진 입자가 분명합니다. 이러한 새로운 입자는 나중에 모델링 섹션에서 논의되는 수상돌기 조각화 메커니즘에 의해 잠재적으로 발생합니다. 결정립이 강한 열 구배에서 핵을 생성하고 성장한 결과, 대부분의 흩어진 결정립은 모든 방향에서 동일한 크기를 갖기보다는 장축이 열 구배 방향과 정렬된 길쭉한 모양을 갖습니다. 그림 3 의 전도 모드 용융 풀 흩어진 입자가 없는 것으로 입증되는 더 나은 단결정 품질을 나타냅니다. 상대적으로 낮은 출력과 높은 속도의 스캐닝 레이저에 의해 생성된 이러한 더 얕은 용융 풀에서 최소한의 결정립 핵형성이 발생한다는 것은 명백합니다. 더 큰 면적 분율을 가진 스트레이 그레인은 고출력 및 저속으로 생성된 깊은 용융 풀에서 더 자주 관찰됩니다. 국부 응고 조건에 대한 동력 및 속도의 영향은 후속 모델링 섹션에서 조사할 것입니다.

응고 모델링

서론에서 언급한 바와 같이 연구자들은 단결정 용접 중에 표류 결정립 형성의 가능한 메커니즘을 평가했습니다. ^{[ 12 , 13 , 14 , 15 , 55 ]}논의된 가장 인기 있는 두 가지 메커니즘은 (1) 응고 전단에 앞서 구성적 과냉각에 의해 도움을 받는 이종 핵형성 및 (2) 용융물 풀의 유체 흐름으로 인한 덴드라이트 조각화입니다. 첫 번째 메커니즘은 광범위하게 연구되었습니다. 이원 합금을 예로 들면, 고체는 액체만큼 많은 용질을 수용할 수 없으므로 응고 중에 용질을 액체로 거부합니다. 결과적으로, 성장하는 수상돌기 앞에서 용질 분할은 실제 온도가 국부 평형 액상선보다 낮은 과냉각 액체를 생성합니다. 충분히 광범위한 체질적으로 과냉각된 구역의 존재는 새로운 결정립의 핵형성 및 성장을 촉진합니다. ^{[ 56 ]}전체 과냉각은 응고 전면에서의 구성, 동역학 및 곡률 과냉각을 포함한 여러 기여의 합입니다. 일반적인 가정은 동역학 및 곡률 과냉각이 합금에 대한 용질 과냉각의 더 큰 기여와 관련하여 무시될 수 있다는 것입니다. ^{[ 57 ]}

서로 다른 기본 메커니즘을 더 잘 이해하려면피- 브이피-V조건에서 응고 모델링이 수행됩니다. 첫 번째 목적은 스트레이 그레인의 전체 범위를 평가하는 것입니다(Φ¯¯¯¯Φ¯) 처리 매개 변수의 함수로 국부적 표류 입자 비율의 변화를 조사하기 위해 (ΦΦ) 용융 풀의 위치 함수로. 두 번째 목적은 금속 AM의 빠른 응고 동안 응고 미세 구조와 표류 입자 형성 메커니즘 사이의 관계를 이해하는 것입니다.

그림 4 는 해석적으로 시뮬레이션된 표류 입자 비율을 보여줍니다.Φ¯¯¯¯Φ¯세 가지 레이저 흡수율 값에서 다양한 레이저 스캐닝 속도 및 레이저 출력에 대해. 결과는 스트레이 그레인 면적 비율이 흡수된 에너지에 민감하다는 것을 보여줍니다. 흡수율을 0.30에서 0.80으로 증가시키면Φ¯¯¯¯Φ¯약 3배이며, 이 효과는 저속 및 고출력 영역에서 더욱 두드러집니다. 다른 모든 조건이 같다면, 흡수된 전력의 큰 영향은 평균 열 구배 크기의 일반적인 감소와 용융 풀 내 평균 응고율의 증가에 기인합니다. 스캐닝 속도가 증가하고 전력이 감소함에 따라 평균 스트레이 그레인 비율이 감소합니다. 이러한 일반적인 경향은 Vitek의 작업에서 채택된 그림 5 의 파란색 영역에서 시뮬레이션된 용접 결과와 일치합니다 . ^{[ 3 ]} 더 큰 과냉각 구역( 즉, 지 /V티G/V티영역)은 용접 풀의 표유 입자의 면적 비율이 분홍색 영역에 해당하는 LPBF 조건의 면적 비율보다 훨씬 더 크다는 것을 의미합니다. 그럼에도 불구하고 두 데이터 세트의 일반적인 경향은 유사합니다. 즉 , 레이저 출력이 감소하고 레이저 속도가 증가함에 따라 표류 입자의 비율이 감소합니다. 또한 그림 5 에서 스캐닝 속도가 LPBF 영역으로 증가함에 따라 표유 입자 면적 분율에 대한 레이저 매개변수의 변화 효과가 감소한다는 것을 추론할 수 있습니다. 그림 6 (a)는 그림 3 의 EBSD 분석에서 나온 실험적 표류 결정립 면적 분율 과 그림 4 의 해석 시뮬레이션 결과를 비교합니다.. 열쇠 구멍 모양의 FZ에서 정확한 값이 다르지만 추세는 시뮬레이션과 실험 데이터 모두에서 일관되었습니다. 키홀 모양의 용융 풀, 특히 전력이 300W인 2개는 분석 시뮬레이션 예측보다 훨씬 더 많은 양의 흩어진 입자를 가지고 있습니다. Rosenthal 방정식은 일반적으로 열 전달이 순전히 전도에 의해 좌우된다는 가정으로 인해 열쇠 구멍 체제의 열 흐름을 적절하게 반영하지 못하기 때문에 이러한 불일치가 실제로 예상됩니다. ^{[ 39 , 40 ]} 그것은 또한 그림 4 의 발견 , 즉 키홀 모드 동안 흡수된 전력의 증가가 표류 입자 형성에 더 이상적인 조건을 초래한다는 것을 검증합니다. 그림 6 (b)는 실험을 비교Φ¯¯¯¯Φ¯수치 CFD 시뮬레이션Φ¯¯¯¯Φ¯. CFD 모델이 약간 초과 예측하지만Φ¯¯¯¯Φ¯전체적으로피- 브이피-V조건에서 열쇠 구멍 조건에서의 예측은 분석 모델보다 정확합니다. 전도 모드 용융 풀의 경우 실험 값이 분석 시뮬레이션 값과 더 가깝게 정렬됩니다.

모의 온도 구배 G 분포 및 응고율 검사V티V티분석 모델링의 쌍은 그림 7 (a)의 CMSX-4 미세 구조 선택 맵에 표시됩니다. 제공지 /V티G/V티( 즉 , 형태 인자)는 형태를 제어하고지 ×V티G×V티( 즉 , 냉각 속도)는 응고된 미세 구조의 규모를 제어하고 , ^{[ 58 , 59 ]}지 -V티G-V티플롯은 전통적인 제조 공정과 AM 공정 모두에서 미세 구조 제어를 지원합니다. 이 플롯의 몇 가지 분명한 특징은 등축, 주상, 평면 전면 및 이러한 경계 근처의 전이 영역을 구분하는 경계입니다. 그림 7 (a)는 몇 가지 선택된 분석 열 시뮬레이션에 대한 미세 구조 선택 맵을 나타내는 반면 그림 7 (b)는 수치 열 모델의 결과와 동일한 맵을 보여줍니다. 등축 미세구조의 형성은 낮은 G 이상 에서 명확하게 선호됩니다.V티V티정황. 이 플롯에서 각 곡선의 평면 전면에 가장 가까운 지점은 용융 풀의 최대 너비 위치에 해당하는 반면 등축 영역에 가까운 지점의 끝은 용융 풀의 후면 꼬리에 해당합니다. 그림 7 (a)에서 대부분의지 -V티G-V티응고 전면의 쌍은 원주형 영역에 속하고 점차 CET 영역으로 위쪽으로 이동하지만 용융 풀의 꼬리는 다음에 따라 완전히 등축 영역에 도달하거나 도달하지 않을 수 있습니다.피- 브이피-V조합. 그림 7 (a) 의 곡선 중 어느 것도 평면 전면 영역을 통과하지 않지만 더 높은 전력의 경우에 가까워집니다. 저속 레이저 용융 공정을 사용하는 이전 작업에서는 곡선이 평면 영역을 통과할 수 있습니다. 레이저 속도가 증가함에 따라 용융 풀 꼬리는 여전히 CET 영역에 있지만 완전히 등축 영역에서 멀어집니다. CET 영역으로 떨어지는 섹션의 수도 감소합니다.Φ¯¯¯¯Φ¯응고된 물질에서.

그만큼지 -V티G-V티CFD 모델을 사용하여 시뮬레이션된 응고 전면의 쌍이 그림 7 (b)에 나와 있습니다. 세 방향 모두에서 각 점 사이의 일정한 간격으로 미리 정의된 좌표에서 수행된 해석 시뮬레이션과 달리, 고충실도 CFD 모델의 출력은 불규칙한 사면체 좌표계에 있었고 G 를 추출하기 전에 일반 3D 그리드에 선형 보간되었습니다. 그리고V티V티그런 다음 미세 구조 선택 맵에 플롯됩니다. 일반적인 경향은 그림 7 (a)의 것과 일치하지만 이 방법으로 모델링된 매우 동적인 유체 흐름으로 인해 결과에 더 많은 분산이 있었습니다. 그만큼지 -V티G-V티분석 열 모델의 쌍 경로는 더 연속적인 반면 수치 시뮬레이션의 경로는 용융 풀 꼬리 모양의 차이를 나타내는 날카로운 굴곡이 있습니다(이는 G 및V티V티) 두 모델에 의해 시뮬레이션됩니다.

유체 흐름을 통합한 응고 모델링

수치 CFD 모델을 사용하여 유동 입자 형성 정도에 대한 유체 흐름의 영향을 이해하고 시뮬레이션 결과를 분석 Rosenthal 솔루션과 비교했습니다. 그림 8 은 응고 매개변수 G 의 분포를 보여줍니다.V티V티,지 /V티G/V티, 그리고지 ×V티G×V티yz 단면에서 x 는 FLOW-3D에서 (a1–d1) 분석 열 모델링 및 (a2–d2) FVM 방법을 사용하여 시뮬레이션된 용융 풀의 최대 폭입니다. 그림 8 의 값은 응고 전선이 특정 위치에 도달할 때 정확한 값일 수도 있고 아닐 수도 있지만 일반적인 추세를 반영한다는 의미의 임시 가상 값입니다. 이 프로파일은 출력 300W 및 속도 400mm/s의 레이저 빔에서 시뮬레이션됩니다. 용융 풀 경계는 흰색 곡선으로 표시됩니다. (a2–d2)의 CFD 시뮬레이션 용융 풀 깊이는 342입니다. μμm, 측정 깊이 352와 잘 일치 μμ일치하는 길쭉한 열쇠 구멍 모양과 함께 그림 1 에 표시된 실험 FZ의 m . 그러나 분석 모델은 반원 모양의 용융 풀을 출력하고 용융 풀 깊이는 264에 불과합니다. μμ열쇠 구멍의 경우 현실과는 거리가 멀다. CFD 시뮬레이션 결과에서 열 구배는 레이저 반사 증가와 불안정한 액체-증기 상호 작용이 발생하는 증기 함몰의 동적 부분 근처에 있기 때문에 FZ 하단에서 더 높습니다. 대조적으로 해석 결과의 열 구배 크기는 경계를 따라 균일합니다. 두 시뮬레이션 결과 모두 그림 8 (a1) 및 (a2) 에서 응고가 용융 풀의 상단 중심선을 향해 진행됨에 따라 열 구배가 점차 감소합니다 . 응고율은 그림 8 과 같이 경계 근처에서 거의 0입니다. (b1) 및 (b2). 이는 경계 영역이 응고되기 시작할 때 국부 응고 전면의 법선 방향이 레이저 스캐닝 방향에 수직이기 때문입니다. 이것은 드라이브θ → π/ 2θ→파이/2그리고V티→ 0V티→0식에서 [ 3 ]. 대조적으로 용융 풀의 상단 중심선 근처 영역에서 응고 전면의 법선 방향은 레이저 스캐닝 방향과 잘 정렬되어 있습니다.θ → 0θ→0그리고V티→ 브이V티→V, 빔 스캐닝 속도. G 와 _V티V티값이 얻어지면 냉각 속도지 ×V티G×V티및 형태 인자지 /V티G/V티계산할 수 있습니다. 그림 8 (c2)는 용융 풀 바닥 근처의 온도 구배가 매우 높고 상단에서 더 빠른 성장 속도로 인해 냉각 속도가 용융 풀의 바닥 및 상단 중심선 근처에서 더 높다는 것을 보여줍니다. 지역. 그러나 이러한 추세는 그림 8 (c1)에 캡처되지 않았습니다. 그림 8 의 형태 요인 (d1) 및 (d2)는 중심선에 접근함에 따라 눈에 띄게 감소합니다. 경계에서 큰 값은 열 구배를 거의 0인 성장 속도로 나누기 때문에 발생합니다. 이 높은 형태 인자는 주상 미세구조 형성 가능성이 높음을 시사하는 반면, 중앙 영역의 값이 낮을수록 등축 미세구조의 가능성이 더 크다는 것을 나타냅니다. Tanet al. 또한 키홀 모양의 용접 풀 ^{[ 59 ]} 에서 이러한 응고 매개변수의 분포 를 비슷한 일반적인 경향으로 보여주었습니다. 그림 3 에서 볼 수 있듯이 용융 풀의 상단 중심선에 있는 흩어진 입자는 낮은 특징을 나타내는 영역과 일치합니다.지 /V티G/V티그림 8 (d1) 및 (d2)의 값. 시뮬레이션과 실험 간의 이러한 일치는 용융 풀의 상단 중심선에 축적된 흩어진 입자의 핵 생성 및 성장이 등온선 속도의 증가와 온도 구배의 감소에 의해 촉진됨을 보여줍니다.

그림 9 는 유체 속도 및 국부적 핵형성 성향을 보여줍니다.ΦΦ300W의 일정한 레이저 출력과 400, 800 및 1200mm/s의 세 가지 다른 레이저 속도에 의해 생성된 3D 용융 풀 전체에 걸쳐. 그림 9 (d)~(f)는 로컬ΦΦ해당 3D 보기에서 밝은 회색 평면으로 표시된 특정 yz 단면의 분포. 이 yz 섹션은 가장 높기 때문에 선택되었습니다.Φ¯¯¯¯Φ¯용융 풀 내의 값은 각각 23.40, 11.85 및 2.45pct입니다. 이들은 그림 3 의 실험 데이터와 비교하기에 적절하지 않을 수 있는 액체 용융 풀의 과도 값이며Φ¯¯¯¯Φ¯그림 6 의 값은 이 값이 고체-액체 계면에 가깝지 않고 용융 풀의 중간에서 취해졌기 때문입니다. 온도가 훨씬 낮아서 핵이 생존하고 성장할 수 있기 때문에 핵 형성은 용융 풀의 중간이 아닌 고체-액체 계면에 더 가깝게 발생할 가능성이 있습니다.

그림 3 (a), (d), (g), (h)에서 위쪽 중심선에서 멀리 떨어져 있는 흩어진 결정립이 있었습니다. 그들은 훨씬 더 높은 열 구배와 더 낮은 응고 속도 필드에 위치하기 때문에 과냉각 이론은 이러한 영역에서 표류 입자의 형성에 대한 만족스러운 설명이 아닙니다. 이것은 떠돌이 결정립의 형성을 야기할 수 있는 두 번째 메커니즘, 즉 수상돌기의 팁을 가로지르는 유체 흐름에 의해 유발되는 수상돌기 조각화를 고려하도록 동기를 부여합니다. 유체 흐름이 열 구배를 따라 속도 성분을 갖고 고체-액체 계면 속도보다 클 때, 주상 수상돌기의 국지적 재용융은 용질이 풍부한 액체가 흐물흐물한 구역의 깊은 곳에서 액상선 등온선까지 이동함으로써 발생할 수 있습니다. . ^{[ 55]} 분리된 수상돌기는 대류에 의해 열린 액체로 운반될 수 있습니다. 풀이 과냉각 상태이기 때문에 이러한 파편은 고온 조건에서 충분히 오래 생존하여 길 잃은 입자의 핵 생성 사이트로 작용할 수 있습니다. 결과적으로 수상 돌기 조각화 과정은 활성 핵의 수를 효과적으로 증가시킬 수 있습니다.N0N0) 용융 풀 ^{[ 15 , 60 , 61 ]} 에서 생성된 미세 구조에서 표류 입자의 면적을 증가시킵니다.

그림 9 (a) 및 (b)에서 반동 압력은 용융 유체를 아래쪽으로 흐르게 하여 결과 흐름을 지배합니다. 유체 속도의 역방향 요소는 V = 400 및 800mm/s에 대해 각각 최대값 1.0 및 1.6m/s로 더 느려집니다 . 그림 9 (c)에서 레이저 속도가 더 증가함에 따라 증기 침하가 더 얕고 넓어지고 반동 압력이 더 고르게 분포되어 증기 침강에서 주변 영역으로 유체를 밀어냅니다. 역류는 최대값 3.5m/s로 더 빨라집니다. 용융 풀의 최대 너비에서 yz 단면 의 키홀 아래 평균 유체 속도는 그림에 표시된 경우에 대해 0.46, 0.45 및 1.44m/s입니다.9 (a), (b) 및 (c). 키홀 깊이의 변동은 각 경우의 최대 깊이와 최소 깊이의 차이로 정의되는 크기로 정량화됩니다. 240 범위의 강한 증기 내림 변동 μμm은 그림 9 (a)의 V = 400mm/s 경우에서 발견 되지만 이 변동은 그림 9 (c)에서 16의 범위로 크게 감소합니다.μμ미디엄. V = 400mm/s인 경우 의 유체장과 높은 변동 범위는 이전 키홀 동역학 시뮬레이션과 일치합니다. ^{[ 34 ]}

따라서 V = 400mm/s 키홀 케이스의 무질서한 변동 흐름이 용융 풀 경계를 따라 응고된 주상 수상돌기에서 분리된 조각을 구동할 가능성이 있습니다. V = 1200mm/s의 경우 강한 역류 는 그림 3 에서 관찰되지 않았지만 동일한 효과를 가질 수 있습니다. . 덴드라이트 조각화에 대한 유체 유동장의 영향에 대한 이 경험적 설명은 용융 풀 경계 근처에 떠돌이 입자의 존재에 대한 그럴듯한 설명을 제공합니다. 분명히 하기 위해, 우리는 이 가설을 검증하기 위해 이 현상에 대한 직접적인 실험적 관찰을 하지 않았습니다. 이 작업에서 표유 입자 면적 분율을 계산할 때 단순화를 위해 핵 생성 모델링에 일정한 핵 생성 수 밀도가 적용되었습니다. 이는 그림 9 의 표류 입자 영역 비율 이 수지상정 조각화가 발생하는 경우 이러한 높은 유체 흐름 용융 풀에서 발생할 수 있는 것, 즉 강화된 핵 생성 밀도를 반영하지 않는다는 것을 의미합니다.

위의 이유로 핵 형성에 대한 수상 돌기 조각화의 영향을 아직 배제할 수 없습니다. ^{그러나 단편화 이론은 용접 문헌 [ 62 ]} 에서 검증될 만큼 충분히 개발되지 않았 으므로 부차적인 중요성만 고려된다는 점에 유의해야 합니다. 1200mm/s를 초과하는 레이저 스캐닝 속도는 최소한의 표류 결정립 면적 분율을 가지고 있음에도 불구하고 분명한 볼링을 나타내기 때문에 단결정 수리 및 AM 처리에 적합하지 않습니다. 따라서 낮은 P 및 높은 V 에 의해 생성된 응고 전면 근처에서 키홀 변동이 최소화되고 유체 속도가 완만해진 용융 풀이 생성된다는 결론을 내릴 수 있습니다., 처리 창의 극한은 아니지만 흩어진 입자를 나타낼 가능성이 가장 적습니다.

마지막으로 단일 레이저 트랙의 응고 거동을 조사하면 에피택셜 성장 동안 표류 입자 형성을 더 잘 이해할 수 있다는 점에 주목하는 것이 중요합니다. 우리의 현재 결과는 최적의 레이저 매개변수에 대한 일반적인 지침을 제공하여 최소 스트레이 그레인을 달성하고 단결정 구조를 유지합니다. 이 가이드라인은 250W 정도의 전력과 600~800mm/s의 스캔 속도로 최소 흩어진 입자에 적합한 공정 창을 제공합니다. 각 처리 매개변수를 신중하게 선택하면 과거에 스테인리스강에 대한 거의 단결정 미세 구조를 인쇄하는 데 성공했으며 이는 CMSX-4 AM 빌드에 대한 가능성을 보여줍니다. ^{[ 63 ]}신뢰성을 보장하기 위해 AM 수리 프로세스를 시작하기 전에 보다 엄격한 실험 테스트 및 시뮬레이션이 여전히 필요합니다. 둘 이상의 레이저 트랙 사이의 상호 작용도 고려해야 합니다. 또한 레이저, CMSX-4 분말 및 벌크 재료 간의 상호 작용이 중요하며, 수리 중에 여러 층의 CMSX-4 재료를 축적해야 하는 경우 다른 스캔 전략의 효과도 중요한 역할을 할 수 있습니다. 분말이 포함된 경우 Lopez-Galilea 등 의 연구에서 제안한 바와 같이 분말이 주로 완전히 녹지 않았을 때 추가 핵 생성 사이트를 도입하기 때문에 단순히 레이저 분말과 속도를 조작하여 흩어진 입자 형성을 완화하기 어려울 수 있습니다 . ^{[ 22 ]}결과적으로 CMSX-4 단결정을 수리하기 위한 레이저 AM의 가능성을 다루기 위해서는 기판 재료, 레이저 출력, 속도, 해치 간격 및 층 두께의 조합을 모두 고려해야 하며 향후 연구에서 다루어야 합니다. CFD 모델링은 2개 이상의 레이저 트랙 사이의 상호작용과 열장에 미치는 영향을 통합할 수 있으며, 이는 AM 빌드 시나리오 동안 핵 생성 조건으로 단일 비드 연구의 지식 격차를 해소할 것입니다.

결론

LPBF 제조의 특징적인 조건 하에서 CMSX-4 단결정 의 에피택셜(기둥형) 대 등축 응고 사이의 경쟁을 실험적 및 이론적으로 모두 조사했습니다. 이 연구는 고전적인 응고 개념을 도입하여 빠른 레이저 용융의 미세 구조 특징을 설명하고 응고 조건과 표유 결정 성향을 예측하기 위해 해석적 및 수치적 고충실도 CFD 열 모델 간의 비교를 설명했습니다. 본 연구로부터 다음과 같은 주요 결론을 도출할 수 있다.

단일 레이저 트랙의 레이저 가공 조건은 용융 풀 형상, 레이저 흡수율, 유체 흐름 및 키홀 요동, 입자 구조 및 표류 입자 형성 민감성에 강한 영향을 미치는 것으로 밝혀졌습니다.

레이저 용접을 위해 개발된 이론적인 표유 결정립 핵형성 분석이 레이저 용융 AM 조건으로 확장되었습니다. 분석 모델링 결과와 단일 레이저 트랙의 미세구조 특성화를 비교하면 예측이 전도 및 볼링 조건에서 실험적 관찰과 잘 일치하는 반면 키홀 조건에서는 예측이 약간 과소하다는 것을 알 수 있습니다. 이러한 불일치는 레이저 트랙의 대표성이 없는 섹션이나 유체 속도 필드의 변화로 인해 발생할 수 있습니다. CFD 모델에서 추출한 열장에 동일한 표유 입자 계산 파이프라인을 적용하면 연구된 모든 사례에서 과대평가가 발생하지만 분석 모델보다 연장된 용융 풀의 실험 데이터와 더 정확하게 일치합니다.

이 연구에서 두 가지 표류 결정립 형성 메커니즘인 불균일 핵형성 및 수상돌기 조각화가 평가되었습니다. 우리의 결과는 불균일 핵형성이 용융 풀의 상단 중심선에서 새로운 결정립의 형성으로 이어지는 주요 메커니즘임을 시사합니다.지 /V티G/V티정권.

용융 풀 경계 근처의 흩어진 입자는 깊은 키홀 모양의 용융 풀에서 독점적으로 관찰되며, 이는 강한 유체 흐름으로 인한 수상 돌기 조각화의 영향이 이러한 유형의 용융 풀에서 고려하기에 충분히 강력할 수 있음을 시사합니다.

일반적으로 더 높은 레이저 스캐닝 속도와 더 낮은 전력 외에도 안정적인 키홀과 최소 유체 속도는 또한 흩어진 입자 형성을 완화하고 레이저 단일 트랙에서 에피택셜 성장을 보존합니다.

References

R.C. Reed: The Superalloys: Fundamentals and Applications, Cambridge University Press, Cambridge, 2006, pp.17–20.BookGoogle Scholar

A. Basak, R. Acharya, and S. Das: Metall. Mater. Trans. A, 2016, vol. 47A, pp. 3845–59.ArticleGoogle Scholar

R. Vilar and A. Almeida: J. Laser Appl., 2015, vol. 27, p. S17004.ArticleGoogle Scholar

T. Kalfhaus, M. Schneider, B. Ruttert, D. Sebold, T. Hammerschmidt, J. Frenzel, R. Drautz, W. Theisen, G. Eggeler, O. Guillon, and R. Vassen: Mater. Des., 2019, vol. 168, p. 107656.ArticleCASGoogle Scholar

S.S. Babu, S.A. David, J.W. Park, and J.M. Vitek: Sci. Technol. Weld. Join., 2004, vol. 9, pp. 1–12.ArticleCASGoogle Scholar

L. Felberbaum, K. Voisey, M. Gäumann, B. Viguier, and A. Mortensen: Mater. Sci. Eng. A, 2001, vol. 299, pp. 152–56.ArticleGoogle Scholar

S. Mokadem, C. Bezençon, J.M. Drezet, A. Jacot, J.D. Wagnière, and W. Kurz: TMS Annual Meeting, 2004, pp. 67–76.

J.M. Vitek: ASM Proc. Int. Conf. Trends Weld. Res., vol. 2005, pp. 773–79.

B. Kianian: Wohlers Report 2017: 3D Printing and Additive Manufacturing State of the Industry, Annual Worldwide Progress Report, Wohlers Associates, Inc., Fort Collins, 2017.Google Scholar

M. Ramsperger, L. Mújica Roncery, I. Lopez-Galilea, R.F. Singer, W. Theisen, and C. Körner: Adv. Eng. Mater., 2015, vol. 17, pp. 1486–93.ArticleCASGoogle Scholar

A.B. Parsa, M. Ramsperger, A. Kostka, C. Somsen, C. Körner, and G. Eggeler: Metals, 2016, vol. 6, pp. 258-1–17.ArticleGoogle Scholar

C. Körner, M. Ramsperger, C. Meid, D. Bürger, P. Wollgramm, M. Bartsch, and G. Eggeler: Metall. Mater. Trans. A, 2018, vol. 49A, pp. 3781–92.ArticleGoogle Scholar

D. Bürger, A. Parsa, M. Ramsperger, C. Körner, and G. Eggeler: Mater. Sci. Eng. A, 2019, vol. 762, p. 138098,ArticleGoogle Scholar

R. Cunningham, C. Zhao, N. Parab, C. Kantzos, J. Pauza, K. Fezzaa, T. Sun, and A.D. Rollett: Science, 2019, vol. 363, pp. 849–52.ArticleCASGoogle Scholar

B. Fotovvati, S.F. Wayne, G. Lewis, and E. Asadi: Adv. Mater. Sci. Eng., 2018, vol. 2018, p. 4920718.ArticleGoogle Scholar

P.-J. Chiang, R. Jiang, R. Cunningham, N. Parab, C. Zhao, K. Fezzaa, T. Sun, and A.D. Rollett: in Advanced Real Time Imaging II, pp. 77–85.

J. Ye, S.A. Khairallah, A.M. Rubenchik, M.F. Crumb, G. Guss, J. Belak, and M.J. Matthews: Adv. Eng. Mater., 2019, vol. 21, pp. 1–9.ArticleGoogle Scholar

C. Zhao, Q. Guo, X. Li, N. Parab, K. Fezzaa, W. Tan, L. Chen, and T. Sun: Phys. Rev. X, 2019, vol. 9, p. 021052.CASGoogle Scholar

S.A. Khairallah, A.T. Anderson, A. Rubenchik, and W.E. King: Acta Mater., 2016, vol. 108, pp. 36–45.ArticleCASGoogle Scholar

N. Kouraytem, X. Li, R. Cunningham, C. Zhao, N. Parab, T. Sun, A.D. Rollett, A.D. Spear, and W. Tan: Appl. Phys. Rev., 2019, vol. 11, p. 064054.ArticleCASGoogle Scholar

T. DebRoy, H. Wei, J. Zuback, T. Mukherjee, J. Elmer, J. Milewski, A. Beese, A. Wilson-Heid, A. De, and W. Zhang: Prog. Mater. Sci., 2018, vol. 92, pp. 112–224.ArticleCASGoogle Scholar

I. Yadroitsev, A. Gusarov, I. Yadroitsava, and I. Smurov: J. Mater. Process. Technol., 2010, vol. 210, pp. 1624–31.ArticleCASGoogle Scholar

S. Ghosh, L. Ma, L.E. Levine, R.E. Ricker, M.R. Stoudt, J.C. Heigel, and J.E. Guyer: JOM, 2018, vol. 70, pp. 1011–16.ArticleCASGoogle Scholar

Y. He, C. Montgomery, J. Beuth, and B. Webler: Mater. Des., 2019, vol. 183, p. 108126.ArticleCASGoogle Scholar

D. Rosenthal: Weld. J., 1941, vol. 20, pp. 220–34.Google Scholar

M. Tang, P.C. Pistorius, and J.L. Beuth: Addit. Manuf., 2017, vol. 14, pp. 39–48.CASGoogle Scholar

R.E. Aune, L. Battezzati, R. Brooks, I. Egry, H.J. Fecht, J.P. Garandet, M. Hayashi, K.C. Mills, A. Passerone, P.N. Quested, E. Ricci, F. Schmidt-Hohagen, S. Seetharaman, B. Vinet, and R.K. Wunderlich: Proc. Int.Symp. Superalloys Var. Deriv., 2005, pp. 467–76.

B.C. Wilson, J.A. Hickman, and G.E. Fuchs: JOM, 2003, vol. 55, pp. 35–40.ArticleCASGoogle Scholar

J.J. Valencia and P.N. Quested: ASM Handb., 2008, vol. 15, pp. 468–81.Google Scholar

H.L. Wei, J. Mazumder, and T. DebRoy: Sci. Rep., 2015, vol. 5, pp. 1–7.Google Scholar

N. Raghavan, R. Dehoff, S. Pannala, S. Simunovic, M. Kirka, J. Turner, N. Carlson, and S.S. Babu: Acta Mater., 2016, vol. 112, pp. 303–14.ArticleCASGoogle Scholar

R. Lin, H. Wang, F. Lu, J. Solomon, and B.E. Carlson: Int. J. Heat Mass Transf., 2017, vol. 108, pp. 244–56.ArticleCASGoogle Scholar

M. Bayat, A. Thanki, S. Mohanty, A. Witvrouw, S. Yang, J. Thorborg, N.S. Tiedje, and J.H. Hattel: Addit. Manuf., 2019, vol. 30, p. 100835.CASGoogle Scholar

K. Higuchi, H.-J. Fecht, and R.K. Wunderlich: Adv. Eng. Mater., 2007, vol. 9, pp. 349–54.ArticleCASGoogle Scholar

Q. Guo, C. Zhao, M. Qu, L. Xiong, L.I. Escano, S.M.H. Hojjatzadeh, N.D. Parab, K. Fezzaa, W. Everhart, T. Sun, and L. Chen: Addit. Manuf., 2019, vol. 28, pp. 600–09.Google Scholar

J. Trapp, A.M. Rubenchik, G. Guss, and M.J. Matthews: Appl. Mater. Today, 2017, vol. 9, pp. 341–49.ArticleGoogle Scholar

M. Schneider, L. Berthe, R. Fabbro, and M. Muller: J. Phys. D, 2008, vol. 41, p. 155502.ArticleGoogle Scholar

Z. Gan, O.L. Kafka, N. Parab, C. Zhao, L. Fang, O. Heinonen, T. Sun, and W.K. Liu: Nat. Commun., 2021, vol. 12, p. 2379.ArticleCASGoogle Scholar

B.J. Simonds, E.J. Garboczi, T.A. Palmer, and P.A. Williams: Appl. Phys. Rev., 2020, vol. 13, p. 024057.ArticleCASGoogle Scholar

J. Dantzig and M. Rappaz: Solidification, 2^{nd} ed., EPFL Press, Lausanne, 2016, pp. 483–532.Google Scholar

W. Tiller, K. Jackson, J. Rutter, and B. Chalmers: Acta Metall., 1953, vol. 1, pp. 428–37.ArticleCASGoogle Scholar

D. Zhang, A. Prasad, M.J. Bermingham, C.J. Todaro, M.J. Benoit, M.N. Patel, D. Qiu, D.H. StJohn, M. Qian, and M.A. Easton: Metall. Mater. Trans. A, 2020, vol. 51A, pp. 4341–59.ArticleGoogle Scholar

F. Yan, W. Xiong, and E.J. Faierson: Materials, 2017, vol. 10, p. 1260.ArticleGoogle Scholar

J.M. Vitek, S.A. David, and L.A. Boatner: Sci. Technol. Weld. Join., 1997, vol. 2, pp. 109–18.ArticleCASGoogle Scholar

X. Wang, J.A. Muñiz-Lerma, O. Sanchez-Mata, S.E. Atabay, M.A. Shandiz, and M. Brochu: Prog. Addit. Manuf., 2020, vol. 5, pp. 41–49.ArticleGoogle Scholar

2014년 2월 영국 해협(영국)과 특히 Dawlish에 영향을 미친 온대 저기압 폭풍 사슬은 남서부 지역과 영국의 나머지 지역을 연결하는 주요 철도에 심각한 피해를 입혔습니다.

이 사건으로 라인이 두 달 동안 폐쇄되어 5천만 파운드의 피해와 12억 파운드의 경제적 손실이 발생했습니다. 이 연구에서는 폭풍의 파괴력을 해독하기 위해 목격자 계정을 수집하고 해수면 데이터를 분석하며 수치 모델링을 수행합니다.

우리의 분석에 따르면 이벤트의 재난 관리는 성공적이고 효율적이었으며 폭풍 전과 도중에 인명과 재산을 구하기 위해 즉각적인 조치를 취했습니다. 파도 부이 분석에 따르면 주기가 4–8, 8–12 및 20–25초인 복잡한 삼중 봉우리 바다 상태가 존재하는 반면, 조위계 기록에 따르면 최대 0.8m의 상당한 파도와 최대 1.5m의 파도 성분이 나타났습니다.

이벤트에서 가능한 기여 요인으로 결합된 진폭. 최대 286 KN의 상당한 임펄스 파동이 손상의 시작 원인일 가능성이 가장 높았습니다. 수직 벽의 반사는 파동 진폭의 보강 간섭을 일으켜 파고가 증가하고 최대 16.1m3/s/m(벽의 미터 너비당)의 상당한 오버탑핑을 초래했습니다.

이 정보와 우리의 공학적 판단을 통해 우리는 이 사고 동안 다중 위험 계단식 실패의 가장 가능성 있는 순서는 다음과 같다고 결론을 내립니다. 조적 파괴로 이어지는 파도 충격력, 충전물 손실 및 연속적인 조수에 따른 구조물 파괴.

The February 2014 extratropical cyclonic storm chain, which impacted the English Channel (UK) and Dawlish in particular, caused significant damage to the main railway connecting the south-west region to the rest of the UK. The incident caused the line to be closed for two months, £50 million of damage and an estimated £1.2bn of economic loss. In this study, we collate eyewitness accounts, analyse sea level data and conduct numerical modelling in order to decipher the destructive forces of the storm. Our analysis reveals that the disaster management of the event was successful and efficient with immediate actions taken to save lives and property before and during the storm. Wave buoy analysis showed that a complex triple peak sea state with periods at 4–8, 8–12 and 20–25 s was present, while tide gauge records indicated that significant surge of up to 0.8 m and wave components of up to 1.5 m amplitude combined as likely contributing factors in the event. Significant impulsive wave force of up to 286 KN was the most likely initiating cause of the damage. Reflections off the vertical wall caused constructive interference of the wave amplitudes that led to increased wave height and significant overtopping of up to 16.1 m^{3}/s/m (per metre width of wall). With this information and our engineering judgement, we conclude that the most probable sequence of multi-hazard cascading failure during this incident was: wave impact force leading to masonry failure, loss of infill and failure of the structure following successive tides.

Introduction

The progress of climate change and increasing sea levels has started to have wide ranging effects on critical engineering infrastructure (Shakou et al. 2019). The meteorological effects of increased atmospheric instability linked to warming seas mean we may be experiencing more frequent extreme storm events and more frequent series or chains of events, as well as an increase in the force of these events, a phenomenon called storminess (Mölter et al. 2016; Feser et al. 2014). Features of more extreme weather events in extratropical latitudes (30°–60°, north and south of the equator) include increased gusting winds, more frequent storm squalls, increased prolonged precipitation and rapid changes in atmospheric pressure and more frequent and significant storm surges (Dacre and Pinto 2020). A recent example of these events impacting the UK with simultaneous significant damage to coastal infrastructure was the extratropical cyclonic storm chain of winter 2013/2014 (Masselink et al. 2016; Adams and Heidarzadeh 2021). The cluster of storms had a profound effect on both coastal and inland infrastructure, bringing widespread flooding events and large insurance claims (RMS 2014).

The extreme storms of February 2014, which had a catastrophic effect on the seawall of the south Devon stretch of the UK’s south-west mainline, caused a two-month closure of the line and significant disruption to the local and regional economy (Fig. 1b) (Network Rail 2014; Dawson et al. 2016; Adams and Heidarzadeh 2021). Restoration costs were £35 m, and economic effects to the south-west region of England were estimated up to £1.2bn (Peninsula Rail Taskforce 2016). Adams and Heidarzadeh (2021) investigated the disparate cascading failure mechanisms which played a part in the failure of the railway through Dawlish and attempted to put these in the context of the historical records of infrastructure damage on the line. Subsequent severe storms in 2016 in the region have continued to cause damage and disruption to the line in the years since 2014 (Met Office 2016). Following the events of 2014, Network Rail ^{Footnote1} who owns the network has undertaken a resilience study. As a result, it has proposed a £400 m refurbishment of the civil engineering assets that support the railway (Fig. 1) (Network Rail 2014). The new seawall structure (Fig. 1a,c), which is constructed of pre-cast concrete sections, encases the existing Brunel seawall (named after the project lead engineer, Isambard Kingdom Brunel) and has been improved with piled reinforced concrete foundations. It is now over 2 m taller to increase the available crest freeboard and incorporates wave return features to minimise wave overtopping. The project aims to increase both the resilience of the assets to extreme weather events as well as maintain or improve amenity value of the coastline for residents and visitors.

In this work, we return to the Brunel seawall and the damage it sustained during the 2014 storms which affected the assets on the evening of the 4th and daytime of the 5th of February and eventually resulted in a prolonged closure of the line. The motivation for this research is to analyse and model the damage made to the seawall and explain the damage mechanisms in order to improve the resilience of many similar coastal structures in the UK and worldwide. The innovation of this work is the multidisciplinary approach that we take comprising a combination of analysis of eyewitness accounts (social science), sea level and wave data analysis (physical science) as well as numerical modelling and engineering judgement (engineering sciences). We investigate the contemporary wave climate and sea levels by interrogating the real-time tide gauge and wave buoys installed along the south-west coast of the English Channel. We then model a typical masonry seawall (Fig. 2), applying the computational fluid dynamics package FLOW3D-Hydro,^{Footnote2} to quantify the magnitude of impact forces that the seawall would have experienced leading to its failure. We triangulate this information to determine the probable sequence of failures that led to the disaster in 2014.

Data and methods

Our data comprise eyewitness accounts, sea level records from coastal tide gauges and offshore wave buoys as well as structural details of the seawall. As for methodology, we analyse eyewitness data, process and investigate sea level records through Fourier transform and conduct numerical simulations using the Flow3D-Hydro package (Flow Science 2022). Details of the data and methodology are provided in the following.

Eyewitness data

The scale of damage to the seawall and its effects led the local community to document the first-hand accounts of those most closely affected by the storms including residents, local businesses, emergency responders, politicians and engineering contractors involved in the post-storm restoration work. These records now form a permanent exhibition in the local museum in Dawlish^{Footnote3}, and some of these accounts have been transcribed into a DVD account of the disaster (Dawlish Museum 2015). We have gathered data from the Dawlish Museum, national and international news reports, social media tweets and videos. Table 1 provides a summary of the eyewitness accounts. Overall, 26 entries have been collected around the time of the incident. Our analysis of the eyewitness data is provided in the third column of Table 1 and is expanded in Sect. 3.Table 1 Eyewitness accounts of damage to the Dawlish railway due to the February 2014 storm and our interpretations

Our sea level data are a collection of three tide gauge stations (Newlyn, Devonport and Swanage Pier—Fig. 5a) owned and operated by the UK National Tide and Sea Level Facility^{Footnote4} for the Environment Agency and four offshore wave buoys (Dawlish, West Bay, Torbay and Chesil Beach—Fig. 6a). The tide gauge sites are all fitted with POL-EKO (www.pol-eko.com.pl) data loggers. Newlyn has a Munro float gauge with one full tide and one mid-tide pneumatic bubbler system. Devonport has a three-channel data pneumatic bubbler system, and Swanage Pier consists of a pneumatic gauge. Each has a sampling interval of 15 min, except for Swanage Pier which has a sampling interval of 10 min. The tide gauges are located within the port areas, whereas the offshore wave buoys are situated approximately 2—3.3 km from the coast at water depths of 10–15 m. The wave buoys are all Datawell Wavemaker Mk III units^{Footnote5} and come with sampling interval of 0.78 s. The buoys have a maximum saturation amplitude of 20.5 m for recording the incident waves which implies that every wave larger than this threshold will be recorded at 20.5 m. The data are provided by the British Oceanographic Data Centre^{Footnote6} for tide gauges and the Channel Coastal Observatory^{Footnote7} for wave buoys.

Sea level analysis

The sea level data underwent quality control to remove outliers and spikes as well as gaps in data (e.g. Heidarzadeh et al. 2022; Heidarzadeh and Satake 2015). We processed the time series of the sea level data using the Matlab signal processing tool (MathWorks 2018). For calculations of the tidal signals, we applied the tidal package TIDALFIT (Grinsted 2008), which is based on fitting tidal harmonics to the observed sea level data. To calculate the surge signals, we applied a 30-min moving average filter to the de-tided data in order to remove all wind, swell and infra-gravity waves from the time series. Based on the surge analysis and the variations of the surge component before the time period of the incident, an error margin of approximately ± 10 cm is identified for our surge analysis. Spectral analysis of the wave buoy data is performed using the fast Fourier transform (FFT) of Matlab package (Mathworks 2018).

Numerical modelling

Numerical modelling of wave-structure interaction is conducted using the computational fluid dynamics package Flow3D-Hydro version 1.1 (Flow Science 2022). Flow3D-Hydro solves the transient Navier–Stokes equations of conservation of mass and momentum using a finite difference method and on Eulerian and Lagrangian frameworks (Flow Science 2022). The aforementioned governing equations are:

∇.u=0∇.u=0

(1)

∂u∂t+u.∇u=−∇Pρ+υ∇2u+g∂u∂t+u.∇u=−∇Pρ+υ∇2u+g

(2)

where uu is the velocity vector, PP is the pressure, ρρ is the water density, υυ is the kinematic viscosity and gg is the gravitational acceleration. A Fractional Area/Volume Obstacle Representation (FAVOR) is adapted in Flow3D-Hydro, which applies solid boundaries within the Eulerian grid and calculates the fraction of areas and volume in partially blocked volume in order to compute flows on corresponding boundaries (Hirt and Nichols 1981). We validated the numerical modelling through comparing the results with Sainflou’s analytical equation for the design of vertical seawalls (Sainflou 1928; Ackhurst 2020), which is as follows:

where pdpd is the hydrodynamic pressure, ρρ is the water density, gg is the gravitational acceleration, HH is the wave height, dd is the water depth, kk is the wavenumber, zz is the difference in still water level and mean water level, σσ is the angular frequency and tt is the time. The Sainflou’s equation (Eq. 3) is used to calculate the dynamic pressure from wave action, which is combined with static pressure on the seawall.

Using Flow3D-Hydro, a model of the Dawlish seawall was made with a computational domain which is 250.0 m in length, 15.0 m in height and 0.375 m in width (Fig. 3a). The computational domain was discretised using a single uniform grid with a mesh size of 0.125 m. The model has a wave boundary at the left side of the domain (x-min), an outflow boundary on the right side (x-max), a symmetry boundary at the bottom (z-min) and a wall boundary at the top (z-max). A wall boundary implies that water or waves are unable to pass through the boundary, whereas a symmetry boundary means that the two edges of the boundary are identical and therefore there is no flow through it. The water is considered incompressible in our model. For volume of fluid advection for the wave boundary (i.e. the left-side boundary) in our simulations, we utilised the “Split Lagrangian Method”, which guarantees the best accuracy (Flow Science, 2022).

The stability of the numerical scheme is controlled and maintained through checking the Courant number (CC) as given in the following:

C=VΔtΔxC=VΔtΔx

(4)

where VV is the velocity of the flow, ΔtΔt is the time step and ΔxΔx is the spatial step (i.e. grid size). For stability and convergence of the numerical simulations, the Courant number must be sufficiently below one (Courant et al. 1928). This is maintained by a careful adjustment of the ΔxΔx and ΔtΔt selections. Flow3D-Hydro applies a dynamic Courant number, meaning the program adjusts the value of time step (ΔtΔt) during the simulations to achieve a balance between accuracy of results and speed of simulation. In our simulation, the time step was in the range ΔtΔt = 0.0051—0.051 s.

In order to achieve the most efficient mesh resolution, we varied cell size for five values of ΔxΔx = 0.1 m, 0.125 m, 0.15 m, 0.175 m and 0.20 m. Simulations were performed for all mesh sizes, and the results were compared in terms of convergence, stability and speed of simulation (Fig. 3). A linear wave with an amplitude of 1.5 m and a period of 6 s was used for these optimisation simulations. We considered wave time histories at two gauges A and B and recorded the waves from simulations using different mesh sizes (Fig. 3). Although the results are close (Fig. 3), some limited deviations are observed for larger mesh sizes of 0.20 m and 0.175 m. We therefore selected mesh size of 0.125 m as the optimum, giving an extra safety margin as a conservative solution.

The pressure from the incident waves on the vertical wall is validated in our model by comparing them with the analytical equation of Sainflou (1928), Eq. (3), which is one of the most common set of equations for design of coastal structures (Fig. 4). The model was tested by running a linear wave of period 6 s and wave amplitude of 1.5 m against the wall, with a still water level of 4.5 m. It can be seen that the model results are very close to those from analytical equations of Sainflou (1928), indicating that our numerical model is accurately modelling the wave-structure interaction (Fig. 4).

Eyewitness account analysis

Contemporary reporting of the 4th and 5th February 2014 storms by the main national news outlets in the UK highlights the extreme nature of the events and the significant damage and disruption they were likely to have on the communities of the south-west of England. In interviews, this was reinforced by Network Rail engineers who, even at this early stage, were forecasting remedial engineering works to last for at least 6 weeks. One week later, following subsequent storms the cascading nature of the events was obvious. Multiple breaches of the seawall had taken place with up to 35 separate landslide events and significant damage to parapet walls along the coastal route also were reported. Residents of the area reported extreme effects of the storm, one likening it to an earthquake and reporting water ingress through doors windows and even through vertical chimneys (Table 1). This suggests extreme wave overtopping volumes and large wave impact forces. One resident described the structural effects as: “the house was jumping up and down on its footings”.

Disaster management plans were quickly and effectively put into action by the local council, police service and National Rail. A major incident was declared, and decisions regarding evacuation of the residents under threat were taken around 2100 h on the night of 4th February when reports of initial damage to the seawall were received (Table 1). Local hotels were asked to provide short-term refuge to residents while local leisure facilities were prepared to accept residents later that evening. Initial repair work to the railway line was hampered by successive high spring tides and storms in the following days although significant progress was still made when weather conditions permitted (Table 1).

Sea level observations and spectral analysis

The results of surge and wave analyses are presented in Figs. 5 and 6. A surge height of up to 0.8 m was recorded in the examined tide gauge stations (Fig. 5b-d). Two main episodes of high surge heights are identified: the first surge started on 3rd February 2014 at 03:00 (UTC) and lasted until 4th of February 2014 at 00:00; the second event occurred in the period 4th February 2014 15:00 to 5th February 2014 at 17:00 (Fig. 5b-d). These data imply surge durations of 21 h and 26 h for the first and the second events, respectively. Based on the surge data in Fig. 5, we note that the storm event of early February 2014 and the associated surges was a relatively powerful one, which impacted at least 230 km of the south coast of England, from Land’s End to Weymouth, with large surge heights.

Based on wave buoy records, the maximum recorded amplitudes are at least 20.5 m in Dawlish and West Bay, 1.9 m in Tor Bay and 4.9 m in Chesil (Fig. 6a-b). The buoys at Tor Bay and Chesil recorded dual peak period bands of 4–8 and 8–12 s, whereas at Dawlish and West Bay registered triple peak period bands at 4–8, 8–12 and 20–25 s (Fig. 6c, d). It is important to note that the long-period waves at 20–25 s occur with short durations (approximately 2 min) while the waves at the other two bands of 4–8 and 8–12 s appear to be present at all times during the storm event.

The wave component at the period band of 4–8 s can be most likely attributed to normal coastal waves while the one at 8–12 s, which is longer, is most likely the swell component of the storm. Regarding the third component of the waves with long period of 20 -25 s, which occurs with short durations of 2 min, there are two hypotheses; it is either the result of a local (port and harbour) and regional (the Lyme Bay) oscillations (eg. Rabinovich 1997; Heidarzadeh and Satake 2014; Wang et al. 1992), or due to an abnormally long swell. To test the first hypothesis, we consider various water bodies such as Lyme Bay (approximate dimensions of 70 km × 20 km with an average water depth of 30 m; Fig. 6), several local bays (approximate dimensions of 3.6 km × 0.6 km with an average water depth of 6 m) and harbours (approximate dimensions of 0.5 km × 0.5 km with an average water depth of 4 m). Their water depths are based on the online Marine navigation website.^{Footnote8} According to Rabinovich (2010), the oscillation modes of a semi-enclosed rectangle basin are given by the following equation:

where TmnTmn is the oscillation period, gg is the gravitational acceleration, dd is the water depth, LL is the length of the basin, WW is the width of the basin, m=1,2,3,…m=1,2,3,… and n=0,1,2,3,…n=0,1,2,3,…; mm and nn are the counters of the different modes. Applying Eq. (5) to the aforementioned water bodies results in oscillation modes of at least 5 min, which is far longer than the observed period of 20–25 s. Therefore, we rule out the first hypothesis and infer that the long period of 20–25 s is most likely a long swell wave coming from distant sources. As discussed by Rabinovich (1997) and Wang et al. (2022), comparison between sea level spectra before and after the incident is a useful method to distinguish the spectrum of the weather event. A visual inspection of Fig. 6 reveals that the forcing at the period band of 20–25 s is non-existent before the incident.

Numerical simulations of wave loading and overtopping

Based on the results of sea level data analyses in the previous section (Fig. 6), we use a dual peak wave spectrum with peak periods of 10.0 s and 25.0 s for numerical simulations because such a wave would be comprised of the most energetic signals of the storm. For variations of water depth (2.0–4.0 m), coastal wave amplitude (0.5–1.5 m) (Fig. 7) and storm surge height (0.5–0.8 m) (Fig. 5), we developed 20 scenarios (Scn) which we used in numerical simulations (Table 2). Data during the incident indicated that water depth was up to the crest level of the seawall (approximately 4 m water depth); therefore, we varied water depth from 2 to 4 m in our simulation scenarios. Regarding wave amplitudes, we referred to the variations at a nearby tide gauge station (West Bay) which showed wave amplitude up to 1.2 m (Fig. 7). Therefore, wave amplitude was varied from 0.5 m to 1.5 m by considering a factor a safety of 25% for the maximum wave amplitude. As for the storm surge component, time series of storm surges calculated at three coastal stations adjacent to Dawlish showed that it was in the range of 0.5 m to 0.8 m (Fig. 5). These 20 scenarios would help to study uncertainties associated with wave amplitudes and pressures. Figure 8 shows snapshots of wave propagation and impacts on the seawall at different times.

Table 2 The 20 scenarios considered for numerical simulations in this study

Large wave amplitudes can induce significant wave forcing on the structure and cause overtopping of the seawall, which could eventually cascade to other hazards such as erosion of the backfill and scour (Adams and Heidarzadeh, 2021). The first 10 scenarios of our modelling efforts are for the same incident wave amplitudes of 0.5 m, which occur at different water depths (2.0–4.0 m) and storm surge heights (0.5–0.8 m) (Table 2 and Fig. 9). This is because we aim at studying the impacts of effective water depth (d_{eff}—the sum of mean sea level and surge height) on the time histories of wave amplitudes as the storm evolves. As seen in Fig. 9a, by decreasing effective water depth, wave amplitude increases. For example, for Scn-1 with effective depth of 4.5 m, the maximum amplitude of the first wave is 1.6 m, whereas it is 2.9 m for Scn-2 with effective depth of 3.5 m. However, due to intensive reflections and interferences of the waves in front of the vertical seawall, such a relationship is barely seen for the second and the third wave peaks. It is important to note that the later peaks (second or third) produce the largest waves rather than the first wave. Extraordinary wave amplifications are seen for the Scn-2 (d_{eff} = 3.5 m) and Scn-7 (d_{eff} = 3.3 m), where the corresponding wave amplitudes are 4.5 m and 3.7 m, respectively. This may indicate that the effective water depth of d_{eff} = 3.3–3.5 m is possibly a critical water depth for this structure resulting in maximum wave amplitudes under similar storms. In the second wave impact, the combined wave height (i.e. the wave amplitude plus the effective water depth), which is ultimately an indicator of wave overtopping, shows that the largest wave heights are generated by Scn-2, 7 and 8 (Fig. 9a) with effective water depths of 3.5 m, 3.3 m and 3.8 m and combined heights of 8.0 m, 7.0 m and 6.9 m (Fig. 9b). Since the height of seawall is 5.4 m, the combined wave heights for Scn-2, 7 and 8 are greater than the crest height of the seawall by 2.6 m, 1.6 m and 1.5 m, respectively, which indicates wave overtopping.

For scenarios 11–20 (Fig. 10), with incident wave amplitudes of 1.5 m (Table 2), the largest wave amplitudes are produced by Scn-17 (d_{eff} = 3.3 m), Scn-13 (d_{eff} = 2.5 m) and Scn-12 (d_{eff} = 3.5 m), which are 5.6 m, 5.1 m and 4.5 m. The maximum combined wave heights belong to Scn-11 (d_{eff} = 4.5 m) and Scn-17 (d_{eff} = 3.3 m), with combined wave heights of 9.0 m and 8.9 m (Fig. 10b), which are greater than the crest height of the seawall by 4.6 m and 3.5 m, respectively.

Our simulations for all 20 scenarios reveal that the first wave is not always the largest and wave interactions, reflections and interferences play major roles in amplifying the waves in front of the seawall. This is primarily because the wall is fully vertical and therefore has a reflection coefficient of close to one (i.e. full reflection). Simulations show that the combined wave height is up to 4.6 m higher than the crest height of the wall, implying that severe overtopping would be expected.

Results of wave loading calculations

The pressure calculations for scenarios 1–10 are given in Fig. 11 and those of scenarios 11–20 in Fig. 12. The total pressure distribution in Figs. 11, 12 mostly follows a triangular shape with maximum pressure at the seafloor as expected from the Sainflou (1928) design equations. These pressure plots comprise both static (due to mean sea level in front of the wall) and dynamic (combined effects of surge and wave) pressures. For incident wave amplitudes of 0.5 m (Fig. 11), the maximum wave pressure varies in the range of 35–63 kPa. At the sea surface, it is in the range of 4–20 kPa (Fig. 11). For some scenarios (Scn-2 and 7), the pressure distribution deviates from a triangular shape and shows larger pressures at the top, which is attributed to the wave impacts and partial breaking at the sea surface. This adds an additional triangle-shaped pressure distribution at the sea surface elevation consistent with the design procedure developed by Goda (2000) for braking waves. The maximum force on the seawall due to scenarios 1–10, which is calculated by integrating the maximum pressure distribution over the wave-facing surface of the seawall, is in the range of 92–190 KN (Table 2).

For scenarios 11–20, with incident wave amplitude of 1.5 m, wave pressures of 45–78 kPa and 7–120 kPa, for the bottom and top of the wall, respectively, were observed (Fig. 12). Most of the plots show a triangular pressure distribution, except for Scn-11 and 15. A significant increase in wave impact pressure is seen for Scn-15 at the top of the structure, where a maximum pressure of approximately 120 kPa is produced while other scenarios give a pressure of 7–32 kPa for the sea surface. In other words, the pressure from Scn-15 is approximately four times larger than the other scenarios. Such a significant increase of the pressure at the top is most likely attributed to the breaking wave impact loads as detailed by Goda (2000) and Cuomo et al. (2010). The wave simulation snapshots in Fig. 8 show that the wave breaks before reaching the wall. The maximum force due to scenarios 11–20 is 120–286 KN.

The breaking wave impacts peaking at 286 KN in our simulations suggest destabilisation of the upper masonry blocks, probably by grout malfunction. This significant impact force initiated the failure of the seawall which in turn caused extensive ballast erosion. Wave impact damage was proposed by Adams and Heidarzadeh (2021) as one of the primary mechanisms in the 2014 Dawlish disaster. In the multi-hazard risk model proposed by these authors, damage mechanism III (failure pathway 5 in Adams and Heidarzadeh, 2021) was characterised by wave impact force causing damage to the masonry elements, leading to failure of the upper sections of the seawall and loss of infill material. As blocks were removed, access to the track bed was increased for inbound waves allowing infill material from behind the seawall to be fluidised and subsequently removed by backwash. The loss of infill material critically compromised the stability of the seawall and directly led to structural failure. In parallel, significant wave overtopping (discussed in the next section) led to ballast washout and cascaded, in combination with masonry damage, to catastrophic failure of the wall and suspension of the rails in mid-air (Fig. 1b), leaving the railway inoperable for two months.

Wave Overtopping

The two most important factors contributing to the 2014 Dawlish railway catastrophe were wave impact forces and overtopping. Figure 13 gives the instantaneous overtopping rates for different scenarios, which experienced overtopping. It can be seen that the overtopping rates range from 0.5 m^{3}/s/m to 16.1 m^{3}/s/m (Fig. 13). Time histories of the wave overtopping rates show that the phenomenon occurs intermittently, and each time lasts 1.0–7.0 s. It is clear that the longer the overtopping time, the larger the volume of the water poured on the structure. The largest wave overtopping rates of 16.1 m^{3}/s/m and 14.4 m^{3}/s/m belong to Scn-20 and 11, respectively. These are the two scenarios that also give the largest combined wave heights (Fig. 10b).

The cumulative overtopping curves (Figs. 14, 15) show the total water volume overtopped the structure during the entire simulation time. This is an important hazard factor as it determines the level of soil saturation, water pore pressure in the soil and soil erosion (Van der Meer et al. 2018). The maximum volume belongs to Scn-20, which is 65.0 m^{3}/m (m-cubed of water per metre length of the wall). The overtopping volumes are 42.7 m^{3}/m for Scn-11 and 28.8 m^{3}/m for Scn-19. The overtopping volume is in the range of 0.7–65.0 m^{3}/m for all scenarios.

For comparison, we compare our modelling results with those estimated using empirical equations. For the case of the Dawlish seawall, we apply the equation proposed by Van Der Meer et al. (2018) to estimate wave overtopping rates, based on a set of decision criteria which are the influence of foreshore, vertical wall, possible breaking waves and low freeboard:

where qq is the mean overtopping rate per metre length of the seawall (m^{3}/s/m), gg is the acceleration due to gravity, HmHm is the incident wave height at the toe of the structure, RcRc is the wall crest height above mean sea level, hshs is the deep-water significant wave height and e(x)e(x) is the exponential function. It is noted that Eq. (6) is valid for 0.1<RcHm<1.350.1<RcHm<1.35. For the case of the Dawlish seawall and considering the scenarios with larger incident wave amplitude of 1.5 m (hshs= 1.5 m), the incident wave height at the toe of the structure is HmHm = 2.2—5.6 m, and the wall crest height above mean sea level is RcRc = 0.6–2.9 m. As a result, Eq. (6) gives mean overtopping rates up to approximately 2.9 m^{3}/s/m. A visual inspection of simulated overtopping rates in Fig. 13 for Scn 11–20 shows that the mean value of the simulated overtopping rates (Fig. 13) is close to estimates using Eq. (6).

Discussion and conclusions

We applied a combination of eyewitness account analysis, sea level data analysis and numerical modelling in combination with our engineering judgement to explain the damage to the Dawlish railway seawall in February 2014. Main findings are:

Eyewitness data analysis showed that the extreme nature of the event was well forecasted in the hours prior to the storm impact; however, the magnitude of the risks to the structures was not well understood. Multiple hazards were activated simultaneously, and the effects cascaded to amplify the damage. Disaster management was effective, exemplified by the establishment of an emergency rendezvous point and temporary evacuation centre during the storm, indicating a high level of hazard awareness and preparedness.

Based on sea level data analysis, we identified triple peak period bands at 4–8, 8–12 and 20–25 s in the sea level data. Storm surge heights and wave oscillations were up to 0.8 m and 1.5 m, respectively.

Based on the numerical simulations of 20 scenarios with different water depths, incident wave amplitudes, surge heights and peak periods, we found that the wave oscillations at the foot of the seawall result in multiple wave interactions and interferences. Consequently, large wave amplitudes, up to 4.6 m higher than the height of the seawall, were generated and overtopped the wall. Extreme impulsive wave impact forces of up to 286 KN were generated by the waves interacting with the seawall.

We measured maximum wave overtopping rates of 0.5–16.1 m^{3}/s/m for our scenarios. The cumulative overtopping water volumes per metre length of the wall were 0.7–65.0 m^{3}/m.

Analysis of all the evidence combined with our engineering judgement suggests that the most likely initiating cause of the failure was impulsive wave impact forces destabilising one or more grouted joints between adjacent masonry blocks in the wall. Maximum observed pressures of 286 KN in our simulations are four times greater in magnitude than background pressures leading to block removal and initiating failure. Therefore, the sequence of cascading events was :1) impulsive wave impact force causing damage to masonry, 2) failure of the upper sections of the seawall, 3) loss of infill resulting in a reduction of structural strength in the landward direction, 4) ballast washout as wave overtopping and inbound wave activity increased and 5) progressive structural failure following successive tides.

From a risk mitigation point of view, the stability of the seawall in the face of future energetic cyclonic storm events and sea level rise will become a critical factor in protecting the rail network. Mitigation efforts will involve significant infrastructure investment to strengthen the civil engineering assets combined with improved hazard warning systems consisting of meteorological forecasting and real-time wave observations and instrumentation. These efforts must take into account the amenity value of coastal railway infrastructure to local communities and the significant number of tourists who visit every year. In this regard, public awareness and active engagement in the planning and execution of the project will be crucial in order to secure local stakeholder support for the significant infrastructure project that will be required for future resilience.

Ackhurst M (2020) Design of Vertical Gravity Sea and Quay Walls. ICE Publishing, Westminster, London

Adams K, Heidarzadeh M (2021) A multi-hazard risk model with cascading failure pathways for the Dawlish (UK) railway using historical and contemporary data. Int J Disaster Risk Reduc 56:102082. https://doi.org/10.1016/j.ijdrr.2021.102082ArticleGoogle Scholar

Courant R, Friedrichs K, Lewy H (1928) Über die partiellen Differenzengleichungen der mathematischen Physik. Math Ann 100(1):32–74ArticleGoogle Scholar

Cuomo G, Allsop W, Bruce T, Pearson J (2010) Breaking wave loads at vertical seawalls and breakwaters. Coastal Eng 57(4):424–439ArticleGoogle Scholar

Dawson D, Shaw J, Gehrels WR (2016) Sea-level rise impacts on transport infrastructure: the notorious case of the coastal railway line at Dawlish, England. J Transport Geog 51:97–109ArticleGoogle Scholar

Feser F, Barcikowska M, Krueger O, Schenk F, Weisse R, Xia L (2015) Storminess over the North Atlantic and northwestern Europe—A review. Quarter J R Meteorol Soc 141:350–382. https://doi.org/10.1002/qj.2364ArticleGoogle Scholar

Heidarzadeh M, Satake K (2015) Source properties of the 1998 July 17 Papua New Guinea tsunami based on tide gauge records. Geophys J Int 202(1):361–369ArticleGoogle Scholar

Heidarzadeh M, Gusman A, Ishibe T, Sabeti R, Šepić J (2022) Estimating the eruption-induced water displacement source of the 15 January 2022 Tonga volcanic tsunami from tsunami spectra and numerical modelling. Ocean Eng 261:112165. https://doi.org/10.1016/j.oceaneng.2022.112165ArticleGoogle Scholar

Masselink G, Castelle B, Scott T, Dodet G, Suanez S, Jackson D, Floc’h F (2016) Extreme wave activity during 2013/2014 winter and morphological impacts along the Atlantic coast of Europe. Geophys Res Lett 43(5):2135–2143. https://doi.org/10.1002/2015GL067492ArticleGoogle Scholar

Mathworks (2018) MATLAB, 2018. 9.7.0.1190202 (R2019b), Natick, Massachusetts: The MathWorks Inc.

Van der Meer JW, Allsop NWH, Bruce T, De Rouck J, Kortenhaus A, Pullen T, Schüttrumpf H, Troch P and Zanuttigh B (2018) EurOtop, Manual on wave overtopping of sea defences and related structures. An overtopping manual largely based on European research, but for worldwide application. Available online at: www.overtopping-manual.com.

Wang Y, Su HY, Ren Z, Ma Y (2022) Source properties and resonance characteristics of the tsunami generated by the 2021 M 8.2 Alaska earthquake. J Geophys Res Oceans, 127(3):e2021JC018308. https://doi.org/10.1029/2021JC018308ArticleGoogle Scholar

We are grateful to Brunel University London for administering the scholarship awarded to KA. The Flow3D-Hydro used in this research for numerical modelling is licenced to Brunel University London through an academic programme contract. We sincerely thank Prof Harsh Gupta (Editor-in-Chief) and two anonymous reviewers for their constructive review comments.

Funding

This project was funded by the UK Engineering and Physical Sciences Research Council (EPSRC) through a PhD scholarship to Keith Adams.

Author information

Authors and Affiliations

Department of Civil and Environmental Engineering, Brunel University London, Uxbridge, UB8 3PH, UKKeith Adams

Department of Architecture and Civil Engineering, University of Bath, Bath, BA2 7AY, UKMohammad Heidarzadeh

The authors have no relevant financial or non-financial interests to disclose.

Availability of data

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

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Adams, K., Heidarzadeh, M. Extratropical cyclone damage to the seawall in Dawlish, UK: eyewitness accounts, sea level analysis and numerical modelling. Nat Hazards (2022). https://doi.org/10.1007/s11069-022-05692-2

Abstract해저 협곡에서 탁도의 장거리 이동은 많은 양의 퇴적물을 심해 평원으로 운반할 수 있습니다. 이전 연구에서는 5.9~28.0m/s 범위의 다중 케이블 손상 이벤트에서 파생된 탁도 전류 속도와 0.15~7.2m/s 사이의 현장 관찰 결과에서 명백한 차이가 있음을 보여줍니다. 따라서 해저 환경의 탁한 유체가 해저 협곡을 고속으로 장거리로 흐를 수 있는지에 대한 질문이 남아 있습니다. 연구실 시험의 결합을 통해 해저협곡의 탁류의 고속 및 장거리 운동을 설명하기 위해 약안정 퇴적물 기반의 새로운 모델(약안정 퇴적물에 대한 파손 전파 모델 제안, 줄여서 WSS-PFP 모델)을 제안합니다. 및 수치 아날로그. 이 모델은 두 가지 메커니즘을 기반으로 합니다. 1) 원래 탁도류는 약하게 안정한 퇴적층의 불안정화를 촉발하고 연질 퇴적물의 불안정화 및 하류 방향으로의 이동을 촉진하고 2) 원래 탁도류가 협곡으로 이동할 때 형성되는 여기파가 불안정화로 이어진다. 하류 방향으로 약하게 안정한 퇴적물의 수송. 제안된 모델은 심해 퇴적, 오염 물질 이동 및 광 케이블 손상 연구를 위한 동적 프로세스 해석을 제공할 것입니다.

The long-distance movement of turbidity currents in submarine canyons can transport large amounts of sediment to deep-sea plains. Previous studies show obvious differences in the turbidity current velocities derived from the multiple cables damage events ranging from 5.9 to 28.0 m/s and those of field observations between 0.15 and 7.2 m/s. Therefore, questions remain regarding whether a turbid fluid in an undersea environment can flow through a submarine canyon for a long distance at a high speed. A new model based on weakly stable sediment is proposed (proposed failure propagation model for weakly stable sediments, WSS-PFP model for short) to explain the high-speed and long-range motion of turbidity currents in submarine canyons through the combination of laboratory tests and numerical analogs. The model is based on two mechanisms: 1) the original turbidity current triggers the destabilization of the weakly stable sediment bed and promotes the destabilization and transport of the soft sediment in the downstream direction and 2) the excitation wave that forms when the original turbidity current moves into the canyon leads to the destabilization and transport of the weakly stable sediment in the downstream direction. The proposed model will provide dynamic process interpretation for the study of deep-sea deposition, pollutant transport, and optical cable damage.

Keyword

turbidity current

excitation wave

dense basal layer

velocity

WSS-PFP model

References

Azpiroz-Zabala M, Cartigny M J B, Talling P J et al. 2017. Newly recognized turbidity current structure can explain prolonged flushing of submarine canyons. Science Advances, 3(10): e1700200, https://doi.org/10.1126/sciadv.1700200.ArticleGoogle Scholar

Bagnold R A. 1962. Auto-suspension of transported sediment; turbidity currents. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 265(1322): 315–319, https://doi.org/10.1098/rspa.1962.0012.Google Scholar

Carter L, Milliman J D, Talling P J et al. 2012. Near-synchronous and delayed initiation of long run-out submarine sediment flows from a record-breaking river flood, offshore Taiwan. Geophysical Research Letters, 39(12): L12603, https://doi.org/10.1029/2012gl051172.ArticleGoogle Scholar

Cooper C, Wood J, Imran J et al. 2016. Designing for turbidity currents in the Congo Canyon. In: Offshore Technology Conference. OTC, Houston, TX. OTC-26919-MSp, https://doi.org/10.4043/26919-ms.Google Scholar

Gavey R, Carter L, Liu J T et al. 2017. Frequent sediment density flows during 2006 to 2015, triggered by competing seismic and weather events: observations from subsea cable breaks off southern Taiwan. Marine Geology, 384: 147–158, https://doi.org/10.1016/j.margeo.2016.06.001.ArticleGoogle Scholar

Nie X, Luo W D, Zhou J. 2017. Depositional characteristics of the Penghu submarine canyon in the northeastern South China Sea. Marine Geology Frontiers, 33(8): 18–23, https://doi.org/10.16028/j.1009-2722.2017.08003. (in Chinese with English abstract)Google Scholar

Paull C K, Caress D W, Ussler III B et al. 2011. High-resolution bathymetry of the axial channels within Monterey and Soquel submarine canyons, offshore central California. Geosphere, 7(5): 1077–1101, https://doi.org/10.1130/GES00636.1.ArticleGoogle Scholar

Piper D J W, Shor A N, Clarke J E H. 1988. The 1929 “Grand banks” earthquake, slump, and turbidity current. In: Clifton H E ed. Sedimentologic Consequences of Convulsive Geologic Events. Geological Society of America. p.77–92, https://doi.org/10.1130/SPE229-p77.

Shepard F P. 1954. High-velocity turbidity currents, a discussion. Proceedings of the Royal Society of Series A: Mathematical, Physical and Engineering Sciences, 222(1150): 323–326, https://doi.org/10.1098/rspa.1954.0072.Google Scholar

Symons W Q, Sumner E J, Paull C K et al. 2017. A new model for turbidity current behavior based on integration of flow monitoring and precision coring in a submarine canyon. Geology, 45(4): 367–370, https://doi.org/10.1130/g38764.1.ArticleGoogle Scholar

Talling P J, Allin J, Armitage D A et al. 2015. Key future directions for research on turbidity currents and their deposits. Journal of Sedimentary Research, 85(2): 153–169, https://doi.org/10.2110/jsr.2015.03.ArticleGoogle Scholar

Wang Z W, Xu J P, Talling P J et al. 2020. Direct evidence of a high-concentration basal layer in a submarine turbidity current. Deep Sea Research Part I: Oceanographic Research Papers, 161: 103300, https://doi.org/10.1016/j.dsr.2020.103300.ArticleGoogle Scholar

We thank Hanru WU from Ocean University of China for his help in thesis writing, and Hao TIAN and Chenxi WANG from Ocean University of China for their helps in the preparation of the experimental materials. Guohui XU is responsible for the development of the initial concept, processing of test data, and management of coauthor contributions to the paper; Yupeng REN for the experiment setup and drafting of the paper; Yi ZHANG and Xingbei XU for the simulation part of the experiment; Houjie WANG for writing guidance; Zhiyuan CHEN for the experiment setup.

Author information

Authors and Affiliations

Shandong Provincial Key Laboratory of Marine Environment and Geological Engineering, Qingdao, 266100, ChinaYupeng Ren, Yi Zhang, Guohui Xu, Xingbei Xu & Zhiyuan Chen

Shandong Provincial Key Laboratory of Marine Environment and Geological Engineering, Ocean University of China, Qingdao, 266100, ChinaYupeng Ren & Houjie Wang

Key Laboratory of Marine Environment and Ecology, Ocean University of China, Ministry of Education, Qingdao, 266100, ChinaYi Zhang, Guohui Xu, Xingbei Xu & Zhiyuan Chen

Supported by the National Natural Science Foundation of China (Nos. 41976049, 41720104001) and the Taishan Scholar Project of Shandong Province (No. TS20190913), and the Fundamental Research Funds for the Central Universities (No. 202061028)

Data Availability Statement

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Ren, Y., Zhang, Y., Xu, G. et al. The failure propagation of weakly stable sediment: A reason for the formation of high-velocity turbidity currents in submarine canyons. J. Ocean. Limnol. (2022). https://doi.org/10.1007/s00343-022-1285-0

TianLi^{ab}J.M.T.Davies^{a}XiangzhenZhu^{c} ^{a}University of Birmingham, Birmingham B15 2TT, United Kingdom ^{b}Grainger and Worrall Ltd, Bridgnorth WV15 5HP, United Kingdom ^{c}Brunel 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., SF_{6}/CO_{2}, SF_{6}/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.

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 Al_{8}Mn_{5} 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.

SF_{6} 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 SF_{6} cover gas (i.e., the reaction between liquid Mg-alloy and SF_{6} 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 SF_{6} was MgO mainly with traces of fluorides, and suggested that SF_{6} was absorbed in the Mg-alloy surface film. Couling [29] further noticed that the absorbed SF_{6} reacted with the Mg-alloy melt to form MgF_{2}. 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 MgF_{2}. 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 MgF_{2} 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 MgF_{2} 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 MgF_{2} layers.(4)

Oxide film consisted of discrete particles. Wang et al [36] stirred the Mg-alloy surface film into the melt under a SF_{6} 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 SF_{6} 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%SF_{6}/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 MgF_{2}. However, the nucleation sites (such as the MgF_{2} spots observed by Aarstad [32] and Xiong [25,33]) were not observed. Mirak also speculated that the MgF_{2} 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 MgF_{2}). 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. SF_{6}/air required a higher content of SF_{6} than did a SF_{6}/CO_{2} 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 CO_{2} was used as a carrier gas, which was different from the films formed in SF_{6}/air. An investigation into Mg combustion [40] reported a detection of Mg_{2}C_{3} in the Mg-alloy sample after burning in CO_{2}, 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., SF_{6}/air and SF_{6}/CO_{2}). 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%SF_{6}/air or 0.5%SF_{6}/CO_{2} (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.

Al

Zn

Mn

Si

Fe

Ni

Mg

9.4

0.61

0.15

0.02

0.005

0.0017

Residual

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).

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.% Na_{2}SiF_{6} 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 SF_{6} 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).

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 SF_{6}/air

The structure and composition of the entrainment defect formed in the AZ91 castings under a cover gas of 0.5%SF_{6}/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. 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 SF_{6} cover gas. Sulphur could not be clearly recognized in the element map, but there was a S-peak in the corresponding ESD spectrum.

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 SF_{6} 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.

Figs. 4 and 5 show cross sectional observations of the entrainment defects formed in the AZ91 alloy sample cast under a cover gas of SF_{6}/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 SF_{6}/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].

Table 2. EDS results (wt.%) corresponding to the regions shown in Fig. 6 (cover gas: SF_{6}/air).

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 SF_{6}/CO_{2}

Similar to the entrainment defect formed in SF_{6}/air, the defects formed under a cover gas of 0.5%SF_{6}/CO_{2} 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 SF_{6}/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. 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%SF_{6}/CO_{2}. 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.

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.

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 Mg_{2}AlO_{4} because it also contains abundant magnesium and oxygen elements. The existing of such Mg_{2}AlO_{4} 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 O_{2}, thus only a few of MgO particles formed in our work. An enrichment of carbon was also found in the film, revealing that CO_{2} 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: SF_{6}/ CO_{2}).

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 SF_{6}/CO_{2} and SF_{6}/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 SF_{6}/air and SF_{6}/CO_{2}. 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%SF_{6}/air. There was only one single layer consisting of fluoride and oxide (MgF_{2} 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.

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 SF_{6}/air, the oxide films formed in SF_{6}/CO_{2} 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%SF_{6}/CO_{2} for 5 min. This film had a single-layered structure consisting of MgF_{2}. 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 MgF_{2}, while the outer layer is the mixture of MgF_{2} and MgO. Sulphur was not detected in this film, which was different from the surface film formed in 0.5%SF_{6}/air. Therefore, fluorides in the cover gas of 0.5%SF_{6}/CO_{2} were also preferentially consumed at an early stage of the film growth process. Compared with the film formed in SF_{6}/air, the MgO in film formed in SF_{6}/CO_{2} appeared later and sulphide did not appear within 30 min. It may mean that the formation and evolution of film in SF_{6}/air is faster than SF_{6}/CO_{2.} CO_{2} 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).

4. Discussion

4.1. Evolution of entrainment defects formed in SF_{6}/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 cm^{3} (3.14 × 10^{−8} kmol) for 0.5%SF_{6}/air, and 0.35 cm^{3} (3.12 × 10^{−8} kmol) for 0.5%SF_{6}/CO_{2}. 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 SF_{6} were SF_{5}, SF_{4}, SF_{3}, SF_{2}, F_{2}, S(g), S_{2}(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%SF_{6}/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 SF_{6} present (≈ 1.57 × 10^{−10} kmol) and therefore would not affect the observed process in a practical way.

This reaction process could be divided into 3 stages.

Stage 1: The formation of fluorides. the AZ91 melt preferentially reacted with SF_{6} and its decomposition products, producing MgF_{2}, AlF_{3}, and ZnF_{2}. However, the amount of ZnF_{2} may have been too small to be detected practically (1.25 × 10^{−12} kmol of ZnF_{2} compared with 3 × 10^{−10} kmol of MgF_{2}), 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 SO_{2}.

Stage 2: The formation of oxides. After the liquid AZ91 alloy had depleted all the available fluorides in the entrapped gas, the amount of AlF_{3} and ZnF_{2} quickly reduced due to a reaction with Mg. O_{2}(g) and SO_{2} reacted with the AZ91 melt, forming MgO, Al_{2}O_{3}, MgAl_{2}O_{4}, ZnO, ZnSO_{4} and MgSO_{4}. However, the amount of ZnO and ZnSO_{4} 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 ZnSO_{4}, in contrast to 4.68 × 10^{−10} kmol of MgF_{2}, 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 MgAl_{2}O_{4} and Al_{2}O_{3} 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)

Mg + MgAl_{2}O_{4} = 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)

Mg_{3}N_{2} + 6H_{2}O =3Mg(OH)_{2} + 2NH_{3}↑(4)

AlN+ 3H_{2}O =Al(OH)_{3} + NH_{3}↑

In addition, Schmidt et al. [61] found that Mg_{3}N_{2} and AlN could react to form ternary nitrides (Mg_{3}Al_{n}N_{n+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 SF_{6}/CO_{2}

Fig. 13 shows the results of the thermodynamic calculation between AZ91 alloy and 0.5%SF_{6}/CO_{2}. This reaction processes can also be divided into three stages.

Stage 1: The formation of fluorides. SF_{6} and its decomposition products were consumed by the AZ91 melt, forming MgF_{2}, AlF_{3}, and ZnF_{2}. As in the reaction of AZ91 in 0.5%SF_{6}/air, the amount of ZnF_{2} was too small to be detected practically (1.51 × 10^{−13} kmol of ZnF_{2} compared with 2.67 × 10^{−10} kmol of MgF_{2}). Sulphur accumulated in the residual trapped gas as S_{2}(g) and a portion of the S_{2}(g) reacted with CO_{2}, to form SO_{2} 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. AlF_{3} and ZnF_{2} reacted with the Mg in the AZ91 melt, forming MgF_{2}, Al and Zn. The SO_{2} began to be consumed, producing oxides in the surface film and S_{2}(g) in the cover gas. Meanwhile, the CO_{2} directly reacted with the AZ91 melt, forming CO, MgO, ZnO, and Al_{2}O_{3}. 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 SO_{2}/CO_{2}. The produced Al_{2}O_{3} may be further combined with MgO, forming MgAl_{2}O_{4}[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 S_{2}(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 SF_{6}/air and SF_{6}/CO_{2}, 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%SF_{6}/air and 0.5%SF_{6}/CO_{2}). 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%SF_{6}/air grew faster than those formed in 0.5%SF_{6}/CO_{2}. The different growth rates suggested that the entrained-gas consumption rate of 0.5%SF_{6}/air was higher than that of 0.5%SF_{6}/CO_{2}, which was more beneficial for the consumption of the entrained gas.

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%SF_{6}/air and 0.5%SF_{6}/CO_{2}) 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%SF_{6}/air and 0.5%SF_{6}/CO_{2} 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 SF_{6}/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 SF_{6}/CO_{2} are 7.7 and 2.7 respectively, suggesting that the reproducibility of the casting protected by SF_{6}/CO_{2} were much lower than that produced in SF_{6}/air.

In addition, the author’s previous publication [69] demonstrated a shortcoming of the linearized Weibull plots, which may cause a higher bias and incorrect R^{2} 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 SF_{6}/air casting was 20.8, while the casting produced under SF_{6}/CO_{2} had a lower UTS Weibull modulus of 11.4, showing a clear difference in their reproducibility. In addition, the SF_{6}/air elongation (El%) dataset also had a Weibull modulus (shape = 5.8) higher than the elongation dataset of SF_{6}/CO_{2} (shape = 3.1). Therefore, both the LLS and Non-LLS estimations suggested that the SF_{6}/air casting has a higher reproducibility than the SF_{6}/CO_{2} casting. It supports the method that the use of air instead of CO_{2} contributes to a quicker consumption of the entrained gas, which may reduce the void volume within the defects. Therefore, the use of 0.5%SF_{6}/air instead of 0.5%SF_{6}/CO_{2} (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 SF_{6} 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 SO_{2}/air, rather than the protective gas.

Therefore, although the results in present work have shown that using air instead of CO_{2} 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 SF_{6}/air had a similar structure to those formed in SF_{6}/CO_{2}, 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 MgF_{2}).3.

The growth rate of the oxide film was demonstrated to be greater under SF_{6}/air than SF_{6}/CO_{2}, contributing to a quicker consumption of the damaging entrapped gas. The reproducibility of an AZ91 alloy casting improved when using SF_{6}/air instead of SF_{6}/CO_{2}.

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.

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.

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.

316-L 스테인리스강의 레이저 분말 베드 융합 중 콜드 스패터 형성의 충실도 높은 수치 모델링

W.E. ALPHONSO^{1}*, M. BAYAT^{1} and J.H. HATTEL^{1} *Corresponding author 1Technical University of Denmark (DTU), 2800, Kgs, Lyngby, Denmark

ABSTRACT

L-PBF(Laser Powder Bed Fusion)는 금속 적층 제조(MAM) 기술로, 기존 제조 공정에 비해 부품 설계 자유도, 조립품 통합, 부품 맞춤화 및 낮은 툴링 비용과 같은 여러 이점을 산업에 제공합니다.

전기 코일 및 열 관리 장치는 일반적으로 높은 전기 및 열 전도성 특성으로 인해 순수 구리로 제조됩니다. 따라서 순동의 L-PBF가 가능하다면 기하학적으로 최적화된 방열판과 자유형 전자코일을 제작할 수 있습니다.

그러나 L-PBF로 조밀한 순동 부품을 생산하는 것은 적외선에 대한 낮은 광 흡수율과 높은 열전도율로 인해 어렵습니다. 기존의 L-PBF 시스템에서 조밀한 구리 부품을 생산하려면 적외선 레이저의 출력을 500W 이상으로 높이거나 구리의 광흡수율이 높은 녹색 레이저를 사용해야 합니다.

적외선 레이저 출력을 높이면 후면 반사로 인해 레이저 시스템의 광학 구성 요소가 손상되고 렌즈의 열 광학 현상으로 인해 공정이 불안정해질 수 있습니다. 이 작업에서 FVM(Finite Volume Method)에 기반한 다중 물리학 중간 규모 수치 모델은 Flow-3D에서 개발되어 용융 풀 역학과 궁극적으로 부품 품질을 제어하는 물리적 현상 상호 작용을 조사합니다.

녹색 레이저 열원과 적외선 레이저 열원은 기판 위의 순수 구리 분말 베드에 단일 트랙 증착을 생성하기 위해 개별적으로 사용됩니다.

용융 풀 역학에 대한 레이저 열원의 유사하지 않은 광학 흡수 특성의 영향이 탐구됩니다. 수치 모델을 검증하기 위해 단일 트랙이 구리 분말 베드에 증착되고 시뮬레이션된 용융 풀 모양과 크기가 비교되는 실험이 수행되었습니다.

녹색 레이저는 광흡수율이 높아 전도 및 키홀 모드 용융이 가능하고 적외선 레이저는 흡수율이 낮아 키홀 모드 용융만 가능하다. 레이저 파장에 대한 용융 모드의 변화는 궁극적으로 기계적, 전기적 및 열적 특성에 영향을 미치는 열 구배 및 냉각 속도에 대한 결과를 가져옵니다.

Laser Powder Bed Fusion (L-PBF) is a Metal Additive Manufacturing (MAM) technology which offers several advantages to industries such as part design freedom, consolidation of assemblies, part customization and low tooling cost over conventional manufacturing processes. Electric coils and thermal management devices are generally manufactured from pure copper due to its high electrical and thermal conductivity properties. Therefore, if L-PBF of pure copper is feasible, geometrically optimized heat sinks and free-form electromagnetic coils can be manufactured. However, producing dense pure copper parts by L-PBF is difficult due to low optical absorptivity to infrared radiation and high thermal conductivity. To produce dense copper parts in a conventional L-PBF system either the power of the infrared laser must be increased above 500W, or a green lase