Sketch of approach channel and spillway of the Kamal-Saleh dam

CFD modeling of flow pattern in spillway’s approach channel

Sustainable Water Resources Management volume 1, pages245–251 (2015)Cite this article

Abstract

Analysis of behavior and hydraulic characteristics of flow over the dam spillway is a complicated task that takes lots of money and time in water engineering projects planning. To model those hydraulic characteristics, several methods such as physical and numerical methods can be used. Nowadays, by utilizing new methods in computational fluid dynamics (CFD) and by the development of fast computers, the numerical methods have become accessible for use in the analysis of such sophisticated flows. The CFD softwares have the capability to analyze two- and three-dimensional flow fields. In this paper, the flow pattern at the guide wall of the Kamal-Saleh dam was modeled by Flow 3D. The results show that the current geometry of the left wall causes instability in the flow pattern and making secondary and vortex flow at beginning approach channel. This shape of guide wall reduced the performance of weir to remove the peak flood discharge.

댐 여수로 흐름의 거동 및 수리학적 특성 분석은 물 공학 프로젝트 계획에 많은 비용과 시간이 소요되는 복잡한 작업입니다. 이러한 수력학적 특성을 모델링하기 위해 물리적, 수치적 방법과 같은 여러 가지 방법을 사용할 수 있습니다. 요즘에는 전산유체역학(CFD)의 새로운 방법을 활용하고 빠른 컴퓨터의 개발로 이러한 정교한 흐름의 해석에 수치 방법을 사용할 수 있게 되었습니다. CFD 소프트웨어에는 2차원 및 3차원 유동장을 분석하는 기능이 있습니다. 본 논문에서는 Kamal-Saleh 댐 유도벽의 흐름 패턴을 Flow 3D로 모델링하였다. 결과는 왼쪽 벽의 현재 형상이 흐름 패턴의 불안정성을 유발하고 시작 접근 채널에서 2차 및 와류 흐름을 만드는 것을 보여줍니다. 이러한 형태의 안내벽은 첨두방류량을 제거하기 위해 둑의 성능을 저하시켰다.

Introduction

Spillways are one of the main structures used in the dam projects. Design of the spillway in all types of dams, specifically earthen dams is important because the inability of the spillway to remove probable maximum flood (PMF) discharge may cause overflow of water which ultimately leads to destruction of the dam (Das and Saikia et al. 2009; E 2013 and Novak et al. 2007). So study on the hydraulic characteristics of this structure is important. Hydraulic properties of spillway including flow pattern at the entrance of the guide walls and along the chute. Moreover, estimating the values of velocity and pressure parameters of flow along the chute is very important (Chanson 2004; Chatila and Tabbara 2004). The purpose of the study on the flow pattern is the effect of wall geometry on the creation transverse waves, flow instability, rotating and reciprocating flow through the inlet of spillway and its chute (Parsaie and Haghiabi 2015ab; 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 characteristics of this structure has been done with these techniques (Chatila and Tabbara 2004; Zhenwei et al. 2012). Using the CFD as a powerful technique for modeling the hydraulic structures can reduce the time and cost of experiments (Tabbara et al. 2005). In CFD field, the Navier–Stokes equation is solved by powerful numerical methods such as finite element method and finite volumes (Kim and Park 2005; Zhenwei et al. 2012). In order to obtain closed-form Navier–Stokes equations turbulence models, such k − ε and Re-Normalisation Group (RNG) models have been presented. To use the technique of computational fluid dynamics, software packages such as Fluent and Flow 3D, etc., are provided. Recently, these two software packages have been widely used in hydraulic engineering because the performance and their accuracy are very suitable (Gessler 2005; Kim 2007; Kim et al. 2012; Milési and Causse 2014; Montagna et al. 2011). In this paper, to assess the flow pattern at Kamal-Saleh guide wall, numerical method has been used. All the stages of numerical modeling were conducted in the Flow 3D software.

Materials and methods

Firstly, a three-dimensional model was constructed according to two-dimensional map that was prepared for designing the spillway. Then a small model was prepared with scale of 1:80 and entered into the Flow 3D software; all stages of the model construction was conducted in AutoCAD 3D. Flow 3D software numerically solved the Navier–Stokes equation by finite volume method. Below is a brief reference on the equations that used in the software. Figure 1 shows the 3D sketch of Kamal-Saleh spillway and Fig. 2 shows the uploading file of the Kamal-Saleh spillway in Flow 3D software.

figure 1
Fig. 1
figure 2
Fig. 2

Review of the governing equations in software Flow 3D

Continuity equation at three-dimensional Cartesian coordinates is given as Eq (1).

vf∂ρ∂t+∂∂x(uAx)+∂∂x(vAy)+∂∂x(wAz)=PSORρ,vf∂ρ∂t+∂∂x(uAx)+∂∂x(vAy)+∂∂x(wAz)=PSORρ,

(1)

where uvz are velocity component in the x, y, z direction; A xA yA z cross-sectional area of the flow; ρ fluid density; PSOR the source term; v f is the volume fraction of the fluid and three-dimensional momentum equations given in Eq (2).

∂u∂t+1vf(uAx∂u∂x+vAy∂u∂y+wAz∂u∂z)=−1ρ∂P∂x+Gx+fx∂v∂t+1vf(uAx∂v∂x+vAy∂v∂y+wAz∂v∂z)=−1ρ∂P∂y+Gy+fy∂w∂t+1vf(uAx∂w∂x+vAy∂w∂y+wAz∂w∂z)=−1ρ∂P∂y+Gz+fz,∂u∂t+1vf(uAx∂u∂x+vAy∂u∂y+wAz∂u∂z)=−1ρ∂P∂x+Gx+fx∂v∂t+1vf(uAx∂v∂x+vAy∂v∂y+wAz∂v∂z)=−1ρ∂P∂y+Gy+fy∂w∂t+1vf(uAx∂w∂x+vAy∂w∂y+wAz∂w∂z)=−1ρ∂P∂y+Gz+fz,

(2)

where P is the fluid pressure; G xG yG z the acceleration created by body fluids; f xf yf z viscosity acceleration in three dimensions and v f is related to the volume of fluid, defined by Eq. (3). For modeling of free surface profile the VOF technique based on the volume fraction of the computational cells has been used. Since the volume fraction F represents the amount of fluid in each cell, it takes value between 0 and 1.

∂F∂t+1vf[∂∂x(FAxu)+∂∂y(FAyv)+∂∂y(FAzw)]=0∂F∂t+1vf[∂∂x(FAxu)+∂∂y(FAyv)+∂∂y(FAzw)]=0

(3)

Turbulence models

Flow 3D offers five types of turbulence models: Prantl mixing length, k − ε equation, RNG models, Large eddy simulation model. Turbulence models that have been proposed recently are based on Reynolds-averaged Navier–Stokes equations. This approach involves statistical methods to extract an averaged equation related to the turbulence quantities.

Steps of solving a problem in Flow 3D software

(1) Preparing the 3D model of spillway by AutoCAD software. (2) Uploading the file of 3D model in Flow 3D software and defining the problem in the software and checking the final mesh. (3) Choosing the basic equations that should be solved. (4) Defining the characteristics of fluid. (5) Defining the boundary conditions; it is notable that this software has a wide range of boundary conditions. (6) Initializing the flow field. (7) Adjusting the output. (8) Adjusting the control parameters, choice of the calculation method and solution formula. (9) Start of calculation. Figure 1 shows the 3D model of the Kamal-Saleh spillway; in this figure, geometry of the left and right guide wall is shown.

Figure 2 shows the uploading of the 3D spillway dam in Flow 3D software. Moreover, in this figure the considered boundary condition in software is shown. At the entrance and end of spillway, the flow rate or fluid elevation and outflow was considered as BC. The bottom of spillway was considered as wall and left and right as symmetry.

Model calibration

Calibration of the Flow 3D for modeling the effect of geometry of guide wall on the flow pattern is included for comparing the results of Flow 3D with measured water surface profile. Calibration the Flow 3D software could be conducted in two ways: first, changing the value of upstream boundary conditions is continued until the results of water surface profile of the Flow 3D along the spillway successfully covered the measurement water surface profile; second is the assessment the mesh sensitivity. Analyzing the size of mesh is a trial-and-error process where the size of mesh is evaluated form the largest to the smallest. With fining the size of mesh the accuracy of model is increased; whereas, the cost of computation is increased. In this research, the value of upstream boundary condition was adjusted with measured data during the experimental studies on the scaled model and the mesh size was equal to 1 × 1 × 1 cm3.

Results and discussion

The behavior of water in spillway is strongly affected by the flow pattern at the entrance of the spillway, the flow pattern formation at the entrance is affected by the guide wall, and choice of an optimized form for the guide wall has a great effect on rising the ability of spillway for easy passing the PMF, so any nonuniformity in flow in the approach channel can cause reduction of spillway capacity, reduction in discharge coefficient of spillway, and even probability of cavitation. Optimizing the flow guiding walls (in terms of length, angle and radius) can cause the loss of turbulence and flow disturbances on spillway. For this purpose, initially geometry proposed for model for the discharge of spillway dam, Kamal-Saleh, 80, 100, and 120 (L/s) were surveyed. These discharges of flow were considered with regard to the flood return period, 5, 100 and 1000 years. Geometric properties of the conducting guidance wall are given in Table 1.Table 1 Characteristics and dimensions of the guidance walls tested

Full size table

Results of the CFD simulation for passing the flow rate 80 (L/s) are shown in Fig. 3. Figure 3 shows the secondary flow and vortex at the left guide wall.

figure 3
Fig. 3

For giving more information about flow pattern at the left and right guide wall, Fig. 4 shows the flow pattern at the right side guide wall and Fig. 5 shows the flow pattern at the left side guide wall.

figure 4
Fig. 4
figure 5
Fig. 5

With regard to Figs. 4 and 5 and observing the streamlines, at discharge equal to 80 (L/s), the right wall has suitable performance but the left wall has no suitable performance and the left wall of the geometric design creates a secondary and circular flow, and vortex motion in the beginning of the entrance of spillway that creates cross waves at the beginning of spillway. By increasing the flow rate (Q = 100 L/s), at the inlet spillway secondary flows and vortex were removed, but the streamline is severely distorted. Results of the guide wall performances at the Q = 100 (L/s) are shown in Fig. 6.

figure 6
Fig. 6

Also more information about the performance of each guide wall can be derived from Figs. 7 and 8. These figures uphold that the secondary and vortex flows were removed, but the streamlines were fully diverted specifically near the left side guide wall.

figure 7
Fig. 7
figure 8
Fig. 8

As mentioned in the past, these secondary and vortex flows and diversion in streamline cause nonuniformity and create cross wave through the spillway. Figure 9 shows the cross waves at the crest of the spillway.

figure 9
Fig. 9

The performance of guide walls at the Q = 120 (L/s) also was assessed. The result of simulation is shown in Fig. 10. Figures 11 and 12 show a more clear view of the streamlines near to right and left side guide wall, respectively. As seen in Fig. 12, the left side wall still causes vortex flow and creation of and diversion in streamline.

figure 10
Fig. 10
figure 11
Fig. 11
figure 12
Fig. 12

The results of the affected left side guide wall shape on the cross wave creation are shown in Fig. 13. As seen from Fig. 3, the left side guide wall also causes cross wave at the spillway crest.

figure 13
Fig. 13

As can be seen clearly in Figs. 9 and 13, by moving from the left side to the right side of the spillway, the cross waves and the nonuniformity in flow is removed. By reviewing Figs. 9 and 13, it is found that the right side guide wall removes the cross waves and nonuniformity. With this point as aim, a geometry similar to the right side guide wall was considered instead of the left side guide wall. The result of simulation for Q = 120 (L/s) is shown in Fig. 14. As seen from this figure, the proposed geometry for the left side wall has suitable performance smoothly passing the flow through the approach channel and spillway.

figure 14
Fig. 14

More information about the proposed shape for the left guide wall is shown in Fig. 15. As seen from this figure, this shape has suitable performance for removing the cross waves and vortex flows.

figure 15
Fig. 15

Figure 16 shows the cross section of flow at the crest of spillway. As seen in this figure, the proposed shape for the left side guide wall is suitable for removing the cross waves and secondary flows.

figure 16
Fig. 16

Conclusion

Analysis of behavior and hydraulic properties of flow over the spillway dam is a complicated task which is cost and time intensive. Several techniques suitable to the purposes of study have been undertaken in this research. Physical modeling, usage of expert experience, usage of mathematical models on simulation flow in one-dimensional, two-dimensional and three-dimensional techniques, are some of the techniques utilized to study this phenomenon. The results of the modeling show that the CFD technique is a suitable tool for simulating the flow pattern in the guide wall. Using this tools helps the designer for developing the optimal shape for hydraulic structure which the flow pattern through them are important.

References

  • Chanson H (2004) 19—Design of weirs and spillways. In: Chanson H (ed) Hydraulics of open channel flow, 2nd edn. Butterworth-Heinemann, Oxford, pp 391–430Chapter Google Scholar 
  • Chatila J, Tabbara M (2004) Computational modeling of flow over an ogee spillway. Comput Struct 82:1805–1812Article Google Scholar 
  • Das MM, Saikia MD (2009) Irrigation and water power engineering. PHI Learning, New DelhiGoogle Scholar 
  • E, Department Of Army: U.S. Army Corps (2013) Hydraulic Design of Spillways. BiblioBazaar, CharlestonGoogle Scholar 
  • Fattor C, Bacchiega J (2009) Design conditions for morning-glory spillways: application to potrerillos dam spillway. Adv Water Res Hydraul Eng Springer, Berlin, pp 2123–2128Google Scholar 
  • Gessler D (2005) CFD modeling of spillway performance. Impacts Glob Clim Change. doi:10.1061/40792(173)398
  • Kim D-G (2007) Numerical analysis of free flow past a sluice gate. KSCE J Civ Eng 11:127–132Article Google Scholar 
  • Kim D, Park J (2005) Analysis of flow structure over ogee-spillway in consideration of scale and roughness effects by using CFD model. KSCE J Civ Eng 9:161–169Article Google Scholar 
  • Kim S, Yu K, Yoon B, Lim Y (2012) A numerical study on hydraulic characteristics in the ice Harbor-type fishway. KSCE J Civ Eng 16:265–272Article Google Scholar 
  • Ma X-D, Dai G-Q, Yang Q, Li G-J, Zhao L (2010) Analysis of influence factors of cavity length in the spillway tunnel downstream of middle gate chamber outlet with sudden lateral enlargement and vertical drop aerator. J Hydrodyn Ser B 22:680–686Article Google Scholar 
  • Milési G, Causse S (2014) 3D numerical modeling of a side-channel spillway. In: Gourbesville P, Cunge J, Caignaert G (eds) Advances in hydroinformatics. Springer, Singapore, pp 487–498Chapter Google Scholar 
  • Montagna F, Bellotti G, Di Risio M (2011) 3D numerical modeling of landslide-generated tsunamis around a conical island. Nat Hazards 58:591–608Article Google Scholar 
  • Novak P, Moffat AIB, Nalluri C, Narayanan R (2007) Hydraulic structures. Taylor & Francis, LondonGoogle Scholar 
  • Parsaie A, Haghiabi A (2015a) Computational modeling of pollution transmission in rivers. Appl Water Sci. doi:10.1007/s13201-015-0319-6
  • Parsaie A, Haghiabi A (2015b) The effect of predicting discharge coefficient by neural network on increasing the numerical modeling accuracy of flow over side weir. Water Res Manag 29:973–985Article Google Scholar 
  • Parsaie A, Yonesi H, Najafian S (2015) Predictive modeling of discharge in compound open channel by support vector machine technique. Model Earth Syst Environ 1:1–6Article Google Scholar 
  • Su P-L, Liao H-S, Qiu Y, Li CJ (2009) Experimental study on a new type of aerator in spillway with low Froude number and mild slope flow. J Hydrodyn Ser B 21:415–422Article Google Scholar 
  • Suprapto M (2013) Increase spillway capacity using Labyrinth Weir. Procedia Eng 54:440–446Article Google Scholar 
  • Tabbara M, Chatila J, Awwad R (2005) Computational simulation of flow over stepped spillways. Comput Struct 83:2215–2224Article Google Scholar 
  • Wang J, Chen H (2009) Experimental study of elimination of vortices along guide wall of bank spillway. Adv Water Res Hydraul Eng Springer, Berlin, pp 2059–2063Google Scholar 
  • Wang Y, Jiang C (2010) Investigation of the surface vortex in a spillway tunnel intake. Tsinghua Sci Technol 15:561–565Article Google Scholar 
  • Zhenwei MU, Zhiyan Z, Tao Z (2012) Numerical simulation of 3-D flow field of spillway based on VOF method. Procedia Eng 28:808–812Article Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Water Engineering, Lorestan University, Khorram Abad, IranAbbas Parsaie, Amir Hamzeh Haghiabi & Amir Moradinejad

Corresponding author

Correspondence to Abbas Parsaie.

Reprints and Permissions

About this article

Cite this article

Parsaie, A., Haghiabi, A.H. & Moradinejad, A. CFD modeling of flow pattern in spillway’s approach channel. Sustain. Water Resour. Manag. 1, 245–251 (2015). https://doi.org/10.1007/s40899-015-0020-9

Download citation

  • Received28 April 2015
  • Accepted28 August 2015
  • Published15 September 2015
  • Issue DateSeptember 2015
  • DOIhttps://doi.org/10.1007/s40899-015-0020-9

Share this article

Anyone you share the following link with will be able to read this content:Get shareable link

Provided by the Springer Nature SharedIt content-sharing initiative

Keywords

  • Approach channel
  • Kamal-Saleh dam
  • Guide wall
  • Flow pattern
  • Numerical modeling
  • Flow 3D software
    Extratropical cyclone damage to the seawall in Dawlish, UK: eyewitness accounts, sea level analysis and numerical modelling

    영국 Dawlish의 방파제에 대한 온대 저기압 피해: 목격자 설명, 해수면 분석 및 수치 모델링

    Extratropical cyclone damage to the seawall in Dawlish, UK: eyewitness accounts, sea level analysis and numerical modelling

    Natural Hazards (2022)Cite this article

    Abstract

    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 m3/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.

    figure 1
    Fig. 1

    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.

    figure 2
    Fig. 2

    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 DawlishFootnote3, 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

    Full size table

    Sea level data and wave environment

    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 FacilityFootnote4 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 unitsFootnote5 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 CentreFootnote6 for tide gauges and the Channel Coastal ObservatoryFootnote7 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:

    pd=ρgHcoshk(d+z)coshkdcosσtpd=ρgHcoshk(d+z)coshkdcosσt

    (3)

    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).

    figure 3
    Fig. 3

    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).

    figure 4
    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.

    figure 5
    Fig. 5
    figure 6
    Fig. 6

    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:

    Tmn=2gd−−√[(m2L)2+(nW)2]−1/2Tmn=2gd[(m2L)2+(nW)2]−1/2

    (5)

    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.

    figure 7
    Fig. 7

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

    Full size table

    figure 8
    Fig. 8

    Results of wave amplitude simulations

    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 (deff—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 (deff = 3.5 m) and Scn-7 (deff = 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 deff = 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.

    figure 9
    Fig. 9

    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 (deff = 3.3 m), Scn-13 (deff = 2.5 m) and Scn-12 (deff = 3.5 m), which are 5.6 m, 5.1 m and 4.5 m. The maximum combined wave heights belong to Scn-11 (deff = 4.5 m) and Scn-17 (deff = 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.

    figure 10
    Fig. 10

    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. 1112 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).

    figure 11
    Fig. 11
    figure 12
    Fig. 12

    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 m3/s/m to 16.1 m3/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 m3/s/m and 14.4 m3/s/m belong to Scn-20 and 11, respectively. These are the two scenarios that also give the largest combined wave heights (Fig. 10b).

    figure 13
    Fig. 13

    The cumulative overtopping curves (Figs. 1415) 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 m3/m (m-cubed of water per metre length of the wall). The overtopping volumes are 42.7 m3/m for Scn-11 and 28.8 m3/m for Scn-19. The overtopping volume is in the range of 0.7–65.0 m3/m for all scenarios.

    figure 14
    Fig. 14
    figure 15
    Fig. 15

    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:

    qgH3m−−−−√=0.0155(Hmhs)12e(−2.2RcHm)qgHm3=0.0155(Hmhs)12e(−2.2RcHm)

    (6)

    where qq is the mean overtopping rate per metre length of the seawall (m3/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 m3/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 m3/s/m for our scenarios. The cumulative overtopping water volumes per metre length of the wall were 0.7–65.0 m3/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.

    Notes

    1. https://www.networkrail.co.uk/..
    2. https://www.flow3d.com/products/flow-3d-hydro/.
    3. https://www.devonmuseums.net/Dawlish-Museum/Devon-Museums/.
    4. https://ntslf.org/.
    5. https://www.datawell.nl/Products/Buoys/DirectionalWaveriderMkIII.aspx.
    6. https://www.bodc.ac.uk/.
    7. https://coastalmonitoring.org/cco/.
    8. https://webapp.navionics.com/#boating@8&key=iactHlwfP.

    References

    Download references

    Acknowledgements

    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

    1. Department of Civil and Environmental Engineering, Brunel University London, Uxbridge, UB8 3PH, UKKeith Adams
    2. Department of Architecture and Civil Engineering, University of Bath, Bath, BA2 7AY, UKMohammad Heidarzadeh

    Corresponding author

    Correspondence to Keith Adams.

    Ethics declarations

    Conflict of interest

    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.

    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/.

    Reprints and Permissions

    About this article

    Verify currency and authenticity via CrossMark

    Cite this article

    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

    Download citation

    • Received17 May 2022
    • Accepted17 October 2022
    • Published14 November 2022
    • DOIhttps://doi.org/10.1007/s11069-022-05692-2

    Share this article

    Anyone you share the following link with will be able to read this content:Get shareable link

    Provided by the Springer Nature SharedIt content-sharing initiative

    Keywords

    • Storm surge
    • Cyclone
    • Railway
    • Climate change
    • Infrastructure
    • Resilience
    Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.

    플라즈마 회전 전극 공정 중 분말 형성에 대한 공정 매개변수 및 냉각 가스의 영향

    Effects of process parameters and cooling gas on powder formation during the plasma rotating electrode process

    Yujie Cuia Yufan Zhaoa1 Haruko Numatab Kenta Yamanakaa Huakang Biana Kenta Aoyagia AkihikoChibaa
    aInstitute for Materials Research, Tohoku University, Sendai 980-8577, JapanbDepartment of Materials Processing, Graduate School of Engineering, Tohoku University, Sendai 980-8577, Japan

    Highlights

    •The limitation of increasing the rotational speed in decreasing powder size was clarified.

    •Cooling and disturbance effects varied with the gas flowing rate.

    •Inclined angle of the residual electrode end face affected powder formation.

    •Additional cooling gas flowing could be applied to control powder size.

    Abstract

    The plasma rotating electrode process (PREP) is rapidly becoming an important powder fabrication method in additive manufacturing. However, the low production rate of fine PREP powder limits the development of PREP. Herein, we investigated different factors affecting powder formation during PREP by combining experimental methods and numerical simulations. The limitation of increasing the rotation electrode speed in decreasing powder size is attributed to the increased probability of adjacent droplets recombining and the decreased tendency of granulation. The effects of additional Ar/He gas flowing on the rotational electrode on powder formation is determined through the cooling effect, the disturbance effect, and the inclined effect of the residual electrode end face simultaneously. A smaller-sized powder was obtained in the He atmosphere owing to the larger inclined angle of the residual electrode end face compared to the Ar atmosphere. Our research highlights the route for the fabrication of smaller-sized powders using PREP.

    플라즈마 회전 전극 공정(PREP)은 적층 제조 에서 중요한 분말 제조 방법으로 빠르게 자리잡고 있습니다. 그러나 미세한 PREP 분말의 낮은 생산율은 PREP의 개발을 제한합니다. 여기에서 우리는 실험 방법과 수치 시뮬레이션을 결합하여 PREP 동안 분말 형성에 영향을 미치는 다양한 요인을 조사했습니다. 분말 크기 감소에서 회전 전극 속도 증가의 한계는 인접한 액적 재결합 확률 증가 및 과립화 경향 감소에 기인합니다.. 회전 전극에 흐르는 추가 Ar/He 가스가 분말 형성에 미치는 영향은 냉각 효과, 외란 효과 및 잔류 전극 단면의 경사 효과를 통해 동시에 결정됩니다. He 분위기에서는 Ar 분위기에 비해 잔류 전극 단면의 경사각이 크기 때문에 더 작은 크기의 분말이 얻어졌다. 우리의 연구는 PREP를 사용하여 더 작은 크기의 분말을 제조하는 경로를 강조합니다.

    Keywords

    Plasma rotating electrode process

    Ti-6Al-4 V alloy, Rotating speed, Numerical simulation, Gas flowing, Powder size

    Introduction

    With the development of additive manufacturing, there has been a significant increase in high-quality powder production demand [1,2]. The initial powder characteristics are closely related to the uniform powder spreading [3,4], packing density [5], and layer thickness observed during additive manufacturing [6], thus determining the mechanical properties of the additive manufactured parts [7,8]. Gas atomization (GA) [9–11], centrifugal atomization (CA) [12–15], and the plasma rotating electrode process (PREP) are three important powder fabrication methods.

    Currently, GA is the dominant powder fabrication method used in additive manufacturing [16] for the fabrication of a wide range of alloys [11]. GA produces powders by impinging a liquid metal stream to droplets through a high-speed gas flow of nitrogen, argon, or helium. With relatively low energy consumption and a high fraction of fine powders, GA has become the most popular powder manufacturing technology for AM.

    The entrapped gas pores are generally formed in the powder after solidification during GA, in which the molten metal is impacted by a high-speed atomization gas jet. In addition, satellites are formed in GA powder when fine particles adhere to partially molten particles.

    The gas pores of GA powder result in porosity generation in the additive manufactured parts, which in turn deteriorates its mechanical properties because pores can become crack initiation sites [17]. In CA, a molten metal stream is poured directly onto an atomizer disc spinning at a high rotational speed. A thin film is formed on the surface of the disc, which breaks into small droplets due to the centrifugal force. Metal powder is obtained when these droplets solidify.

    Compared with GA powder, CA powder exhibits higher sphericity, lower impurity content, fewer satellites, and narrower particle size distribution [12]. However, very high speed is required to obtain fine powder by CA. In PREP, the molten metal, melted using the plasma arc, is ejected from the rotating rod through centrifugal force. Compared with GA powder, PREP-produced powders also have higher sphericity and fewer pores and satellites [18].

    For instance, PREP-fabricated Ti6Al-4 V alloy powder with a powder size below 150 μm exhibits lower porosity than gas-atomized powder [19], which decreases the porosity of additive manufactured parts. Furthermore, the process window during electron beam melting was broadened using PREP powder compared to GA powder in Inconel 718 alloy [20] owing to the higher sphericity of the PREP powder.

    In summary, PREP powder exhibits many advantages and is highly recommended for powder-based additive manufacturing and direct energy deposition-type additive manufacturing. However, the low production rate of fine PREP powder limits the widespread application of PREP powder in additive manufacturing.

    Although increasing the rotating speed is an effective method to decrease the powder size [21,22], the reduction in powder size becomes smaller with the increased rotating speed [23]. The occurrence of limiting effects has not been fully clarified yet.

    Moreover, the powder size can be decreased by increasing the rotating electrode diameter [24]. However, these methods are quite demanding for the PREP equipment. For instance, it is costly to revise the PREP equipment to meet the demand of further increasing the rotating speed or electrode diameter.

    Accordingly, more feasible methods should be developed to further decrease the PREP powder size. Another factor that influences powder formation is the melting rate [25]. It has been reported that increasing the melting rate decreases the powder size of Inconel 718 alloy [26].

    In contrast, the powder size of SUS316 alloy was decreased by decreasing the plasma current within certain ranges. This was ascribed to the formation of larger-sized droplets from fluid strips with increased thickness and spatial density at higher plasma currents [27]. The powder size of NiTi alloy also decreases at lower melting rates [28]. Consequently, altering the melting rate, varied with the plasma current, is expected to regulate the PREP powder size.

    Furthermore, gas flowing has a significant influence on powder formation [27,29–31]. On one hand, the disturbance effect of gas flowing promotes fluid granulation, which in turn contributes to the formation of smaller-sized powder [27]. On the other hand, the cooling effect of gas flowing facilitates the formation of large-sized powder due to increased viscosity and surface tension. However, there is a lack of systematic research on the effect of different gas flowing on powder formation during PREP.

    Herein, the authors systematically studied the effects of rotating speed, electrode diameter, plasma current, and gas flowing on the formation of Ti-6Al-4 V alloy powder during PREP as additive manufactured Ti-6Al-4 V alloy exhibits great application potential [32]. Numerical simulations were conducted to explain why increasing the rotating speed is not effective in decreasing powder size when the rotation speed reaches a certain level. In addition, the different factors incited by the Ar/He gas flowing on powder formation were clarified.

    Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.
    Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.

    References

    [1] W. Ding, G. Chen, M. Qin, Y. He, X. Qu, Low-cost Ti powders for additive manufacturing treated by fluidized bed, Powder Technol. 350 (2019) 117–122, https://doi.org/
    10.1016/j.powtec.2019.03.042.
    [2] W.S.W. Harun, M.S.I.N. Kamariah, N. Muhamad, S.A.C. Ghani, F. Ahmad, Z. Mohamed,
    A review of powder additive manufacturing processes for metallic biomaterials,
    Powder Technol. 327 (2018) 128–151, https://doi.org/10.1016/j.powtec.2017.12.
    058.
    [3] M. Ahmed, M. Pasha, W. Nan, M. Ghadiri, A simple method for assessing powder
    spreadability for additive manufacturing, Powder Technol. 367 (2020) 671–679,
    https://doi.org/10.1016/j.powtec.2020.04.033.
    [4] W. Nan, M. Pasha, M. Ghadiri, Numerical simulation of particle flow and segregation
    during roller spreading process in additive manufacturing, Powder Technol. 364
    (2020) 811–821, https://doi.org/10.1016/j.powtec.2019.12.023.
    [5] A. Averardi, C. Cola, S.E. Zeltmann, N. Gupta, Effect of particle size distribution on the
    packing of powder beds : a critical discussion relevant to additive manufacturing,
    Mater. Today Commun. 24 (2020) 100964, https://doi.org/10.1016/j.mtcomm.
    2020.100964.
    [6] K. Riener, N. Albrecht, S. Ziegelmeier, R. Ramakrishnan, L. Haferkamp, A.B. Spierings,
    G.J. Leichtfried, Influence of particle size distribution and morphology on the properties of the powder feedstock as well as of AlSi10Mg parts produced by laser powder bed fusion (LPBF), Addit. Manuf. 34 (2020) 101286, https://doi.org/10.1016/j.
    addma.2020.101286.
    [7] W.S.W. Harun, N.S. Manam, M.S.I.N. Kamariah, S. Sharif, A.H. Zulkifly, I. Ahmad, H.
    Miura, A review of powdered additive manufacturing techniques for Ti-6Al-4V biomedical applications, Powder Technol. 331 (2018) 74–97, https://doi.org/10.1016/j.
    powtec.2018.03.010.
    [8] A.T. Sutton, C.S. Kriewall, M.C. Leu, J.W. Newkirk, Powder characterisation techniques and effects of powder characteristics on part properties in powder-bed fusion processes, Virtual Phys. Prototyp. 12 (2017) (2017) 3–29, https://doi.org/10.
    1080/17452759.2016.1250605.
    [9] G. Chen, Q. Zhou, S.Y. Zhao, J.O. Yin, P. Tan, Z.F. Li, Y. Ge, J. Wang, H.P. Tang, A pore
    morphological study of gas-atomized Ti-6Al-4V powders by scanning electron microscopy and synchrotron X-ray computed tomography, Powder Technol. 330
    (2018) 425–430, https://doi.org/10.1016/j.powtec.2018.02.053.
    [10] Y. Feng, T. Qiu, Preparation, characterization and microwave absorbing properties of
    FeNi alloy prepared by gas atomization method, J. Alloys Compd. 513 (2012)
    455–459, https://doi.org/10.1016/j.jallcom.2011.10.079.

    [11] I.E. Anderson, R.L. Terpstra, Progress toward gas atomization processing with increased uniformity and control, Mater. Sci. Eng. A 326 (2002) 101–109, https://
    doi.org/10.1016/S0921-5093(01)01427-7.
    [12] P. Phairote, T. Plookphol, S. Wisutmethangoon, Design and development of a centrifugal atomizer for producing zinc metal powder, Int. J. Appl. Phys. Math. 2 (2012)
    77–82, https://doi.org/10.7763/IJAPM.2012.V2.58.
    [13] L. Tian, I. Anderson, T. Riedemann, A. Russell, Production of fine calcium powders by
    centrifugal atomization with rotating quench bath, Powder Technol. 308 (2017)
    84–93, https://doi.org/10.1016/j.powtec.2016.12.011.
    [14] M. Eslamian, J. Rak, N. Ashgriz, Preparation of aluminum/silicon carbide metal matrix composites using centrifugal atomization, Powder Technol. 184 (2008) 11–20,
    https://doi.org/10.1016/j.powtec.2007.07.045.
    [15] T. Plookphol, S. Wisutmethangoon, S. Gonsrang, Influence of process parameters on
    SAC305 lead-free solder powder produced by centrifugal atomization, Powder
    Technol. 214 (2011) 506–512, https://doi.org/10.1016/j.powtec.2011.09.015.
    [16] M.Z. Gao, B. Ludwig, T.A. Palmer, Impact of atomization gas on characteristics of austenitic stainless steel powder feedstocks for additive manufacturing, Powder
    Technol. 383 (2021) 30–42, https://doi.org/10.1016/j.powtec.2020.12.005.
    [17] X. Shui, K. Yamanaka, M. Mori, Y. Nagata, A. Chiba, Effects of post-processing on cyclic fatigue response of a titanium alloy additively manufactured by electron beam
    melting, Mater. Sci. Eng. A 680 (2017) 239–248, https://doi.org/10.1016/j.msea.
    2016.10.059.
    [18] C. Wang, X.H. Zhao, Y.C. Ma, Q.X. Wang, Y.J. Lai, S.J. Liang, Study of the spherical
    HoCu powders prepared by supreme-speed plasma rotating electrode process,
    Powder Metallurgy Technology 38 (3) (2020) 227–233, https://doi.org/10.19591/
    j.cnki.cn11-1974/tf.2020.03.011 (in Chinese).
    [19] G. Chen, S.Y. Zhao, P. Tan, J. Wang, C.S. Xiang, H.P. Tang, A comparative study of Ti6Al-4V powders for additive manufacturing by gas atomization, plasma rotating
    electrode process and plasma atomization, Powder Technol. 333 (2018) 38–46,
    https://doi.org/10.1016/j.powtec.2018.04.013.
    [20] Y. Zhao, K. Aoyagi, Y. Daino, K. Yamanaka, A. Chiba, Significance of powder feedstock
    characteristics in defect suppression of additively manufactured Inconel 718, Addit.
    Manuf. 34 (2020) 101277, https://doi.org/10.1016/j.addma.2020.101277.
    [21] Y. Nie, J. Tang, B. Yang, Q. Lei, S. Yu, Y. Li, Comparison in characteristic and atomization behavior of metallic powders produced by plasma rotating electrode process,
    Adv. Powder Technol. 31 (2020) 2152–2160, https://doi.org/10.1016/j.apt.2020.03.
    006.
    [22] Y. Cui, Y. Zhao, H. Numata, H. Bian, K. Wako, K. Yamanaka, K. Aoyagi, C. Zhang, A.
    Chiba, Effects of plasma rotating electrode process parameters on the particle size
    distribution and microstructure of Ti-6Al-4 V alloy powder, Powder Technol 376
    (2020) 363–372, https://doi.org/10.1016/j.powtec.2020.08.027.
    [23] J. Tang, Y. Nie, Q. Lei, Y. Li, Characteristics and atomization behavior of Ti-6Al-4V
    powder produced by plasma rotating electrode process Adv, Powder Technol. 10
    (2019) 2330–2337, https://doi.org/10.1016/j.apt.2019.07.015.
    [24] M. Zdujić, D. Uskoković, Production of atomized metal and alloy powders by the rotating electrode process, Sov. Powder Metall. Met. Ceram. 29 (1990) 673–683,
    https://doi.org/10.1007/BF00795571.
    [25] L. Zhang, Y. Zhao, Particle size distribution of tin powder produced by centrifugal
    atomisation using rotating cups, Powder Technol. 318 (2017) 62–67, https://doi.
    org/10.1016/j.powtec.2017.05.038.
    [26] Y. Liu, S. Liang, Z. Han, J. Song, Q. Wang, A novel model of calculating particle sizes in
    plasma rotating electrode process for superalloys, Powder Technol. 336 (2018)
    406–414, https://doi.org/10.1016/j.powtec.2018.06.002.
    [27] Y. Zhao, Y. Cui, H. Numata, H. Bian, K. Wako, K. Yamanaka, Centrifugal granulation
    behavior in metallic powder fabrication by plasma rotating electrode process, Sci.
    Rep. (2020) 1–15, https://doi.org/10.1038/s41598-020-75503-w.
    [28] T. Hsu, C. Wei, L. Wu, Y. Li, A. Chiba, M. Tsai, Nitinol powders generate from plasma
    rotation electrode process provide clean powder for biomedical devices used with
    suitable size, spheroid surface and pure composition, Sci. Rep. 8 (2018) 1–8,
    https://doi.org/10.1038/s41598-018-32101-1.
    [29] M. Wei, S. Chen, M. Sun, J. Liang, C. Liu, M. Wang, Atomization simulation and preparation of 24CrNiMoY alloy steel powder using VIGA technology at high gas pressure, Powder Technol. 367 (2020) 724–739, https://doi.org/10.1016/j.powtec.
    2020.04.030.
    [30] Y. Tan, X. Zhu, X.Y. He, B. Ding, H. Wang, Q. Liao, H. Li, Granulation characteristics of
    molten blast furnace slag by hybrid centrifugal-air blast technique, Powder Technol.
    323 (2018) 176–185, https://doi.org/10.1016/j.powtec.2017.09.040.
    [31] P. Xu, D.H. Liu, J. Hu, G.Y. Lin, Synthesis of Ni-Ti composite powder by radio frequency plasma spheroidization process, Nonferrous Metals Science and Engineering
    39 (1) (2020) 67–71 , (in Chinese) 10.13264/j.cnki.ysjskx.2020.01.011.
    [32] H. Mehboob, F. Tarlochan, A. Mehboob, S.H. Chang, S. Ramesh, W.S.W. Harun, K.
    Kadirgama, A novel design, analysis and 3D printing of Ti-6Al-4V alloy bioinspired porous femoral stem, J. Mater. Sci. Mater. Med. 31 (2020) 78, https://doi.
    org/10.1007/s10856-020-06420-7.
    [33] FLOW-3D® Version 11.2 [Computer software]. , Flow Science, Inc., Santa Fe, NM,
    2017https://www.flow3d.com.
    [34] M. Boivineau, C. Cagran, D. Doytier, V. Eyraud, M.H. Nadal, B. Wilthan, G. Pottlacher,
    Thermophysical properties of solid and liquid Ti-6Al-4V (TA6V) alloy, Int. J.
    Thermophys. 27 (2006) 507–529, https://doi.org/10.1007/PL00021868.
    [35] J. Liu, Q. Qin, Q. Yu, The effect of size distribution of slag particles obtained in dry
    granulation on blast furnace slag cement strength, Powder Technol. 362 (2020)
    32–36, https://doi.org/10.1016/j.powtec.2019.11.115.
    [36] M. Tanaka, S. Tashiro, A study of thermal pinch effect of welding arcs, J. Japan Weld.
    Soc. 25 (2007) 336–342, https://doi.org/10.2207/qjjws.25.336 (in Japanese).
    [37] T. Kamiya, A. Kayano, Disintegration of viscous fluid in the ligament state purged
    from a rotating disk, J. Chem. Eng. JAPAN. 4 (1971) 364–369, https://doi.org/10.
    1252/jcej.4.364.
    [38] T. Kamiya, An analysis of the ligament-type disintegration of thin liquid film at the
    edge of a rotating disk, J. Chem. Eng. Japan. 5 (1972) 391–396, https://doi.org/10.
    1252/jcej.5.391.
    [39] J. Burns, C. Ramshaw, R. Jachuck, Measurement of liquid film thickness and the determination of spin-up radius on a rotating disc using an electrical resistance technique, Chem. Eng. Sci. 58 (2003) 2245–2253, https://doi.org/10.1016/S0009-2509
    (03)00091-5.
    [40] J. Rauscher, R. Kelly, J. Cole, An asymptotic solution for the laminar flow of a thin film
    on a rotating disk, J. Appl. Mech. Trans. ASME 40 (1973) 43–47, https://doi.org/10.
    1115/1.3422970

    Fig. 1. Model geometry with the computational domain, extrusion nozzle, toolpath, and boundary conditions. The model is presented while printing the fifth layer.

    재료 압출 적층 제조에서 증착된 층의 안정성 및 변형

    Md Tusher Mollah Raphaël 사령관 Marcin P. Serdeczny David B. Pedersen Jon Spangenberg덴마크 공과 대학 기계 공학과, Kgs. 덴마크 링비

    2020년 12월 22일 접수, 2021년 5월 1일 수정, 2021년 7월 15일 수락, 2021년 7월 21일 온라인 사용 가능, 기록 버전 2021년 8월 17일 .

    Abstract

    이 문서는 재료 압출 적층 제조 에서 여러 레이어를 인쇄하는 동안 증착 흐름의 전산 유체 역학 시뮬레이션 을 제공합니다 개발된 모델은 증착된 레이어의 형태를 예측하고 점소성 재료 를 인쇄하는 동안 레이어 변형을 캡처합니다 . 물리학은 일반화된 뉴턴 유체 로 공식화된 Bingham 구성 모델의 연속성 및 운동량 방정식에 의해 제어됩니다. . 증착된 층의 단면 모양이 예측되고 재료의 다양한 구성 매개변수에 대해 층의 변형이 연구됩니다. 층의 변형은 인쇄물의 정수압과 압출시 압출압력으로 인한 것임을 알 수 있다. 시뮬레이션에 따르면 항복 응력이 높을수록 변형이 적은 인쇄물이 생성되는 반면 플라스틱 점도 가 높을수록 증착된 레이어에서변형이 커 집니다 . 또한, 인쇄 속도, 압출 속도 의 영향, 층 높이 및 인쇄된 층의 변형에 대한 노즐 직경을 조사합니다. 마지막으로, 이 모델은 후속 인쇄된 레이어의 정수압 및 압출 압력을 지원하기 위해 증착 후 점소성 재료가 요구하는 항복 응력의 필요한 증가에 대한 보수적인 추정치를 제공합니다.

    This paper presents computational fluid dynamics simulations of the deposition flow during printing of multiple layers in material extrusion additive manufacturing. The developed model predicts the morphology of the deposited layers and captures the layer deformations during the printing of viscoplastic materials. The physics is governed by the continuity and momentum equations with the Bingham constitutive model, formulated as a generalized Newtonian fluid. The cross-sectional shapes of the deposited layers are predicted, and the deformation of layers is studied for different constitutive parameters of the material. It is shown that the deformation of layers is due to the hydrostatic pressure of the printed material, as well as the extrusion pressure during the extrusion. The simulations show that a higher yield stress results in prints with less deformations, while a higher plastic viscosity leads to larger deformations in the deposited layers. Moreover, the influence of the printing speed, extrusion speed, layer height, and nozzle diameter on the deformation of the printed layers is investigated. Finally, the model provides a conservative estimate of the required increase in yield stress that a viscoplastic material demands after deposition in order to support the hydrostatic and extrusion pressure of the subsequently printed layers.

    Fig. 1. Model geometry with the computational domain, extrusion nozzle, toolpath, and boundary conditions. The model is presented while printing the fifth layer.
    Fig. 1. Model geometry with the computational domain, extrusion nozzle, toolpath, and boundary conditions. The model is presented while printing the fifth layer.

    키워드

    점성 플라스틱 재료, 재료 압출 적층 제조(MEX-AM), 다층 증착, 전산유체역학(CFD), 변형 제어
    Viscoplastic Materials, Material Extrusion Additive Manufacturing (MEX-AM), Multiple-Layers Deposition, Computational Fluid Dynamics (CFD), Deformation Control

    Introduction

    Three-dimensional printing of viscoplastic materials has grown in popularity over the recent years, due to the success of Material Extrusion Additive Manufacturing (MEX-AM) [1]. Viscoplastic materials, such as ceramic pastes [2,3], hydrogels [4], thermosets [5], and concrete [6], behave like solids when the applied load is below their yield stress, and like a fluid when the applied load exceeds their yield stress [7]. Viscoplastic materials are typically used in MEX-AM techniques such as Robocasting [8], and 3D concrete printing [9,10]. The differences between these technologies lie in the processing of the material before the extrusion and in the printing scale (from microscale to big area additive manufacturing). In these extrusion-based technologies, the structure is fabricated in a layer-by-layer approach onto a solid surface/support [11, 12]. During the process, the material is typically deposited on top of the previously printed layers that may be already solidified (wet-on-dry printing) or still deformable (wet-on-wet printing) [1]. In wet-on-wet printing, control over the deformation of layers is important for the stability and geometrical accuracy of the prints. If the material is too liquid after the deposition, it cannot support the pressure of the subsequently deposited layers. On the other hand, the material flowability is a necessity during extrusion through the nozzle. Several experimental studies have been performed to analyze the physics of the extrusion and deposition of viscoplastic materials, as reviewed in Refs. [13–16]. The experimental measurements can be supplemented with Computational Fluid Dynamics (CFD) simulations to gain a more complete picture of MEX-AM. A review of the CFD studies within the material processing and deposition in 3D concrete printing was presented by Roussel et al. [17]. Wolfs et al. [18] predicted numerically the failure-deformation of a cylindrical structure due to the self-weight by calculating the stiffness and strength of the individual layers. It was found that the deformations can take place in all layers, however the most critical deformation occurs in the bottom layer. Comminal et al. [19,20] presented three-dimensional simulations of the material deposition in MEX-AM, where the fluid was approximated as Newtonian. Subsequently, the model was experimentally validated in Ref. [21] for polymer-based MEX-AM, and extended to simulate the deposition of multiple layers in Ref. [22], where the previously printed material was assumed solid. Xia et al. [23] simulated the influence of the viscoelastic effects on the shape of deposited layers in MEX-AM. A numerical model for simulating the deposition of a viscoplastic material was recently presented and experimentally validated in Refs. [24] and [25]. These studies focused on predicting the cross-sectional shape of a single printed layer for different processing conditions (relative printing speed, and layer height). Despite these research efforts, a limited number of studies have focused on investigating the material deformations in wet-on-wet printing when multiple layers are deposited on top of each other. This paper presents CFD simulations of the extrusion-deposition flow of a viscoplastic material for several subsequent layers (viz. three- and five-layers). The material is continuously printed one layer over another on a fixed solid surface. The rheology of the viscoplastic material is approximated by the Bingham constitutive equation that is formulated using the Generalized Newtonian Fluid (GNF) model. The CFD model is used to predict the cross-sectional shapes of the layers and their deformations while printing the next layers on top. Moreover, the simulations are used to quantify the extrusion pressure applied by the deposited material on the substrate, and the previously printed layers. Numerically, it is investigated how the process parameters (i.e., the extrusion speed, printing speed, nozzle diameter, and layer height) and the material rheology affect the deformations of the deposited layers. Section 2 describes the methodology of the study. Section 3 presents and discusses the results. The study is summarized and concluded in Section 4.

    References

    [1] R.A. Buswell, W.R. Leal De Silva, S.Z. Jones, J. Dirrenberger, 3D printing using
    concrete extrusion: a roadmap for research, Cem. Concr. Res. 112 (2018) 37–49.
    [2] Z. Chen, Z. Li, J. Li, C. Liu, C. Lao, Y. Fu, C. Liu, Y. Li, P. Wang, Y. He, 3D printing of
    ceramics: a review, J. Eur. Ceram. Soc. 39 (4) (2019) 661–687.
    [3] A. Bellini, L. Shor, S.I. Guceri, New developments in fused deposition modeling of
    ceramics, Rapid Prototyp. J. 11 (4) (2005) 214–220.
    [4] S. Aktas, D.M. Kalyon, B.M. Marín-Santib´
    anez, ˜ J. P´erez-Gonzalez, ´ Shear viscosity
    and wall slip behavior of a viscoplastic hydrogel, J. Rheol. 58 (2) (2014) 513–535.
    [5] J. Lindahl, A. Hassen, S. Romberg, B. Hedger, P. Hedger Jr., M. Walch, T. Deluca,
    W. Morrison, P. Kim, A. Roschli, D. Nuttall, Large-scale Additive Manufacturing
    with Reactive Polymers, Oak Ridge National Lab.(ORNL), Oak Ridge, TN (United
    States), 2018.
    [6] V.N. Nerella, V. Mechtcherine, Studying the printability of fresh concrete for
    formwork-free Concrete onsite 3D Printing Technology (CONPrint3D), 3D Concr.
    Print. Technol. (2019) 333–347.
    [7] C. Tiu, J. Guo, P.H.T. Uhlherr, Yielding behaviour of viscoplastic materials, J. Ind.
    Eng. Chem. 12 (5) (2006) 653–662.
    [8] B. Dietemann, F. Bosna, M. Lorenz, N. Travitzky, H. Kruggel-Emden, T. Kraft,
    C. Bierwisch, Modeling robocasting with smoothed particle hydrodynamics:
    printing gapspanning filaments, Addit. Manuf. 36 (2020), 101488.
    [9] B. Khoshnevis, R. Russell, H. Kwon, S. Bukkapatnam, Contour crafting – a layered
    fabrication, Spec. Issue IEEE Robot. Autom. Mag. 8 (3) (2001) 33–42.
    [10] D. Asprone, F. Auricchio, C. Menna, V. Mercuri, 3D printing of reinforced concrete
    elements: technology and design approach, Constr. Build. Mater. 165 (2018)
    218–231.
    [11] J. Jiang, Y. Ma, Path planning strategies to optimize accuracy, quality, build time
    and material use in additive manufacturing: a review, Micromachines 11 (7)
    (2020) 633.
    [12] J. Jiang, A novel fabrication strategy for additive manufacturing processes,
    J. Clean. Prod. 272 (2020), 122916.
    [13] F. Bos, R. Wolfs, Z. Ahmed, T. Salet, Additive manufacturing of concrete in
    construction: potentials and challenges, Virtual Phys. Prototyp. 11 (3) (2016)
    209–225.
    [14] P. Wu, J. Wang, X. Wang, A critical review of the use of 3-D printing in the
    construction industry, Autom. Constr. 68 (2016) 21–31.
    [15] T.D. Ngo, A. Kashani, G. Imbalzano, K.T. Nguyen, D. Hui, Additive manufacturing
    (3D printing): a review of materials, methods, applications and challenges,
    Compos. Part B: Eng. 143 (2018) 172–196.
    [16] M. Valente, A. Sibai, M. Sambucci, Extrusion-based additive manufacturing of
    concrete products: revolutionizing and remodeling the construction industry,
    J. Compos. Sci. 3 (3) (2019) 88.
    [17] N. Roussel, J. Spangenberg, J. Wallevik, R. Wolfs, Numerical simulations of
    concrete processing: from standard formative casting to additive manufacturing,
    Cem. Concr. Res. 135 (2020), 106075.
    [18] R.J.M. Wolfs, F.P. Bos, T.A.M. Salet, Early age mechanical behaviour of 3D printed
    concrete: numerical modelling and experimental testing, Cem. Concr. Res. 106
    (2018) 103–116.
    [19] R. Comminal, M.P. Serdeczny, D.B. Pedersen, J. Spangenberg, Numerical modeling
    of the strand deposition flow in extrusion-based additive manufacturing, Addit.
    Manuf. 20 (2018) 68–76.
    [20] R. Comminal, M.P. Serdeczny, D.B. Pedersen, J. Spangenberg, Numerical modeling
    of the material deposition and contouring precision in fused deposition modeling,
    in Proceedings of the 29th Annual International Solid Freeform Fabrication
    Symposium, Austin, TX, USA, 2018, pp. 1855–1864.
    [21] M.P. Serdeczny, R. Comminal, D.B. Pedersen, J. Spangenberg, Experimental
    validation of a numerical model for the strand shape in material extrusion additive
    manufacturing, Addit. Manuf. 24 (2018) 145–153.
    [22] M.P. Serdeczny, R. Comminal, D.B. Pedersen, J. Spangenberg, Numerical
    simulations of the mesostructure formation in material extrusion additive
    manufacturing, Addit. Manuf. 28 (2019) 419–429.
    [23] H. Xia, J. Lu, G. Tryggvason, A numerical study of the effect of viscoelastic stresses
    in fused filament fabrication, Comput. Methods Appl. Mech. Eng. 346 (2019)
    242–259.
    [24] R. Comminal, W.R.L. da Silva, T.J. Andersen, H. Stang, J. Spangenberg, Influence
    of processing parameters on the layer geometry in 3D concrete printing:
    experiments and modelling, in: Proceedings of the Second RILEM International
    Conference on Concrete and Digital Fabrication, vol. 28, 2020, pp. 852–862.
    [25] R. Comminal, W.R.L. da Silva, T.J. Andersen, H. Stang, J. Spangenberg, Modelling
    of 3D concrete printing based on computational fluid dynamics, Cem. Concr. Res.
    38 (2020), 106256.
    [26] E.C. Bingham, An investigation of the laws of plastic flow, US Bur. Stand. Bull. 13
    (1916) 309–352.
    [27] N. Casson, A flow equation for pigment-oil suspensions of the printing ink type,
    Rheol. Disperse Syst. (1959) 84–104.
    [28] W.H. Herschel, R. Bulkley, Konsistenzmessungen von Gummi-Benzollosungen, ¨
    Kolloid Z. 39 (1926) 291–300.
    [29] “FLOW-3D | We solve The World’s Toughest CFD Problems,” FLOW SCIENCE.
    〈https://www.flow3d.com/〉. (Accessed 27 June 2020).
    [30] S. Jacobsen, R. Cepuritis, Y. Peng, M.R. Geiker, J. Spangenberg, Visualizing and
    simulating flow conditions in concrete form filling using pigments, Constr. Build.
    Mater. 49 (2013) 328–342.
    [31] E.J. O’Donovan, R.I. Tanner, Numerical study of the Bingham squeeze film
    problem, J. Non-Newton. Fluid Mech. 15 (1) (1984) 75–83.
    [32] C.W. Hirt, B.D. Nichols, Volume of fluid (VOF) method for the dynamics of free
    boundaries, J. Comput. Phys. 39 (1) (1981) 201–225.
    [33] R. Comminal, J. Spangenberg, J.H. Hattel, Cellwise conservative unsplit advection
    for the volume of fluid method, J. Comput. Phys. 283 (2015) 582–608.
    [34] A. Negar, S. Nazarian, N.A. Meisel, J.P. Duarte, Experimental prediction of material
    deformation in large-scale additive manufacturing of concrete, Addit. Manuf. 37
    (2021), 101656.

    Fig 2(b) Observed velocity field for aspect ratio 0.25(Sukhodolov 2002)

    고정 베드의 불침투성 토양에서 흐름 패턴의 수치 시뮬레이션

    NUMERICAL SIMULATION OF FLOW PATTERN IN SERIES OF IMPERMEABLE GROYNES IN FIXED BED

    Kafle, Mukesh Raj1
    1Asst. Professor, Department of Civil Engineering, Institute of Engineering, Pulchowk Campus, Nepal
    Email: mkafle@pcampus.edu.np

    Abstract

    This paper presents a numerical simulation of recirculating flow patterns in groyne fields. Moreover, it entails the concept determination of proper spacing of vertical unsubmerged and impermeable groynesin seriesto control the bank erosion. Flow pattern between the groynes varies along their space. The flow in groyne field may significantly affect the flow change, bed change, bank erosion and condition of habitat. In this regard, an assessment of flow along the space of groynes will yield important data needed to diversify the object of groyne installation. So, knowledge about determination of the proper spacing of groynes in groyne field is important. Space of vertical groynes was set from 1.5 to 10 times the length of groynes. The velocity field between groynes was simulated by using Computational Fluid Dynamics (CFD) model Nays 2D. Simulated velocity field was compared with existing experimentaldata for the same parameter, which agreed satisfactorily. Based on simulated results,the optimal spacing of vertical groynes to control the bank erosion was recommended.

    이 논문은 groyne 필드에서 재순환 흐름 패턴의 수치 시뮬레이션을 제공합니다. 더욱이, 그것은 제방 침식을 제어하기 위해 수직 비침수 및 불침투성 그로이네신 시리즈의 적절한 간격의 개념 결정을 수반합니다. groynes 사이의 흐름 패턴은 공간에 따라 다릅니다. groyne field의 흐름은 흐름 변화, 하상 변화, 제방 침식 및 서식지 상태에 중대한 영향을 미칠 수 있습니다. 이와 관련하여, groyne 공간을 따른 흐름의 평가는 groyne 설치 대상을 다양화하는 데 필요한 중요한 데이터를 산출할 것입니다. 따라서, groyne field에서 groyne의 적절한 간격 결정에 대한 지식이 중요합니다. 수직 여백의 간격은 여아 길이의 1.5배에서 10배 사이로 설정하였다. groyne 사이의 속도장은 CFD(Computational Fluid Dynamics) 모델 Nays 2D를 사용하여 시뮬레이션되었습니다. 시뮬레이션된 속도장은 동일한 매개변수에 대해 기존 실험 데이터와 비교되었으며 만족스럽게 일치했습니다. 모의 결과를 바탕으로 제방 침식을 억제하기 위한 최적의 수직 제방 간격을 제안하였다.

    1. Introduction
      Spur dikes or groynes are used to protect river banks from erosion and also keep the channel
      navigable.Depending upon the flow characteristics, spur-dikes may be classified as submerged and unsubmerged. Also, based on the permeability, spur dikes are further classified as permeable and
      impermeable. Herein, un-submerged !impermeable spur dikes are dealt. These structures are built from the river bank into the stream flow and usually built in group. Construction of groyne against the flow causes significant changes in flow pattern in channel. Those changes may result in scour phenomenon around groynes which may lead structure instability and changes in river morphology. Moreover, in series of groynes, spacing of groynes leads different types of recirculating flow patterns.Therefore, investigating the characteristics of flow pattern around groynes have been a great interest in river engineering. Numerous researchers like Sukhodolov et al. (2002), Hao Zhang et al.(2009), Beheshti (2010), Duan (2009), Naji(2010), Karami(2011) made a variety of experiments in order to determine the flow pattern around groynes. Most of these researchers studied effect of single groyne, while using series of groynes is more effective in protection of rivers. Besides experimental studies, variety of CFD models have been developed for computing flow pattern around hydraulic structures; like Fluent, Flow 3D, Nays 2D, Nays CUBE and SSIIM. In this study, Nays 2D numerical modelling has been used to investigate flow and recirculating pattern around a series of groynes and streamlines including components of velocities.
    1. Flow pattern in groyne fields
      Under conditions where the groynes are not submerged, the groyne fields are not really part of the wetted cross section of a river. Because of that, the flow pattern in the groyne-field is not directly the result of the discharge in the main channel. Reducing the main stream velocity has no effect on the flow pattern itself, whereas lowering the water level does (Uijttewaal et al.2001). Moreover, the flow pattern inside a groyne field may change with the change of its geometry, location along the river (inner curve, outer curve, or straight part), and/ or the groynes orientation( Przedwojski et al.1995). However, there is an indirect effect of the discharge on the flow pattern in the groyne field. Because of the flow that is diverted from the main channel into the groyne fields, water flows into the groyne field with low velocity through the downstream half of the interfacial section between the groyne field and the main channel. This water flows back to the main channel through a small width of, just downstream the upstream groyne of the groyne field ( Termes et al.1991). Flow separates on a groyne head and forms a secondary flow represented by a large scale vortex with a vertical axis of rotation called primary gyre. Deflection of the flow inside the groyne field by banks and upstream groynes leads to the development of a secondary gyre with an opposite direction of rotation to the primary gyre. Location, mutual interactions, and energy exchange between gyres are the factors that create a specific recirculation pattern, and, consequently assuming correspondence with sedimentation processes, they define deposition patterns.
    2. Model Formulation
      The CFD model selected for this study is the publically available software NAYS 2D (iRIC 2.0), which is an analytical solver for calculation of unsteady two-dimensional plane flow and riverbed deformation using boundary-fitted coordinates within general curvilinear coordinates. A numerical channel of length 8.0m and width 0.9m was created with grid size of 0.01m im stream wise and 0.03m in cross stream directions. Groynes or spur dikes of length 0.15 and width 0.01m were chosen in series. Groyne field with various aspect ratio (b/x) 0.7, 0.25, 0.17, 0.125 and 0.10, where b=length of spur dike, x=spacing of two dikes. Discharge of 0.0175 m3 /s was applied. For boundary conditions, water surface at downstream and velocity at upstream were considered as uniform flow. Relaxation coefficient for water surface calculation was considered as 0.8. For the finite-difference method, the CIP method was applied to the advection terms in equations of motion. For the turbulent field calculation, Constant eddy viscosity, Zero-equation model and k-G models were applied and compared. The model!s accuracy in predicting the velocity magnitudes is evaluated using statistical parameters- mean absolute error (MAE), mean square error(MSE), and root mean square error (RMSE). The comparison of results shows the importance of selecting an appropriate turbulence model in simulating flow field around a spur dike. From the comparison, k-I model is found superior over zero energy model and eddy viscosity model. So, k-I model is chosen as appropriate turbulence closure model.
    3. Model!s Validation
      The capability of CFD model Nays 2D to simulate the velocity field and recirculation pattern in groyne field was compared with experimental data of laboratory experiments by Sukhodolov et al. (2002). The numerical simulation was validated for aspect ratio (R=b/x=0.7) and R=0.25. For aspect ratio R=0.7, one gyre system occupies the whole area of the groyne field. The areas with lower-than-average velocity values are clearly seen in the central part of the gyre and near its corners. Velocities increase towards the margins of the gyre. For aspect ratio R=0.25, two gyre velocity fields were observed in the groyne field. In the downstream part of the groyne field a large gyre, covering two-thirds of the area is clearly visible. The left part(upstream) contains second gyre rotating much more slowly and in the direction opposed to the primary gyre. The simulated and observed velocity field pattern and gyre found satisfactorily agreed. Now, after validation, the model was used for further analysis of velocity field for various aspect ratios.
    Fig 2(b) Observed velocity field for aspect ratio 0.25(Sukhodolov 2002)
    Fig 2(b) Observed velocity field for aspect ratio 0.25(Sukhodolov 2002)
    1. Results and Discussions
      The calibrated model was applied to five different cases of un-submerged and impermeable groyne fields with aspect ratios R=0.70,0.25,0.17,0.125 & 0.10 and flow pattern was numerically simulated. For aspect ratio R=0.7 i.e x/b=1.5, Fig 1(a) only one lateral primary gyre was formed inside the groyne field. The circulation pattern in this case is distinguished by the main flow that is deflected outside the groyne field. The developed primary gyre prevents the main flow from penetrating the groyne field. Therefore, this pattern is desirable for navigation purposes as a continuous deep channel is maintained along the face of the groyne field. Simulated velocity pattern satisfactorily agrees with the observed velocity field Fig 1(b) for the same aspect ratio by Sukhodolov (2002). The spacing of the groyne was further increased maintaining aspect ratio R= 0.25 i. e x/b=4 Fig 2(a) and flow pattern inside the groyne field was simulated. In this case, in the downstream part of the groyne field, a primary gyre occupying almost two-third area was formed. In addition, deflection of the flow inside the groyne field by banks and upstream groynes leads to the development of a secondary gyre with an opposite direction of rotation to the primary gyre covering almost one-third part of the groyne field. Likewise in the first case, the main current is maintained deflected outside the groyne field. Simulated velocity pattern satisfactorily agrees with the observed velocity field Fig 2(b) for the same aspect ratio by Sukhodolov (2002). The spacing of the groyne was further increased maintaining aspect ratio R=0.17 i.e x/b=6. In this case the flow pattern was similar to the aspect ratio R=0.25. The spacing of the groynes was further increased maintaining aspect ratio R=0.125 i. e x/b=8. In this case, similar to the previous scenarios two longitudinal gyres but with different positions are formed. The main current is directed in to the groyne field (Fig 3) creating a much more stronger eddy near the upstream groyne and greater turbulence along the upstream face and at the groyne lower head. As the spacing between groynes increased maintaining aspect ratio R=0.10 i. e x/b=10 (Fig 4), still primary and secondary gyres are generated. The formed gyres deflect the main flow thus preventing to enter in to the groyne field in upstream part. However, in the downstream of the primary gyre and just upstream of the second groyne, the flow attacks the bank directly. The resultant velocity profiles at the deflected region y/b=3 were plotted and how the spacing of second groyne affect the result was analyzed. Spacing of groynes makes little change in upstream resultant velocity. However, in the deflected region, its effect is significant. Higher value of spacing of groyne leads higher average deviation in resultant velocity. For aspect ratio R=0.7, the average deviation estimated as 0.02%. In the case of aspect ratio R=0.25, this value was reached to 1.57%. Further increment of spacing i. e decreasing the aspect ratio R=0.17, average deviation was found 3.82%. For the aspect ratio R=0.125, that value was estimated as 4.16%.
    2. Conclusions
      Geometry of the groyne fields; width and length of the groyne field mainly cause the specific flow patterns including number and shape of eddies or gyres. Eddies developed inside the groyne field deflects the main flow preventing it entering into the dead zone. An aspect ratio close to unity gives rise to a single eddy. A smaller aspect ratio (higher spacing between groynes) gives room to two stationary eddies, a large one called primary eddy, in the downstream part of the groyne field, and a smaller secondary eddy emerges near the upstream groyne. The extreme long groyne field -case of length to width ratio of larger thaneight shows penetration of main flow into the groyne field. The two eddies remain in a relatively stable position, while the main flow zone starts to penetrate into groyne field further downstream. In all cases, there is an eddy detaches from the upstream groyne tip that travels along the main channel groyne field interface and eventually merges with the primary eddy. The simulated results indicate that the spacing of groynes or spur dikes from the controlling of bank erosion point of view should be limited within six times the length of groyne.
    Fig 3 Computed velocity field for aspect ratio 0.125
    Fig 3 Computed velocity field for aspect ratio 0.125
    Fig 4 Computed velocity field for aspect ratio 0.10
    Fig 4 Computed velocity field for aspect ratio 0.10
    Fig 5 Resultant velocity profiles at y/b=3
    Fig 5 Resultant velocity profiles at y/b=3
    Fig 5 Resultant velocity profiles at y/b=3
    Fig 5 Resultant velocity profiles at y/b=3

    References

    1. Holtz, K.P  Numerical simulation of recirculating flow at groynes.Å Computer Methods in Water Resources, Vol 2, No 2 (1991).
    2. Hossein, Bassar; Abdollah, Ardeshir; Hojat, Karami.  Numerical simulation of flow pattern around spur dikes series in rigid bed.Å 9th international congress on civil engineering, May 8- 10,2012, Isfahan University of Technology (IUT) , Isfahan, Iran (2012).
    3. Kang, J.G; Yeo, H.K; Kim,S.J An experimental study on a characteristics of flow around groyne area by install conditions.Å www.SciRP.org/journal/eng(2012).
    4. Shimizu,Y; Nelson,JIntroduction of Nays solver in iRIC.Åwww.i-ric.org(2012).
    5. Sukhodolov, A. Uijttewaal, W. S. J., and Engelhardt, C. On the correspondence between morphological and hydro dynamical patterns of groyne fields.Å Earth Surf. Processes Landforms, 27(3) (2002).
    6. Uijttewall, W.S.J; Lehman,D; VanMazijk,A.  Exchange process between a river and its groyne fields-model experiments.Å Journal of Hydraulic Engineering, ASCE, 127(11) (2001).
    7. Uijttewall, W.S.J Groyne field velocity patterns determined with particle tracking
      velocimetryÅ.28th IAHR congress, Graz, Austria (1999).
    8. Yossef, Mohamed  Flow details near groynes: Experimental investigations.Å Journal of Hydraulic Engineering, ASCE, 137 (2011).
    Fig. 5. The predicted shapes of initial breach (a) Rectangular (b) V-notch. Fig. 6. Dam breaching stages.

    Investigating the peak outflow through a spatial embankment dam breach

    공간적 제방댐 붕괴를 통한 최대 유출량 조사

    Mahmoud T.GhonimMagdy H.MowafyMohamed N.SalemAshrafJatwaryFaculty of Engineering, Zagazig University, Zagazig 44519, Egypt

    Abstract

    Investigating the breach outflow hydrograph is an essential task to conduct mitigation plans and flood warnings. In the present study, the spatial dam breach is simulated by using a three-dimensional computational fluid dynamics model, FLOW-3D. The model parameters were adjusted by making a comparison with a previous experimental model. The different parameters (initial breach shape, dimensions, location, and dam slopes) are studied to investigate their effects on dam breaching. The results indicate that these parameters have a significant impact. The maximum erosion rate and peak outflow for the rectangular shape are higher than those for the V-notch by 8.85% and 5%, respectively. Increasing breach width or decreasing depth by 5% leads to increasing maximum erosion rate by 11% and 15%, respectively. Increasing the downstream slope angle by 4° leads to an increase in both peak outflow and maximum erosion rate by 2.0% and 6.0%, respectively.

    유출 유출 수문곡선을 조사하는 것은 완화 계획 및 홍수 경보를 수행하는 데 필수적인 작업입니다. 본 연구에서는 3차원 전산유체역학 모델인 FLOW-3D를 사용하여 공간 댐 붕괴를 시뮬레이션합니다. 이전 실험 모델과 비교하여 모델 매개변수를 조정했습니다.

    다양한 매개변수(초기 붕괴 형태, 치수, 위치 및 댐 경사)가 댐 붕괴에 미치는 영향을 조사하기 위해 연구됩니다. 결과는 이러한 매개변수가 상당한 영향을 미친다는 것을 나타냅니다. 직사각형 형태의 최대 침식율과 최대 유출량은 V-notch보다 각각 8.85%, 5% 높게 나타났습니다.

    위반 폭을 늘리거나 깊이를 5% 줄이면 최대 침식률이 각각 11% 및 15% 증가합니다. 하류 경사각을 4° 증가시키면 최대 유출량과 최대 침식률이 각각 2.0% 및 6.0% 증가합니다.

    Keywords

    Spatial dam breach; FLOW-3D; Overtopping erosion; Computational fluid dynamics (CFD)

    1. Introduction

    There are many purposes for dam construction, such as protection from flood disasters, water storage, and power generationEmbankment failures may have a catastrophic impact on lives and infrastructure in the downstream regions. One of the most common causes of embankment dam failure is overtopping. Once the overtopping of the dam begins, the breach formation will start in the dam body then end with the dam failure. This failure occurs within a very short time, which threatens to be very dangerous. Therefore, understanding and modeling the embankment breaching processes is essential for conducting mitigation plans, flood warnings, and forecasting flood damage.

    The analysis of the dam breaching process is implemented by different techniques: comparative methods, empirical models with dimensional and dimensionless solutions, physical-based models, and parametric models. These models were described in detail [1]Parametric modeling is commonly used to simulate breach growth as a time-dependent linear process and calculate outflow discharge from the breach using hydraulics principles [2]. Alhasan et al. [3] presented a simple one-dimensional mathematical model and a computer code to simulate the dam breaching process. These models were validated by small dams breaching during the floods in 2002 in the Czech Republic. Fread [4] developed an erosion model (BREACH) based on hydraulics principles, sediment transport, and soil mechanics to estimate breach size, time of formation, and outflow discharge. Říha et al. [5] investigated the dam break process for a cascade of small dams using a simple parametric model for piping and overtopping erosion, as well as a 2D shallow-water flow model for the flood in downstream areas. Goodarzi et al. [6] implemented mathematical and statistical methods to assess the effect of inflows and wind speeds on the dam’s overtopping failure.

    Dam breaching studies can be divided into two main modes of erosion. The first mode is called “planar dam breach” where the flow overtops the whole dam width. While the second mode is called “spatial dam breach” where the flow overtops through the initial pilot channel (i.e., a channel created in the dam body). Therefore, the erosion will be in both vertical and horizontal directions [7].

    The erosion process through the embankment dams occurs due to the shear stress applied by water flows. The dam breaching evolution can be divided into three stages [8][9], but Y. Yang et al. [10] divided the breach development into five stages: Stage I, the seepage erosion; Stage II, the initial breach formation; Stage III, the head erosion; Stage IV, the breach expansion; and Stage V, the re-equilibrium of the river channel through the breach. Many experimental tests have been carried out on non-cohesive embankment dams with an initial breach to examine the effect of upstream inflow discharges on the longitudinal profile evolution and the time to inflection point [11].

    Zhang et al. [12] studied the effect of changing downstream slope angle, sediment grain size, and dam crest length on erosion rates. They noticed that increasing dam crest length and decreasing downstream slope angle lead to decreasing sediment transport rate. While the increase in sediment grain size leads to an increased sediment transport rate at the initial stages. Höeg et al. [13] presented a series of field tests to investigate the stability of embankment dams made of various materials. Overtopping and piping were among the failure tests carried out for the dams composed of homogeneous rock-fill, clay, or gravel with a height of up to 6.0 m. Hakimzadeh et al. [14] constructed 40 homogeneous cohesive and non-cohesive embankment dams to study the effect of changing sediment diameter and dam height on the breaching process. They also used genetic programming (GP) to estimate the breach outflow. Refaiy et al. [15] studied different scenarios for the downstream drain geometry, such as length, height, and angle, to minimize the effect of piping phenomena and therefore increase dam safety.

    Zhu et al. [16] examined the effect of headcut erosion on dam breach growth, especially in the case of cohesive dams. They found that the breach growth in non-cohesive embankments is slower than cohesive embankments due to the little effect of headcut. Schmocker and Hager [7] proposed a relationship for estimating peak outflow from the dam breach process.(1)QpQin-1=1.7exp-20hc23d5013H0

    where: Qp = peak outflow discharge.

    Qin = inflow discharge.

    hc = critical flow depth.

    d50 = mean sediment diameter.

    Ho = initial dam height.

    Yu et al. [17] carried out an experimental study for homogeneous non-cohesive embankment dams in a 180° bending rectangular flume to determine the effect of overtopping flows on breaching formation. They found that the main factors influencing breach formation are water level, river discharge, and embankment material diameter.

    Wu et al. [18] carried out a series of experiments to investigate the effect of breaching geometry on both non-cohesive and cohesive embankment dams in a U-bend flume due to overtopping flows. In the case of non-cohesive embankments, the non-symmetrical lateral expansion was noticed during the breach formation. This expansion was described by a coefficient ranging from 2.7 to 3.3.

    The numerical models of the dam breach can be categorized according to different parameters, such as flow dimensions (1D, 2D, or 3D), flow governing equations, and solution methods. The 1D models are mainly used to predict the outflow hydrograph from the dam breach. Saberi et al. [19] applied the 1D Saint-Venant equation, which is solved by the finite difference method to investigate the outflow hydrograph during dam overtopping failure. Because of the ability to study dam profile evolution and breach formation, 2D models are more applicable than 1D models. Guan et al. [20] and Wu et al. [21] employed both 2D shallow water equations (SWEs) and sediment erosion equations, which are solved by the finite volume method to study the effect of the dam’s geometry parameters on outflow hydrograph and dam profile evolution. Wang et al. [22] also proposed a second-order hybrid-type of total variation diminishing (TVD) finite-difference to estimate the breach outflow by solving the 2D (SWEs). The accuracy of (SWEs) for both vertical flow contraction and surface roughness has been assessed [23]. They noted that the accuracy of (SWEs) is acceptable for milder slopes, but in the case of steeper slopes, modelers should be more careful. Generally, the accuracy of 2D models is still low, especially with velocity distribution over the flow depth, lateral momentum exchange, density-driven flows, and bottom friction [24]. Therefore, 3D models are preferred. Larocque et al. [25] and Yang et al. [26] started to use three-dimensional (3D) models that depend on the Reynolds-averaged Navier-Stokes (RANS) equations.

    Previous experimental studies concluded that there is no clear relationship between the peak outflow from the dam breach and the initial breach characteristics. Some of these studies depend on the sharp-crested weir fixed at the end of the flume to determine the peak outflow from the breach, which leads to a decrease in the accuracy of outflow calculations at the microscale. The main goals of this study are to carry out a numerical simulation for a spatial dam breach due to overtopping flows by using (FLOW-3D) software to find an empirical equation for the peak outflow discharge from the breach and determine the worst-case that leads to accelerating the dam breaching process.

    2. Numerical simulation

    The current study for spatial dam breach is simulated by using (FLOW-3D) software [27], which is a powerful computational fluid dynamics (CFD) program.

    2.1. Geometric presentations

    A stereolithographic (STL) file is prepared for each change in the initial breach geometry and dimensions. The CAD program is useful for creating solid objects and converting them to STL format, as shown in Fig. 1.

    2.2. Governing equations

    The governing equations for water flow are three-dimensional Reynolds Averaged Navier-Stokes equations (RANS).

    The continuity equation:(2)∂ui∂xi=0

    The momentum equation:(3)∂ui∂t+1VFuj∂ui∂xj=1ρ∂∂xj-pδij+ν∂ui∂xj+∂uj∂xi-ρu`iu`j¯

    where u is time-averaged velocity,ν is kinematic viscosity, VF is fractional volume open to flow, p is averaged pressure and -u`iu`j¯ are components of Reynold’s stress. The Volume of Fluid (VOF) technique is used to simulate the free surface profile. Hirt et al. [28] presented the VOF algorithm, which employs the function (F) to express the occupancy of each grid cell with fluid. The value of (F) varies from zero to unity. Zero value refers to no fluid in the grid cell, while the unity value refers to the grid cell being fully occupied with fluid. The free surface is formed in the grid cells having (F) values between zero and unity.(4)∂F∂t+1VF∂∂xFAxu+∂∂yFAyv+∂∂zFAzw=0

    where (u, v, w) are the velocity components in (x, y, z) coordinates, respectively, and (AxAyAz) are the area fractions.

    2.3. Boundary and initial conditions

    To improve the accuracy of the results, the boundary conditions should be carefully determined. In this study, two mesh blocks are used to minimize the time consumed in the simulation. The boundary conditions for mesh block 1 are as follows: The inlet and sides boundaries are defined as a wall boundary condition (wall boundary condition is usually used for bound fluid by solid regions. In the case of viscous flows, no-slip means that the tangential velocity is equal to the wall velocity and the normal velocity is zero), the outlet is defined as a symmetry boundary condition (symmetry boundary condition is usually used to reduce computational effort during CFD simulation. This condition allows the flow to be transferred from one mesh block to another. No inputs are required for this boundary condition except that its location should be defined accurately), the bottom boundary is defined as a uniform flow rate boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure. The boundary conditions for mesh block 2 are as follows: The inlet is defined as a symmetry boundary condition, the outlet is defined as a free flow boundary condition, the bottom and sides boundaries are defined as a wall boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure as shown in Fig. 2. The initial conditions required to be set for the fluid (i.e., water) inside of the domain include configuration, temperature, velocities, and pressure distribution. The configuration of water depends on the dimensions and shape of the dam reservoir. While the other conditions have been assigned as follows: temperature is normal water temperature (25 °c) and pressure distribution is hydrostatic with no initial velocity.

    2.4. Numerical method

    FLOW-3D uses the finite volume method (FVM) to solve the governing equation (Reynolds-averaged Navier-Stokes) over the computational domain. A finite-volume method is an Eulerian approach for representing and evaluating partial differential equations in algebraic equations form [29]. At discrete points on the mesh geometry, values are determined. Finite volume expresses a small volume surrounding each node point on a mesh. In this method, the divergence theorem is used to convert volume integrals with a divergence term to surface integrals. After that, these terms are evaluated as fluxes at each finite volume’s surfaces.

    2.5. Turbulent models

    Turbulence is the chaotic, unstable motion of fluids that occurs when there are insufficient stabilizing viscous forces. In FLOW-3D, there are six turbulence models available: the Prandtl mixing length model, the one-equation turbulent energy model, the two-equation (k – ε) model, the Renormalization-Group (RNG) model, the two-equation (k – ω) models, and a large eddy simulation (LES) model. For simulating flow motion, the RNG model is adopted to simulate the motion behavior better than the k – ε and k – ω.

    models [30]. The RNG model consists of two main equations for the turbulent kinetic energy KT and its dissipation.εT(5)∂kT∂t+1VFuAx∂kT∂x+vAy∂kT∂y+wAz∂kT∂z=PT+GT+DiffKT-εT(6)∂εT∂t+1VFuAx∂εT∂x+vAy∂εT∂y+wAz∂εT∂z=C1.εTKTPT+c3.GT+Diffε-c2εT2kT

    where KT is the turbulent kinetic energy, PT is the turbulent kinetic energy production, GT is the buoyancy turbulence energy, εT is the turbulent energy dissipation rate, DiffKT and Diffε are terms of diffusion, c1, c2 and c3 are dimensionless parameters, in which c1 and c3 have a constant value of 1.42 and 0.2, respectively, c2 is computed from the turbulent kinetic energy (KT) and turbulent production (PT) terms.

    2.6. Sediment scour model

    The sediment scour model available in FLOW-3D can calculate all the sediment transport processes including Entrainment transport, Bedload transport, Suspended transport, and Deposition. The erosion process starts once the water flows remove the grains from the packed bed and carry them into suspension. It happens when the applied shear stress by water flows exceeds critical shear stress. This process is represented by entrainment transport in the numerical model. After entrained, the grains carried by water flow are represented by suspended load transport. After that, some suspended grains resort to settling because of the combined effect of gravity, buoyancy, and friction. This process is described through a deposition. Finally, the grains sliding motions are represented by bedload transport in the model. For the entrainment process, the shear stress applied by the fluid motion on the packed bed surface is calculated using the standard wall function as shown in Eq.7.(7)ks,i=Cs,i∗d50

    where ks,i is the Nikuradse roughness and Cs,i is a user-defined coefficient. The critical bed shear stress is defined by a dimensionless parameter called the critical shields number as expressed in Eq.8.(8)θcr,i=τcr,i‖g‖diρi-ρf

    where θcr,i is the critical shields number, τcr,i is the critical bed shear stress, g is the absolute value of gravity acceleration, di is the diameter of the sediment grain, ρi is the density of the sediment species (i) and ρf is the density of the fluid. The value of the critical shields number is determined according to the Soulsby-Whitehouse equation.(9)θcr,i=0.31+1.2d∗,i+0.0551-exp-0.02d∗,i

    where d∗,i is the dimensionless diameter of the sediment, given by Eq.10.(10)d∗,i=diρfρi-ρf‖g‖μf213

    where μf is the fluid dynamic viscosity. For the sloping bed interface, the value of the critical shields number is modified according to Eq.11.(11)θ`cr,i=θcr,icosψsinβ+cos2βtan2φi-sin2ψsin2βtanφi

    where θ`cr,i is the modified critical shields number, φi is the angle of repose for the sediment, β is the angle of bed slope and ψ is the angle between the flow and the upslope direction. The effects of the rolling, hopping, and sliding motions of grains along the packed bed surface are taken by the bedload transport process. The volumetric bedload transport rate (qb,i) per width of the bed is expressed in Eq.12.(12)qb,i=Φi‖g‖ρi-ρfρfdi312

    where Φi is the dimensionless bedload transport rate is calculated by using Meyer Peter and Müller equation.(13)Φi=βMPM,iθi-θ`cr,i1.5cb,i

    where βMPM,i is the Meyer Peter and Müller user-defined coefficient and cb,i is the volume fraction of species i in the bed material. The suspended load transport is calculated as shown in Eq.14.(14)∂Cs,i∂t+∇∙Cs,ius,i=∇∙∇DCs,i

    where Cs,i is the suspended sediment mass concentration, D is the diffusivity, and us,i is the grain velocity of species i. Entrainment and deposition are two opposing processes that take place at the same time. The lifting and settling velocities for both entrainment and deposition processes are calculated according to Eq.15 and Eq.16, respectively.(15)ulifting,i=αid∗,i0.3θi-θ`cr,igdiρiρf-1(16)usettling,i=υfdi10.362+1.049d∗,i3-10.36

    where αi is the entrainment coefficient of species i and υf is the kinematic viscosity of the fluid.

    2.7. Grid type

    Using simple rectangular orthogonal elements in planes and hexahedral in volumes in the (FLOW-3D) program makes the mesh generation process easier, decreases the required memory, and improves numerical accuracy. Two mesh blocks were used in a joined form with a size ratio of 2:1. The first mesh block is coarser, which contains the reservoir water, and the second mesh block is finer, which contains the dam. For achieving accuracy and efficiency in results, the mesh size is determined by using a grid convergence test. The optimum uniform cell size for the first mesh block is 0.012 m and for the second mesh block is 0.006 m.

    2.8. Time step

    The maximum time step size is determined by using a Courant number, which controls the distance that the flow will travel during the simulation time step. In this study, the Courant number was taken equal to 0.25 to prevent the flow from traveling through more than one cell in the time step. Based on the Courant number, a maximum time step value of 0.00075 s was determined.

    2.9. Numerical model validation

    The numerical model accuracy was achieved by comparing the numerical model results with previous experimental results. The experimental study of Schmocker and Hager [7] was based on 31 tests with changes in six parameters (d50, Ho, Bo, Lk, XD, and Qin). All experimental tests were conducted in a straight open glass-sided flume. The horizontal flume has a rectangular cross-section with a width of 0.4 m and a height of 0.7 m. The flume was provided with a flow straightener and an intake with a length of 0.66 m. All tested dams were inserted at various distances (XD) from the intake. Test No.1 from this experimental program was chosen to validate the numerical model. The different parameters used in test No.1 are as follows:

    (1) uniform sediment with a mean diameter (d50 = 0.31 mm), (2) Ho = 0.2 m, (3) Bo = 0.2 m, (4) Lk = 0.1 m,

    (5) XD = 1.0 m, (6) Qin = 6.0 lit/s, (7) Su and Sd = 2:1, (8) mass density (ρs = 2650 kg/m3(9) Homogenous and non-cohesive embankment dam. As shown in Fig. 2, the simulation is contained within a rectangular grid with dimensions: 3.56 m in the x-direction (where 0.66 m is used as inlet, 0.9 m as dam base width, and 1.0 m as outlet), in y-direction 0.2 m (dam length), and in the z-direction 0.3 m, which represents the dam height (0.2 m) with a free distance (0.1 m) above the dam. There are two main reasons that this experimental program is preferred for the validation process. The first reason is that this program deals with homogenous, non-cohesive soil, which is available in FLOW-3D. The second reason is that this program deals with small-scale models which saves time for numerical simulation. Finally, some important assumptions were considered during the validation process. The flow is assumed to be incompressible, viscous, turbulent, and three-dimensional.

    By comparing dam profiles at different time instants for the experimental test with the current numerical model, it appears that the numerical model gives good agreement as shown in Fig. 3 and Fig. 4, with an average error percentage of 9% between the experimental results and the numerical model.

    3. Analysis and discussions

    The current model is used to study the effects of different parameters such as (initial breach shapes, dimensions, locations, upstream and downstream dam slopes) on the peak outflow discharge, QP, time of peak outflow, tP, and rate of erosion, E.

    This study consists of a group of scenarios. The first scenario is changing the shapes of the initial breach according to Singh [1], the most predicted shapes are rectangular and V-notch as shown in Fig. 5. The second scenario is changing the initial breach dimensions (i.e., width and depth). While the third scenario is changing the location of the initial breach. Eventually, the last scenario is changing the upstream and downstream dam slopes.

    All scenarios of this study were carried out under the same conditions such as inflow discharge value (Qin=1.0lit/s), dimensions of the tested dam, where dam height (Ho=0.20m), crest width.

    (Lk=0.1m), dam length (Bo=0.20m), and homogenous & non-cohesive soil with a mean diameter (d50=0.31mm).

    3.1. Dam breaching process evolution

    The dam breaching process is a very complex process due to the quick changes in hydrodynamic conditions during dam failure. The dam breaching process starts once water flows reach the downstream face of the dam. During the initial stage of dam breaching, the erosion process is relatively quiet due to low velocities of flow. As water flows continuously, erosion rates increase, especially in two main zones: the crest and the downstream face. As soon as the dam crest is totally eroded, the water levels in the dam reservoir decrease rapidly, accompanied by excessive erosion in the dam body. The erosion process continues until the water levels in the dam reservoir equal the remaining height of the dam.

    According to Zhou et al. [11], the breaching process consists of three main stages. The first stage starts with beginning overtopping flow, then ends when the erosion point directed upstream and reached the inflection point at the inflection time (ti). The second stage starts from the end of the stage1 until the occurrence of peak outflow discharge at the peak outflow time (tP). The third stage starts from the end of the stage2 until the value of outflow discharge becomes the same as the value of inflow discharge at the final time (tf). The outflow discharge from the dam breach increases rapidly during stage1 and stage2 because of the large dam storage capacity (i.e., the dam reservoir is totally full of water) and excessive erosion. While at stage3, the outflow values start to decrease slowly because most of the dam’s storage capacity was run out. The end of stage3 indicates that the dam storage capacity was totally run out, so the outflow equalized with the inflow discharge as shown in Fig. 6 and Fig. 7.

    3.2. The effect of initial breach shape

    To identify the effect of the initial breach shape on the evolution of the dam breaching process. Three tests were carried out with different cross-section areas for each shape. The initial breach is created at the center of the dam crest. Each test had an ID to make the process of arranging data easier. The rectangular shape had an ID (Rec5h & 5b), which means that its depth and width are equal to 5% of the dam height, and the V-notch shape had an ID (V-noch5h & 1:1) which means that its depth is equal to 5% of the dam height and its side slope is equal to 1:1. The comparison between rectangular and V-notch shapes is done by calculating the ratio between maximum dam height at different times (ZMax) to the initial dam height (Ho), rate of erosion, and hydrograph of outflow discharge for each test. The rectangular shape achieves maximum erosion rate and minimum inflection time, in addition to a rapid decrease in the dam reservoir levels. Therefore, the dam breaching is faster in the case of a rectangular shape than in a V-notch shape, which has the same cross-section area as shown in Fig. 8.

    Also, by comparing the hydrograph for each test, the peak outflow discharge value in the case of a rectangular shape is higher than the V-notch shape by 5% and the time of peak outflow for the rectangular shape is shorter than the V-notch shape by 9% as shown in Fig. 9.

    3.3. The effect of initial breach dimensions

    The results of the comparison between the different initial breach shapes indicate that the worst initial breach shape is rectangular, so the second scenario from this study concentrated on studying the effect of a change in the initial rectangular breach dimensions. Groups of tests were carried out with different depths and widths for the rectangular initial breach. The first group had a depth of 5% from the dam height and with three different widths of 5,10, and 15% from the dam height, the second group had a depth of 10% with three different widths of 5,10, and 15%, the third group had a depth of 15% with three different widths of 5,10, and 15% and the final group had a width of 15% with three different heights of 5, 10, and 15% for a rectangular breach shape. The comparison was made as in the previous section to determine the worst case that leads to the quick dam failure as shown in Fig. 10.

    The results show that the (Rec 5 h&15b) test achieves a maximum erosion rate for a shorter period of time and a minimum ratio for (Zmax / Ho) as shown in Fig. 10, which leads to accelerating the dam failure process. The dam breaching process is faster with the minimum initial breach depth and maximum initial breach width. In the case of a minimum initial breach depth, the retained head of water in the dam reservoir is high and the crest width at the bottom of the initial breach (L`K) is small, so the erosion point reaches the inflection point rapidly. While in the case of the maximum initial breach width, the erosion perimeter is large.

    3.4. The effect of initial breach location

    The results of the comparison between the different initial rectangular breach dimensions indicate that the worst initial breach dimension is (Rec 5 h&15b), so the third scenario from this study concentrated on studying the effect of a change in the initial breach location. Three locations were checked to determine the worst case for the dam failure process. The first location is at the center of the dam crest, which was named “Center”, the second location is at mid-distance between the dam center and dam edge, which was named “Mid”, and the third location is at the dam edge, which was named “Edge” as shown in Fig. 11. According to this scenario, the results indicate that the time of peak outflow discharge (tP) is the same in the three cases, but the maximum value of the peak outflow discharge occurs at the center location. The difference in the peak outflow values between the three cases is relatively small as shown in Fig. 12.

    The rates of erosion were also studied for the three cases. The results show that the maximum erosion rate occurs at the center location as shown in Fig. 13. By making a comparison between the three cases for the dam storage volume. The results show that the center location had the minimum values for the dam storage volume, which means that a large amount of water has passed to the downstream area as shown in Fig. 14. According to these results, the center location leads to increased erosion rate and accelerated dam failure process compared with the two other cases. Because the erosion occurs on both sides, but in the case of edge location, the erosion occurs on one side.

    3.5. The effect of upstream and downstream dam slopes

    The results of the comparison between the different initial rectangular breach locations indicate that the worst initial breach location is the center location, so the fourth scenario from this study concentrated on studying the effect of a change in the upstream (Su) and downstream (Sd) dam slopes. Three slopes were checked individually for both upstream and downstream slopes to determine the worst case for the dam failure process. The first slope value is (2H:1V), the second slope value is (2.5H:1V), and the third slope value is (3H:1V). According to this scenario, the results show that the decreasing downstream slope angle leads to increasing time of peak outflow discharge (tP) and decreasing value of peak outflow discharge. The difference in the peak outflow values between the three cases for the downstream slope is 2%, as shown in Fig. 15, but changing the upstream slope has a negligible impact on the peak outflow discharge and its time as shown in Fig. 16.

    The rates of erosion were also studied in the three cases for both upstream and downstream slopes. The results show that the maximum erosion rate increases by 6.0% with an increasing downstream slope angle by 4°, as shown in Fig. 17. The results also indicate that the erosion rates aren’t affected by increasing or decreasing the upstream slope angle, as shown in Fig. 18. According to these results, increasing the downstream slope angle leads to increased erosion rate and accelerated dam failure process compared with the upstream slope angle. Because of increasing shear stress applied by water flows in case of increasing downstream slope.

    According to all previous scenarios, the dimensionless peak outflow discharge QPQin is presented for a fixed dam height (Ho) and inflow discharge (Qin). Fig. 19 illustrates the relationship between QP∗=QPQin and.

    Lr=ho2/3∗bo2/3Ho. The deduced relationship achieves R2=0.96.(17)QP∗=2.2807exp-2.804∗Lr

    4. Conclusions

    A spatial dam breaching process was simulated by using FLOW-3D Software. The validation process was performed by making a comparison between the simulated results of dam profiles and the dam profiles obtained by Schmocker and Hager [7] in their experimental study. And also, the peak outflow value recorded an error percentage of 12% between the numerical model and the experimental study. This model was used to study the effect of initial breach shape, dimensions, location, and dam slopes on peak outflow discharge, time of peak outflow, and the erosion process. By using the parameters obtained from the validation process, the results of this study can be summarized in eight points as follows.1.

    The rectangular initial breach shape leads to an accelerating dam failure process compared with the V-notch.2.

    The value of peak outflow discharge in the case of a rectangular initial breach is higher than the V-notch shape by 5%.3.

    The time of peak outflow discharge for a rectangular initial breach is shorter than the V-notch shape by 9%.4.

    The minimum depth and maximum width for the initial breach achieve maximum erosion rates (increasing breach width, b0, or decreasing breach depth, h0, by 5% from the dam height leads to an increase in the maximum rate of erosion by 11% and 15%, respectively), so the dam failure is rapid.5.

    The center location of the initial breach leads to an accelerating dam failure compared with the edge location.6.

    The initial breach location has a negligible effect on the peak outflow discharge value and its time.7.

    Increasing the downstream slope angle by 4° leads to an increase in both peak outflow discharge and maximum rate of erosion by 2.0% and 6.0%, respectively.8.

    The upstream slope has a negligible effect on the dam breaching process.

    References

    [1]V. SinghDam breach modeling technologySpringer Science & Business Media (1996)Google Scholar[2]Wahl TL. Prediction of embankment dam breach parameters: a literature review and needs assessment. 1998.Google Scholar[3]Z. Alhasan, J. Jandora, J. ŘíhaStudy of dam-break due to overtopping of four small dams in the Czech RepublicActa Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 63 (3) (2015), pp. 717-729 View PDFCrossRefView Record in ScopusGoogle Scholar[4]D. FreadBREACH, an erosion model for earthen dam failures: Hydrologic Research LaboratoryNOAA, National Weather Service (1988)Google Scholar[5]J. Říha, S. Kotaška, L. PetrulaDam Break Modeling in a Cascade of Small Earthen Dams: Case Study of the Čižina River in the Czech RepublicWater, 12 (8) (2020), p. 2309, 10.3390/w12082309 View PDFView Record in ScopusGoogle Scholar[6]E. Goodarzi, L. Teang Shui, M. ZiaeiDam overtopping risk using probabilistic concepts–Case study: The Meijaran DamIran Ain Shams Eng J, 4 (2) (2013), pp. 185-197ArticleDownload PDFView Record in ScopusGoogle Scholar[7]L. Schmocker, W.H. HagerPlane dike-breach due to overtopping: effects of sediment, dike height and dischargeJ Hydraul Res, 50 (6) (2012), pp. 576-586 View PDFCrossRefView Record in ScopusGoogle Scholar[8]J.S. Walder, R.M. Iverson, J.W. Godt, M. Logan, S.A. SolovitzControls on the breach geometry and flood hydrograph during overtopping of noncohesive earthen damsWater Resour Res, 51 (8) (2015), pp. 6701-6724View Record in ScopusGoogle Scholar[9]H. Wei, M. Yu, D. Wang, Y. LiOvertopping breaching of river levees constructed with cohesive sedimentsNat Hazards Earth Syst Sci, 16 (7) (2016), pp. 1541-1551 View PDFCrossRefView Record in ScopusGoogle Scholar[10]Y. Yang, S.-Y. Cao, K.-J. Yang, W.-P. LiYang K-j, Li W-p. Experimental study of breach process of landslide dams by overtopping and its initiation mechanismsJ Hydrodynamics, 27 (6) (2015), pp. 872-883ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[11]G.G.D. Zhou, M. Zhou, M.S. Shrestha, D. Song, C.E. Choi, K.F.E. Cui, et al.Experimental investigation on the longitudinal evolution of landslide dam breaching and outburst floodsGeomorphology, 334 (2019), pp. 29-43ArticleDownload PDFView Record in ScopusGoogle Scholar[12]J. Zhang, Z.-x. Guo, S.-y. CaoYang F-g. Experimental study on scour and erosion of blocked damWater Sci Eng, 5 (2012), pp. 219-229ArticleDownload PDFView Record in ScopusGoogle Scholar[13]K. Höeg, A. Løvoll, K. VaskinnStability and breaching of embankment dams: Field tests on 6 m high damsInt J Hydropower Dams, 11 (2004), pp. 88-92View Record in ScopusGoogle Scholar[14]H. Hakimzadeh, V. Nourani, A.B. AminiGenetic programming simulation of dam breach hydrograph and peak outflow dischargeJ Hydrol Eng, 19 (4) (2014), pp. 757-768View Record in ScopusGoogle Scholar[15]A.R. Refaiy, N.M. AboulAtta, N.Y. Saad, D.A. El-MollaModeling the effect of downstream drain geometry on seepage through earth damsAin Shams Eng J, 12 (3) (2021), pp. 2511-2531ArticleDownload PDFView Record in ScopusGoogle Scholar[16]Y. Zhu, P.J. Visser, J.K. Vrijling, G. WangExperimental investigation on breaching of embankmentsScience China Technological Sci, 54 (1) (2011), pp. 148-155 View PDFCrossRefView Record in ScopusGoogle Scholar[17]M.-H. Yu, H.-Y. Wei, Y.-J. Liang, Y. ZhaoInvestigation of non-cohesive levee breach by overtopping flowJ Hydrodyn, 25 (4) (2013), pp. 572-579ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[18]S. Wu, M. Yu, H. Wei, Y. Liang, J. ZengNon-symmetrical levee breaching processes in a channel bend due to overtoppingInt J Sedim Res, 33 (2) (2018), pp. 208-215ArticleDownload PDFView Record in ScopusGoogle Scholar[19]O. Saberi, G. ZenzNumerical investigation on 1D and 2D embankment dams failure due to overtopping flowInt J Hydraulic Engineering, 5 (2016), pp. 9-18View Record in ScopusGoogle Scholar[20]M. Guan, N.G. Wright, P.A. Sleigh2D Process-Based Morphodynamic Model for Flooding by Noncohesive Dyke BreachJ Hydraul Eng, 140 (7) (2014), p. 04014022, 10.1061/(ASCE)HY.1943-7900.0000861 View PDFView Record in ScopusGoogle Scholar[21]W. Wu, R. Marsooli, Z. HeDepth-Averaged Two-Dimensional Model of Unsteady Flow and Sediment Transport due to Noncohesive Embankment Break/BreachingJ Hydraul Eng, 138 (6) (2012), pp. 503-516View Record in ScopusGoogle Scholar[22]Z. Wang, D.S. BowlesThree-dimensional non-cohesive earthen dam breach model. Part 1: Theory and methodologyAdv Water Resour, 29 (10) (2006), pp. 1528-1545ArticleDownload PDFView Record in ScopusGoogle Scholar[23]Říha J, Duchan D, Zachoval Z, Erpicum S, Archambeau P, Pirotton M, et al. Performance of a shallow-water model for simulating flow over trapezoidal broad-crested weirs. J Hydrology Hydromechanics. 2019;67:322-8.Google Scholar[24]C.B. VreugdenhilNumerical methods for shallow-water flowSpringer Science & Business Media (1994)Google Scholar[25]L.A. Larocque, J. Imran, M.H. Chaudhry3D numerical simulation of partial breach dam-break flow using the LES and k–∊ turbulence modelsJ Hydraul Res, 51 (2) (2013), pp. 145-157 View PDFCrossRefView Record in ScopusGoogle Scholar[26]C. Yang, B. Lin, C. Jiang, Y. LiuPredicting near-field dam-break flow and impact force using a 3D modelJ Hydraul Res, 48 (6) (2010), pp. 784-792 View PDFCrossRefView Record in ScopusGoogle Scholar[27]FLOW-3D. Version 11.1.1 Flow Science, Inc., Santa Fe, NM. https://wwwflow3dcom.Google Scholar[28]C.W. Hirt, B.D. NicholsVolume of fluid (VOF) method for the dynamics of free boundariesJ Comput Phys, 39 (1) (1981), pp. 201-225ArticleDownload PDFGoogle Scholar[29]S.V. PatankarNumerical heat transfer and fluid flow, Hemisphere PublCorp, New York, 58 (1980), p. 288View Record in ScopusGoogle Scholar[30]M. Alemi, R. MaiaNumerical simulation of the flow and local scour process around single and complex bridge piersInt J Civil Eng, 16 (5) (2018), pp. 475-487 View PDFCrossRefView Record in ScopusGoogle Scholar

    Effect of roughness on separation zone dimensions.

    Experimental and numerical study of flow at a 90 degree lateral turnout with enhanced roughness coefficient and invert level changes

    조도 계수 및 역전 수준 변화가 개선된 90도 측면 분출구에서의 유동에 대한 실험적 및 수치적 연구

    Maryam BagheriSeyed M. Ali ZomorodianMasih ZolghadrH. Md. AzamathullaC. Venkata Siva Rama Prasad

    Abstract

    측면 분기기(흡입구)의 상류 측에서 흐름 분리는 분기기 입구에서 와류를 일으키는 중요한 문제입니다. 이는 흐름의 유효 폭, 출력 용량 및 효율성을 감소시킵니다. 따라서 분리지대의 크기를 파악하고 크기를 줄이기 위한 방안을 제시하는 것이 필수적이다. 본 연구에서는 분리 구역의 치수를 줄이기 위한 방법으로 7가지 유형의 거칠기 요소를 분기구 입구에 설치하고 4가지 다른 배출(총 84번의 실험을 수행)과 함께 3개의 서로 다른 베드 반전 레벨을 조사했습니다. 또한 3D CFD(Computational Fluid Dynamics) 모델을 사용하여 분리 영역의 흐름 패턴과 치수를 평가했습니다. 결과는 거칠기 계수를 향상시키면 분리 영역 치수를 최대 38%까지 줄일 수 있는 반면, 드롭 구현 효과는 사용된 거칠기 계수를 기반으로 이 영역을 다르게 축소할 수 있음을 보여주었습니다. 두 가지 방법을 결합하면 분리 영역 치수를 최대 63%까지 줄일 수 있습니다.

    Flow separation at the upstream side of lateral turnouts (intakes) is a critical issue causing eddy currents at the turnout entrance. It reduces the effective width of flow, turnout capacity and efficiency. Therefore, it is essential to identify the dimensions of the separation zone and propose remedies to reduce its dimensions. Installation of 7 types of roughening elements at the turnout entrance and 3 different bed invert levels, with 4 different discharges (making a total of 84 experiments) were examined in this study as a method to reduce the dimensions of the separation zone. Additionally, a 3-D Computational Fluid Dynamic (CFD) model was utilized to evaluate the flow pattern and dimensions of the separation zone. Results showed that enhancing the roughness coefficient can reduce the separation zone dimensions up to 38% while the drop implementation effect can scale down this area differently based on the roughness coefficient used. Combining both methods can reduce the separation zone dimensions up to 63%.

    HIGHLIGHTS

    Listen

    • Flow separation at the upstream side of lateral turnouts (intakes) is a critical issue causing eddy currents at the turnout entrance.
    • Installation of 7 types of roughening elements at the turnout entrance and 3 different bed level inverts were investigated.
    • Additionally, a 3-D Computational Fluid Dynamic (CFD) model was utilized to evaluate the flow.
    • Combining both methods can reduce the separation zone dimensions by up to 63%.
    Experimental and numerical study of flow at a 90 degree lateral turnout with enhanced roughness coefficient and invert level changes
    Experimental and numerical study of flow at a 90 degree lateral turnout with enhanced roughness coefficient and invert level changes

    Keywords

    discharge ratioflow separation zoneintakethree dimensional simulation

    INTRODUCTION

    Listen

    Turnouts or intakes are amongst the oldest and most widely used hydraulic structures in irrigation networks. Turnouts are also used in water distribution, transmission networks, power generation facilities, and waste water treatment plants etc. The flows that enter a turnout have a strong momentum in the direction of the main waterway and that is why flow separation occurs inside the turnout. The horizontal vortex formed in the separation area is a suitable place for accumulation and deposition of sediments. The separation zone is a vulnerable area for sedimentation and for reduction of effective flow due to a contracted flow region in the lateral channel. Sedimentaion in the entrance of the intake can gradually be transfered into the lateral channel and decrease the capacity of the higher order channels over time (Jalili et al. 2011). On the other hand, the existence of coarse-grained materials causes erosion and destruction of the waterway side walls and bottom. In addition, sedimentation creates conditions for vegetation to take root and damage the waterway cover, which causes water to leak from its perimeter. Therefore, it is important to investigate the pattern of the flow separation area in turnouts and provide solutions to reduce the dimensions of this area.

    The three-dimensional flow structure at turnouts is quite complex. In an experimental study by Neary & Odgaard (1993) in a 90-degree water turnout it was found that the secondary currents and separation zone varies from the bed to the water surface. They also found that at a 90-degree water turnout, the bed roughness and discharge ratio play a critical role in flow structure. They asserted that an explanation of sediment behavior at a diversion entrance requires a comprehensive understanding of 3D flow patterns around the lateral-channel entrance. In addition, they suggested that there is a strong similarity between flow in a channel bend and a diversion channel, and that this similarity can rationalize the use of bend flow models for estimation of 3D flow structures in diversion channels.

    Some of the distinctive characteristics of dividing flow in a turnout include a zone of separation immediately near the entrance of the lateral turnout (separation zone), a contracted flow region in the branch channel (contracted flow), and a stagnation point near the downstream corner of the junction (stagnation zone). In the region downstream of the junction, along the continuous far wall, separation due to flow expansion may occur (Ramamurthy et al. 2007), that is, a separation zone. This can both reduce the turnout efficiency and the effective width of flow while increasing the sediment deposition in the turnout entrance (Jalili et al. 2011). Installation of submerged vanes in the turnout entrance is a method which is already applied to reduce the size of flow separation zones. The separation zone draws sediments and floating materials into themselves. This reduces effective cross-section area and reduces transmission capacity. These results have also been obtained in past studies, including by Ramamurthy et al. (2007) and in Jalili et al. (2011). Submerged vanes (Iowa vanes) are designed in order to modify the near-bed flow pattern and bed-sediment motion in the transverse direction of the river. The vanes are installed vertically on the channel bed, at an angle of attack which is usually oriented at 10–25 degrees to the local primary flow direction. Vane height is typically 0.2–0.5 times the local water depth during design flow conditions and vane length is 2–3 times its height (Odgaard & Wang 1991). They are vortex-generating devices that generate secondary circulation, thereby redistributing sediment within the channel cross section. Several factors affect the flow separation zone such as the ratio of lateral turnout discharge to main channel discharge, angle of lateral channel with respect to the main channel flow direction and size of applied submerged vanes. Nakato et al. (1990) found that sediment management using submerged vanes in the turnout entrance to Station 3 of the Council Bluffs plant, located on the Missouri River, is applicable and efficient. The results show submerged vanes are an appropriate solution for reduction of sediment deposition in a turnout entrance. The flow was treated as 3D and tests results were obtained for the flow characteristics of dividing flows in a 90-degree sharp-edged, junction. The main and lateral channel were rectangular with the same dimensions (Ramamurthy et al., 2007).

    Keshavarzi & Habibi (2005) carried out experiments on intake with angles of 45, 67, 79 and 90 degrees in different discharge ratios and reported the optimum angle for inlet flow with the lowest flow separation area to be about 55 degrees. The predicted flow characteristics were validated using experimental data. The results indicated that the width and length of the separation zone increases with the increase in the discharge ratio Qr (ratio of outflow per unit width in the turnout to inflow per unit width in the main channel).

    Abbasi et al. (2004) performed experiments to investigate the dimensions of the flow separation zone at a lateral turnout entrance. They demonstrated that the length and width of the separation zone decreases with the increasing ratio of lateral turn-out discharge. They also found that with a reducing angle of lateral turnout, the length of the separation zone scales up and width of separation zone reduces. Then they compared their observations with results of Kasthuri & Pundarikanthan (1987) who conducted some experiments in an open-channel junction formed by channels of equal width and an angle of lateral 90 degree turnout, which showed the dimensions of the separation zone in their experiments to be smaller than in previous studies. Kasthuri & Pundarikanthan (1987) studied vortex and flow separation dimensions at the entrance of a 90 degree channel. Results showed that increasing the diversion discharge ratio can reduce the length and width of the vortex area. They also showed that the length and width of the vortex area remain constant at diversion ratios greater than 0.7. Karami Moghaddam & Keshavarzi (2007) analyzed the flow characteristics in turnouts with angles of 55 and 90 degrees. They reported that the dimensions of the separation zone decrease by increasing the discharge ratio and reducing the turnout angle with respect to the main channel. Studies about flow separation zone can be found in Jalili et al. (2011)Nikbin & Borghei (2011)Seyedian et al. (2008).

    Jamshidi et al. (2016) measured the dimensions of a flow separation zone in the presence of submerged vanes with five arrangements (parallel, stagger, compound, piney and butterflies). Results showed that the ratio of the width to the length of the separation zone (shape index) was between 0.2 and 0.28 for all arrangements.

    Karami et al. (2017) developed a 3D computational fluid dynamic (CFD) code which was calibrated by measured data. They used the model to evaluate flow pattern, diversion ratio of discharge, strength of the secondary flow, and dimensions of the vortex inside the channel in various dikes and submerged vane installation scenarios. Results showed that the diversion ratio of discharge in the diversion channel is dependent on the width of the flow separation area in the main channel. A dike, perpendicular to the flow, doubles the ratio of diverted discharge and reduces the suspended sediment load compared with the base-line situation by creating outer arch conditions. In addition, increasing the longitudinal distance between vanes increases the velocity gradient between the vanes and leads to a more severe erosion of the bed near the vanes.Figure 1VIEW LARGEDOWNLOAD SLIDE

    Laboratory channel dimensions.

    Al-Zubaidy & Hilo (2021) used the Navier–Stokes equation to study the flow of incompressible fluids. Using the CFD software ANSYS Fluent 19.2, 3D flow patterns were simulated at a diversion channel. Their results showed good agreement using the comparison between the experimental and numerical results when the k-omega turbulence viscous model was employed. Simulation of the flow pattern was then done at the lateral channel junction using a variety of geometry designs. These improvements included changing the intake’s inclination angle and chamfering and rounding the inner corner of the intake mouth instead of the sharp edge. Flow parameters at the diversion including velocity streamlines, bed shear stress, and separation zone dimensions were computed in their study. The findings demonstrated that changing the 90° lateral intake geometry can improve the flow pattern and bed shear stress at the intake junction. Consequently, sedimentation and erosion problems are reduced. According to the conclusions of their study, a branching angle of 30° to 45° is the best configuration for increasing branching channel discharge, lowering branching channel sediment concentration.

    The review of the literature shows that most of the studies deal with turnout angle, discharge ratio and implementation of vanes as techniques to reduce the area of the separation zone. This study examines the effect of roughness coefficient and drop implementation at the entrance of a 90-degree lateral turnout on the dimensions of the separation zone. As far as the authors are aware, these two variables have never been studied as a remedy to decrease the separation zone dimensions whilst enhancing turnout efficiency. Additionally, a three-dimensional numerical model is applied to simulate the flow pattern around the turnout. The numerical results are verified against experimental data.

    METHOD

    Experimental setup

    Listen

    The experiments were conducted in a 90 degree dividing flow laboratory channel. The main channel is 15 m long, 0.5 m wide and 0.4 m high and the branch channel is 3 m long, 0.35 m wide and 0.4 m high, as shown in Figure 1. The tests were carried out at 9.65 m from the beginning of the flume and were far enough from the inlet, so we were sure that the flow was fully developed. According to Kirkgöz & Ardiçlioğlu (1997) the length of the developing region would be approximantly 65 and 72 times the flow depth. In this study, the depth is 9 cm, which makes this condition.

    Both the main and lateral channel had a slope of 0.0003 with side walls of concrete. A 100 hp pump discharged the water into a stilling basin at the entrance of the main flume. The discharge was measured using an ultrasonic discharge meter around the discharge pipe. Eighty-four experiments in total were carried out at range of 0.1<Fr<0.4 (Froude numbers in main channel and upstream of turnout). The depth of water in the main channel in the experiments was 9 cm, in which case the effect of surface tension can be considered; according to research by Zolghadr & Shafai Bejestan (2020) and Zolghadr et al. (2021), when the water depth is more than 6 cm, the effect of surface tension is reduced and can be ignored given that the separation phenomenon occurs in the boundary layer, the height of the roughness creates disturbances in growth and development of the boundary layer and, as a result, separation growth is also faced with disruption and its dimensions grow less compared to smooth surfaces. Similar conditions occur in case of drop implementation. A disturbance occurs in the growth of the boundary layer and as a result the separation zone dimensions decrease. In order to investigate the effect of roughness coefficient and drop implementation on the separation zone dimensions, four different discharges (16, 18, 21, 23 l/s) in subcritical conditions, seven Manning (Strickler) roughness coefficients (0.009, 0.011, 0.017, 0.023, 0.028, 0.030, 0.032) as shown in Figure 2 and three invert elevation differences between the main channel and lateral turnout invert (0, 5 and 10 cm) at the entrance of the turnout were considered. The Manning roughness coefficient values were selected based on available and feasible values for real conditions, so that 0.009 is equivalent to galvanized sheet roughness and selected for the baseline tests. 0.011 is for concrete with neat surface, 0.017 and 0.023 are for unfinished and gunite concrete respectively. 0.030 and 0.032 values are for concrete on irregular excavated rock (Chow 1959). The roughness coefficients were created by gluing sediment particles on a thin galvanized sheet which was installed at the upstream side of the lateral turnout. The values of roughness coefficients were calculated based on the Manning-Strickler formula. For this purpose, some uniformly graded sediment samples were prepared and the Manning roughness coefficient of each sample was determined with respect to the median size (D50) value pasted into the Manning-Strickler formula. Some KMnO4 was sifted in the main channel upstream to visualize and measure the dimensions of the separation zone. Consequently, when KMnO4 approached the lateral turnout a photo of the separation zone was taken from a top view. All the experiments were recorded and several photos were taken during the experiment after stablishment of steady flow conditions. The photos were then imported to AutoCAD to measure the separation zone dimensions. Because all the shooting was done with a high-definition camera and it was possible to zoom in, the results are very accurate.Figure 2VIEW LARGEDOWNLOAD SLIDE

    Roughness plates.

    The velocity values were also recorded by a one-dimensional velocity meter at 15 cm distance from the turnout entrance and in transverse direction (perpendicular to the flow direction).

    The water level was also measured by depth gauges with a accuracy of 0.1 mm, and velocity in one direction with a single-dimensional KENEK LP 1100 with an accuracy of ±0.02 m/s (0–1 m/s), ± 0.04 m/s (1–2 m/s), ± 0.08 m/s (2–4 m/s), ±0.10 m/s (4–5 m/s).

    Numerical simulation

    ListenA FLOW-3D numerical model was utilized as a solver of the Navier-Stokes equation to simulate the three-dimensional flow field at the entrance of the turnout. The governing equations included continuity momentum equations. The continuity equation, regardless of the density of the fluid in the form of Cartesian coordinates x, y, and z, is as follows:

    formula

    (1)where uv, and w represent the velocity components in the x, y, and z directions, respectively; AxAy, and Az are the surface flow fractions in the xy, and z directions, respectively; VF denotes flow volume fraction; r is the density of the fluid; t is time; and Rsor refers to the source of the mass. Equations (2)–(4) show momentum equations in xy and z dimensions respectively :

    formula

    (2)

    formula

    (3)

    formula

    (4)where GxGy, and Gz are the accelerations caused by gravity in the xy, and z directions, respectively; and fxfy, and fz are the accelerations caused by viscosity in the xy, and z directions, respectively.

    The turbulence models used in this study were the renormalized group (RNG) models. Evaluation of the concordance of the mentioned models with experimental studies showed that the RNG model provides more accurate results.

    Two blocks of mesh were used to simulate the main channels and lateral turnout. The meshes were denser in the vicinity of the entrance of the turnout in order to increase the accuracy of computations. Boundary conditions for the main mesh block included inflow for the channel entrance (volumetric flow rate), outflow for the channel exit, ‘wall’ for the bed and the right boundary and ‘symmetry’ for the top (free surface) and left boundaries (turnout). The side wall roughness coefficient was given to the software as the Manning number in surface roughness of any component. Considering the restrictions in the available processor, a main mesh block with appropriate mesh size was defined to simulate the main flow field in the channel, while the nested mesh-block technique was utilized to create a very dense solution field near the roughness plate in order to provide accurate results around the plates and near the entrance of the lateral turnout. This technique reduced the number of required mesh elements by up to 60% in comparison with the method in which the mesh size of the main solution field was decreased to the required extent.

    The numerical outputs are verified against experimental data. The hydraulic characteristics of the experiment are shown in Table 1.Table 1

    Hydraulic conditions of the flow

    Q(L/s)FrY1 (m)Q2/Q1
    16 0.449 0.09 0.22 
    18 0.335 0.09 0.61 
    21 0.242 0.09 0.71 
    23 0.180 0.09 1.04 

    RESULTS AND DISCUSSION

    Experimental results

    Listen

    During the experiments, the dimensions of the separation zone were recorded with an HD camera. Some photos were imported to AutoCad software. Then, the separation zones dimensions were measured and compared in different scenarios.

    At the beginning, the flow pattern in the separation zone for four different hydraulic conditions was studied for seven different Manning roughness coefficients from 0.009 to 0.032. To compare the obtained results, roughness of 0.009 was considered as the base line. The percentage of reduction in separation zone area in different roughness coefficients is shown in Figure 3. According to this figure, by increasing the roughness of the turnout side wall, the separation zone area ratio reduces (ratio of separation zone area to turnout area). In other words, in any desired Froud number, the highest dimensions of the separation zone area are related to the lowest roughness coefficients. In Figure 3, ‘A’ is the area of the separation zone and ‘Ai’ represents the total area of the turnout.Figure 3VIEW LARGEDOWNLOAD SLIDE

    Effect of roughness on separation zone dimensions.Figure 4VIEW LARGEDOWNLOAD SLIDE

    Effect of roughness on separation zone dimensions.

    It should be mentioned that the separation zone dimensions change with depth, so that the area is larger at the surface than near the bed. This study measured the dimensions of this area at the surface. Figure 4 show exactly where the roughness elements were located.Figure 5VIEW LARGEDOWNLOAD SLIDE

    Comparison of separation zone for n=0.023 and n=0.032.

    Figure 5 shows images of the separation zone at n=0.023 and n=0.032 as examples, and show that the separation area at n=0.032 is smaller than that of n=0.023.

    The difference between the effect of the two 0.032 and 0.030 roughnesses is minor. In other words, the dimensions of the separation zone decreased by increasing roughness up to 0.030 and then remained with negligable changes.

    In the next step, the effect of intake invert relative to the main stream (drop) on the dimensions of the separation zone was investigated. To do this, three different invert levels were considered: (1) without drop; (2) a 5 cm drop between the main canal and intake canal; and (3) a 10 cm drop between the main canal and intake canal. The without drop mode was considered as the control state. Figure 6 shows the effect of drop implementation on separation zone dimensions. Tables 2 and 3 show the reduced percentage of separation zone areas in 5 and 10 cm drop compared to no drop conditions as the base line. It was found that the best results were obtained when a 10 cm drop was implemented.Table 2

    Decrease percentage of separation zone area in 5 cm drop

    Frn=0.011n=0.017n=0.023n=0.028n=0.030n=0.032
    0.08 10.56 11.06 25.27 33.03 35.57 36.5 
    0.121 7.66 11.14 11.88 15.93 34.59 36.25 
    0.353 1.38 2.63 8.17 14.39 31.20 31.29 
    0.362 11.54 19.56 25.73 37.89 38.31 

    Table 3

    Decrease percentage of separation zone area in 10 cm drop

    Frn=0.011n=0.017n=0.023n=0.028n=0.030n=0.032
    0.047 4.30 8.75 23.47 31.22 34.96 35.13 
    0.119 11.01 13.16 15.02 21.48 39.45 40.68 
    0.348 3.89 5.71 9.82 16.09 29 30.96 
    0.354 2.84 10.44 18.42 25.45 35.68 35.76 

    Figure 6VIEW LARGEDOWNLOAD SLIDE

    Effect of drop implementation on separation zone dimensions.

    The combined effect of drop and roughness is shown in Figure 7. According to this figure, by installing a drop structure at the entrance of the intake, the dimensions of the separation zone scales down in any desired roughness coefficient. Results indicated that by increasing the roughness coefficient or drop implementation individually, the separation zone area decreases up to 38 and 25% respectively. However, employing both techniques simultaneously can reduce the separation zone area up to 63% (Table 4). The reason for the reduction of the dimensions of the separation zone area by drop implementation can be attributed to the increase of discharge ratio. This reduces the dimensions of the separation zone area.Table 4

    Reduction in percentage of combined effect of roughness and 10 cm drop

    Qin=0.011n=0.017n=0.023n=0.028n=0.030n=0.032
    16 32.3 35.07 37.2 45.7 58.01 59.1 
    18 44.5 34.15 36.18 48.13 54.2 56.18 
    21 43.18 32.33 42.30 37.79 57.16 63.2 
    23 40.56 34.5 34.09 46.25 50.12 57.2 

    Figure 7VIEW LARGEDOWNLOAD SLIDE

    Combined effect of roughness and drop on separation zone dimensions.

    This method increases the discharge ratio (ratio of turnout to main channel discharge). The results are compatible with the literature. Some other researchers reported that increasing the discharge ratio can scale down the separation zone dimensions (Karami Moghaddam & Keshavarzi 2007Ramamurthy et al. 2007). However, these researchers employed other methods to enhance the discharge ratio. Drop implementation is simple and applicable in practice, since there is normally an elevation difference between the main and lateral canal in irrigation networks to ensure gravity flow occurance.

    Table 4 depicts the decrease in percentage of the separation zone compared to base line conditions in different arrangements of the combined tests.Figure 8VIEW LARGEDOWNLOAD SLIDE

    Velocity profiles for various roughness coefficients along turnout width.

    A comparison between the proposed methods introduced in this paper and traditional methods such as installation of submerged vanes, and changing the inlet geometry (angle, radius) was performed. Figure 8 shows the comparison of the results. The comparison shows that the new techniques can be highly influential and still practical. In this research, with no change in structural geometry (enhancement of roughness coefficient) or minor changes with respect to drop implementation, the dimensions of the separation zone are decreased noticeably. The velocity values were also recorded by a one-dimensional velocity meter at 15 cm distance from the turnout entrance and in a transverse direction (perpendicular to the flow direction). The results are shown in Figure 9.Figure 9VIEW LARGEDOWNLOAD SLIDE

    Effect of roughness on separation zone dimensions in numerical study.

    Numerical results

    Listen

    This study examined the flow patterns around the entrance of a diversion channel due to various wall roughnesses in the diversion channel. Results indicated that increasing the discharge ratio in the main channel and diversion channel reduces the area of the separation zone in the diversion channel.Figure 10VIEW LARGEDOWNLOAD SLIDE

    Comparision of the vortex area (software output) for three roughnesses (0.009, 0.023 and 0.032).A laboratory and numerical error rate of 0.2605 was calculated from the following formula,

    formula

    where Uexp is the experimental result, Unum is the numerical result, and N is the number of data.

    Figure 9 shows the effect of roughness on separation zone dimensions in numerical study. Figure 10 compares the vortex area (software output) for three roughnesses, 0.009, 0.023 and 0.032 and Figure 11 shows the flow lines (tecplot output) that indicate the effect of roughness on flow in the separation zone. Numerical analysis shows that by increasing the roughness coefficient, the dimensions of the separation zone area decrease, as shown in Figure 10 where the separation zone area at n=0.032 is less than the separation zone area at n=0.009.Figure 11VIEW LARGEDOWNLOAD SLIDE

    Comparison of vortex area in 3D mode (tecplot output) with two roughnesses (a) 0.009 and (b) 0.032.Figure 12VIEW LARGEDOWNLOAD SLIDE

    Velocity vector for flow condition Q1/422 l/s, near surface.

    The velocities intensified moving midway toward the turnout showing that the effective area is scaled down. The velocity values were almost equal to zero near the side walls as expected. As shown in Figure 12 the approach vortex area velocity decreases. Experimental and numerical measured velocity at x=0.15 m of the diversion channel compared in Figure 13 shows that away from the separation zone area, the velocity increases. All longitudinal velocity contours near the vortex area are distinctly different between different roughnesses. The separation zone is larger at less roughness both in length and width.Figure 13VIEW LARGEDOWNLOAD SLIDE

    Exprimental and numerical measured velocity.

    CONCLUSION

    Listen

    This study introduces practical and feasible methods for enhancing turnout efficiency by reducing the separation zone dimensions. Increasing the roughness coefficient and implementation of inlet drop were considered as remedies for reduction of separation zone dimensions. A data set has been compiled that fully describes the complex, 3D flow conditions present in a 90 degree turnout channel for selected flow conditions. The aim of this numerical model was to compare the results of a laboratory model in the area of the separation zone and velocity. Results showed that enhancing roughness coefficient reduce the separation zone dimensions up to 38% while the drop implementation effect can scale down this area differently based on roughness coefficient used. Combining both methods can reduce the separation zone dimensions up to 63%. Further research is proposed to investigate the effect of roughness and drop implementation on sedimentation pattern at lateral turnouts. The dimensions of the separation zone decreases with the increase of the non-dimensional parameter, due to the reduction ratio of turnout discharge increasing in all the experiments.

    This method increases the discharge ratio (ratio of turnout to main channel discharge). The results are compatible with the literature. Other researchers have reported that intensifying the discharge ratio can scale down the separation zone dimensions (Karami Moghaddam & Keshavarzi 2007Ramamurthy et al. 2007). However, they employed other methods to enhance the discharge ratio. Employing both techniques simultaneously can decrease the separation zone dimensions up to 63%. A comparison between the new methods introduced in this paper and traditional methods such as installation of submerged vanes, and changing the inlet geometry (angle, radius) was performed. The comparison shows that the new techniques can be highly influential and still practical. The numerical and laboratory models are in good agreement and show that the method used in this study has been effective in reducing the separation area. This method is simple, economical and can prevent sediment deposition in the intake canal. Results show that CFD prediction of the fluid through the separation zone at the canal intake can be predicted reasonably well and the RNG model offers the best results in terms of predictability.

    DATA AVAILABILITY STATEMENT

    Listen

    All relevant data are included in the paper or its Supplementary Information.

    REFERENCES

    Abbasi A., Ghodsian M., Habibi M. & Salehi Neishabouri S. A. 2004 Experimental investigation on dimensions of flow separation zone at lateral intakeentrance. Research & Construction; Pajouhesh va Sazandegi 62, 38–44. (In Persian).Google Scholar Al-Zubaidy R. & Hilo A. 2021 Numerical investigation of flow behavior at the lateral intake using Computational Fluid Dynamics (CFD). Materials Today: Proceedings. https://doi.org/10.1016/j.matpr.2021.11.172.Google Scholar Chow V. T. 1959 Open Channel Hydraulics. McGraw-Hill, New York.Jalili H., Hosseinzadeh Dalir A. & Farsadizadeh D. 2011 Effect of intake geometry on the sediment transport and lateral flow pattern. Iranian Water Research Journal 5 (9), 1–10. (In Persian).Google Scholar Jamshidi A., Farsadizadeh D. & Hosseinzadeh Dalir A. 2016 Variations of flow separation zone at lateral intake entrance using submerged vanes. Journal of Civil Engineering Urban 6 (3), 54–63. Journal homepage. Available from: www.ojceu.ir/main.Google Scholar Karami Moghaddam K. & Keshavarzi A. 2007 Investigation of flow structure in lateral intakes of 55° and 90° with rounded entrance edge. In: 03 National Congress on Civil Engineering University of Tabriz. Available from: https://civilica.com/doc/16317. (In Persian).Google Scholar Karami H., Farzin S., Sadrabadi M. T. & Moazeni H. 2017 Simulation of flow pattern at rectangular lateral intake with different dike and submerged vane scenarios. Journal of Water Science and Engineering 10 (3), 246–255. https://doi.org/10.1016/j.wse.2017.10.001.Google ScholarCrossref  Kasthuri B. & Pundarikanthan N. V. 1987 Discussion on separation zone at open- channel junction. Journal of Hydraulic Engineering 113 (4), 543–548.Google ScholarCrossref  Keshavarzi A. & Habibi L. 2005 Optimizing water intake angle by flow separation analysis. Journal of Irrigation and Drain 54, 543–552. https://doi.org/10.1002/ird.207.Google ScholarCrossref  Kirkgöz M. S. & Ardiçlioğlu M. 1997 Velocity profiles of developing and developed open channel flow. Journal of Hydraulic Engineering 1099–1105. 10.1061/(ASCE)0733-9429(1997)123:12(1099).Google Scholar Nakato T., Kennedy J. F. & Bauerly D. 1990 Pumpstation intake-shoaling control with submerge vanes. Journal of Hydraulic Engineering. https://doi.org/10.1061/(ASCE)0733-9429(1990)116:1(119).Google Scholar Neary V. S. & Odgaard J. A. 1993 Three-dimensional flow structure at open channel diversions. Journal of Hydraulic Engineering. ASCE 119 (11), 1224–1230. https://doi.org/10.1061/(ASCE)0733-9429(1993)119:11(1223).Google ScholarCrossref  Nikbin S. & Borghei S. M. 2011 Experimental investigation of submerged vanes effect on dimensions of flow separation zone at a 90° openchannel junction. In: 06rd National Congress on Civil Engineering University of Semnan. (In Persian). Available from: https://civilica.com/doc/120494.Google Scholar Odgaard J. A. & Wang Y. 1991 Sediment management with submerged vanes, I: theory. Journal of Hydraulic Engineering 117 (3), 267–283.Google ScholarCrossref  Ramamurthy A. S., Junying Q. & Diep V. 2007 Numerical and experimental study of dividing open-channel flows. Journal of Hydraulic Engineering. See: https://doi.org/10.1061/(ASCE)0733-9429(2007)133:10(1135).Google Scholar Seyedian S., Karami Moghaddam K. & Shafai Begestan M. 2008 Determining the optimal radius in lateral intakes of 55° and 90° using variation of flow velocity. In: 07th Iranian Hydraulic Conference. Power & Water University of Technology (PWUT). (In Persian). Available from: https://civilica.com/doc/56251.Google Scholar Zolghadr M. & Shafai Bejestan M. 2020 Six legged concrete (SLC) elements as scour countermeasures at wing wall bridge abutments. International Journal of River Basin Management. doi: 10.1080/15715124.2020.1726357.Google Scholar Zolghadr M., Zomorodian S. M. A., Shabani R. & Azamatulla H.Md. 2021 Migration of sand mining pit in rivers: an experimental, numerical and case study. Measurement. https://doi.org/10.1016/j.measurement.2020.108944.Google Scholar © 2022 The AuthorsThis is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).

    Figure 6. Circular section of the viscosity and shear-rate clouds.

    Simulation and Visual Tester Verification of Solid Propellant Slurry Vacuum Plate Casting

    Wu Yue,Li Zhuo,Lu RongFirst published: 26 February 2020 https://doi.org/10.1002/prep.201900411Citations: 3

    Abstract

    Using an improved Carreau constitutive model, a numerical simulation of the casting process of a type of solid propellant slurry vacuum plate casting was carried out using the Flow3D software. Through the flow process in the orifice flow channel and the combustion chamber, the flow velocity of the slurry passing through the plate flow channel was quantitatively analyzed, and the viscosity, shear rate, and leveling characteristics of the slurry in the combustion chamber were qualitatively analyzed and predicted. The pouring time, pouring quality, and flow state predicted by the numerical simulation were verified using a visual tester consisting of a vacuum plate casting system in which a pouring experiment was carried out. Studies have shown that HTPB three-component propellant slurry is a typical yielding pseudoplastic fluid. When the slurry flows through the flower plate and the airfoil, the fluid shear rate reaches its maximum value and the viscosity of the slurry decreases. The visual pouring platform was built and the experiment was controlled according to the numerically-calculated parameters, ensuring the same casting speed. The comparison between the predicted casting quality and the one obtained in the verification test resulted in an error less than 10 %. Moreover, the error between the simulated casting completion time and the process verification test result was also no more than 10 %. Last, the flow state of the slurry during the simulation was consistent with the one during the experimental test. The overall leveling of the slurry in the combustion chamber was adequate and no relatively large holes and flaws developed during the pouring process.

    개선된 Carreau 구성 모델을 사용하여 FLOW-3D 소프트웨어를 사용하여 고체 추진제 슬러리 진공판 유형의 Casting Process에 대한 수치 시뮬레이션을 수행했습니다. 오리피스 유로와 연소실에서의 유동과정을 통해 판 유로를 통과하는 슬러리의 유속을 정량적으로 분석하고, 연소실에서 슬러리의 점도, 전단율, 레벨링 특성을 정성적으로 분석하하고, 예측하였습니다.

    타설시간, 타설품질, 수치해석으로 예측된 ​​유동상태는 타설실험을 수행한 진공판주조시스템으로 구성된 비주얼 테스터를 이용하여 검증하였습니다.

    연구에 따르면 HTPB 3성분 추진제 슬러리는 전형적인 생성 가소성 유체입니다. 슬러리가 플라워 플레이트와 에어포일을 통과할 때 유체 전단율이 최대값에 도달하고 슬러리의 점도가 감소합니다.

    시각적 주입 플랫폼이 구축되었고 동일한 주조 속도를 보장하기 위해 수치적으로 계산된 매개변수에 따라 실험이 제어되었습니다. 예측된 casting 품질과 검증 테스트에서 얻은 품질을 비교한 결과 10 % 미만의 오류가 발생했습니다.

    또한 모의 casting 완료시간과 공정검증시험 결과의 오차도 10 % 이하로 나타났습니다.

    마지막으로 시뮬레이션 중 슬러리의 흐름 상태는 실험 테스트 시와 일치하였다. 연소실에서 슬러리의 전체 레벨링은 적절했으며 주입 과정에서 상대적으로 큰 구멍과 결함이 발생하지 않았습니다.

    Figure 1. The equipment used in the vacuum flower-plate pouring process.
    Figure 1. The equipment used in the vacuum flower-plate pouring process.
    Figure 2. Calculation model.
    Figure 2. Calculation model.
    Figure 3. Grid block division unit.
    Figure 3. Grid block division unit.
    Figure 4. Circular section of the speed cloud.
    Figure 4. Circular section of the speed cloud.
    Figure 5. Viscosity and shear rate distribution cloud pattern flowing through the plate holes.
    Figure 5. Viscosity and shear rate distribution cloud pattern flowing through the plate holes.
    Figure 6. Circular section of the viscosity and shear-rate clouds.
    Figure 6. Circular section of the viscosity and shear-rate clouds.
    Figure 7. Volume fraction cloud chart at different time.
    Figure 7. Volume fraction cloud chart at different time.
    Figure 8. Experimental program.
    Figure 8. Experimental program.
    Figure 9. Emulation experimental device.
    Figure 9. Emulation experimental device.
    Figure 10. Visualization of the flow state of the pulp inside the tester.
    Figure 10. Visualization of the flow state of the pulp inside the tester.

    References

    [1] B. M. Bandgar, V. N. Krishnamurthy, T. Mukundan, K. C. Sharma,
    Mathematical Modeling of Rheological Properties of HydroxylTerminated Polybutadiene Binder and Dioctyl Adipate Plasticizer, J. Appl. Polym. Sci. 2002, 85, 1002–1007.
    [2] B. Thiyyarkandy, M. Jain, G. S. Dombe, M. Mehilal, P. P. Singh, B.
    Bhattacharya, Numerical Studies on Flow Behavior of Composite Propellant Slurry during Vacuum Casting, J.Aerosp.Technol.
    Manage. 2012, 4, 197–203.
    [3] T. Shimada, H. Habu, Y. Seike, S. Ooya, H. Miyachi, M. Ishikawa,
    X-Ray Visualization Measurement of Slurry Flow in Solid Propellant Casting, Flow Meas. Instrum. 2007, 18, 235–240.
    [4] Y. Damianou, G. C. Georgiou, On Poiseuille Flows of a Bingham
    Plastic with Pressure-Dependent Rheological Parameters, J.
    Non-Newtonian Fluid Mech. 2017, 250, 1–7.
    [5] S. Sadasivan, S. K. Arumugam, M. Aggarwal, Numerical Simulation of Diffuser of a Gas Turbine using the Actuator Disc
    Model, J.Appl. Fluid Mech. 2019, 12, 77–84.
    [6] M. Acosta, V. L. Wiesner, C. J. Martinez, R. W. Trice, J. P. Youngblood, Effect of Polyvinylpyrrolidone Additions on the Rheology of Aqueous, Highly Loaded Alumina Suspensions, J. Am.
    Ceram. Soc. 2013, 96, 1372–1382.
    [7] Y. Wu, Numerical Simulation and Experiment Study of Flower
    Plate Pouring System for Solid Propellant, Chin. J. Expl. Propell.
    2017, 41, 506–511.
    [8] T. M. G. Chu, J. W. Halloran, High-Temperature Flow Behavior
    of Ceramic Suspensions, J. Am. Ceram. Soc. 2004, 83, 2189–
    2195.
    [9] T. Kaully, A. Siegmann, D. Shacham, Rheology of Highly Filled
    Natural CaCO3 Composites. I. Effects of Solid Loading and Particle Size Distribution on Capillary Rheometry, Polym. Compos.
    2007, 28, 512–523.
    [10] M. M. Rueda, M.-C. Auscher, R. Fulchiron, T. Périé, G. Martin, P.
    Sonntag, P. Cassagnau, Rheology and Applications of Highly
    Filled Polymers: A Review of Current Understanding, Prog. Polym. Sci. 2017, 66, 22–53.
    [11] F. Soltani, Ü. Yilmazer, Slip Velocity and Slip Layer Thickness in
    Flow of Concentrated Suspensions, J. Appl. Polym. Sci. 1998,
    70, 515–522.

    [12] E. Landsem, T. L. Jensen, F. K. Hansen. E. Unneberg, T. E. Kristensen, Neutral Polymeric Bonding Agents (NPBA) and Their
    Use in Smokeless Composite Rocket Propellants Based on
    HMX-GAP-BuNENA. Propellants, Explos., Pyrotech.. 2012, 37,
    581–589.
    [13] J. Mewis, N. J. Wagner, Colloidal Suspension Rheology, Cambridge University Press, 2011.
    [14] D. M. Kalyon, An Overview of the Rheological Behavior and
    Characterization of Energetic Formulations: Ramifications on
    Safety and Product Quality, J. Energ. Mater. 2006, 24, 213–245.
    [15] H. Ohshima, Effective Viscosity of a Concentrated Suspension
    of Uncharged Spherical Soft Particles, Langmuir 2010, 26,
    6287–6294.

    Figure 7. The simulated velocity (a) and simulated pressure pattern (b) across the Parshall flume. The patterns match the physical behavior of actual Parshall flumes [7].

    Application of Numerical and Experimental Modeling to Improve the Efficiency of Parshall Flumes: A Review of the State-of-the-Art

    Parshall Flumes의 효율성 향상을 위한 수치 및 실험 모델링의 적용: 최신 기술 검토

    Mehdi Heyrani 1,* , Abdolmajid Mohammadian 1, Ioan Nistor 1 and Omerul Faruk Dursun 2

    Abstract

    열린 채널에서 흐름을 관리하는 기본 단계 중 하나는 속성을 결정하는 것입니다. 개방 수로의 흐름에 관한 추가 정보를 제공하기 위해 경험적 방정식이 개발되었습니다. 이러한 실험 방정식을 얻는 것은 비용과 시간이 많이 소요됩니다. 따라서 대체 솔루션이 모색되었습니다.

    지난 세기 동안 움직이는 부분이 없는 정적 측정 장치인 Parshall 수로가 개방 수로의 흐름을 측정하는 데 중요한 역할을 했습니다. 많은 연구자들이 관개 및 폐수 관리와 같은 다양한 분야에서 Parshall 수로의 적용을 연구하는 데 관심을 집중해 왔습니다.

    여러 학자들이 실험 결과를 사용하여 Parshall 수로의 등급 방정식을 향상시켰지만 다른 학자들은 수치 시뮬레이션을 사용하여 높이-방전 관계 방정식을 재보정하기 위해 대체 데이터 소스를 사용했습니다. 컴퓨팅 하드웨어가 지난 수십 년 동안 크게 발전하여 과거에 경험했던 제한된 해상도를 뛰어넘는 것이 가능해짐에 따라 CFD(Computational Fluid Dynamic) 소프트웨어가 오늘날 대중화되고 있습니다.

    여러 CFD 모델은 가용성에 따라 오픈 소스 또는 상업적으로 허가되어 수위 결과를 생성하기 위해 다양한 구성의 수로, 특히 Parshall 수로에 대한 수치 시뮬레이션을 수행하는 데 사용되었습니다.

    FLOW-3D, Ansys Fluent, OpenFOAM 등 지금까지 사용되어 온 다양한 CFD 도구에 대해 실험 데이터로 정밀 교정한 결과, 출력이 안정적이고 실제 시나리오에 구현할 수 있음이 확인되었습니다.

    결과를 생성하기 위해 이 기술을 사용하는 이점은 필요한 경우 유속 또는 구조적 형상과 같은 초기 조건을 조정하는 CFD 접근 방식의 능력입니다. 수로 크기와 수로가 위치한 부지의 조건과 관련하여 상황에 적합한 특정 Parshall 수로로 선택이 좁혀집니다.

    표준 Parshall 수로를 선택하는 것이 항상 가능한 것은 아닙니다. 따라서 엔지니어는 가장 가까운 수로 크기에 약간의 수정을 제공하고 정확한 유량을 생성하기 위해 새로운 등급 곡선을 제공합니다.

    이 검토는 기존 등급 방정식을 향상시키거나 구조의 기하학에 대한 추가 수정을 제안하기 위해 Parshall 수로에서 수치 시뮬레이션 및 물리적 실험 데이터의 적용을 목표로 하는 여러 학자의 작업에 대해 수행되었습니다.

    One of the primary steps in managing the flow in an open channel is determining its properties. Empirical equations are developed to provide further information regarding the flow in open channels. Obtaining such experimental equations is expensive and time consuming; therefore, alternative solutions have been sought. Over the last century, the Parshall flume, a static measuring device with no moving parts, has played a significant role in measuring the flow in open channels. Many researchers have focused their interest on studying the application of Parshall flumes in various fields like irrigation and wastewater management. Although various scholars used experimental results to enhance the rating equation of the Parshall flume, others used an alternative source of data to recalibrate the height–discharge relation equation using numerical simulation. Computational Fluid Dynamic (CFD) software is becoming popular nowadays as computing hardware has advanced significantly within the last few decades, making it possible to go beyond the limited resolution that was experienced in the past. Multiple CFD models, depending on their availability, either open-source or commercially licensed, have been used to perform numerical simulations on different configurations of flumes, especially Parshall flumes, to produce water level results. Regarding various CFD tools that have been used, i.e., FLOW-3D, Ansys Fluent, or OpenFOAM, after precise calibration with experimental data, it has been determined that the output is reliable and can be implemented to the actual scenarios. The benefit of using this technique to produce results is the ability of the CFD approach to adjust the initial conditions, like flow velocity or structural geometry, where necessary. With respect to channel size and the condition of the site where the flume is located, the choices are narrowed to the specific Parshall flume suitable to the situation. It is not always possible to select the standard Parshall flume; therefore, engineers provide some modification to the closest flume size and provide a new rating curve to produce accurate flowrates. This review has been performed on the works of a number of scholars who targeted the application of numerical simulation and physical experimental data in Parshall flumes to either enhance the existing rating equation or propose further modification to the structure’s geometry.

    Keywords

    Parshall flume; CFD; OpenFOAM; FLOW-3D; numerical simulation; turbulence model

    Figure 1. Parshall flume measuring structure, installed [2].
    Figure 1. Parshall flume measuring structure, installed [2].
    Figure 2. Parshall flume measuring structure, uninstalled [3]
    Figure 2. Parshall flume measuring structure, uninstalled [3]
    Figure 4. Mesh sensitivity analysis: top view and side view of the Parshall flume: (a) contains 27,000 cells; (b) 52,000 cells; (c) 75,000 cells; (d) 270,000 cells. The C setup was used in their simulation [7].
    Figure 4. Mesh sensitivity analysis: top view and side view of the Parshall flume: (a) contains 27,000 cells; (b) 52,000 cells; (c) 75,000 cells; (d) 270,000 cells. The C setup was used in their simulation [7].
    Figure 7. The simulated velocity (a) and simulated pressure pattern (b) across the Parshall flume. The patterns match the physical behavior of actual Parshall flumes [7].
    Figure 7. The simulated velocity (a) and simulated pressure pattern (b) across the Parshall flume. The patterns match the physical behavior of actual Parshall flumes [7].
    Figure 8. Computational grid system in the Side A flume. (a) contains a triangular grid system (b) demonstrates the rectangular grid system. (c) and (d) are three-dimensional schematics showing the superimposed grid system. (e) magnifies the dashed section in (b). (Reprinted with permission from Ref. [11]. 2020 ELSEVIER). ).
    Figure 8. Computational grid system in the Side A flume. (a) contains a triangular grid system (b) demonstrates the rectangular grid system. (c) and (d) are three-dimensional schematics showing the superimposed grid system. (e) magnifies the dashed section in (b). (Reprinted with permission from Ref. [11]. 2020 ELSEVIER). ).
    Figure 10. The results of flow patterns in different flumes; (a) Cutthroat flume, (b) airfoil-shaped flume, (c) airfoil pillar-shaped flume, (d) optimized airfoil-shaped flume [23]
    Figure 10. The results of flow patterns in different flumes; (a) Cutthroat flume, (b) airfoil-shaped flume, (c) airfoil pillar-shaped flume, (d) optimized airfoil-shaped flume [23]
    Figure 11. Experimental setup: contraction ratio used on each flume [23].
    Figure 11. Experimental setup: contraction ratio used on each flume [23].
    Figure 12. Entire flume geometry [25]
    Figure 12. Entire flume geometry [25]

    References

    1. Cone, V.M. The Venturi Flume; U.S. Government Printing Office: Washington, DC, USA, 1917.
    2. 20-Foot Concrete Parshall Flume with Radius Wing Walls. Available online: https://www.openchannelflow.com/assets/uploads/
      media/_large/20-foot-parshall-flume-curved-wing-walls.jpg (accessed on 12 January 2021).
    3. Fiberglass 6-Inch Parshall Flume with Gauge. Available online: https://www.openchannelflow.com/assets/uploads/media/
      _large/flume-parshall-6-inch-fiberglass.png (accessed on 12 January 2021).
    4. Parshall, R.L. The Parshall Measuring Flume; Colorado State College, Colorado Experiment Station: Fort Collins, CO, USA, 1936.
    5. Selecting Between a Weir and a Flume. 2022. Available online: https://www.openchannelflow.com/blog/selecting-a-primarydevice-part-1-choosing-between-a-weir-and-a-flume (accessed on 29 December 2021).
    6. Parshall, R.L. The Improved Venturi Flume. Trans. Am. Soc. Civ. Eng. 1928, 89, 841–851. [CrossRef]
    7. Heyrani, M.; Mohammadian, A.; Nistor, I. Numerical Simulation of Flow in Parshall Flume Using Selected Nonlinear Turbulence
      Models. Hydrology 2021, 8, 151. [CrossRef]
    8. Heyrani, M.; Mohammadian, A.; Nistor, I.; Dursun, O.F. Numerical Modeling of Venturi Flume. Hydrology 2021, 8, 27. [CrossRef]
    9. Alfonsi, G. Reynolds-Averaged Navier–Stokes Equations for Turbulence Modeling. Appl. Mech. Rev. 2009, 62, 040802. [CrossRef]
    10. Imanian, H.; Mohammadian, A. Numerical Simulation of Flow over Ogee Crested Spillways under High Hydraulic Head Ratio.
      Eng. Appl. Comput. Fluid Mech. 2019, 13, 983–1000. [CrossRef]
    11. Khosronejad, A.; Herb, W.; Sotiropoulos, F.; Kang, S.; Yang, X. Assessment of Parshall Flumes for Discharge Measurement of
      Open-Channel Flows: A Comparative Numerical and Field Case Study. Measurement 2020, 167, 108292. [CrossRef]
    12. Dursun, O.F. An Experimental Investigation of the Aeration Performance of Parshall Flume and Venturi Flumes. KSCE J. Civ. Eng.
      2016, 20, 943–950. [CrossRef]
    13. Shih, T.-H.; Liu, N.-S.; Chen, K.-H. A Non-Linear k-Epsilon Model for Turbulent Shear Flows. In Proceedings of the 34th
      AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Cleveland, OH, USA, 13 July 1998; p. 3983.
    14. Lien, F.S. Low-Reynolds-Number Eddy-Viscosity Modelling Based on Non-Linear Stress-Strain/Vorticity Relations. In Proceedings of the 3rd Symposium on Engineering Turbulence Modelling and Measurement, Heraklion, Greece, 27 May 1996.
    15. Davis, R.W.; Deutsch, S. A Numerical-Experimental Study of Parhall Flumes. J. Hydraul. Res. 1980, 18, 135–152. [CrossRef]
    16. Xiao, Y.; Wang, W.; Hu, X.; Zhou, Y. Experimental and Numerical Research on Portable Short-Throat Flume in the Field. Flow
      Meas. Instrum. 2016, 47, 54–61. [CrossRef]
    17. Wright, S.J.; Tullis, B.P.; Long, T.M. Recalibration of Parshall Flumes at Low Discharges. J. Irrig. Drain. Eng. 1994, 120, 348–362.
      [CrossRef]
    18. Heiner, B.; Barfuss, S.L. Parshall Flume Discharge Corrections: Wall Staff Gauge and Centerline Measurements. J. Irrig. Drain.
      Eng. 2011, 137, 779–792. [CrossRef]
    19. Savage, B.M.; Heiner, B.; Barfuss, S. Parshall Flume Discharge Correction Coefficients through Modelling. Proc. ICE Water Manag.
      2013, 167, 279–287. [CrossRef]
    20. Zerihun, Y.T. A Numerical Study on Curvilinear Free Surface Flows in Venturi Flumes. Fluids 2016, 1, 21. [CrossRef]
    21. Sun, B.; Zhu, S.; Yang, L.; Liu, Q.; Zhang, C.; Zhang, J. ping Experimental and Numerical Investigation of Flow Measurement
      Mechanism and Hydraulic Performance on Curved Flume in Rectangular Channel. Arab. J. Sci. Eng. 2020. [CrossRef]
    22. Hu, H.; Huang, J.; Qian, Z.; Huai, W.; Yu, G. Hydraulic Analysis of Parabolic Flume for Flow Measurement. Flow Meas. Instrum.
      2014, 37, 54–64. [CrossRef]
    23. Sun, B.; Yang, L.; Zhu, S.; Liu, Q.; Wang, C.; Zhang, C. Study on the Applicability of Four Flumes in Small Rectangular Channels.
      Flow Meas. Instrum. 2021, 80, 101967. [CrossRef]
    24. Willeitner, R.P.; Barfuss, S.L.; Johnson, M.C. Using Numerical Modeling to Correct Flow Rates for Submerged Montana Flumes. J.
      Irrig. Drain. Eng. 2013, 139, 586–592. [CrossRef]
    25. Ran, D.; Wang, W.; Hu, X. Three-Dimensional Numerical Simulation of Flow in Trapezoidal Cutthroat Flumes Based on FLOW-3D.
      Front. Agric. Sci. Eng. 2018, 5, 168–176. [CrossRef]
    26. Kim, S.-Y.; Lee, J.-H.; Hong, N.-K.; Lee, S.-O. Numerical Simulation for Determining Scale of Parshall Flume. Proc. Korea Water
      Resour. Assoc. Conf. 2010, 719–723.
    27. Tekade, S.A.; Vasudeo, A.D.; Ghare, A.D.; Ingle, R.N. Measurement of Flow in Supercritical Flow Regime Using Cutthroat Flumes.
      Sadhana 2016, 41, 265–272. [CrossRef]
    28. Wahl, T.L.; Replogle, J.A.; Wahlin, B.T.; Higgs, J.A. New Developments in Design and Application of Long-Throated Flumes. In
      Proceedings of the Joint Conference on Water Resource Engineering and Water Resources Planning and Management, Minneapolis,
      MN, USA, 30 July–2 August 2000.
    29. Howes, D.J.; Burt, C.M.; Sanders, B.F. Subcritical Contraction for Improved Open-Channel Flow Measurement Accuracy with an
      Upward-Looking ADVM. J. Irrig. Drain. Eng. 2010, 136, 617–626. [CrossRef]
    30. Tiwari, N.K.; Sihag, P. Prediction of Oxygen Transfer at Modified Parshall Flumes Using Regression Models. ISH J. Hydraul. Eng.
      2020, 26, 209–220. [CrossRef]
    31. Thornton, C.I.; Smith, B.A.; Abt, S.R.; Robeson, M.D. Supercritical Flow Measurement Using a Small Parshall Flume. J. Irrig.
      Drain. Eng. 2009, 135, 683–692. [CrossRef]
    32. Cox, A.L.; Thornton, C.I.; Abt, S.R. Supercritical Flow Measurement Using a Large Parshall Flume. J. Irrig. Drain. Eng. 2013, 139,
      655–662. [CrossRef]
    1. Ribeiro, Á.S.; Sousa, J.A.; Simões, C.; Martins, L.L.; Dias, L.; Mendes, R.; Martins, C. Parshall Flumes Flow Rate Uncertainty
      Including Contributions of the Model Parameters and Correlation Effects. Meas. Sens. 2021, 18, 100108. [CrossRef]
    2. Singh, J.; Mittal, S.K.; Tiwari, H.L. Discharge Relation for Small Parshall Flume in Free Flow Condition. Int. J. Res. Eng. Technol.
      2014, 3, 317–321.
    3. Kim, S.-D.; Lee, H.-J.; Oh, B.-D. Investigation on Application of Parshall Flume for Flow Measurement of Low-Flow Season in
      Korea. Meas. Sci. Rev. 2010, 10, 111. [CrossRef]
    4. Willeitner, R.P.; Barfuss, S.L.; Johnson, M.C. Montana Flume Flow Corrections under Submerged Flow. J. Irrig. Drain. Eng. 2012,
      138, 685–689. [CrossRef]
    5. Dufresne, M.; Vazquez, J. Head–Discharge Relationship of Venturi Flumes: From Long to Short Throats. J. Hydraul. Res. 2013, 51,
      465–468. [CrossRef]
    Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

    다공성 미디어 및 나노유체에 의해 강화된 수집기로 태양광 CCHP 시스템의 최적화

    Optimization of Solar CCHP Systems with Collector Enhanced by Porous Media and Nanofluid


    Navid Tonekaboni,1Mahdi Feizbahr,2 Nima Tonekaboni,1Guang-Jun Jiang,3,4 and Hong-Xia Chen3,4

    Abstract

    태양열 집열기의 낮은 효율은 CCHP(Solar Combined Cooling, Heating, and Power) 사이클의 문제점 중 하나로 언급될 수 있습니다. 태양계를 개선하기 위해 나노유체와 다공성 매체가 태양열 집열기에 사용됩니다.

    다공성 매질과 나노입자를 사용하는 장점 중 하나는 동일한 조건에서 더 많은 에너지를 흡수할 수 있다는 것입니다. 이 연구에서는 평균 일사량이 1b인 따뜻하고 건조한 지역의 600 m2 건물의 전기, 냉방 및 난방을 생성하기 위해 다공성 매질과 나노유체를 사용하여 태양열 냉난방 복합 발전(SCCHP) 시스템을 최적화했습니다.

    본 논문에서는 침전물이 형성되지 않는 lb = 820 w/m2(이란) 정도까지 다공성 물질에서 나노유체의 최적량을 계산하였다. 이 연구에서 태양열 집열기는 구리 다공성 매체(95% 다공성)와 CuO 및 Al2O3 나노 유체로 향상되었습니다.

    나노유체의 0.1%-0.6%가 작동 유체로 물에 추가되었습니다. 나노유체의 0.5%가 태양열 집열기 및 SCCHP 시스템에서 가장 높은 에너지 및 엑서지 효율 향상으로 이어지는 것으로 밝혀졌습니다.

    본 연구에서 포물선형 집열기(PTC)의 최대 에너지 및 엑서지 효율은 각각 74.19% 및 32.6%입니다. 그림 1은 태양 CCHP의 주기를 정확하게 설명하기 위한 그래픽 초록으로 언급될 수 있습니다.

    The low efficiency of solar collectors can be mentioned as one of the problems in solar combined cooling, heating, and power (CCHP) cycles. For improving solar systems, nanofluid and porous media are used in solar collectors. One of the advantages of using porous media and nanoparticles is to absorb more energy under the same conditions. In this research, a solar combined cooling, heating, and power (SCCHP) system has been optimized by porous media and nanofluid for generating electricity, cooling, and heating of a 600 m2 building in a warm and dry region with average solar radiation of Ib = 820 w/m2 in Iran. In this paper, the optimal amount of nanofluid in porous materials has been calculated to the extent that no sediment is formed. In this study, solar collectors were enhanced with copper porous media (95% porosity) and CuO and Al2O3 nanofluids. 0.1%–0.6% of the nanofluids were added to water as working fluids; it is found that 0.5% of the nanofluids lead to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Maximum energy and exergy efficiency of parabolic thermal collector (PTC) riches in this study are 74.19% and 32.6%, respectively. Figure 1 can be mentioned as a graphical abstract for accurately describing the cycle of solar CCHP.

    1. Introduction

    Due to the increase in energy consumption, the use of clean energy is one of the important goals of human societies. In the last four decades, the use of cogeneration cycles has increased significantly due to high efficiency. Among clean energy, the use of solar energy has become more popular due to its greater availability [1]. Low efficiency of energy production, transmission, and distribution system makes a new system to generate simultaneously electricity, heating, and cooling as an essential solution to be widely used. The low efficiency of the electricity generation, transmission, and distribution system makes the CCHP system a basic solution to eliminate waste of energy. CCHP system consists of a prime mover (PM), a power generator, a heat recovery system (produce extra heating/cooling/power), and thermal energy storage (TES) [2]. Solar combined cooling, heating, and power (SCCHP) has been started three decades ago. SCCHP is a system that receives its propulsive force from solar energy; in this cycle, solar collectors play the role of propulsive for generating power in this system [3].

    Increasing the rate of energy consumption in the whole world because of the low efficiency of energy production, transmission, and distribution system causes a new cogeneration system to generate electricity, heating, and cooling energy as an essential solution to be widely used. Building energy utilization fundamentally includes power required for lighting, home electrical appliances, warming and cooling of building inside, and boiling water. Domestic usage contributes to an average of 35% of the world’s total energy consumption [4].

    Due to the availability of solar energy in all areas, solar collectors can be used to obtain the propulsive power required for the CCHP cycle. Solar energy is the main source of energy in renewable applications. For selecting a suitable area to use solar collectors, annual sunshine hours, the number of sunny days, minus temperature and frosty days, and the windy status of the region are essentially considered [5]. Iran, with an average of more than 300 sunny days, is one of the suitable countries to use solar energy. Due to the fact that most of the solar radiation is in the southern regions of Iran, also the concentration of cities is low in these areas, and transmission lines are far apart, one of the best options is to use CCHP cycles based on solar collectors [6]. One of the major problems of solar collectors is their low efficiency [7]. Low efficiency increases the area of collectors, which increases the initial cost of solar systems and of course increases the initial payback period. To increase the efficiency of solar collectors and improve their performance, porous materials and nanofluids are used to increase their workability.

    There are two ways to increase the efficiency of solar collectors and mechanical and fluid improvement. In the first method, using porous materials or helical filaments inside the collector pipes causes turbulence of the flow and increases heat transfer. In the second method, using nanofluids or salt and other materials increases the heat transfer of water. The use of porous materials has grown up immensely over the past twenty years. Porous materials, especially copper porous foam, are widely used in solar collectors. Due to the high contact surface area, porous media are appropriate candidates for solar collectors [8]. A number of researchers investigated Solar System performance in accordance with energy and exergy analyses. Zhai et al. [9] reviewed the performance of a small solar-powered system in which the energy efficiency was 44.7% and the electrical efficiency was 16.9%.

    Abbasi et al. [10] proposed an innovative multiobjective optimization to optimize the design of a cogeneration system. Results showed the CCHP system based on an internal diesel combustion engine was the applicable alternative at all regions with different climates. The diesel engine can supply the electrical requirement of 31.0% and heating demand of 3.8% for building.

    Jiang et al. [11] combined the experiment and simulation together to analyze the performance of a cogeneration system. Moreover, some research focused on CCHP systems using solar energy. It integrated sustainable and renewable technologies in the CCHP, like PV, Stirling engine, and parabolic trough collector (PTC) [21215].

    Wang et al. [16] optimized a cogeneration solar cooling system with a Rankine cycle and ejector to reach the maximum total system efficiency of 55.9%. Jing et al. analyzed a big-scale building with the SCCHP system and auxiliary heaters to produced electrical, cooling, and heating power. The maximum energy efficiency reported in their work is 46.6% [17]. Various optimization methods have been used to improve the cogeneration system, minimum system size, and performance, such as genetic algorithm [1819].

    Hirasawa et al. [20] investigated the effect of using porous media to reduce thermal waste in solar systems. They used the high-porosity metal foam on top of the flat plate solar collector and observed that thermal waste decreased by 7% due to natural heat transfer. Many researchers study the efficiency improvement of the solar collector by changing the collector’s shapes or working fluids. However, the most effective method is the use of nanofluids in the solar collector as working fluid [21]. In the experimental study done by Jouybari et al. [22], the efficiency enhancement up to 8.1% was achieved by adding nanofluid in a flat plate collector. In this research, by adding porous materials to the solar collector, collector efficiency increased up to 92% in a low flow regime. Subramani et al. [23] analyzed the thermal performance of the parabolic solar collector with Al2O3 nanofluid. They conducted their experiments with Reynolds number range 2401 to 7202 and mass flow rate 0.0083 to 0.05 kg/s. The maximum efficiency improvement in this experiment was 56% at 0.05 kg/s mass flow rate.

    Shojaeizadeh et al. [24] investigated the analysis of the second law of thermodynamic on the flat plate solar collector using Al2O3/water nanofluid. Their research showed that energy efficiency rose up to 1.9% and the exergy efficiency increased by a maximum of 0.72% compared to pure water. Tiwari et al. [25] researched on the thermal performance of solar flat plate collectors for working fluid water with different nanofluids. The result showed that using 1.5% (optimum) particle volume fraction of Al2O3 nanofluid as an absorbing medium causes the thermal efficiency to enhance up to 31.64%.

    The effect of porous media and nanofluids on solar collectors has already been investigated in the literature but the SCCHP system with a collector embedded by both porous media and nanofluid for enhancing the ratio of nanoparticle in nanofluid for preventing sedimentation was not discussed. In this research, the amount of energy and exergy of the solar CCHP cycles with parabolic solar collectors in both base and improved modes with a porous material (copper foam with 95% porosity) and nanofluid with different ratios of nanoparticles was calculated. In the first step, it is planned to design a CCHP system based on the required load, and, in the next step, it will analyze the energy and exergy of the system in a basic and optimize mode. In the optimize mode, enhanced solar collectors with porous material and nanofluid in different ratios (0.1%–0.7%) were used to optimize the ratio of nanofluids to prevent sedimentation.

    2. Cycle Description

    CCHP is one of the methods to enhance energy efficiency and reduce energy loss and costs. The SCCHP system used a solar collector as a prime mover of the cogeneration system and assisted the boiler to generate vapor for the turbine. Hot water flows from the expander to the absorption chiller in summer or to the radiator or fan coil in winter. Finally, before the hot water wants to flow back to the storage tank, it flows inside a heat exchanger for generating domestic hot water [26].

    For designing of solar cogeneration system and its analysis, it is necessary to calculate the electrical, heating (heating load is the load required for the production of warm water and space heating), and cooling load required for the case study considered in a residential building with an area of 600 m2 in the warm region of Iran (Zahedan). In Table 1, the average of the required loads is shown for the different months of a year (average of electrical, heating, and cooling load calculated with CARRIER software).Table 1 The average amount of electric charges, heating load, and cooling load used in the different months of the year in the city of Zahedan for a residential building with 600 m2.

    According to Table 1, the maximum magnitude of heating, cooling, and electrical loads is used to calculate the cogeneration system. The maximum electric load is 96 kW, the maximum amount of heating load is 62 kW, and the maximum cooling load is 118 kW. Since the calculated loads are average, all loads increased up to 10% for the confidence coefficient. With the obtained values, the solar collector area and other cogeneration system components are calculated. The cogeneration cycle is capable of producing 105 kW electric power, 140 kW cooling capacity, and 100 kW heating power.

    2.1. System Analysis Equations

    An analysis is done by considering the following assumptions:(1)The system operates under steady-state conditions(2)The system is designed for the warm region of Iran (Zahedan) with average solar radiation Ib = 820 w/m2(3)The pressure drops in heat exchangers, separators, storage tanks, and pipes are ignored(4)The pressure drop is negligible in all processes and no expectable chemical reactions occurred in the processes(5)Potential, kinetic, and chemical exergy are not considered due to their insignificance(6)Pumps have been discontinued due to insignificance throughout the process(7)All components are assumed adiabatic

    Schematic shape of the cogeneration cycle is shown in Figure 1 and all data are given in Table 2.

    Figure 1 Schematic shape of the cogeneration cycle.Table 2 Temperature and humidity of different points of system.

    Based on the first law of thermodynamic, energy analysis is based on the following steps.

    First of all, the estimated solar radiation energy on collector has been calculated:where α is the heat transfer enhancement coefficient based on porous materials added to the collector’s pipes. The coefficient α is increased by the porosity percentage, the type of porous material (in this case, copper with a porosity percentage of 95), and the flow of fluid to the collector equation.

    Collector efficiency is going to be calculated by the following equation [9]:

    Total energy received by the collector is given by [9]

    Also, the auxiliary boiler heat load is [2]

    Energy consumed from vapor to expander is calculated by [2]

    The power output form by the screw expander [9]:

    The efficiency of the expander is 80% in this case [11].

    In this step, cooling and heating loads were calculated and then, the required heating load to reach sanitary hot water will be calculated as follows:

    First step: calculating the cooling load with the following equation [9]:

    Second step: calculating heating loads [9]:

    Then, calculating the required loud for sanitary hot water will be [9]

    According to the above-mentioned equations, efficiency is [9]

    In the third step, calculated exergy analysis as follows.

    First, the received exergy collector from the sun is calculated [9]:

    In the previous equation, f is the constant of air dilution.

    The received exergy from the collector is [9]

    In the case of using natural gas in an auxiliary heater, the gas exergy is calculated from the following equation [12]:

    Delivering exergy from vapor to expander is calculated with the following equation [9]:

    In the fourth step, the exergy in cooling and heating is calculated by the following equation:

    Cooling exergy in summer is calculated [9]:

    Heating exergy in winter is calculated [9]:

    In the last step based on thermodynamic second law, exergy efficiency has been calculated from the following equation and the above-mentioned calculated loads [9]:

    3. Porous Media

    The porous medium that filled the test section is copper foam with a porosity of 95%. The foams are determined in Figure 2 and also detailed thermophysical parameters and dimensions are shown in Table 3.

    Figure 2 Copper foam with a porosity of 95%.Table 3 Thermophysical parameters and dimensions of copper foam.

    In solar collectors, copper porous materials are suitable for use at low temperatures and have an easier and faster manufacturing process than ceramic porous materials. Due to the high coefficient conductivity of copper, the use of copper metallic foam to increase heat transfer is certainly more efficient in solar collectors.

    Porous media and nanofluid in solar collector’s pipes were simulated in FLOW-3D software using the finite-difference method [27]. Nanoparticles Al2O3 and CUO are mostly used in solar collector enhancement. In this research, different concentrations of nanofluid are added to the parabolic solar collectors with porous materials (copper foam with porosity of 95%) to achieve maximum heat transfer in the porous materials before sedimentation. After analyzing PTC pipes with the nanofluid flow in FLOW-3D software, for energy and exergy efficiency analysis, Carrier software results were used as EES software input. Simulation PTC with porous media inside collector pipe and nanofluids sedimentation is shown in Figure 3.

    Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

    3.1. Nano Fluid

    In this research, copper and silver nanofluids (Al2O3, CuO) have been added with percentages of 0.1%–0.7% as the working fluids. The nanoparticle properties are given in Table 4. Also, system constant parameters are presented in Table 4, which are available as default input in the EES software.Table 4 Properties of the nanoparticles [9].

    System constant parameters for input in the software are shown in Table 5.Table 5 System constant parameters.

    The thermal properties of the nanofluid can be obtained from equations (18)–(21). The basic fluid properties are indicated by the index (bf) and the properties of the nanoparticle silver with the index (np).

    The density of the mixture is shown in the following equation [28]:where ρ is density and ϕ is the nanoparticles volume fraction.

    The specific heat capacity is calculated from the following equation [29]:

    The thermal conductivity of the nanofluid is calculated from the following equation [29]:

    The parameter β is the ratio of the nanolayer thickness to the original particle radius and, usually, this parameter is taken equal to 0.1 for the calculated thermal conductivity of the nanofluids.

    The mixture viscosity is calculated as follows [30]:

    In all equations, instead of water properties, working fluids with nanofluid are used. All of the above equations and parameters are entered in the EES software for calculating the energy and exergy of solar collectors and the SCCHP cycle. All calculation repeats for both nanofluids with different concentrations of nanofluid in the solar collector’s pipe.

    4. Results and Discussion

    In the present study, relations were written according to Wang et al. [16] and the system analysis was performed to ensure the correctness of the code. The energy and exergy charts are plotted based on the main values of the paper and are shown in Figures 4 and 5. The error rate in this simulation is 1.07%.

    Figure 4 Verification charts of energy analysis results.

    Figure 5 Verification charts of exergy analysis results.

    We may also investigate the application of machine learning paradigms [3141] and various hybrid, advanced optimization approaches that are enhanced in terms of exploration and intensification [4255], and intelligent model studies [5661] as well, for example, methods such as particle swarm optimizer (PSO) [6062], differential search (DS) [63], ant colony optimizer (ACO) [616465], Harris hawks optimizer (HHO) [66], grey wolf optimizer (GWO) [5367], differential evolution (DE) [6869], and other fusion and boosted systems [4146485054557071].

    At the first step, the collector is modified with porous copper foam material. 14 cases have been considered for the analysis of the SCCHP system (Table 6). It should be noted that the adding of porous media causes an additional pressure drop inside the collector [922263072]. All fourteen cases use copper foam with a porosity of 95 percent. To simulate the effect of porous materials and nanofluids, the first solar PTC pipes have been simulated in the FLOW-3D software and then porous media (copper foam with porosity of 95%) and fluid flow with nanoparticles (AL2O3 and CUO) are generated in the software. After analyzing PTC pipes in FLOW-3D software, for analyzing energy and exergy efficiency, software outputs were used as EES software input for optimization ratio of sedimentation and calculating energy and exergy analyses.Table 6 Collectors with different percentages of nanofluids and porous media.

    In this research, an enhanced solar collector with both porous media and Nanofluid is investigated. In the present study, 0.1–0.5% CuO and Al2O3 concentration were added to the collector fully filled by porous media to achieve maximum energy and exergy efficiencies of solar CCHP systems. All steps of the investigation are shown in Table 6.

    Energy and exergy analyses of parabolic solar collectors and SCCHP systems are shown in Figures 6 and 7.

    Figure 6 Energy and exergy efficiencies of the PTC with porous media and nanofluid.

    Figure 7 Energy and exergy efficiency of the SCCHP.

    Results show that the highest energy and exergy efficiencies are 74.19% and 32.6%, respectively, that is achieved in Step 12 (parabolic collectors with filled porous media and 0.5% Al2O3). In the second step, the maximum energy efficiency of SCCHP systems with fourteen steps of simulation are shown in Figure 7.

    In the second step, where 0.1, −0.6% of the nanofluids were added, it is found that 0.5% leads to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Using concentrations more than 0.5% leads to sediment in the solar collector’s pipe and a decrease of porosity in the pipe [73]. According to Figure 7, maximum energy and exergy efficiencies of SCCHP are achieved in Step 12. In this step energy efficiency is 54.49% and exergy efficiency is 18.29%. In steps 13 and 14, with increasing concentration of CUO and Al2O3 nanofluid solution in porous materials, decreasing of energy and exergy efficiency of PTC and SCCHP system at the same time happened. This decrease in efficiency is due to the formation of sediment in the porous material. Calculations and simulations have shown that porous materials more than 0.5% nanofluids inside the collector pipe cause sediment and disturb the porosity of porous materials and pressure drop and reduce the coefficient of performance of the cogeneration system. Most experience showed that CUO and AL2O3 nanofluids with less than 0.6% percent solution are used in the investigation on the solar collectors at low temperatures and discharges [74]. One of the important points of this research is that the best ratio of nanofluids in the solar collector with a low temperature is 0.5% (AL2O3 and CUO); with this replacement, the cost of solar collectors and SCCHP cycle is reduced.

    5. Conclusion and Future Directions

    In the present study, ways for increasing the efficiency of solar collectors in order to enhance the efficiency of the SCCHP cycle are examined. The research is aimed at adding both porous materials and nanofluids for estimating the best ratio of nanofluid for enhanced solar collector and protecting sedimentation in porous media. By adding porous materials (copper foam with porosity of 95%) and 0.5% nanofluids together, high efficiency in solar parabolic collectors can be achieved. The novelty in this research is the addition of both nanofluids and porous materials and calculating the best ratio for preventing sedimentation and pressure drop in solar collector’s pipe. In this study, it was observed that, by adding 0.5% of AL2O3 nanofluid in working fluids, the energy efficiency of PTC rises to 74.19% and exergy efficiency is grown up to 32.6%. In SCCHP cycle, energy efficiency is 54.49% and exergy efficiency is 18.29%.

    In this research, parabolic solar collectors fully filled by porous media (copper foam with a porosity of 95) are investigated. In the next step, parabolic solar collectors in the SCCHP cycle were simultaneously filled by porous media and different percentages of Al2O3 and CuO nanofluid. At this step, values of 0.1% to 0.6% of each nanofluid were added to the working fluid, and the efficiency of the energy and exergy of the collectors and the SCCHP cycle were determined. In this case, nanofluid and the porous media were used together in the solar collector and maximum efficiency achieved. 0.5% of both nanofluids were used to achieve the biggest efficiency enhancement.

    In the present study, as expected, the highest efficiency is for the parabolic solar collector fully filled by porous material (copper foam with a porosity of 95%) and 0.5% Al2O3. Results of the present study are as follows:(1)The average enhancement of collectors’ efficiency using porous media and nanofluids is 28%.(2)Solutions with 0.1 to 0.5% of nanofluids (CuO and Al2O3) are used to prevent collectors from sediment occurrence in porous media.(3)Collector of solar cogeneration cycles that is enhanced by both porous media and nanofluid has higher efficiency, and the stability of output temperature is more as well.(4)By using 0.6% of the nanofluids in the enhanced parabolic solar collectors with copper porous materials, sedimentation occurs and makes a high-pressure drop in the solar collector’s pipe which causes decrease in energy efficiency.(5)Average enhancement of SCCHP cycle efficiency is enhanced by both porous media and nanofluid 13%.

    Nomenclature

    :Solar radiation
    a:Heat transfer augmentation coefficient
    A:Solar collector area
    Bf:Basic fluid
    :Specific heat capacity of the nanofluid
    F:Constant of air dilution
    :Thermal conductivity of the nanofluid
    :Thermal conductivity of the basic fluid
    :Viscosity of the nanofluid
    :Viscosity of the basic fluid
    :Collector efficiency
    :Collector energy receives
    :Auxiliary boiler heat
    :Expander energy
    :Gas energy
    :Screw expander work
    :Cooling load, in kilowatts
    :Heating load, in kilowatts
    :Solar radiation energy on collector, in Joule
    :Sanitary hot water load
    Np:Nanoparticle
    :Energy efficiency
    :Heat exchanger efficiency
    :Sun exergy
    :Collector exergy
    :Natural gas exergy
    :Expander exergy
    :Cooling exergy
    :Heating exergy
    :Exergy efficiency
    :Steam mass flow rate
    :Hot water mass flow rate
    :Specific heat capacity of water
    :Power output form by the screw expander
    Tam:Average ambient temperature
    :Density of the mixture.

    Greek symbols

    ρ:Density
    ϕ:Nanoparticles volume fraction
    β:Ratio of the nanolayer thickness.

    Abbreviations

    CCHP:Combined cooling, heating, and power
    EES:Engineering equation solver.

    Data Availability

    For this study, data were generated by CARRIER software for the average electrical, heating, and cooling load of a residential building with 600 m2 in the city of Zahedan, Iran.

    Conflicts of Interest

    The authors declare that they have no conflicts of interest.

    Acknowledgments

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

    References

    1. A. Fudholi and K. Sopian, “Review on solar collector for agricultural produce,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 9, no. 1, p. 414, 2018.View at: Publisher Site | Google Scholar
    2. G. Yang and X. Zhai, “Optimization and performance analysis of solar hybrid CCHP systems under different operation strategies,” Applied Thermal Engineering, vol. 133, pp. 327–340, 2018.View at: Publisher Site | Google Scholar
    3. J. Wang, Z. Han, and Z. Guan, “Hybrid solar-assisted combined cooling, heating, and power systems: a review,” Renewable and Sustainable Energy Reviews, vol. 133, p. 110256, 2020.View at: Publisher Site | Google Scholar
    4. Y. Tian and C. Y. Zhao, “A review of solar collectors and thermal energy storage in solar thermal applications,” Applied Energy, vol. 104, pp. 538–553, 2013.View at: Publisher Site | Google Scholar
    5. J. M. Hassan, Q. J. Abdul-Ghafour, and M. F. Mohammed, “CFD simulation of enhancement techniques in flat plate solar water collectors,” Al-Nahrain Journal for Engineering Sciences, vol. 20, no. 3, pp. 751–761, 2017.View at: Google Scholar
    6. M. Jahangiri, O. Nematollahi, A. Haghani, H. A. Raiesi, and A. Alidadi Shamsabadi, “An optimization of energy cost of clean hybrid solar-wind power plants in Iran,” International Journal of Green Energy, vol. 16, no. 15, pp. 1422–1435, 2019.View at: Publisher Site | Google Scholar
    7. I. H. Yılmaz and A. Mwesigye, “Modeling, simulation and performance analysis of parabolic trough solar collectors: a comprehensive review,” Applied Energy, vol. 225, pp. 135–174, 2018.View at: Google Scholar
    8. F. Wang, J. Tan, and Z. Wang, “Heat transfer analysis of porous media receiver with different transport and thermophysical models using mixture as feeding gas,” Energy Conversion and Management, vol. 83, pp. 159–166, 2014.View at: Publisher Site | Google Scholar
    9. H. Zhai, Y. J. Dai, J. Y. Wu, and R. Z. Wang, “Energy and exergy analyses on a novel hybrid solar heating, cooling and power generation system for remote areas,” Applied Energy, vol. 86, no. 9, pp. 1395–1404, 2009.View at: Publisher Site | Google Scholar
    10. M. H. Abbasi, H. Sayyaadi, and M. Tahmasbzadebaie, “A methodology to obtain the foremost type and optimal size of the prime mover of a CCHP system for a large-scale residential application,” Applied Thermal Engineering, vol. 135, pp. 389–405, 2018.View at: Google Scholar
    11. R. Jiang, F. G. F. Qin, X. Yang, S. Huang, and B. Chen, “Performance analysis of a liquid absorption dehumidifier driven by jacket-cooling water of a diesel engine in a CCHP system,” Energy and Buildings, vol. 163, pp. 70–78, 2018.View at: Publisher Site | Google Scholar
    12. F. A. Boyaghchi and M. Chavoshi, “Monthly assessments of exergetic, economic and environmental criteria and optimization of a solar micro-CCHP based on DORC,” Solar Energy, vol. 166, pp. 351–370, 2018.View at: Publisher Site | Google Scholar
    13. F. A. Boyaghchi and M. Chavoshi, “Multi-criteria optimization of a micro solar-geothermal CCHP system applying water/CuO nanofluid based on exergy, exergoeconomic and exergoenvironmental concepts,” Applied Thermal Engineering, vol. 112, pp. 660–675, 2017.View at: Publisher Site | Google Scholar
    14. B. Su, W. Han, Y. Chen, Z. Wang, W. Qu, and H. Jin, “Performance optimization of a solar assisted CCHP based on biogas reforming,” Energy Conversion and Management, vol. 171, pp. 604–617, 2018.View at: Publisher Site | Google Scholar
    15. F. A. Al-Sulaiman, F. Hamdullahpur, and I. Dincer, “Performance assessment of a novel system using parabolic trough solar collectors for combined cooling, heating, and power production,” Renewable Energy, vol. 48, pp. 161–172, 2012.View at: Publisher Site | Google Scholar
    16. J. Wang, Y. Dai, L. Gao, and S. Ma, “A new combined cooling, heating and power system driven by solar energy,” Renewable Energy, vol. 34, no. 12, pp. 2780–2788, 2009.View at: Publisher Site | Google Scholar
    17. Y.-Y. Jing, H. Bai, J.-J. Wang, and L. Liu, “Life cycle assessment of a solar combined cooling heating and power system in different operation strategies,” Applied Energy, vol. 92, pp. 843–853, 2012.View at: Publisher Site | Google Scholar
    18. J.-J. Wang, Y.-Y. Jing, and C.-F. Zhang, “Optimization of capacity and operation for CCHP system by genetic algorithm,” Applied Energy, vol. 87, no. 4, pp. 1325–1335, 2010.View at: Publisher Site | Google Scholar
    19. L. Ali, “LDA–GA–SVM: improved hepatocellular carcinoma prediction through dimensionality reduction and genetically optimized support vector machine,” Neural Computing and Applications, vol. 87, pp. 1–10, 2020.View at: Google Scholar
    20. S. Hirasawa, R. Tsubota, T. Kawanami, and K. Shirai, “Reduction of heat loss from solar thermal collector by diminishing natural convection with high-porosity porous medium,” Solar Energy, vol. 97, pp. 305–313, 2013.View at: Publisher Site | Google Scholar
    21. E. Bellos, C. Tzivanidis, and Z. Said, “A systematic parametric thermal analysis of nanofluid-based parabolic trough solar collectors,” Sustainable Energy Technologies and Assessments, vol. 39, p. 100714, 2020.View at: Publisher Site | Google Scholar
    22. H. J. Jouybari, S. Saedodin, A. Zamzamian, M. E. Nimvari, and S. Wongwises, “Effects of porous material and nanoparticles on the thermal performance of a flat plate solar collector: an experimental study,” Renewable Energy, vol. 114, pp. 1407–1418, 2017.View at: Publisher Site | Google Scholar
    23. J. Subramani, P. K. Nagarajan, S. Wongwises, S. A. El-Agouz, and R. Sathyamurthy, “Experimental study on the thermal performance and heat transfer characteristics of solar parabolic trough collector using Al2O3 nanofluids,” Environmental Progress & Sustainable Energy, vol. 37, no. 3, pp. 1149–1159, 2018.View at: Publisher Site | Google Scholar
    24. E. Shojaeizadeh, F. Veysi, and A. Kamandi, “Exergy efficiency investigation and optimization of an Al2O3-water nanofluid based Flat-plate solar collector,” Energy and Buildings, vol. 101, pp. 12–23, 2015.View at: Publisher Site | Google Scholar
    25. A. K. Tiwari, P. Ghosh, and J. Sarkar, “Solar water heating using nanofluids–a comprehensive overview and environmental impact analysis,” International Journal of Emerging Technology and Advanced Engineering, vol. 3, no. 3, pp. 221–224, 2013.View at: Google Scholar
    26. D. R. Rajendran, E. Ganapathy Sundaram, P. Jawahar, V. Sivakumar, O. Mahian, and E. Bellos, “Review on influencing parameters in the performance of concentrated solar power collector based on materials, heat transfer fluids and design,” Journal of Thermal Analysis and Calorimetry, vol. 140, no. 1, pp. 33–51, 2020.View at: Publisher Site | Google Scholar
    27. M. Feizbahr, C. Kok Keong, F. Rostami, and M. Shahrokhi, “Wave energy dissipation using perforated and non perforated piles,” International Journal of Engineering, vol. 31, no. 2, pp. 212–219, 2018.View at: Google Scholar
    28. K. Khanafer and K. Vafai, “A critical synthesis of thermophysical characteristics of nanofluids,” International Journal of Heat and Mass Transfer, vol. 54, no. 19-20, pp. 4410–4428, 2011.View at: Publisher Site | Google Scholar
    29. K. Farhana, K. Kadirgama, M. M. Rahman et al., “Improvement in the performance of solar collectors with nanofluids – a state-of-the-art review,” Nano-Structures & Nano-Objects, vol. 18, p. 100276, 2019.View at: Publisher Site | Google Scholar
    30. M. Turkyilmazoglu, “Condensation of laminar film over curved vertical walls using single and two-phase nanofluid models,” European Journal of Mechanics-B/Fluids, vol. 65, pp. 184–191, 2017.View at: Publisher Site | Google Scholar
    31. X. Zhang, J. Wang, T. Wang, R. Jiang, J. Xu, and L. Zhao, “Robust feature learning for adversarial defense via hierarchical feature alignment,” Information Sciences, vol. 2020, 2020.View at: Google Scholar
    32. X. Zhang, T. Wang, W. Luo, and P. Huang, “Multi-level fusion and attention-guided CNN for image dehazing,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
    33. X. Zhang, M. Fan, D. Wang, P. Zhou, and D. Tao, “Top-k feature selection framework using robust 0-1 integer programming,” IEEE Transactions on Neural Networks and Learning Systems, vol. 1, pp. 1–15, 2020.View at: Publisher Site | Google Scholar
    34. X. Zhang, D. Wang, Z. Zhou, and Y. Ma, “Robust low-rank tensor recovery with rectification and alignment,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 238–255, 2019.View at: Google Scholar
    35. X. Zhang, R. Jiang, T. Wang, and J. Wang, “Recursive neural network for video deblurring,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
    36. X. Zhang, T. Wang, J. Wang, G. Tang, and L. Zhao, “Pyramid channel-based feature attention network for image dehazing,” Computer Vision and Image Understanding, vol. 1, 2020.View at: Google Scholar
    37. M. Mirmozaffari, “Machine learning algorithms based on an optimization model,” 2020.View at: Google Scholar
    38. M. Mirmozaffari, M. Yazdani, A. Boskabadi, H. Ahady Dolatsara, K. Kabirifar, and N. Amiri Golilarz, “A novel machine learning approach combined with optimization models for eco-efficiency evaluation,” Applied Sciences, vol. 10, no. 15, p. 5210, 2020.View at: Publisher Site | Google Scholar
    39. M. Vosoogha and A. Addeh, “An intelligent power prediction method for wind energy generation based on optimized fuzzy system,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 5, pp. 34–43, 2019.View at: Google Scholar
    40. A. Javadi, N. Mikaeilvand, and H. Hosseinzdeh, “Presenting a new method to solve partial differential equations using a group search optimizer method (GSO),” Computational Research Progress in Applied Science and Engineering, vol. 4, no. 1, pp. 22–26, 2018.View at: Google Scholar
    41. F. J. Golrokh, Gohar Azeem, and A. Hasan, “Eco-efficiency evaluation in cement industries: DEA malmquist productivity index using optimization models,” ENG Transactions, vol. 1, pp. 1–8, 2020.View at: Google Scholar
    42. H. Yu, “Dynamic Gaussian bare-bones fruit fly optimizers with abandonment mechanism: method and analysis,” Engineering with Computers, vol. 1, pp. 1–29, 2020.View at: Google Scholar
    43. C. Yu, “SGOA: annealing-behaved grasshopper optimizer for global tasks,” Engineering with Computers, vol. 1, pp. 1–28, 2021.View at: Google Scholar
    44. W. Shan, Z. Qiao, A. A. Heidari, H. Chen, H. Turabieh, and Y. Teng, “Double adaptive weights for stabilization of moth flame optimizer: balance analysis, engineering cases, and medical diagnosis,” Knowledge-Based Systems, vol. 1, p. 106728, 2020.View at: Google Scholar
    45. J. Tu, H. Chen, J. Liu et al., “Evolutionary biogeography-based whale optimization methods with communication structure: towards measuring the balance,” Knowledge-Based Systems, vol. 212, p. 106642, 2021.View at: Publisher Site | Google Scholar
    46. Y. Zhang, “Towards augmented kernel extreme learning models for bankruptcy prediction: algorithmic behavior and comprehensive analysis,” Neurocomputing, vol. 1, 2020.View at: Google Scholar
    47. Y. Zhang, R. Liu, X. Wang, H. Chen, and C. Li, “Boosted binary Harris hawks optimizer and feature selection,” Engineering with Computers, vol. 1, pp. 1–30, 2020.View at: Google Scholar
    48. H.-L. Chen, G. Wang, C. Ma, Z.-N. Cai, W.-B. Liu, and S.-J. Wang, “An efficient hybrid kernel extreme learning machine approach for early diagnosis of Parkinson’s disease,” Neurocomputing, vol. 184, pp. 131–144, 2016.View at: Publisher Site | Google Scholar
    49. L. Hu, G. Hong, J. Ma, X. Wang, and H. Chen, “An efficient machine learning approach for diagnosis of paraquat-poisoned patients,” Computers in Biology and Medicine, vol. 59, pp. 116–124, 2015.View at: Publisher Site | Google Scholar
    50. L. Shen, H. Chen, Z. Yu et al., “Evolving support vector machines using fruit fly optimization for medical data classification,” Knowledge-Based Systems, vol. 96, pp. 61–75, 2016.View at: Publisher Site | Google Scholar
    51. J. Xia, H. Chen, Q. Li et al., “Ultrasound-based differentiation of malignant and benign thyroid Nodules: an extreme learning machine approach,” Computer Methods and Programs in Biomedicine, vol. 147, pp. 37–49, 2017.View at: Publisher Site | Google Scholar
    52. C. Li, L. Hou, B. Y. Sharma et al., “Developing a new intelligent system for the diagnosis of tuberculous pleural effusion,” Computer Methods and Programs in Biomedicine, vol. 153, pp. 211–225, 2018.View at: Publisher Site | Google Scholar
    53. X. Zhao, X. Zhang, Z. Cai et al., “Chaos enhanced grey wolf optimization wrapped ELM for diagnosis of paraquat-poisoned patients,” Computational Biology and Chemistry, vol. 78, pp. 481–490, 2019.View at: Publisher Site | Google Scholar
    54. M. Wang and H. Chen, “Chaotic multi-swarm whale optimizer boosted support vector machine for medical diagnosis,” Applied Soft Computing Journal, vol. 88, 2020.View at: Publisher Site | Google Scholar
    55. X. Xu and H.-L. Chen, “Adaptive computational chemotaxis based on field in bacterial foraging optimization,” Soft Computing, vol. 18, no. 4, pp. 797–807, 2014.View at: Publisher Site | Google Scholar
    56. R. U. Khan, X. Zhang, R. Kumar, A. Sharif, N. A. Golilarz, and M. Alazab, “An adaptive multi-layer botnet detection technique using machine learning classifiers,” Applied Sciences, vol. 9, no. 11, p. 2375, 2019.View at: Publisher Site | Google Scholar
    57. A. Addeh, A. Khormali, and N. A. Golilarz, “Control chart pattern recognition using RBF neural network with new training algorithm and practical features,” ISA Transactions, vol. 79, pp. 202–216, 2018.View at: Publisher Site | Google Scholar
    58. N. Amiri Golilarz, H. Gao, R. Kumar, L. Ali, Y. Fu, and C. Li, “Adaptive wavelet based MRI brain image de-noising,” Frontiers in Neuroscience, vol. 14, p. 728, 2020.View at: Publisher Site | Google Scholar
    59. N. A. Golilarz, H. Gao, and H. Demirel, “Satellite image de-noising with Harris hawks meta heuristic optimization algorithm and improved adaptive generalized Gaussian distribution threshold function,” IEEE Access, vol. 7, pp. 57459–57468, 2019.View at: Publisher Site | Google Scholar
    60. M. Eisazadeh and J. Rezapour, “Multi-objective optimization of the composite sheets using PSO algorithm,” 2017.View at: Google Scholar
    61. I. Bargegol, M. Nikookar, R. V. Nezafat, E. J. Lashkami, and A. M. Roshandeh, “Timing optimization of signalized intersections using shockwave theory by genetic algorithm,” Computational Research Progress in Applied Science & Engineering, vol. 1, pp. 160–167, 2015.View at: Google Scholar
    62. B. Bai, Z. Guo, C. Zhou, W. Zhang, and J. Zhang, “Application of adaptive reliability importance sampling-based extended domain PSO on single mode failure in reliability engineering,” Information Sciences, vol. 546, pp. 42–59, 2021.View at: Publisher Site | Google Scholar
    63. J. Liu, C. Wu, G. Wu, and X. Wang, “A novel differential search algorithm and applications for structure design,” Applied Mathematics and Computation, vol. 268, pp. 246–269, 2015.View at: Publisher Site | Google Scholar
    64. X. Zhao, D. Li, B. Yang, C. Ma, Y. Zhu, and H. Chen, “Feature selection based on improved ant colony optimization for online detection of foreign fiber in cotton,” Applied Soft Computing, vol. 24, pp. 585–596, 2014.View at: Publisher Site | Google Scholar
    65. D. Zhao, “Chaotic random spare ant colony optimization for multi-threshold image segmentation of 2D Kapur entropy,” Knowledge-Based Systems, vol. 24, p. 106510, 2020.View at: Google Scholar
    66. H. Chen, A. A. Heidari, H. Chen, M. Wang, Z. Pan, and A. H. Gandomi, “Multi-population differential evolution-assisted Harris hawks optimization: framework and case studies,” Future Generation Computer Systems, vol. 111, pp. 175–198, 2020.View at: Publisher Site | Google Scholar
    67. J. Hu, H. Chen, A. A. Heidari et al., “Orthogonal learning covariance matrix for defects of grey wolf optimizer: insights, balance, diversity, and feature selection,” Knowledge-Based Systems, vol. 213, p. 106684, 2021.View at: Publisher Site | Google Scholar
    68. G. Sun, B. Yang, Z. Yang, and G. Xu, “An adaptive differential evolution with combined strategy for global numerical optimization,” Soft Computing, vol. 24, pp. 1–20, 2019.View at: Google Scholar
    69. G. Sun, C. Li, and L. Deng, “An adaptive regeneration framework based on search space adjustment for differential evolution,” Neural Computing and Applications, vol. 24, pp. 1–17, 2021.View at: Google Scholar
    70. A. Addeh and M. Iri, “Brain tumor type classification using deep features of MRI images and optimized RBFNN,” ENG Transactions, vol. 2, pp. 1–7, 2021.View at: Google Scholar
    71. F. J. Golrokh and A. Hasan, “A comparison of machine learning clustering algorithms based on the DEA optimization approach for pharmaceutical companies in developing countries,” Soft Computing, vol. 1, pp. 1–8, 2020.View at: Google Scholar
    72. H. Tyagi, P. Phelan, and R. Prasher, “Predicted efficiency of a low-temperature nanofluid-based direct absorption solar collector,” Journal of Solar Energy Engineering, vol. 131, no. 4, 2009.View at: Publisher Site | Google Scholar
    73. S. Rashidi, M. Bovand, and J. A. Esfahani, “Heat transfer enhancement and pressure drop penalty in porous solar heat exchangers: a sensitivity analysis,” Energy Conversion and Management, vol. 103, pp. 726–738, 2015.View at: Publisher Site | Google Scholar
    74. N. Akram, R. Sadri, S. N. Kazi et al., “A comprehensive review on nanofluid operated solar flat plate collectors,” Journal of Thermal Analysis and Calorimetry, vol. 139, no. 2, pp. 1309–1343, 2020.View at: Publisher Site | Google Scholar
    Fig. 6. Configuration of Johnson (1958) hydraulic experiment.

    전체 수심 범위에서 선박 파고에 대한 방정식

    Equation for ship wave crests in the entire range of water depths

    Byeong Wook Lee a
    , Changhoon Lee b,
    *a Coastal Development and Ocean Energy Research Center, Korea Institute of Ocean Science & Technology, 385 Haeyang-ro, Busan, 49111, Republic of Korea
    b Department of Civil and Environmental Engineering, Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul, 05006, Republic of Korea

    ABSTRACT

    An equation for ship wave crests y/x in the entire range of water depths is developed using the linear dispersion relation. In deep water, the developed equation is reduced to the equation of Kelvin (1906). The locations of ship wave crests in the x – and y -directions are obtained using a dimensionless constant C. The wave ray angle θc at the cusp locus is determined using the condition that θc is maximal at the cusp locus and the cusp locus angle is determined as αc=−tan−1(y/x)max. Numerical experiments are conducted using the FLOW-3D to simulate ship wave propagation. The cusp locus angles of the FLOW-3D are similar to both those of the present theory and Havelock (1908) theory in the entire range of the Froude number. Both the present theory and the FLOW-3D yield that, with the increase of ship speed, the Froude number increases and does the wavelength. For the Froude number equal to or greater than unity, the wavelength becomes infinitely large and the transverse waves disappear. The wavelengths of the FLOW-3D are slightly smaller than those of the present theory because the FLOW-3D considers the decrease of wavelength due to energy dissipation which happens because of viscosity of water and turbulence of high-speed particle velocities.

    Fig. 6. Configuration of Johnson (1958) hydraulic experiment.
    Fig. 6. Configuration of Johnson (1958) hydraulic experiment.
    Fig. 8. Comparison of ship wave crest patterns: (a) Fr ¼ 0:66 (Us ¼ 6:5m=s,  kh � 0:724π), (b) Fr ¼ 0:86 (Us ¼ 8:5m=s, kh � 0:342π), (c) Fr ¼ 1:21 (Us ¼ 12:0m=s, kh � 0:003π). Line definition: red solid line ¼ present theory; yellow  dashed line ¼ Kelvin theory; white dot ¼ FLOW-3D solution. (For interpretation  of the references to colour in this figure legend, the reader is referred to the  Web version of this article.)
    Fig. 8. Comparison of ship wave crest patterns: (a) Fr ¼ 0:66 (Us ¼ 6:5m=s, kh >= 0:724π), (b) Fr ¼ 0:86 (Us ¼ 8:5m=s, kh >= 0:342π), (c) Fr ¼ 1:21 (Us ¼ 12:0m=s, kh >= 0:003π). Line definition: red solid line ¼ present theory; yellow dashed line ¼ Kelvin theory; white dot ¼ FLOW-3D solution. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

    Keywords

    Ship wave crests
    Cusp locus angle
    Entire range of water depths
    Theoretical solution
    Numerical experiment

    References

    kylas, T.R., 1984. On the excitation of long nonlinear water waves by a moving pressure
    distribution. J. Fluid Mech. 141, 455–466.

    Chen, X.N., Sharma, S.D., 1995. A slender ship moving at a near-critical speed in a
    shallow channel. J. Fluid Mech. 291, 263–285.
    David, C.G., Roeber, V., Goseberg, N., Schlurmann, T., 2017. Generation and propagation
    of ship-borne waves – solutions from a Boussinesq-type model. Cost Eng. 127,
    170–187.
    Ersan, D.B., Beji, S., 2013. Numerical simulation of waves generated by a moving
    pressure field. Ocean Eng. 59, 231–239.
    Ertekin, R.C., Webster, W.C., Wehausen, J.V., 1986. Waves caused by a moving
    disturbance in a shallow channel of finite width. J. Fluid Mech. 169, 275–292.
    Fang, M.-C., Yang, R.-Y., Shugan, I.V., 2011. Kelvin ship wake in the wind waves field
    and on the finite sea depth. J. Mech. 27 (1), 71–77.
    Havelock, T.H., 1908. The propagation of groups of waves in dispersive media with
    application to waves on water produced by a travelling disturbance. Proc. Royal Soc.
    London Ser. A 398–430.
    Hennings, I., Romeiser, R., Alpers, W., Viola, A., 1999. Radar imaging of Kelvin arms of
    ship wakes. Int. J. Remote Sens. 20 (13), 2519–2543.
    Hur, D.S., Lee, J., Choi, D.S., Lee, H.W., 2011. On run-up characteristics of revetment
    under interaction among ocean wave, current and ship induced wave in the canal.
    In: Proceedings of the 37th Conference on the Korean Society of Civil Engineers,
    pp. 588–591 (in Korean).
    Johnson, J.W., 1958. Ship waves in navigation channels. In: Proceedings of the 6th
    Conference on Coastal Engineering, pp. 666–690.
    Kang, Y.S., Kim, P.J., Hyun, S.K., Sung, H.K., 2008. Numerical simulation of ship-induced
    wave using FLOW-3D. J. Korean Soc. Coast. Ocean Eng. 20 (3), 255–267 (in Korean).
    Kelvin, 1887. On ship waves. In: Proceedings of the Institution of Mechanical
    Engineering, pp. 409–433.
    Kelvin, 1906. Deep sea ship-waves. Proc. R. Soc. Edinb. 25 (2), 1060–1084.
    Lamb, H., 1945. Hydrodynamics. Dover Publications.
    Lee, C., Lee, B.W., Kim, Y.J., Ko, K.O., 2011. Ship wave crests in intermediate-depth
    water. In: Proceedings of the 6th International Conference on Asian and Pacific
    Coasts, pp. 1818–1825.
    Lee, B.W., Lee, C., Kim, Y.J., Ko, K.O., 2013. Prediction of ship wave crests on varying
    water depths and verification by FLOW-3D. J. Korean Soc. Civil Eng. 33 (4),
    1447–1454 (in Korean).
    Lighthill, M.J., Whitham, G.B., 1955. On kinematic waves: I. Flood movement in long
    rivers; II. Theory of traffic flow on long crowded roads. Proc. R. Soc. A 229, 281–345.
    Newman, J.N., 1970. Recent research on ship waves. In: Proceedings of the 8th
    Symposium on Naval Hydrodynamics, pp. 519–545.
    Newman, J.N., 1977. Marine Hydrodynamics. The MIT Press.
    Reed, A.M., Milgram, J.H., 2002. Ship wakes and their radar images. Annu. Rev. Fluid
    Mech. 34, 469–502.
    Shemdin, O.H., 1990. Synthetic aperture radar imaging of ship wakes in the Gulf of
    Alaska. J. Geophys. Res. 95 (C9), 16319–16338.
    Shi, F., Malej, M., Smith, J.M., Kirby, J.T., 2018. Breaking of ship bores in a Boussinesqtype ship-wake model. Cost Eng. 132, 1–12.
    Sorensen, R.M., 1967. Investigation of ship-generated waves. J. Waterw. Harb. Div.
    85–99. ASCE.
    Sorensen, R.M., 1969. Waves generated by model ship hull. J. Waterw. Harb. Div.
    513–538. ASCE.
    Sorensen, R.M., Weggel, J.R., 1984. Development of ship wave design information. In:
    Proceedings of the 19th Conference on Coastal Engineering, pp. 3227–3243. ASCE.
    Stoker, J.J., 1957. Water Waves: the Mathematical Theory with Applications.
    Interscience Publishers.
    Tuck, E.O., 1966. Shallow-water flows past slender bodies. J. Fluid Mech. 26, 81–95.
    Wu, D.M., Wu, T.Y., 1982. Three-dimensional nonlinear long waves due to moving
    surface pressure. In: Proceedings of the 14th Symposium on Naval Hydrodynamics,
    pp. 103–129.

    Flow velocity profiles for canals with a depth of 3 m and flow velocities of 5–5.3 m/s.

    Optimization Algorithms and Engineering: Recent Advances and Applications

    Mahdi Feizbahr,1 Navid Tonekaboni,2Guang-Jun Jiang,3,4 and Hong-Xia Chen3,4Show moreAcademic Editor: Mohammad YazdiReceived08 Apr 2021Revised18 Jun 2021Accepted17 Jul 2021Published11 Aug 2021

    Abstract

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


    강의 식생은 거칠기를 증가시키고 평균 유속을 감소시키며, 유속 에너지를 감소시키고 강의 단면에서 유속 프로파일을 변경합니다. 자연의 많은 운하와 강은 홍수 동안 초목으로 덮여 있습니다. 운하의 조도는 식물의 영향을 많이 받으므로 홍수시 유동저항에 큰 영향을 미칩니다. 식물로 인한 흐름에 대한 거칠기 저항은 흐름 조건 및 식물에 따라 다르므로 모델은 유속, 흐름 깊이 및 운하를 따라 식생 유형의 영향을 고려하여 현재 속도를 시뮬레이션해야 합니다. 근관의 거칠기의 영향을 조사하기 위해 총 48개의 모델이 시뮬레이션되었습니다. 결과는 유속을 높임으로써 유속을 감소시키는 식생의 영향은 무시할 수 있는 반면, 해류가 더 낮은 유속일 때 유속을 감소시키는 식생의 영향은 분명히 상당함을 나타냈다.

    1. Introduction

    Considering the impact of each variable is a very popular field within the analytical and statistical methods and intelligent systems [114]. This can help research for better modeling considering the relation of variables or interaction of them toward reaching a better condition for the objective function in control and engineering [1527]. Consequently, it is necessary to study the effects of the passive factors on the active domain [2836]. Because of the effect of vegetation on reducing the discharge capacity of rivers [37], pruning plants was necessary to improve the condition of rivers. One of the important effects of vegetation in river protection is the action of roots, which cause soil consolidation and soil structure improvement and, by enhancing the shear strength of soil, increase the resistance of canal walls against the erosive force of water. The outer limbs of the plant increase the roughness of the canal walls and reduce the flow velocity and deplete the flow energy in vicinity of the walls. Vegetation by reducing the shear stress of the canal bed reduces flood discharge and sedimentation in the intervals between vegetation and increases the stability of the walls [3841].

    One of the main factors influencing the speed, depth, and extent of flood in this method is Manning’s roughness coefficient. On the other hand, soil cover [42], especially vegetation, is one of the most determining factors in Manning’s roughness coefficient. Therefore, it is expected that those seasonal changes in the vegetation of the region will play an important role in the calculated value of Manning’s roughness coefficient and ultimately in predicting the flood wave behavior [4345]. The roughness caused by plants’ resistance to flood current depends on the flow and plant conditions. Flow conditions include depth and velocity of the plant, and plant conditions include plant type, hardness or flexibility, dimensions, density, and shape of the plant [46]. In general, the issue discussed in this research is the optimization of flood-induced flow in canals by considering the effect of vegetation-induced roughness. Therefore, the effect of plants on the roughness coefficient and canal transmission coefficient and in consequence the flow depth should be evaluated [4748].

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

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

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

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

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

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

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

    Roughness in river vegetation is simulated in mathematical models with a variable floor slope flume by different densities and discharges. The vegetation considered submerged in the bed of the flume. Results showed that, with increasing vegetation density, canal roughness and flow shear speed increase and with increasing flow rate and depth, Manning’s roughness coefficient decreases. Factors affecting the roughness caused by vegetation include the effect of plant density and arrangement on flow resistance, the effect of flow velocity on flow resistance, and the effect of depth [4555].

    One of the works that has been done on the effect of vegetation on the roughness coefficient is Darby [56] study, which investigates a flood wave model that considers all the effects of vegetation on the roughness coefficient. There are currently two methods for estimating vegetation roughness. One method is to add the thrust force effect to Manning’s equation [475758] and the other method is to increase the canal bed roughness (Manning-Strickler coefficient) [455961]. These two methods provide acceptable results in models designed to simulate floodplain flow. Wang et al. [62] simulate the floodplain with submerged vegetation using these two methods and to increase the accuracy of the results, they suggested using the effective height of the plant under running water instead of using the actual height of the plant. Freeman et al. [49] provided equations for determining the coefficient of vegetation roughness under different conditions. Lee et al. [63] proposed a method for calculating the Manning coefficient using the flow velocity ratio at different depths. Much research has been done on the Manning roughness coefficient in rivers, and researchers [496366] sought to obtain a specific number for n to use in river engineering. However, since the depth and geometric conditions of rivers are completely variable in different places, the values of Manning roughness coefficient have changed subsequently, and it has not been possible to choose a fixed number. In river engineering software, the Manning roughness coefficient is determined only for specific and constant conditions or normal flow. Lee et al. [63] stated that seasonal conditions, density, and type of vegetation should also be considered. Hydraulic roughness and Manning roughness coefficient n of the plant were obtained by estimating the total Manning roughness coefficient from the matching of the measured water surface curve and water surface height. The following equation is used for the flow surface curve:where  is the depth of water change, S0 is the slope of the canal floor, Sf is the slope of the energy line, and Fr is the Froude number which is obtained from the following equation:where D is the characteristic length of the canal. Flood flow velocity is one of the important parameters of flood waves, which is very important in calculating the water level profile and energy consumption. In the cases where there are many limitations for researchers due to the wide range of experimental dimensions and the variety of design parameters, the use of numerical methods that are able to estimate the rest of the unknown results with acceptable accuracy is economically justified.

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

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

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

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

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

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

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

    Figure 1 The simulated model and its boundary conditions.

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

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

    Any control and simulation system has some inputs that we should determine to test any technology [7077]. Determination and true implementation of such parameters is one of the key steps of any simulation [237881] and computing procedure [8286]. The input current is created by applying the flow rate through the VFR (Volume Flow Rate) option and the output flow is considered Output and for other borders the Symmetry option is considered.

    Simulation of the models and checking their action and responses and observing how a process behaves is one of the accepted methods in engineering and science [8788]. For verification of FLOW-3D software, the results of computer simulations are compared with laboratory measurements and according to the values of computational error, convergence error, and the time required for convergence, the most appropriate option for real-time simulation is selected (Figures 2 and 3 ).

    Figure 2 Modeling the plant with cylindrical tubes at the bottom of the canal.

    Figure 3 Velocity profiles in positions 2 and 5.

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

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

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

    2. Modeling Results

    After analyzing the models, the results were shown in graphs (Figures 414 ). The total number of experiments in this study was 48 due to the limitations of modeling.(a)
    (a)(b)
    (b)(c)
    (c)(d)
    (d)(a)
    (a)(b)
    (b)(c)
    (c)(d)
    (d)Figure 4 Flow velocity profiles for canals with a depth of 1 m and flow velocities of 3–3.3 m/s. Canal with a depth of 1 meter and a flow velocity of (a) 3 meters per second, (b) 3.1 meters per second, (c) 3.2 meters per second, and (d) 3.3 meters per second.

    Figure 5 Canal diagram with a depth of 1 meter and a flow rate of 3 meters per second.

    Figure 6 Canal diagram with a depth of 1 meter and a flow rate of 3.1 meters per second.

    Figure 7 Canal diagram with a depth of 1 meter and a flow rate of 3.2 meters per second.

    Figure 8 Canal diagram with a depth of 1 meter and a flow rate of 3.3 meters per second.(a)
    (a)(b)
    (b)(c)
    (c)(d)
    (d)(a)
    (a)(b)
    (b)(c)
    (c)(d)
    (d)Figure 9 Flow velocity profiles for canals with a depth of 2 m and flow velocities of 4–4.3 m/s. Canal with a depth of 2 meters and a flow rate of (a) 4 meters per second, (b) 4.1 meters per second, (c) 4.2 meters per second, and (d) 4.3 meters per second.

    Figure 10 Canal diagram with a depth of 2 meters and a flow rate of 4 meters per second.

    Figure 11 Canal diagram with a depth of 2 meters and a flow rate of 4.1 meters per second.

    Figure 12 Canal diagram with a depth of 2 meters and a flow rate of 4.2 meters per second.

    Figure 13 Canal diagram with a depth of 2 meters and a flow rate of 4.3 meters per second.(a)
    (a)(b)
    (b)(c)
    (c)(d)
    (d)(a)
    (a)(b)
    (b)(c)
    (c)(d)
    (d)Figure 14 Flow velocity profiles for canals with a depth of 3 m and flow velocities of 5–5.3 m/s. Canal with a depth of 2 meters and a flow rate of (a) 4 meters per second, (b) 4.1 meters per second, (c) 4.2 meters per second, and (d) 4.3 meters per second.

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

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

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

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

    According to Figure 11, we see that, with increasing Manning’s coefficient, the canal bed speed decreases. But this deceleration is more noticeable than the deceleration of Figures 510, which can be justified by increasing the speed and, of course, increasing the Froude number.

    With increasing Manning’s coefficient, the canal bed speed decreases (Figure 12). But this deceleration is more noticeable than the deceleration of the higher models (Figures 58 and 1011), which can be justified by increasing the speed and, of course, increasing the Froude number.

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

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

    Figure 15 Canal diagram with a depth of 3 meters and a flow rate of 5 meters per second.

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

    Figure 16 Canal diagram with a depth of 3 meters and a flow rate of 5.1 meters per second.

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

    Figure 17 Canal diagram with a depth of 3 meters and a flow rate of 5.2 meters per second.

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

    Figure 18 Canal diagram with a depth of 3 meters and a flow rate of 5.3 meters per second.

    According to Figure 19, it can be seen that the vegetation placed in front of the flow input velocity has negligible effect on the reduction of velocity, which of course can be justified due to the flexibility of the vegetation. The only unusual thing is the unexpected decrease in floor speed of 3 m/s compared to higher speeds.(a)
    (a)(b)
    (b)(c)
    (c)(a)
    (a)(b)
    (b)(c)
    (c)Figure 19 Comparison of velocity profiles with the same plant densities (depth 1 m). Comparison of velocity profiles with (a) plant densities of 25%, depth 1 m; (b) plant densities of 50%, depth 1 m; and (c) plant densities of 75%, depth 1 m.

    According to Figure 20, by increasing the speed of vegetation, the effect of vegetation on reducing the flow rate becomes more noticeable. And the role of input current does not have much effect in reducing speed.(a)
    (a)(b)
    (b)(c)
    (c)(a)
    (a)(b)
    (b)(c)
    (c)Figure 20 Comparison of velocity profiles with the same plant densities (depth 2 m). Comparison of velocity profiles with (a) plant densities of 25%, depth 2 m; (b) plant densities of 50%, depth 2 m; and (c) plant densities of 75%, depth 2 m.

    According to Figure 21, it can be seen that, with increasing speed, the effect of vegetation on reducing the bed flow rate becomes more noticeable and the role of the input current does not have much effect. In general, it can be seen that, by increasing the speed of the input current, the slope of the profiles increases from the bed to the water surface and due to the fact that, in software, the roughness coefficient applies to the channel floor only in the boundary conditions, this can be perfectly justified. Of course, it can be noted that, due to the flexible conditions of the vegetation of the bed, this modeling can show acceptable results for such grasses in the canal floor. In the next directions, we may try application of swarm-based optimization methods for modeling and finding the most effective factors in this research [27815188994]. In future, we can also apply the simulation logic and software of this research for other domains such as power engineering [9599].(a)
    (a)(b)
    (b)(c)
    (c)(a)
    (a)(b)
    (b)(c)
    (c)Figure 21 Comparison of velocity profiles with the same plant densities (depth 3 m). Comparison of velocity profiles with (a) plant densities of 25%, depth 3 m; (b) plant densities of 50%, depth 3 m; and (c) plant densities of 75%, depth 3 m.

    3. Conclusion

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

    Nomenclature

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

    Data Availability

    All data are included within the paper.

    Conflicts of Interest

    The authors declare that they have no conflicts of interest.

    Acknowledgments

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

    References

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

    Conflict resolution in the multi-stakeholder stepped spillway design under uncertainty by machine learning techniques

    Md TusherMollah, Raphaël Comminal, Marcin P.Serdeczny, David B.Pedersen, Jon Spangenberg
    Department of Mechanical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark

    Abstract

    This paper presents computational fluid dynamics simulations of the deposition flow during printing of multiple layers in material extrusion additive manufacturing. The developed model predicts the morphology of the deposited layers and captures the layer deformations during the printing of viscoplastic materials. The physics is governed by the continuity and momentum equations with the Bingham constitutive model, formulated as a generalized Newtonian fluid. The cross-sectional shapes of the deposited layers are predicted, and the deformation of layers is studied for different constitutive parameters of the material. It is shown that the deformation of layers is due to the hydrostatic pressure of the printed material, as well as the extrusion pressure during the extrusion. The simulations show that a higher yield stress results in prints with less deformations, while a higher plastic viscosity leads to larger deformations in the deposited layers. Moreover, the influence of the printing speed, extrusion speed, layer height, and nozzle diameter on the deformation of the printed layers is investigated. Finally, the model provides a conservative estimate of the required increase in yield stress that a viscoplastic material demands after deposition in order to support the hydrostatic and extrusion pressure of the subsequently printed layers.

    이 논문은 재료 압출 적층 제조에서 여러 레이어를 인쇄하는 동안 증착 흐름의 전산 유체 역학 시뮬레이션을 제공합니다. 개발된 모델은 증착된 레이어의 형태를 예측하고 점소성 재료를 인쇄하는 동안 레이어 변형을 캡처합니다.

    물리학은 일반화된 뉴턴 유체로 공식화된 Bingham 구성 모델의 연속성 및 운동량 방정식에 의해 제어됩니다. 증착된 층의 단면 모양이 예측되고 재료의 다양한 구성 매개변수에 대해 층의 변형이 연구됩니다. 층의 변형은 인쇄물의 정수압과 압출시 압출압력으로 인한 것임을 알 수 있다.

    시뮬레이션에 따르면 항복 응력이 높을수록 변형이 적은 인쇄물이 생성되는 반면 플라스틱 점도가 높을수록 증착된 레이어에서 변형이 커집니다. 또한 인쇄 속도, 압출 속도, 층 높이 및 노즐 직경이 인쇄된 층의 변형에 미치는 영향을 조사했습니다.

    마지막으로, 이 모델은 후속 인쇄된 레이어의 정수압 및 압출 압력을 지원하기 위해 증착 후 점소성 재료가 요구하는 항복 응력의 필요한 증가에 대한 보수적인 추정치를 제공합니다.

    Stability and deformations of deposited layers in material extrusion additive manufacturing
    Stability and deformations of deposited layers in material extrusion additive manufacturing

    Keywords

    Viscoplastic MaterialsMaterial Extrusion Additive Manufacturing (MEX-AM)Multiple-Layers DepositionComputational Fluid Dynamics (CFD)Deformation Control

    Figure 17. Longitudinal turbulent kinetic energy distribution on the smooth and triangular macroroughnesses: (A) Y/2; (B) Y/6.

    Numerical Simulations of the Flow Field of a Submerged Hydraulic Jump over Triangular Macroroughnesses

    Triangular Macroroughnesses 대한 잠긴 수압 점프의 유동장 수치 시뮬레이션

    by Amir Ghaderi 1,2,Mehdi Dasineh 3,Francesco Aristodemo 2 andCostanza Aricò 4,*1Department of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan 537138791, Iran2Department of Civil Engineering, University of Calabria, Arcavacata, 87036 Rende, Italy3Department of Civil Engineering, Faculty of Engineering, University of Maragheh, Maragheh 8311155181, Iran4Department of Engineering, University of Palermo, Viale delle Scienze, 90128 Palermo, Italy*Author to whom correspondence should be addressed.Academic Editor: Anis YounesWater202113(5), 674; https://doi.org/10.3390/w13050674

    Abstract

    The submerged hydraulic jump is a sudden change from the supercritical to subcritical flow, specified by strong turbulence, air entrainment and energy loss. Despite recent studies, hydraulic jump characteristics in smooth and rough beds, the turbulence, the mean velocity and the flow patterns in the cavity region of a submerged hydraulic jump in the rough beds, especially in the case of triangular macroroughnesses, are not completely understood. The objective of this paper was to numerically investigate via the FLOW-3D model the effects of triangular macroroughnesses on the characteristics of submerged jump, including the longitudinal profile of streamlines, flow patterns in the cavity region, horizontal velocity profiles, streamwise velocity distribution, thickness of the inner layer, bed shear stress coefficient, Turbulent Kinetic Energy (TKE) and energy loss, in different macroroughness arrangements and various inlet Froude numbers (1.7 < Fr1 < 9.3). To verify the accuracy and reliability of the present numerical simulations, literature experimental data were considered.

    Keywords: submerged hydraulic jumptriangular macroroughnessesTKEbed shear stress coefficientvelocityFLOW-3D model

    수중 유압 점프는 강한 난류, 공기 동반 및 에너지 손실로 지정된 초임계에서 아임계 흐름으로의 급격한 변화입니다. 최근 연구에도 불구하고, 특히 삼각형 거시적 거칠기의 경우, 평활 및 거친 베드에서의 수압 점프 특성, 거친 베드에서 잠긴 수압 점프의 공동 영역에서 난류, 평균 속도 및 유동 패턴이 완전히 이해되지 않았습니다.

    이 논문의 목적은 유선의 종방향 프로파일, 캐비티 영역의 유동 패턴, 수평 속도 프로파일, 스트림 방향 속도 분포, 두께를 포함하여 서브머지드 점프의 특성에 대한 삼각형 거시 거칠기의 영향을 FLOW-3D 모델을 통해 수치적으로 조사하는 것이었습니다.

    내부 층의 층 전단 응력 계수, 난류 운동 에너지(TKE) 및 에너지 손실, 다양한 거시 거칠기 배열 및 다양한 입구 Froude 수(1.7 < Fr1 < 9.3). 현재 수치 시뮬레이션의 정확성과 신뢰성을 검증하기 위해 문헌 실험 데이터를 고려했습니다.

     Introduction

    격렬한 난류 혼합과 기포 동반이 있는 수압 점프는 초임계에서 아임계 흐름으로의 변화 과정으로 간주됩니다[1]. 자유 및 수중 유압 점프는 일반적으로 게이트, 배수로 및 둑과 같은 수력 구조 아래의 에너지 손실에 적합합니다. 매끄러운 베드에서 유압 점프의 특성은 널리 연구되었습니다[2,3,4,5,6,7,8,9].

    베드의 거칠기 요소가 매끄러운 베드와 비교하여 수압 점프의 특성에 어떻게 영향을 미치는지 예측하기 위해 거시적 거칠기에 대한 자유 및 수중 수력 점프에 대해 여러 실험 및 수치 연구가 수행되었습니다. Ead와 Rajaratnam[10]은 사인파 거대 거칠기에 대한 수리학적 점프의 특성을 조사하고 무차원 분석을 통해 수면 프로파일과 배출을 정규화했습니다.

    Tokyayet al. [11]은 두 사인 곡선 거대 거칠기에 대한 점프 길이 비율과 에너지 손실이 매끄러운 베드보다 각각 35% 더 작고 6% 더 높다는 것을 관찰했습니다. Abbaspur et al. [12]는 6개의 사인파형 거대 거칠기에 대한 수력학적 점프의 특성을 연구했습니다. 그 결과, 꼬리수심과 점프길이는 평상보다 낮았고 Froude 수는 점프길이에 큰 영향을 미쳤습니다.

    Shafai-Bejestan과 Neisi[13]는 수압 점프에 대한 마름모꼴 거대 거칠기의 영향을 조사했습니다. 결과는 마름모꼴 거시 거칠기를 사용하면 매끄러운 침대와 비교하여 꼬리 수심과 점프 길이를 감소시키는 것으로 나타났습니다. Izadjoo와 Shafai-Bejestan[14]은 다양한 사다리꼴 거시 거칠기에 대한 수압 점프를 연구했습니다.

    그들은 전단응력계수가 평활층보다 10배 이상 크고 점프길이가 50% 감소하는 것을 관찰하였습니다. Nikmehr과 Aminpour[15]는 Flow-3D 모델 버전 11.2[16]를 사용하여 사다리꼴 블록이 있는 거시적 거칠기에 대한 수력학적 점프의 특성을 조사했습니다. 결과는 거시 거칠기의 높이와 거리가 증가할수록 전단 응력 계수뿐만 아니라 베드 근처에서 속도가 감소하는 것으로 나타났습니다.

    Ghaderi et al. [17]은 다양한 형태의 거시 거칠기(삼각형, 정사각형 및 반 타원형)에 대한 자유 및 수중 수력 점프 특성을 연구했습니다. 결과는 Froude 수의 증가에 따라 자유 및 수중 점프에서 전단 응력 계수, 에너지 손실, 수중 깊이, 미수 깊이 및 상대 점프 길이가 증가함을 나타냅니다.

    자유 및 수중 점프에서 가장 높은 전단 응력과 에너지 손실은 삼각형의 거시 거칠기가 존재할 때 발생했습니다. Elsebaie와 Shabayek[18]은 5가지 형태의 거시적 거칠기(삼각형, 사다리꼴, 2개의 측면 경사 및 직사각형이 있는 정현파)에 대한 수력학적 점프의 특성을 연구했습니다. 결과는 모든 거시적 거칠기에 대한 에너지 손실이 매끄러운 베드에서보다 15배 이상이라는 것을 보여주었습니다.

    Samadi-Boroujeni et al. [19]는 다양한 각도의 6개의 삼각형 거시 거칠기에 대한 수력 점프를 조사한 결과 삼각형 거시 거칠기가 평활 베드에 비해 점프 길이를 줄이고 에너지 손실과 베드 전단 응력 계수를 증가시키는 것으로 나타났습니다.

    Ahmed et al. [20]은 매끄러운 베드와 삼각형 거시 거칠기에서 수중 수력 점프 특성을 조사했습니다. 결과는 부드러운 침대와 비교할 때 잠긴 깊이와 점프 길이가 감소했다고 밝혔습니다. 표 1은 다른 연구자들이 제시한 과거의 유압 점프에 대한 실험 및 수치 연구의 세부 사항을 나열합니다.

    Table 1. Main characteristics of some past experimental and numerical studies on hydraulic jumps.

    ReferenceShape Bed-Channel Type-
    Jump Type
    Channel Dimension (m)Roughness (mm)Fr1Investigated Flow
    Properties
    Ead and Rajaratnam [10]-Smooth and rough beds-Rectangular channel-Free jumpCL1 = 7.60
    CW2 = 0.44
    CH3 = 0.60
    -Corrugated sheets (RH4 = 13 and 22)4–10-Upstream and tailwater depths-Jump length-Roller length-Velocity-Water surface profile
    Tokyay et al. [11]-Smooth and rough beds-Rectangular channel-Free jumpCL = 10.50
    CW = 0.253
    CH = 0.432
    -Two sinusoidal corrugated (RH = 10 and 13)5–12-Depth ratio-Jump length-Energy loss
    Izadjoo and Shafai-Bejestan [14]-Smooth and rough beds-Two rectangular-channel-Free jumpCL = 1.2, 9
    CW = 0.25, 0.50
    CH = 0.40
    Baffle with trapezoidal cross section
    (RH: 13 and 26)
    6–12-Upstream and tailwater depths-Jump length-Velocity-Bed shear stress coefficient
    Abbaspour et al. [12]-Horizontal bed with slope 0.002-Rectangular channel—smooth and rough beds-Free jumpCL = 10
    CW = 0.25
    CH = 0.50
    -Sinusoidal bed (RH = 15,20, 25 and 35)3.80–8.60-Water surface profile-Depth ratio-Jump length-Energy loss-Velocity profiles-Bed shear stress coefficient
    Shafai-Bejestan and Neisi [13]-Smooth and rough beds-Rectangular channel-Free jumpCL = 7.50
    CW = 0.35
    CH = 0.50
    Lozenge bed4.50–12-Sequent depth-Jump length
    Elsebaie and Shabayek [18]-Smooth and rough beds-Rectangular channel-With side slopes of 45 degrees for two trapezoidal and triangular macroroughnesses and of 60 degrees for other trapezoidal macroroughnesses-Free jumpCL = 9
    CW = 0.295
    CH = 0.32
    -Sinusoidal-Triangular-Trapezoidal with two side-Rectangular-(RH = 18 and corrugation wavelength = 65)50-Water surface profile-Sequent depth-Jump length-Bed shear stress coefficient
    Samadi-Boroujeni et al. [19]-Rectangular channel-Smooth and rough beds-Free jumpCL = 12
    CW = 0.40
    CH = 0.40
    -Six triangular corrugated (RH = 2.5)6.10–13.10-Water surface profile-Sequent depth-Jump length-Energy loss-Velocity profiles-Bed shear stress coefficient
    Ahmed et al. [20]-Smooth and rough beds-Rectangular channel-Submerged jumpCL = 24.50
    CW = 0.75
    CH = 0.70
    -Triangular corrugated sheet (RH = 40)1.68–9.29-Conjugated and tailwater depths-Submerged ratio-Deficit depth-Relative jump length-Jump length-Relative roller jump length-Jump efficiency-Bed shear stress coefficient
    Nikmehr and Aminpour [15]-Horizontal bed with slope 0.002-Rectangular channel-Rough bed-Free jumpCL = 12
    CW = 0.25
    CH = 0.50
    -Trapezoidal blocks (RH = 2, 3 and 4)5.01–13.70-Water surface profile-Sequent depth-Jump length-Roller length-Velocity
    Ghaderi et al. [17]-Smooth and rough beds-Rectangular channel-Free and submerged jumpCL = 4.50
    CW = 0.75
    CH = 0.70
    -Triangular, square and semi-oval macroroughnesses (RH = 40 and distance of roughness of I = 40, 80, 120, 160 and 200)1.70–9.30-Horizontal velocity distributions-Bed shear stress coefficient-Sequent depth ratio and submerged depth ratio-Jump length-Energy loss
    Present studyRectangular channel
    Smooth and rough beds
    Submerged jump
    CL = 4.50
    CW = 0.75
    CH = 0.70
    -Triangular macroroughnesses (RH = 40 and distance of roughness of I = 40, 80, 120, 160 and 200)1.70–9.30-Longitudinal profile of streamlines-Flow patterns in the cavity region-Horizontal velocity profiles-Streamwise velocity distribution-Bed shear stress coefficient-TKE-Thickness of the inner layer-Energy loss

    CL1: channel length, CW2: channel width, CH3: channel height, RH4: roughness height.

    이전에 논의된 조사의 주요 부분은 실험실 접근 방식을 기반으로 하며 사인파, 마름모꼴, 사다리꼴, 정사각형, 직사각형 및 삼각형 매크로 거칠기가 공액 깊이, 잠긴 깊이, 점프 길이, 에너지 손실과 같은 일부 자유 및 수중 유압 점프 특성에 어떻게 영향을 미치는지 조사합니다.

    베드 및 전단 응력 계수. 더욱이, 저자[17]에 의해 다양한 형태의 거시적 거칠기에 대한 수력학적 점프에 대한 이전 발표된 논문을 참조하면, 삼각형의 거대조도는 가장 높은 층 전단 응력 계수 및 에너지 손실을 가지며 또한 가장 낮은 잠긴 깊이, tailwater를 갖는 것으로 관찰되었습니다.

    다른 거친 모양, 즉 정사각형 및 반 타원형과 부드러운 침대에 비해 깊이와 점프 길이. 따라서 본 논문에서는 삼각형 매크로 거칠기를 사용하여(일정한 거칠기 높이가 T = 4cm이고 삼각형 거칠기의 거리가 I = 4, 8, 12, 16 및 20cm인 다른 T/I 비율에 대해), 특정 캐비티 영역의 유동 패턴, 난류 운동 에너지(TKE) 및 흐름 방향 속도 분포와 같은 연구가 필요합니다.

    CFD(Computational Fluid Dynamics) 방법은 자유 및 수중 유압 점프[21]와 같은 복잡한 흐름의 모델링 프로세스를 수행하는 중요한 도구로 등장하며 수중 유압 점프의 특성은 CFD 시뮬레이션을 사용하여 정확하게 예측할 수 있습니다 [22,23 ].

    본 논문은 초기에 수중 유압 점프의 주요 특성, 수치 모델에 대한 입력 매개변수 및 Ahmed et al.의 참조 실험 조사를 제시합니다. [20], 검증 목적으로 보고되었습니다. 또한, 본 연구에서는 유선의 종방향 프로파일, 캐비티 영역의 유동 패턴, 수평 속도 프로파일, 내부 층의 두께, 베드 전단 응력 계수, TKE 및 에너지 손실과 같은 특성을 조사할 것입니다.

    Figure 1. Definition sketch of a submerged hydraulic jump at triangular macroroughnesses.
    Figure 1. Definition sketch of a submerged hydraulic jump at triangular macroroughnesses.

    Table 2. Effective parameters in the numerical model.

    Bed TypeQ
    (l/s)
    I
    (cm)
    T (cm)d (cm)y1
    (cm)
    y4
    (cm)
    Fr1= u1/(gy1)0.5SRe1= (u1y1)/υ
    Smooth30, 4551.62–3.839.64–32.101.7–9.30.26–0.5039,884–59,825
    Triangular macroroughnesses30, 454, 8, 12, 16, 20451.62–3.846.82–30.081.7–9.30.21–0.4439,884–59,825
    Figure 2. Longitudinal profile of the experimental flume (Ahmed et al. [20]).
    Figure 2. Longitudinal profile of the experimental flume (Ahmed et al. [20]).

    Table 3. Main flow variables for the numerical and physical models (Ahmed et al. [20]).

    ModelsBed TypeQ (l/s)d (cm)y1 (cm)u1 (m/s)Fr1
    Numerical and PhysicalSmooth4551.62–3.831.04–3.701.7–9.3
    T/I = 0.54551.61–3.831.05–3.711.7–9.3
    T/I = 0.254551.60–3.841.04–3.711.7–9.3
    Figure 3. The boundary conditions governing the simulations.
    Figure 3. The boundary conditions governing the simulations.
    Figure 4. Sketch of mesh setup.
    Figure 4. Sketch of mesh setup.

    Table 4. Characteristics of the computational grids.

    MeshNested Block Cell Size (cm)Containing Block Cell Size (cm)
    10.551.10
    20.651.30
    30.851.70

    Table 5. The numerical results of mesh convergence analysis.

    ParametersAmounts
    fs1 (-)7.15
    fs2 (-)6.88
    fs3 (-)6.19
    K (-)5.61
    E32 (%)10.02
    E21 (%)3.77
    GCI21 (%)3.03
    GCI32 (%)3.57
    GCI32/rp GCI210.98
    Figure 5. Time changes of the flow discharge in the inlet and outlet boundaries conditions (A): Q = 0.03 m3/s (B): Q = 0.045 m3/s.
    Figure 5. Time changes of the flow discharge in the inlet and outlet boundaries conditions (A): Q = 0.03 m3/s (B): Q = 0.045 m3/s.
    Figure 6. The evolutionary process of a submerged hydraulic jump on the smooth bed—Q = 0.03 m3/s.
    Figure 6. The evolutionary process of a submerged hydraulic jump on the smooth bed—Q = 0.03 m3/s.
    Figure 7. Numerical versus experimental basic parameters of the submerged hydraulic jump. (A): y3/y1; and (B): y4/y1.
    Figure 7. Numerical versus experimental basic parameters of the submerged hydraulic jump. (A): y3/y1; and (B): y4/y1.
    Figure 8. Velocity vector field and flow pattern through the gate in a submerged hydraulic jump condition: (A) smooth bed; (B) triangular macroroughnesses.
    Figure 8. Velocity vector field and flow pattern through the gate in a submerged hydraulic jump condition: (A) smooth bed; (B) triangular macroroughnesses.
    Figure 9. Velocity vector distributions in the x–z plane (y = 0) within the cavity region.
    Figure 9. Velocity vector distributions in the x–z plane (y = 0) within the cavity region.
    Figure 10. Typical vertical distribution of the mean horizontal velocity in a submerged hydraulic jump [46].
    Figure 10. Typical vertical distribution of the mean horizontal velocity in a submerged hydraulic jump [46].
    Figure 11. Typical horizontal velocity profiles in a submerged hydraulic jump on smooth bed and triangular macroroughnesses.
    Figure 11. Typical horizontal velocity profiles in a submerged hydraulic jump on smooth bed and triangular macroroughnesses.
    Figure 12. Horizontal velocity distribution at different distances from the sluice gate for the different T/I for Fr1 = 6.1
    Figure 12. Horizontal velocity distribution at different distances from the sluice gate for the different T/I for Fr1 = 6.1
    Figure 13. Stream-wise velocity distribution for the triangular macroroughnesses with T/I = 0.5 and 0.25.
    Figure 13. Stream-wise velocity distribution for the triangular macroroughnesses with T/I = 0.5 and 0.25.
    Figure 14. Dimensionless horizontal velocity distribution in the submerged hydraulic jump for different Froude numbers in triangular macroroughnesses.
    Figure 14. Dimensionless horizontal velocity distribution in the submerged hydraulic jump for different Froude numbers in triangular macroroughnesses.
    Figure 15. Spatial variations of (umax/u1) and (δ⁄y1).
    Figure 15. Spatial variations of (umax/u1) and (δ⁄y1).
    Figure 16. The shear stress coefficient (ε) versus the inlet Froude number (Fr1).
    Figure 16. The shear stress coefficient (ε) versus the inlet Froude number (Fr1).
    Figure 17. Longitudinal turbulent kinetic energy distribution on the smooth and triangular macroroughnesses: (A) Y/2; (B) Y/6.
    Figure 17. Longitudinal turbulent kinetic energy distribution on the smooth and triangular macroroughnesses: (A) Y/2; (B) Y/6.
    Figure 18. The energy loss (EL/E3) of the submerged jump versus inlet Froude number (Fr1).
    Figure 18. The energy loss (EL/E3) of the submerged jump versus inlet Froude number (Fr1).

    Conclusions

    • 본 논문에서는 유선의 종방향 프로파일, 공동 영역의 유동 패턴, 수평 속도 프로파일, 스트림 방향 속도 분포, 내부 층의 두께, 베드 전단 응력 계수, 난류 운동 에너지(TKE)를 포함하는 수중 유압 점프의 특성을 제시하고 논의했습니다. ) 및 삼각형 거시적 거칠기에 대한 에너지 손실. 이러한 특성은 FLOW-3D® 모델을 사용하여 수치적으로 조사되었습니다. 자유 표면을 시뮬레이션하기 위한 VOF(Volume of Fluid) 방법과 난류 RNG k-ε 모델이 구현됩니다. 본 모델을 검증하기 위해 평활층과 삼각형 거시 거칠기에 대해 수치 시뮬레이션과 실험 결과를 비교했습니다. 본 연구의 다음과 같은 결과를 도출할 수 있다.
    • 개발 및 개발 지역의 삼각형 거시 거칠기의 흐름 패턴은 수중 유압 점프 조건의 매끄러운 바닥과 비교하여 더 작은 영역에서 동일합니다. 삼각형의 거대 거칠기는 거대 거칠기 사이의 공동 영역에서 또 다른 시계 방향 와류의 형성으로 이어집니다.
    • T/I = 1, 0.5 및 0.33과 같은 거리에 대해 속도 벡터 분포는 캐비티 영역에서 시계 방향 소용돌이를 표시하며, 여기서 속도의 크기는 평균 유속보다 훨씬 작습니다. 삼각형 거대 거칠기(T/I = 0.25 및 0.2) 사이의 거리를 늘리면 캐비티 영역에 크기가 다른 두 개의 소용돌이가 형성됩니다.
    • 삼각형 거시조도 사이의 거리가 충분히 길면 흐름이 다음 조도에 도달할 때까지 속도 분포가 회복됩니다. 그러나 짧은 거리에서 흐름은 속도 분포의 적절한 회복 없이 다음 거칠기에 도달합니다. 따라서 거시 거칠기 사이의 거리가 감소함에 따라 마찰 계수의 증가율이 감소합니다.
    • 삼각형의 거시적 거칠기에서, 잠수 점프의 지정된 섹션에서 최대 속도는 자유 점프보다 높은 값으로 이어집니다. 또한, 수중 점프에서 두 가지 유형의 베드(부드러움 및 거친 베드)에 대해 깊이 및 와류 증가로 인해 베드로부터의 최대 속도 거리는 감소합니다. 잠수 점프에서 경계층 두께는 자유 점프보다 얇습니다.
    • 매끄러운 베드의 난류 영역은 게이트로부터의 거리에 따라 생성되고 자유 표면 롤러 영역 근처에서 발생하는 반면, 거시적 거칠기에서는 난류가 게이트 근처에서 시작되어 더 큰 강도와 제한된 스위프 영역으로 시작됩니다. 이는 반시계 방향 순환의 결과입니다. 거시 거칠기 사이의 공간에서 자유 표면 롤러 및 시계 방향 와류.
    • 삼각 거시 거칠기에서 침지 점프의 베드 전단 응력 계수와 에너지 손실은 유입구 Froude 수의 증가에 따라 증가하는 매끄러운 베드에서 발견된 것보다 더 큽니다. T/I = 0.50 및 0.20에서 최고 및 최저 베드 전단 응력 계수 및 에너지 손실이 평활 베드에 비해 거칠기 요소의 거리가 증가함에 따라 발생합니다.
    • 거의 거칠기 요소가 있는 삼각형 매크로 거칠기의 존재에 의해 주어지는 점프 길이와 잠긴 수심 및 꼬리 수심의 감소는 결과적으로 크기, 즉 길이 및 높이가 감소하는 정수조 설계에 사용될 수 있습니다.
    • 일반적으로 CFD 모델은 다양한 수력 조건 및 기하학적 배열을 고려하여 잠수 점프의 특성 예측을 시뮬레이션할 수 있습니다. 캐비티 영역의 흐름 패턴, 흐름 방향 및 수평 속도 분포, 베드 전단 응력 계수, TKE 및 유압 점프의 에너지 손실은 수치적 방법으로 시뮬레이션할 수 있습니다. 그러나 거시적 차원과 유동장 및 공동 유동의 변화에 ​​대한 다양한 배열에 대한 연구는 향후 과제로 남아 있다.

    References

    1. White, F.M. Viscous Fluid Flow, 2nd ed.; McGraw-Hill University of Rhode Island: Montreal, QC, Canada, 1991. [Google Scholar]
    2. Launder, B.E.; Rodi, W. The turbulent wall jet. Prog. Aerosp. Sci. 197919, 81–128. [Google Scholar] [CrossRef]
    3. McCorquodale, J.A. Hydraulic jumps and internal flows. In Encyclopedia of Fluid Mechanics; Cheremisinoff, N.P., Ed.; Golf Publishing: Houston, TX, USA, 1986; pp. 120–173. [Google Scholar]
    4. Federico, I.; Marrone, S.; Colagrossi, A.; Aristodemo, F.; Antuono, M. Simulating 2D open-channel flows through an SPH model. Eur. J. Mech. B Fluids 201234, 35–46. [Google Scholar] [CrossRef]
    5. Khan, S.A. An analytical analysis of hydraulic jump in triangular channel: A proposed model. J. Inst. Eng. India Ser. A 201394, 83–87. [Google Scholar] [CrossRef]
    6. Mortazavi, M.; Le Chenadec, V.; Moin, P.; Mani, A. Direct numerical simulation of a turbulent hydraulic jump: Turbulence statistics and air entrainment. J. Fluid Mech. 2016797, 60–94. [Google Scholar] [CrossRef]
    7. Daneshfaraz, R.; Ghahramanzadeh, A.; Ghaderi, A.; Joudi, A.R.; Abraham, J. Investigation of the effect of edge shape on characteristics of flow under vertical gates. J. Am. Water Works Assoc. 2016108, 425–432. [Google Scholar] [CrossRef]
    8. Azimi, H.; Shabanlou, S.; Kardar, S. Characteristics of hydraulic jump in U-shaped channels. Arab. J. Sci. Eng. 201742, 3751–3760. [Google Scholar] [CrossRef]
    9. De Padova, D.; Mossa, M.; Sibilla, S. SPH numerical investigation of characteristics of hydraulic jumps. Environ. Fluid Mech. 201818, 849–870. [Google Scholar] [CrossRef]
    10. Ead, S.A.; Rajaratnam, N. Hydraulic jumps on corrugated beds. J. Hydraul. Eng. 2002128, 656–663. [Google Scholar] [CrossRef]
    11. Tokyay, N.D. Effect of channel bed corrugations on hydraulic jumps. In Proceedings of the World Water and Environmental Resources Congress 2005, Anchorage, AK, USA, 15–19 May 2005; pp. 1–9. [Google Scholar]
    12. Abbaspour, A.; Dalir, A.H.; Farsadizadeh, D.; Sadraddini, A.A. Effect of sinusoidal corrugated bed on hydraulic jump characteristics. J. Hydro-Environ. Res. 20093, 109–117. [Google Scholar] [CrossRef]
    13. Shafai-Bejestan, M.S.; Neisi, K. A new roughened bed hydraulic jump stilling basin. Asian J. Appl. Sci. 20092, 436–445. [Google Scholar] [CrossRef]
    14. Izadjoo, F.; Shafai-Bejestan, M. Corrugated bed hydraulic jump stilling basin. J. Appl. Sci. 20077, 1164–1169. [Google Scholar] [CrossRef]
    15. Nikmehr, S.; Aminpour, Y. Numerical Simulation of Hydraulic Jump over Rough Beds. Period. Polytech. Civil Eng. 201764, 396–407. [Google Scholar] [CrossRef]
    16. Flow Science Inc. FLOW-3D V 11.2 User’s Manual; Flow Science Inc.: Santa Fe, NM, USA, 2016. [Google Scholar]
    17. Ghaderi, A.; Dasineh, M.; Aristodemo, F.; Ghahramanzadeh, A. Characteristics of free and submerged hydraulic jumps over different macroroughnesses. J. Hydroinform. 202022, 1554–1572. [Google Scholar] [CrossRef]
    18. Elsebaie, I.H.; Shabayek, S. Formation of hydraulic jumps on corrugated beds. Int. J. Civil Environ. Eng. IJCEE–IJENS 201010, 37–47. [Google Scholar]
    19. Samadi-Boroujeni, H.; Ghazali, M.; Gorbani, B.; Nafchi, R.F. Effect of triangular corrugated beds on the hydraulic jump characteristics. Can. J. Civil Eng. 201340, 841–847. [Google Scholar] [CrossRef]
    20. Ahmed, H.M.A.; El Gendy, M.; Mirdan, A.M.H.; Ali, A.A.M.; Haleem, F.S.F.A. Effect of corrugated beds on characteristics of submerged hydraulic jump. Ain Shams Eng. J. 20145, 1033–1042. [Google Scholar] [CrossRef]
    21. Viti, N.; Valero, D.; Gualtieri, C. Numerical simulation of hydraulic jumps. Part 2: Recent results and future outlook. Water 201911, 28. [Google Scholar] [CrossRef]
    22. Gumus, V.; Simsek, O.; Soydan, N.G.; Akoz, M.S.; Kirkgoz, M.S. Numerical modeling of submerged hydraulic jump from a sluice gate. J. Irrig. Drain. Eng. 2016142, 04015037. [Google Scholar] [CrossRef]
    23. Jesudhas, V.; Roussinova, V.; Balachandar, R.; Barron, R. Submerged hydraulic jump study using DES. J. Hydraul. Eng. 2017143, 04016091. [Google Scholar] [CrossRef]
    24. Rajaratnam, N. The hydraulic jump as a wall jet. J. Hydraul. Div. 196591, 107–132. [Google Scholar] [CrossRef]
    25. Hager, W.H. Energy Dissipaters and Hydraulic Jump; Kluwer Academic Publisher: Dordrecht, The Netherlands, 1992; pp. 185–224. [Google Scholar]
    26. Long, D.; Steffler, P.M.; Rajaratnam, N. LDA study of flow structure in submerged Hydraulic jumps. J. Hydraul. Res. 199028, 437–460. [Google Scholar] [CrossRef]
    27. Chow, V.T. Open Channel Hydraulics; McGraw-Hill: New York, NY, USA, 1959. [Google Scholar]
    28. Wilcox, D.C. Turbulence Modeling for CFD, 3rd ed.; DCW Industries, Inc.: La Canada, CA, USA, 2006. [Google Scholar]
    29. Hirt, C.W.; Nichols, B.D. Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 198139, 201–225. [Google Scholar] [CrossRef]
    30. Pourshahbaz, H.; Abbasi, S.; Pandey, M.; Pu, J.H.; Taghvaei, P.; Tofangdar, N. Morphology and hydrodynamics numerical simulation around groynes. ISH J. Hydraul. Eng. 2020, 1–9. [Google Scholar] [CrossRef]
    31. Choufu, L.; Abbasi, S.; Pourshahbaz, H.; Taghvaei, P.; Tfwala, S. Investigation of flow, erosion, and sedimentation pattern around varied groynes under different hydraulic and geometric conditions: A numerical study. Water 201911, 235. [Google Scholar] [CrossRef]
    32. Zhenwei, Z.; Haixia, L. Experimental investigation on the anisotropic tensorial eddy viscosity model for turbulence flow. Int. J. Heat Technol. 201634, 186–190. [Google Scholar]
    33. Carvalho, R.; Lemos Ramo, C. Numerical computation of the flow in hydraulic jump stilling basins. J. Hydraul. Res. 200846, 739–752. [Google Scholar] [CrossRef]
    34. Bayon, A.; Valero, D.; García-Bartual, R.; López-Jiménez, P.A. Performance assessment of Open FOAM and FLOW-3D in the numerical modeling of a low Reynolds number hydraulic jump. Environ. Model. Softw. 201680, 322–335. [Google Scholar] [CrossRef]
    35. Daneshfaraz, R.; Ghaderi, A.; Akhtari, A.; Di Francesco, S. On the Effect of Block Roughness in Ogee Spillways with Flip Buckets. Fluids 20205, 182. [Google Scholar] [CrossRef]
    36. Ghaderi, A.; Abbasi, S. CFD simulation of local scouring around airfoil-shaped bridge piers with and without collar. Sādhanā 201944, 216. [Google Scholar] [CrossRef]
    37. Ghaderi, A.; Daneshfaraz, R.; Dasineh, M.; Di Francesco, S. Energy Dissipation and Hydraulics of Flow over Trapezoidal–Triangular Labyrinth Weirs. Water 202012, 1992. [Google Scholar] [CrossRef]
    38. Ghaderi, A.; Abbasi, S.; Abraham, J.; Azamathulla, H.M. Efficiency of trapezoidal labyrinth shaped stepped spillways. Flow Meas. Instrum. 202072, 101711. [Google Scholar] [CrossRef]
    39. Yakhot, V.; Orszag, S.A. Renormalization group analysis of turbulence. I. basic theory. J. Sci. Comput. 19861, 3–51. [Google Scholar] [CrossRef] [PubMed]
    40. Biscarini, C.; Di Francesco, S.; Ridolfi, E.; Manciola, P. On the simulation of floods in a narrow bending valley: The malpasset dam break case study. Water 20168, 545. [Google Scholar] [CrossRef]
    41. Ghaderi, A.; Daneshfaraz, R.; Abbasi, S.; Abraham, J. Numerical analysis of the hydraulic characteristics of modified labyrinth weirs. Int. J. Energy Water Resour. 20204, 425–436. [Google Scholar] [CrossRef]
    42. Alfonsi, G. Reynolds-averaged Navier–Stokes equations for turbulence modeling. Appl. Mech. Rev. 200962. [Google Scholar] [CrossRef]
    43. Abbasi, S.; Fatemi, S.; Ghaderi, A.; Di Francesco, S. The Effect of Geometric Parameters of the Antivortex on a Triangular Labyrinth Side Weir. Water 202113, 14. [Google Scholar] [CrossRef]
    44. Celik, I.B.; Ghia, U.; Roache, P.J. Procedure for estimation and reporting of uncertainty due to discretization in CFD applications. J. Fluids Eng. 2008130, 0780011–0780013. [Google Scholar]
    45. Khan, M.I.; Simons, R.R.; Grass, A.J. Influence of cavity flow regimes on turbulence diffusion coefficient. J. Vis. 20069, 57–68. [Google Scholar] [CrossRef]
    46. Javanappa, S.K.; Narasimhamurthy, V.D. DNS of plane Couette flow with surface roughness. Int. J. Adv. Eng. Sci. Appl. Math. 2020, 1–13. [Google Scholar] [CrossRef]
    47. Nasrabadi, M.; Omid, M.H.; Farhoudi, J. Submerged hydraulic jump with sediment-laden flow. Int. J. Sediment Res. 201227, 100–111. [Google Scholar] [CrossRef]
    48. Pourabdollah, N.; Heidarpour, M.; Abedi Koupai, J. Characteristics of free and submerged hydraulic jumps in different stilling basins. In Water Management; Thomas Telford Ltd.: London, UK, 2019; pp. 1–11. [Google Scholar]
    49. Rajaratnam, N. Turbulent Jets; Elsevier Science: Amsterdam, The Netherlands, 1976. [Google Scholar]
    50. Aristodemo, F.; Marrone, S.; Federico, I. SPH modeling of plane jets into water bodies through an inflow/outflow algorithm. Ocean Eng. 2015105, 160–175. [Google Scholar] [CrossRef]
    51. Shekari, Y.; Javan, M.; Eghbalzadeh, A. Three-dimensional numerical study of submerged hydraulic jumps. Arab. J. Sci. Eng. 201439, 6969–6981. [Google Scholar] [CrossRef]
    52. Khan, A.A.; Steffler, P.M. Physically based hydraulic jump model for depth-averaged computations. J. Hydraul. Eng. 1996122, 540–548. [Google Scholar] [CrossRef]
    53. De Dios, M.; Bombardelli, F.A.; García, C.M.; Liscia, S.O.; Lopardo, R.A.; Parravicini, J.A. Experimental characterization of three-dimensional flow vortical structures in submerged hydraulic jumps. J. Hydro-Environ. Res. 201715, 1–12. [Google Scholar] [CrossRef]
    Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
    Fig. 12. Comparison of simulation results with experimental data for a flow rate of water = Ql=15 ml/hr and a flow rate of air = Qg =3 ml/hr.

    Simulation of Droplet Dynamics and Mixing in Microfluidic Devices using a VOF-Based Method

    Abstract

    This paper demonstrates that the Volume of Fluid (TruVOF) method in FLOW-3D (a general purpose CFD software) is an effective tool for studying droplet dynamics and mixing in microfluidic devices. The first example studied is a T-junction where flow patterns for both droplet generation and passive mixing are analyzed. The second example studied is a co-flowing device where the formation and breakup of bubbles is simulated. The effect of viscosity on bubble formation is also analyzed. For a T-junction the bubble size is corroborated with experimental data. Both the bubble size and frequency are studied and corroborated with experimental data for a co-flowing device. The third example studied is the electrowetting phenomenon observed in a small water droplet resting on a dielectric material. The steady-state contact angle is plotted against the voltage applied. The results are compared with both the Young-Lippmann curve and experimental results. 

    이 논문은 FLOW-3D (범용 CFD 소프트웨어)의 유체 부피 (TruVOF) 방법이 미세 유체 장치에서 액적 역학 및 혼합을 연구하는데 효과적인 도구임을 보여줍니다.

    연구된 첫 번째 예는 액적 생성 및 수동 혼합에 대한 흐름 패턴이 분석되는 T- 접합입니다. 연구된 두 번째 예는 기포의 형성 및 분해가 시뮬레이션 되는 동시 유동 장치입니다.

    기포 형성에 대한 점도의 영향도 분석됩니다. T 접합의 경우 기포 크기는 실험 데이터로 확증됩니다. 기포 크기와 빈도 모두 공동 유동 장치에 대한 실험 데이터로 연구되고 확증됩니다.

    연구된 세 번째 예는 유전 물질 위에 놓인 작은 물방울에서 관찰 된 전기 습윤 현상입니다. 정상 상태 접촉각은 적용된 전압에 대해 플롯됩니다. 결과는 Young-Lippmann 곡선 및 실험 결과와 비교됩니다.

    Simulation of Droplet Dynamics and Mixing in Microfluidic Devices using a VOF-Based Method Fig 1
    Simulation of Droplet Dynamics and Mixing in Microfluidic Devices using a VOF-Based Method Fig 1
    Simulation of Droplet Dynamics and Mixing in Microfluidic Devices using a VOF-Based Method Fig 2
    Simulation of Droplet Dynamics and Mixing in Microfluidic Devices using a VOF-Based Method Fig 2

    References

    Formation of bubbles in a simple co-flowing micro-channel

    SaveAlertResearch FeedFormation of droplets and bubbles in a microfluidic T-junction-scaling and mechanism of break-up.

    SaveAlertResearch FeedCreating, transporting, cutting, and merging liquid droplets by electrowetting-based actuation for digital microfluidic circuits,

    SaveAlertResearch FeedFLOW DEVELOPMENT OF CO-FLOWING STREAMS IN RECTANGULAR MICRO-CHANNELS

    SaveAlertResearch FeedA microfluidic system for controlling reaction networks in time.

    SaveAlertResearch FeedElectrowetting: from basics to applications

    SaveAlertResearch FeedVolume of fluid (VOF) method for the dynamics of free boundaries

    Fig.4 Schematic of a package structure

    Three-Dimensional Flow Analysis of a Thermosetting Compound during Mold Filling

    Junichi Saeki and Tsutomu Kono
    Production Engineering Research Laboratory, Hitachi Ltd.
    292, Y shida-cho, Totsuka-ku, Yokohama, 244-0817 Japan

    Abstract

    Thermosetting molding compounds are widely used for encapsulating semiconductor devices and electronic modules. In recent years, the number of electronic parts encapsulated in an electronic module has increased, in order to meet the requirements for high performance. As a result, the configuration of inserted parts during molding has become very complicated. Meanwhile, package thickness has been reduced in response to consumer demands for miniaturization. These trends have led to complicated flow patterns of molten compounds in a mold cavity, increasing the difficulty of predicting the occurrence of void formation or gold-wire deformation.

    A method of three-dimensional (3-D) flow analysis of thermosetting compounds has been developed with the objective of minimizing the trial term before mass production and of enhancing the quality of molded products. A constitutive equation model was developed to describe isothermal viscosity changes as a function of time and temperature. This isothermal model was used for predicting non-isothermal viscosity changes. In addition, an empirical model was developed for calculating the amount of wire deformation as a function of viscosity, wire configuration, and other parameters. These models were integrated with FLOW-3D® software, which is used for multipurpose 3-D flow analysis.

    The mold-filling dynamics of an epoxy compound were analyzed using the newly developed modeling software during transfer molding of an actual high performance electronic module. The changes in the 3-D distributions of parameters such as temperature, viscosity, velocity, and pressure were compared with the flow front patterns. The predicted results of cavity filling behavior corresponded well with actual short shot data. As well, the predicted amount of gold-wire deformation at each LSI chip with a substrate connection also corresponded well with observed data obtained by X-ray inspection of the molded product.

    Korea Abstract

    열경화성 몰딩 컴파운드는 반도체 장치 및 전자 모듈을 캡슐화하는 데 널리 사용됩니다. 최근에는 고성능에 대한 요구 사항을 충족시키기 위해 전자 모듈에 캡슐화되는 전자 부품의 수가 증가하고 있습니다.

    그 결과 성형시 삽입 부품의 구성이 매우 복잡해졌습니다. 한편, 소비자의 소형화 요구에 부응하여 패키지 두께를 줄였다. 이러한 경향은 몰드 캐비티에서 용융된 화합물의 복잡한 흐름 패턴을 야기하여 보이드 형성 또는 금선 변형의 발생을 예측하기 어렵게합니다.

    열경화성 화합물의 3 차원 (3-D) 유동 분석 방법은 대량 생산 전에 시험 기간을 최소화하고 성형 제품의 품질을 향상시킬 목적으로 개발되었습니다. 시간과 온도의 함수로서 등온 점도 변화를 설명하기 위해 구성 방정식 모델이 개발되었습니다. 이 등온 모델은 비등 온 점도 변화를 예측하는 데 사용되었습니다.

    또한 점도, 와이어 구성 및 기타 매개 변수의 함수로 와이어 변형량을 계산하기위한 경험적 모델이 개발되었습니다. 이 모델은 다목적 3D 흐름 분석에 사용되는 FLOW-3D® 소프트웨어와 통합되었습니다.

    실제 고성능 전자 모듈의 트랜스퍼 몰딩 과정에서 새로 개발 된 모델링 소프트웨어를 사용하여 에폭시 화합물의 몰드 충전 역학을 분석했습니다. 온도, 점도, 속도 및 압력과 같은 매개 변수의 3D 분포 변화를 유동 선단 패턴과 비교했습니다.

    캐비티 충전 거동의 예측 결과는 실제 미 성형 데이터와 잘 일치했습니다. 또한, 기판 연결이 있는 각 LSI 칩에서 예상되는 금선 변형량은 성형품의 X-ray 검사에서 얻은 관찰 데이터와도 잘 일치했습니다.

    Fig.1 A system of three-dimensional flow analysis for thermosetting compounds
    Fig.1 A system of three-dimensional flow analysis for thermosetting compounds
    Fig.2 Procedure for determining viscosity changes of thermosetting compounds
    Fig.2 Procedure for determining viscosity changes of thermosetting compounds
    Fig.4 Schematic of a package structure
    Fig.4 Schematic of a package structure
    Fig.6 Calculated results of filling behavior and temperature  distribution in the runner
    Fig.6 Calculated results of filling behavior and temperature distribution in the runner
    Fig.8 Comparison of cavity filling
    Fig.8 Comparison of cavity filling

    References

    1)J.Saeki et al. ,6th annual meeting of PPS, 12KN1(1990)
    2)J.Saeki et al. , JSME International Journal Series Ⅱ, 33,486(1990)
    3)J.Saeki et al.,SEIKEI KAKOU,12,67(2000)
    4) J.Saeki et al.,SEIKEI KAKOU,12,788(2000)
    5) J.Saeki et al.,SEIKEI KAKOU,13,49(2001)

    The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.

    Numerical investigation of flow characteristics over stepped spillways

    Güven, Aytaç
    Mahmood, Ahmed Hussein
    Water Supply (2021) 21 (3): 1344–1355.
    https://doi.org/10.2166/ws.2020.283Article history

    Abstract

    Spillways are constructed to evacuate flood discharge safely so that a flood wave does not overtop the dam body. There are different types of spillways, with the ogee type being the conventional one. A stepped spillway is an example of a nonconventional spillway. The turbulent flow over a stepped spillway was studied numerically by using the Flow-3D package. Different fluid flow characteristics such as longitudinal flow velocity, temperature distribution, density and chemical concentration can be well simulated by Flow-3D. In this study, the influence of slope changes on flow characteristics such as air entrainment, velocity distribution and dynamic pressures distribution over a stepped spillway was modelled by Flow-3D. The results from the numerical model were compared with an experimental study done by others in the literature. Two models of a stepped spillway with different discharge for each model were simulated. The turbulent flow in the experimental model was simulated by the Renormalized Group (RNG) turbulence scheme in the numerical model. A good agreement was achieved between the numerical results and the observed ones, which are exhibited in terms of graphics and statistical tables.

    배수로는 홍수가 댐 몸체 위로 넘치지 않도록 안전하게 홍수를 피할 수 있도록 건설되었습니다. 다른 유형의 배수로가 있으며, ogee 유형이 기존 유형입니다. 계단식 배수로는 비 전통적인 배수로의 예입니다. 계단식 배수로 위의 난류는 Flow-3D 패키지를 사용하여 수치적으로 연구되었습니다.

    세로 유속, 온도 분포, 밀도 및 화학 농도와 같은 다양한 유체 흐름 특성은 Flow-3D로 잘 시뮬레이션 할 수 있습니다. 이 연구에서는 계단식 배수로에 대한 공기 혼입, 속도 분포 및 동적 압력 분포와 같은 유동 특성에 대한 경사 변화의 영향을 Flow-3D로 모델링 했습니다.

    수치 모델의 결과는 문헌에서 다른 사람들이 수행한 실험 연구와 비교되었습니다. 각 모델에 대해 서로 다른 배출이 있는 계단식 배수로의 두 모델이 시뮬레이션되었습니다. 실험 모델의 난류 흐름은 수치 모델의 Renormalized Group (RNG) 난류 계획에 의해 시뮬레이션되었습니다. 수치 결과와 관찰 된 결과 사이에 좋은 일치가 이루어졌으며, 이는 그래픽 및 통계 테이블로 표시됩니다.

    HIGHLIGHTS

    ListenReadSpeaker webReader: Listen

    • A numerical model was developed for stepped spillways.
    • The turbulent flow was simulated by the Renormalized Group (RNG) model.
    • Both numerical and experimental results showed that flow characteristics are greatly affected by abrupt slope change on the steps.

    Keyword

    CFDnumerical modellingslope changestepped spillwayturbulent flow

    INTRODUCTION

    댐 구조는 물 보호가 생활의 핵심이기 때문에 물을 저장하거나 물을 운반하는 전 세계에서 가장 중요한 프로젝트입니다. 그리고 여수로는 댐의 가장 중요한 부분 중 하나로 분류됩니다. 홍수로 인한 파괴 나 피해로부터 댐을 보호하기 위해 여수로가 건설됩니다.

    수력 발전, 항해, 레크리에이션 및 어업의 중요성을 감안할 때 댐 건설 및 홍수 통제는 전 세계적으로 매우 중요한 문제로 간주 될 수 있습니다. 많은 유형의 배수로가 있지만 가장 일반적인 유형은 다음과 같습니다 : ogee 배수로, 자유 낙하 배수로, 사이펀 배수로, 슈트 배수로, 측면 채널 배수로, 터널 배수로, 샤프트 배수로 및 계단식 배수로.

    그리고 모든 여수로는 입구 채널, 제어 구조, 배출 캐리어 및 출구 채널의 네 가지 필수 구성 요소로 구성됩니다. 특히 롤러 압축 콘크리트 (RCC) 댐 건설 기술과 더 쉽고 빠르며 저렴한 건설 기술로 분류 된 계단식 배수로 건설과 관련하여 최근 수십 년 동안 많은 계단식 배수로가 건설되었습니다 (Chanson 2002; Felder & Chanson 2011).

    계단식 배수로 구조는 캐비테이션 위험을 감소시키는 에너지 소산 속도를 증가시킵니다 (Boes & Hager 2003b). 계단식 배수로는 다양한 조건에서 더 매력적으로 만드는 장점이 있습니다.

    계단식 배수로의 흐름 거동은 일반적으로 낮잠, 천이 및 스키밍 흐름 체제의 세 가지 다른 영역으로 분류됩니다 (Chanson 2002). 유속이 낮을 때 nappe 흐름 체제가 발생하고 자유 낙하하는 낮잠의 시퀀스로 특징 지워지는 반면, 스키밍 흐름 체제에서는 물이 외부 계단 가장자리 위의 유사 바닥에서 일관된 흐름으로 계단 위로 흐릅니다.

    또한 주요 흐름에서 3 차원 재순환 소용돌이가 발생한다는 것도 분명합니다 (예 : Chanson 2002; Gonzalez & Chanson 2008). 계단 가장자리 근처의 의사 바닥에서 흐름의 방향은 가상 바닥과 가상으로 정렬됩니다. Takahashi & Ohtsu (2012)에 따르면, 스키밍 흐름 체제에서 주어진 유속에 대해 흐름은 계단 가장자리 근처의 수평 계단면에 영향을 미치고 슈트 경사가 감소하면 충돌 영역의 면적이 증가합니다. 전이 흐름 체제는 나페 흐름과 스키밍 흐름 체제 사이에서 발생합니다. 계단식 배수로를 설계 할 때 스키밍 흐름 체계를 고려해야합니다 (예 : Chanson 1994, Matos 2000, Chanson 2002, Boes & Hager 2003a).

    CFD (Computational Fluid Dynamics), 즉 수력 공학의 수치 모델은 일반적으로 물리적 모델에 소요되는 총 비용과 시간을 줄여줍니다. 따라서 수치 모델은 실험 모델보다 빠르고 저렴한 것으로 분류되며 동시에 하나 이상의 목적으로 사용될 수도 있습니다. 사용 가능한 많은 CFD 소프트웨어 패키지가 있지만 가장 널리 사용되는 것은 FLOW-3D입니다. 이 연구에서는 Flow 3D 소프트웨어를 사용하여 유량이 서로 다른 두 모델에 대해 계단식 배수로에서 공기 농도, 속도 분포 및 동적 압력 분포를 시뮬레이션합니다.

    Roshan et al. (2010)은 서로 다른 수의 계단 및 배출을 가진 계단식 배수로의 두 가지 물리적 모델에 대한 흐름 체제 및 에너지 소산 조사를 연구했습니다. 실험 모델의 기울기는 각각 19.2 %, 12 단계와 23 단계의 수입니다. 결과는 23 단계 물리적 모델에서 관찰 된 흐름 영역이 12 단계 모델보다 더 수용 가능한 것으로 간주되었음을 보여줍니다. 그러나 12 단계 모델의 에너지 손실은 23 단계 모델보다 더 많았습니다. 그리고 실험은 스키밍 흐름 체제에서 23 단계 모델의 에너지 소산이 12 단계 모델보다 약 12 ​​% 더 적다는 것을 관찰했습니다.

    Ghaderi et al. (2020a)는 계단 크기와 유속이 다른 정련 매개 변수의 영향을 조사하기 위해 계단식 배수로에 대한 실험 연구를 수행했습니다. 그 결과, 흐름 체계가 냅페 흐름 체계에서 발생하는 최소 scouring 깊이와 같은 scouring 구멍 치수에 영향을 미친다는 것을 보여주었습니다. 또한 테일 워터 깊이와 계단 크기는 최대 scouring깊이에 대한 실제 매개 변수입니다. 테일 워터의 깊이를 6.31cm에서 8.54 및 11.82cm로 늘림으로써 수세 깊이가 각각 18.56 % 및 11.42 % 증가했습니다. 또한 이 증가하는 테일 워터 깊이는 scouring 길이를 각각 31.43 % 및 16.55 % 감소 시킵니다. 또한 유속을 높이면 Froude 수가 증가하고 흐름의 운동량이 증가하면 scouring이 촉진됩니다. 또한 결과는 중간의 scouring이 횡단면의 측벽보다 적다는 것을 나타냅니다. 계단식 배수로 하류의 최대 scouring 깊이를 예측 한 후 실험 결과와 비교하기 위한 실험식이 제안 되었습니다. 그리고 비교 결과 제안 된 공식은 각각 3.86 %와 9.31 %의 상대 오차와 최대 오차 내에서 scouring 깊이를 예측할 수 있음을 보여주었습니다.

    Ghaderi et al. (2020b)는 사다리꼴 미로 모양 (TLS) 단계의 수치 조사를 했습니다. 결과는 이러한 유형의 배수로가 확대 비율 LT / Wt (LT는 총 가장자리 길이, Wt는 배수로의 폭)를 증가시키기 때문에 더 나은 성능을 갖는 것으로 관찰되었습니다. 또한 사다리꼴 미로 모양의 계단식 배수로는 더 큰 마찰 계수와 더 낮은 잔류 수두를 가지고 있습니다. 마찰 계수는 다양한 배율에 대해 0.79에서 1.33까지 다르며 평평한 계단식 배수로의 경우 대략 0.66과 같습니다. 또한 TLS 계단식 배수로에서 잔류 수두의 비율 (Hres / dc)은 약 2.89이고 평평한 계단식 배수로의 경우 약 4.32와 같습니다.

    Shahheydari et al. (2015)는 Flow-3D 소프트웨어, RNG k-ε 모델 및 VOF (Volume of Fluid) 방법을 사용하여 배출 계수 및 에너지 소산과 같은 자유 표면 흐름의 프로파일을 연구하여 스키밍 흐름 체제에서 계단식 배수로에 대한 흐름을 조사했습니다. 실험 결과와 비교했습니다. 결과는 에너지 소산 율과 방전 계수율의 관계가 역으로 실험 모델의 결과와 잘 일치 함을 보여 주었다.

    Mohammad Rezapour Tabari & Tavakoli (2016)는 계단 높이 (h), 계단 길이 (L), 계단 수 (Ns) 및 단위 폭의 방전 (q)과 같은 다양한 매개 변수가 계단식 에너지 ​​소산에 미치는 영향을 조사했습니다. 방수로. 그들은 해석에 FLOW-3D 소프트웨어를 사용하여 계단식 배수로에서 에너지 손실과 임계 흐름 깊이 사이의 관계를 평가했습니다. 또한 유동 난류에 사용되는 방정식과 표준 k-ɛ 모델을 풀기 위해 유한 체적 방법을 적용했습니다. 결과에 따르면 스텝 수가 증가하고 유량 배출량이 증가하면 에너지 손실이 감소합니다. 얻은 결과를 다른 연구와 비교하고 경험적, 수학적 조사를 수행하여 결국 합격 가능한 결과를 얻었습니다.

    METHODOLOGY

    ListenReadSpeaker webReader: ListenFor all numerical models the basic principle is very similar: a set of partial differential equations (PDE) present the physical problems. The flow of fluids (gas and liquid) are governed by the conservation laws of mass, momentum and energy. For Computational Fluid Dynamics (CFD), the PDE system is substituted by a set of algebraic equations which can be worked out by using numerical methods (Versteeg & Malalasekera 2007). Flow-3D uses the finite volume approach to solve the Reynolds Averaged Navier-Stokes (RANS) equation, by applying the technique of Fractional Area/Volume Obstacle Representation (FAVOR) to define an obstacle (Flow Science Inc. 2012). Equations (1) and (2) are RANS and continuity equations with FAVOR variables that are applied for incompressible flows.

    formula

    (1)

    formula

    (2)where  is the velocity in xi direction, t is the time,  is the fractional area open to flow in the subscript directions,  is the volume fraction of fluid in each cell, p is the hydrostatic pressure,  is the density, is the gravitational force in subscript directions and  is the Reynolds stresses.

    Turbulence modelling is one of three key elements in CFD (Gunal 1996). There are many types of turbulence models, but the most common are Zero-equation models, One-equation models, Two-equation models, Reynolds Stress/Flux models and Algebraic Stress/Flux models. In FLOW-3D software, five turbulence models are available. The formulation used in the FLOW-3D software differs slightly from other formulations that includes the influence of the fractional areas/volumes of the FAVORTM method and generalizes the turbulence production (or decay) associated with buoyancy forces. The latter generalization, for example, includes buoyancy effects associated with non-inertial accelerations.

    The available turbulence models in Flow-3D software are the Prandtl Mixing Length Model, the One-Equation Turbulent Energy Model, the Two-Equation Standard  Model, the Two-Equation Renormalization-Group (RNG) Model and large Eddy Simulation Model (Flow Science Inc. 2012).In this research the RNG model was selected because this model is more commonly used than other models in dealing with particles; moreover, it is more accurate to work with air entrainment and other particles. In general, the RNG model is classified as a more widely-used application than the standard k-ɛ model. And in particular, the RNG model is more accurate in flows that have strong shear regions than the standard k-ɛ model and it is defined to describe low intensity turbulent flows. For the turbulent dissipation  it solves an additional transport equation:

    formula

    (3)where CDIS1, CDIS2, and CDIS3 are dimensionless parameters and the user can modify them. The diffusion of dissipation, Diff ɛ, is

    formula

    (4)where uv and w are the x, y and z coordinates of the fluid velocity; ⁠, ⁠,  and ⁠, are FLOW-3D’s FAVORTM defined terms;  and  are turbulence due to shearing and buoyancy effects, respectively. R and  are related to the cylindrical coordinate system. The default values of RMTKE, CDIS1 and CNU differ, being 1.39, 1.42 and 0.085 respectively. And CDIS2 is calculated from turbulent production (⁠⁠) and turbulent kinetic energy (⁠⁠).The kinematic turbulent viscosity is the same in all turbulence transport models and is calculated from

    formula

    (5)where ⁠: is the turbulent kinematic viscosity.  is defined as the numerical challenge between the RNG and the two-equation k-ɛ models, found in the equation below. To avoid an unphysically large result for  in Equation (3), since this equation could produce a value for  very close to zero and also because the physical value of  may approach to zero in such cases, the value of  is calculated from the following equation:

    formula

    (6)where ⁠: the turbulent length scale.

    VOF and FAVOR are classifications of volume-fraction methods. In these two methods, firstly the area should be subdivided into a control volume grid or a small element. Each flow parameter like velocity, temperature and pressure values within the element are computed for each element containing liquids. Generally, these values represent the volumetric average of values in the elements.Numerous methods have been used recently to solve free infinite boundaries in the various numerical simulations. VOF is an easy and powerful method created based on the concept of a fractional intensity of fluid. A significant number of studies have confirmed that this method is more flexible and efficient than others dealing with the configurations of a complex free boundary. By using VOF technology the Flow-3D free surface was modelled and first declared in Hirt & Nichols (1981). In the VOF method there are three ingredients: a planner to define the surface, an algorithm for tracking the surface as a net mediator moving over a computational grid, and application of the boundary conditions to the surface. Configurations of the fluids are defined in terms of VOF function, F (x, y, z, t) (Hirt & Nichols 1981). And this VOF function shows the volume of flow per unit volume

    formula

    (7)

    formula

    (8)

    formula

    (9)where  is the density of the fluid, is a turbulent diffusion term,  is a mass source,  is the fractional volume open to flow. The components of velocity (u, v, w) are in the direction of coordinates (x, y, z) or (r, ⁠).  in the x-direction is the fractional area open to flow,  and  are identical area fractions for flow in the y and z directions. The R coefficient is based on the selection of the coordinate system.

    The FAVOR method is a different method and uses another volume fraction technique, which is only used to define the geometry, such as the volume of liquid in each cell used to determine the position of fluid surfaces. Another fractional volume can be used to define the solid surface. Then, this information is used to determine the boundary conditions of the wall that the flow should be adapted for.

    Case study

    ListenReadSpeaker webReader: Listen

    In this study, the experimental results of Ostad Mirza (2016) was simulated. In a channel composed of two 4 m long modules, with a transparent sidewall of height 0.6 m and 0.5 m width. The upstream chute slope (i.e. pseudo-bottom angle) Ɵ1 = 50°, the downstream chute slope Ɵ2 = 30° or 18.6°, the step heights h = 0.06 m, the total number of steps along the 50° chute 41 steps, the total number of steps along the 30° chute 34 steps and the total number of steps along the 18.6° chute 20 steps.

    The flume inflow tool contained a jetbox with a maximum opening set to 0.12 meters, designed for passing the maximum unit discharge of 0.48 m2/s. The measurements of the flow properties (i.e. air concentration and velocity) were computed perpendicular to the pseudo-bottom as shown in Figure 1 at the centre of twenty stream-wise cross-sections, along the stepped chute, (i.e. in five steps up on the slope change and fifteen steps down on the slope change, namely from step number −09 to +23 on 50°–30° slope change, or from −09 to +15 on 50°–18.6° slope change, respectively).

    Sketch of the air concentration C and velocity V measured perpendicular to the pseudo-bottom used by Mirza (Ostad Mirza 2016).
    Sketch of the air concentration C and velocity V measured perpendicular to the pseudo-bottom used by Mirza (Ostad Mirza 2016).

    Sketch of the air concentration C and velocity V measured perpendicular to the pseudo-bottom used by Mirza (Ostad Mirza 2016).

    Pressure sensors were arranged with the x/l values for different slope change as shown in Table 1, where x is the distance from the step edge, along the horizontal step face, and l is the length of the horizontal step face. The location of pressure sensors is shown in Table 1.Table 1

    Location of pressure sensors on horizontal step faces

    Θ(°)L(m)x/l (–)
    50.0 0.050 0.35 0.64 – – – 
    30.0 0.104 0.17 0.50 0.84 – – 
    18.6 0.178 0.10 0.30 0.50 0.7 0.88 
    Location of pressure sensors on horizontal step faces
    Inlet boundary condition for Q = 0.235 m3/s and fluid elevation 4.21834 m.
    Inlet boundary condition for Q = 0.235 m3/s and fluid elevation 4.21834 m.

    Inlet boundary condition for Q = 0.235 m3/s and fluid elevation 4.21834 m.

    Numerical model set-up

    ListenReadSpeaker webReader: Listen

    A 3D numerical model of hydraulic phenomena was simulated based on an experimental study by Ostad Mirza (2016). The water surcharge and flow pressure over the stepped spillway was computed for two models of a stepped spillway with different discharge for each model. In this study, the package was used to simulate the flow parameters such as air entrainment, velocity distribution and dynamic pressures. The solver uses the finite volume technique to discretize the computational domain. In every test run, one incompressible fluid flow with a free surface flow selected at 20̊ was used for this simulation model. Table 2 shows the variables used in test runs.Table 2

    Variables used in test runs

    Test no.Θ1 (°)Θ2 (°)h(m)d0q (m3s1)dc/h (–)
    50 18.6 0.06 0.045 0.1 2.6 
    50 18.6 0.06 0.082 0.235 4.6 
    50 30.0 0.06 0.045 0.1 2.6 
    50 30.0 0.06 0.082 0.235 4.6 
    Table 2 Variables used in test runs

    For stepped spillway simulation, several parameters should be specified to get accurate simulations, which is the scope of this research. Viscosity and turbulent, gravity and non-inertial reference frame, air entrainment, density evaluation and drift-flux should be activated for these simulations. There are five different choices in the ‘viscosity and turbulent’ option, in the viscosity flow and Renormalized Group (RNG) model. Then a dynamical model is selected as the second option, the ‘gravity and non-inertial reference frame’. Only the z-component was inputted as a negative 9.81 m/s2 and this value represents gravitational acceleration but in the same option the x and y components will be zero. Air entrainment is selected. Finally, in the drift-flux model, the density of phase one is input as (water) 1,000 kg/m3 and the density of phase two (air) as 1.225 kg/m3. Minimum volume fraction of phase one is input equal to 0.1 and maximum volume fraction of phase two to 1 to allow air concentration to reach 90%, then the option allowing gas to escape at free surface is selected, to obtain closer simulation.

    The flow domain is divided into small regions relatively by the mesh in Flow-3D numerical model. Cells are the smallest part of the mesh, in which flow characteristics such as air concentration, velocity and dynamic pressure are calculated. The accuracy of the results and simulation time depends directly on the mesh block size so the cell size is very important. Orthogonal mesh was used in cartesian coordinate systems. A smaller cell size provides more accuracy for results, so we reduced the number of cells whilst including enough accuracy. In this study, the size of cells in x, y and z directions was selected as 0.015 m after several trials.

    Figure 3 shows the 3D computational domain model 50–18.6 slope change, that is 6.0 m length, 0.50 m width and 4.23 m height. The 3D model of the computational domain model 50–30 slope changes this to 6.0 m length, 0.50 m width and 5.068 m height and the size of meshes in x, y, and z directions are 0.015 m. For the 50–18.6 slope change model: both total number of active and passive cells = 4,009,952, total number of active cells = 3,352,307, include real cells (used for solving the flow equations) = 3,316,269, open real cells = 3,316,269, fully blocked real cells equal to zero, external boundary cells were 36,038, inter-block boundary cells = 0 (Flow-3D report). For 50–30 slope change model: both total number of active and passive cells = 4,760,002, total number of active cells equal to 4,272,109, including real cells (used for solving the flow equations) were 3,990,878, open real cells = 3,990,878 fully blocked real cells = zero, external boundary cells were 281,231, inter-block boundary cells = 0 (Flow-3D report).

    The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.
    Figure3 The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.

    Figure 3VIEW LARGEDOWNLOAD SLIDE

    The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.

    When solving the Navier-Stokes equation and continuous equations, boundary conditions should be applied. The most important work of boundary conditions is to create flow conditions similar to physical status. The Flow-3D software has many types of boundary condition; each type can be used for the specific condition of the models. The boundary conditions in Flow-3D are symmetry, continuative, specific pressure, grid overlay, wave, wall, periodic, specific velocity, outflow, and volume flow rate.

    There are two options to input finite flow rate in the Flow-3D software either for inlet discharge of the system or for the outlet discharge of the domain: specified velocity and volume flow rate. In this research, the X-minimum boundary condition, volume flow rate, has been chosen. For X-maximum boundary condition, outflow was selected because there is nothing to be calculated at the end of the flume. The volume flow rate and the elevation of surface water was set for Q = 0.1 and 0.235 m3/s respectively (Figure 2).

    The bottom (Z-min) is prepared as a wall boundary condition and the top (Z-max) is computed as a pressure boundary condition, and for both (Y-min) and (Y-max) as symmetry.

    RESULTS AND DISCUSSION

    ListenReadSpeaker webReader: Listen

    The air concentration distribution profiles in two models of stepped spillway were obtained at an acquisition time equal to 25 seconds in skimming flow for both upstream and downstream of a slope change 50°–18.6° and 50°–30° for different discharge as in Table 2, and as shown in Figure 4 for 50°–18.6° slope change and Figure 5 for 50°–30° slope change configuration for dc/h = 4.6. The simulation results of the air concentration are very close to the experimental results in all curves and fairly close to that predicted by the advection-diffusion model for the air bubbles suggested by Chanson (1997) on a constant sloping chute.

    Figure 4 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6. VIEW LARGEDOWNLOAD SLIDE Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6.
    Figure 4 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6. VIEW LARGEDOWNLOAD SLIDE Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6.

    Figure 4VIEW LARGEDOWNLOAD SLIDE

    Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6.

    Figure5 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +11, +19 and +22 along the 50°–30° slope change, for dc/h = 4.6.
    Figure5 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +11, +19 and +22 along the 50°–30° slope change, for dc/h = 4.6.

    Figure 5VIEW LARGEDOWNLOAD SLIDE

    Experimental and simulated air concentration distribution for steps number −5, +1, +5, +11, +19 and +22 along the 50°–30° slope change, for dc/h = 4.6.

    Figure 6VIEW LARGEDOWNLOAD SLIDE

    Figure 6 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 2.6.
    Figure 6 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 2.6.

    Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 2.6.

    Figure 7 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5. +11, +15 and +22 along the 50°–30° slope change for dc/h = 2.6.
    Figure 7 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5. +11, +15 and +22 along the 50°–30° slope change for dc/h = 2.6.

    Figure 7VIEW LARGEDOWNLOAD SLIDE

    Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5. +11, +15 and +22 along the 50°–30° slope change for dc/h = 2.6.

    But as is shown in all above mentioned figures it is clear that at the pseudo-bottom the CFD results of air concentration are less than experimental ones until the depth of water reaches a quarter of the total depth of water. Also the direction of the curves are parallel to each other when going up towards the surface water and are incorporated approximately near the surface water. For all curves, the cross-section is separate between upstream and downstream steps. Therefore the (-) sign for steps represents a step upstream of the slope change cross-section and the (+) sign represents a step downstream of the slope change cross-section.

    The dimensionless velocity distribution (V/V90) profile was acquired at an acquisition time equal to 25 seconds in skimming flow of the upstream and downstream slope change for both 50°–18.6° and 50°–30° slope change. The simulation results are compared with the experimental ones showing that for all curves there is close similarity for each point between the observed and experimental results. The curves increase parallel to each other and they merge near at the surface water as shown in Figure 6 for slope change 50°–18.6° configuration and Figure 7 for slope change 50°–30° configuration. However, at step numbers +1 and +5 in Figure 7 there are few differences between the simulated and observed results, namely the simulation curves ascend regularly meaning the velocity increases regularly from the pseudo-bottom up to the surface water.

    Figure 8 (50°–18.6° slope change) and Figure 9 (50°–30° slope change) compare the simulation results and the experimental results for the presented dimensionless dynamic pressure distribution for different points on the stepped spillway. The results show a good agreement with the experimental and numerical simulations in all curves. For some points, few discrepancies can be noted in pressure magnitudes between the simulated and the observed ones, but they are in the acceptable range. Although the experimental data do not completely agree with the simulated results, there is an overall agreement.

    Figure 8 Comparison between simulated and experimental results for the dimensionless pressure for steps number  −1, −2, −3 and +1, +2 +3 and +20 on the horizontal step faces of 50°–18.6° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.
    Figure 8 Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 +3 and +20 on the horizontal step faces of 50°–18.6° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

    Figure 8VIEW LARGEDOWNLOAD SLIDE

    Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 +3 and +20 on the horizontal step faces of 50°–18.6° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

    Figure 9 Comparison between simulated and experimental results for the dimensionless pressure for steps number  −1, −2, −3 and +1, +2 and +30, +31 on the horizontal step face of 50°–30° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.
    Figure 9 Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 and +30, +31 on the horizontal step face of 50°–30° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

    Figure 9VIEW LARGEDOWNLOAD SLIDE

    Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 and +30, +31 on the horizontal step face of 50°–30° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

    The pressure profiles were acquired at an acquisition time equal to 70 seconds in skimming flow on 50°–18.6°, where p is the measured dynamic pressure, h is step height and ϒ is water specific weight. A negative sign for steps represents a step upstream of the slope change cross-section and a positive sign represents a step downstream of the slope change cross-section.

    Figure 10 shows the experimental streamwise development of dimensionless pressure on the 50°–18.6° slope change for dc/h = 4.6, x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute compared with the numerical simulation. It is obvious from Figure 10 that the streamwise development of dimensionless pressure before slope change (steps number −1, −2 and −3) both of the experimental and simulated results are close to each other. However, it is clear that there is a little difference between the results of the streamwise development of dimensionless pressure at step numbers +1, +2 and +3. Moreover, from step number +3 to the end, the curves get close to each other.

    Figure 10 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–18.6° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute.
    Figure 10 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–18.6° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute.

    Figure 10VIEW LARGEDOWNLOAD SLIDE

    Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–18.6° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute.

    Figure 11 compares the experimental and the numerical results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute. It is apparent that the outcomes of the experimental work are close to the numerical results, however, the results of the simulation are above the experimental ones before the slope change, but the results of the simulation descend below the experimental ones after the slope change till the end.

    Figure 11 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute.
    Figure 11 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute.

    Figure 11VIEW LARGEDOWNLOAD SLIDE

    Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute.

    CONCLUSION

    ListenReadSpeaker webReader: Listen

    In this research, numerical modelling was attempted to investigate the effect of abrupt slope change on the flow properties (air entrainment, velocity distribution and dynamic pressure) over a stepped spillway with two different models and various flow rates in a skimming flow regime by using the CFD technique. The numerical model was verified and compared with the experimental results of Ostad Mirza (2016). The same domain of the numerical model was inputted as in experimental models to reduce errors as much as possible.

    Flow-3D is a well modelled tool that deals with particles. In this research, the model deals well with air entrainment particles by observing their results with experimental results. And the reason for the small difference between the numerical and the experimental results is that the program deals with particles more accurately than the laboratory. In general, both numerical and experimental results showed that near to the slope change the flow bulking, air entrainment, velocity distribution and dynamic pressure are greatly affected by abrupt slope change on the steps. Although the extent of the slope change was relatively small, the influence of the slope change was major on flow characteristics.

    The Renormalized Group (RNG) model was selected as a turbulence solver. For 3D modelling, orthogonal mesh was used as a computational domain and the mesh grid size used for X, Y, and Z direction was equal to 0.015 m. In CFD modelling, air concentration and velocity distribution were recorded for a period of 25 seconds, but dynamic pressure was recorded for a period of 70 seconds. The results showed that there is a good agreement between the numerical and the physical models. So, it can be concluded that the proposed CFD model is very suitable for use in simulating and analysing the design of hydraulic structures.

    이 연구에서 수치 모델링은 두 가지 다른 모델과 다양한 유속을 사용하여 스키밍 흐름 영역에서 계단식 배수로에 대한 유동 특성 (공기 혼입, 속도 분포 및 동적 압력)에 대한 급격한 경사 변화의 영향을 조사하기 위해 시도되었습니다. CFD 기술. 수치 모델을 검증하여 Ostad Mirza (2016)의 실험 결과와 비교 하였다. 오차를 최대한 줄이기 위해 실험 모형과 동일한 수치 모형을 입력 하였다.

    Flow-3D는 파티클을 다루는 잘 모델링 된 도구입니다. 이 연구에서 모델은 실험 결과를 통해 결과를 관찰하여 공기 혼입 입자를 잘 처리합니다. 그리고 수치와 실험 결과의 차이가 작은 이유는 프로그램이 실험실보다 입자를 더 정확하게 다루기 때문입니다. 일반적으로 수치 및 실험 결과는 경사에 가까워지면 유동 벌킹, 공기 혼입, 속도 분포 및 동적 압력이 계단의 급격한 경사 변화에 크게 영향을받는 것으로 나타났습니다. 사면 변화의 정도는 상대적으로 작았지만 사면 변화의 영향은 유동 특성에 큰 영향을 미쳤다.

    Renormalized Group (RNG) 모델이 난류 솔버로 선택되었습니다. 3D 모델링의 경우 계산 영역으로 직교 메쉬가 사용되었으며 X, Y, Z 방향에 사용 된 메쉬 그리드 크기는 0.015m입니다. CFD 모델링에서 공기 농도와 속도 분포는 25 초 동안 기록되었지만 동적 압력은 70 초 동안 기록되었습니다. 결과는 수치 모델과 물리적 모델간에 좋은 일치가 있음을 보여줍니다. 따라서 제안 된 CFD 모델은 수력 구조물의 설계 시뮬레이션 및 해석에 매우 적합하다는 결론을 내릴 수 있습니다.

    DATA AVAILABILITY STATEMENT

    ListenReadSpeaker webReader: Listen

    All relevant data are included in the paper or its Supplementary Information.

    REFERENCES

    Boes R. M. Hager W. H. 2003a Hydraulic design of stepped spillways. Journal of Hydraulic Engineering 129 (9), 671–679.
    Google Scholar
    Boes R. M. Hager W. H. 2003b Two-Phase flow characteristics of stepped spillways. Journal of Hydraulic Engineering 129 (9), 661–670.
    Google Scholar
    Chanson H. 1994 Hydraulics of skimming flows over stepped channels and spillways. Journal of Hydraulic Research 32 (3), 445–460.
    Google Scholar
    Chanson H. 1997 Air Bubble Entrainment in Free Surface Turbulent Shear Flows. Academic Press, London.
    Google Scholar
    Chanson H. 2002 The Hydraulics of Stepped Chutes and Spillways. Balkema, Lisse, The Netherlands.
    Google Scholar
    Felder S. Chanson H. 2011 Energy dissipation down a stepped spillway with nonuniform step heights. Journal of Hydraulic Engineering 137 (11), 1543–1548.
    Google Scholar
    Flow Science, Inc. 2012 FLOW-3D v10-1 User Manual. Flow Science, Inc., Santa Fe, CA.
    Ghaderi A. Daneshfaraz R. Torabi M. Abraham J. Azamathulla H. M. 2020a Experimental investigation on effective scouring parameters downstream from stepped spillways. Water Supply 20 (5), 1988–1998.
    Google Scholar
    Ghaderi A. Abbasi S. Abraham J. Azamathulla H. M. 2020b Efficiency of trapezoidal labyrinth shaped stepped spillways. Flow Measurement and Instrumentation 72, 101711.
    Google Scholar
    Gonzalez C. A. Chanson H. 2008 Turbulence and cavity recirculation in air-water skimming flows on a stepped spillway. Journal of Hydraulic Research 46 (1), 65–72.
    Google Scholar
    Gunal M. 1996 Numerical and Experimental Investigation of Hydraulic Jumps. PhD Thesis, University of Manchester, Institute of Science and Technology, Manchester, UK.
    Hirt C. W. Nichols B. D. 1981 Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics 39 (1), 201–225.
    Google Scholar
    Matos J. 2000 Hydraulic design of stepped spillways over RCC dams. In: Intl Workshop on Hydraulics of Stepped Spillways (H.-E. Minor & W. Hager, eds). Balkema Publ, Zurich, pp. 187–194.
    Google Scholar
    Mohammad Rezapour Tabari M. Tavakoli S. 2016 Effects of stepped spillway geometry on flow pattern and energy dissipation. Arabian Journal for Science & Engineering (Springer Science & Business Media BV) 41 (4), 1215–1224.
    Google Scholar
    Ostad Mirza M. J. 2016 Experimental Study on the Influence of Abrupt Slope Changes on Flow Characteristics Over Stepped Spillways. Communications du Laboratoire de Constructions Hydrauliques, No. 64 (A. J. Schleiss, ed.). Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland.
    Roshan R. Azamathulla H. M. Marosi M. Sarkardeh H. Pahlavan H. Ab Ghani A. 2010 Hydraulics of stepped spillways with different numbers of steps. Dams and Reservoirs 20 (3), 131–136.
    Google Scholar
    Shahheydari H. Nodoshan E. J. Barati R. Moghadam M. A. 2015 Discharge coefficient and energy dissipation over stepped spillway under skimming flow regime. KSCE Journal of Civil Engineering 19 (4), 1174–1182.
    Google Scholar
    Takahashi M. Ohtsu I. 2012 Aerated flow characteristics of skimming flow over stepped chutes. Journal of Hydraulic Research 50 (4), 427–434.
    Google Scholar
    Versteeg H. K. Malalasekera W. 2007 An Introduction to Computational Fluid Dynamics: The Finite Volume Method. Pearson Education, Harlow.
    Google Scholar
    © 2021 The Authors
    This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).

    Figure 1. Steady-state shear stress a as a function of shear rate y in Sn-Pb alloy [10).

    Numerical Modelling of Semi-Solid Flow under Processing Conditions

    처리조건에서의 반고체유동의 수치모델링

    David H. Kirkwood and Philip J. Ward
    Department of Engineering Materials, University of Sheffield, Sheffield I UK

    Keywords: semi-solid alloys, thixotropy, flow modelling.

    Abstract

    During the industrial process of semi-solid forming (or thixoforming) of alloy slurries, typically the operation of die filling takes around 0.1s.
    During this time period the alloy slug is transformed from a solid-like structure capable of maintaining its shape, into a liquid-like slurry able
    to fill a complex die cavity: this involves a decrease in viscosity of some 6 orders of magnitUde. Many attempts to measure thixotropic breakdown experimentally in alloy slurries have relied on the use of concentric cylindrical viscometers in which viscosity changes have been followed after shear rate changes over times above 1s to in excess of 1000 s, which have little relevance to actual processing conditions and therefore to modelling of flow in industrial practice. The present paper is an attempt to abstract thixotropic breakdown rates from rapid compression tests between parallel plates moving together at velocities of around 1mis, similar to industrial conditions. From this analysis, a model of slurry flow has been developed in which rapid thixotropic breakdown of the slurry occurs at high shear rates.

    합금 슬러리의 반고체 성형 (또는 틱소 성형)의 산업 공정 동안, 일반적으로 다이 충진 작업은 약 0.1 초가 걸립니다.
    이 기간 동안 합금 슬러그는 모양을 유지할 수있는 고체와 같은 구조에서 액체와 같은 슬러리로 변형됩니다.
    복잡한 다이 캐비티를 채우기 위해 : 이것은 약 6 차의 마그 니트 점도 감소를 포함합니다. 합금 슬러리에서 실험적으로 요 변성 파괴를 측정하려는 많은 시도는 전단 속도가 1 초 이상에서 1000 초 이상으로 변화 한 후 점도 변화가 뒤 따르는 동심원 원통형 점도계의 사용에 의존하여 실제 가공 조건과는 거의 관련이 없습니다. 따라서 산업 현장에서 흐름 모델링에. 본 논문은 산업 조건과 유사하게 약 1mis의 속도로 함께 이동하는 평행 판 사이의 빠른 압축 테스트에서 요 변성 파괴 율을 추상화하려는 시도입니다. 이 분석으로부터 슬러리의 급속한 요 변성 분해가 높은 전단 속도에서 발생하는 슬러리 흐름 모델이 개발되었습니다.

    Introduction

    기존의 다이캐스팅을 위한 다이 설계는 과거에 예비 테스트 및 조정과 함께 축적 된 실무 경험의 문제였으며, 단기 실행, 랩, 다공성 등과 같은 결함을 제거하기 위해 다이 캐스트 제품을 검사했습니다. 이것은 모두 비용이 많이 드는 절차입니다.

    시간과 비용, 그리고 프로세스의 컴퓨터 모델링은 이를 줄이거 나 없애기 위해 많은 운영자에 의해 개발되었습니다. 반고체 가공 (thixoforming)에서는 반고체 합금 슬러리의 전단이 내부 구조를 파괴하여 충전 작업 중 시간이 지남에 따라 점도가 낮아짐으로 발생하는 비 뉴턴 점도로 인해 모델링 문제가 더욱 어려워집니다.

    시스템 전체에서 균일하지 않습니다. 충전 중에 발생하는 추가 응고로 인해 문제가 더욱 복잡해집니다. 빠른 충전으로 인해 이 단계에서 매우 작은 것으로 간주되기 때문에 현재 분석에서는 무시되었습니다.

    우리 모델의 또 다른 한계는 슬러리가 균질한 물질로 거동 한다는 가정이며, 이는 어느 지점에서나 단일 점도로 설명될 수 있습니다. 이것은 빠른 전단의 고려 사항과 정상적인 요 변형성 조건 내에서 0.6 미만의 고형분을 분별하는 것으로 제한합니다.

    <중략>……

    Figure 1. Steady-state shear stress a as a function of shear rate y in Sn-Pb alloy [10).
    Figure 1. Steady-state shear stress a as a function of shear rate y in Sn-Pb alloy [10).
    Figure 2. Equilibrium viscosity as a function of shear rate in Sn-Pb alloy, fraction solid:0.36, fitted to Cross Model.
    Figure 2. Equilibrium viscosity as a function of shear rate in Sn-Pb alloy, fraction solid:0.36, fitted to Cross Model.
    Figure 3. Cheng Diagram: shear stress vs. shear rate.
    Figure 3. Cheng Diagram: shear stress vs. shear rate.
    Figure 4. Reciprocal of experimental breakdown time vs. y 1.3 for Sn-Pb alloy
    Figure 4. Reciprocal of experimental breakdown time vs. y 1.3 for Sn-Pb alloy
    Figure 5. Relaxation time, T, as a function of shear rate; see also figure 4, Fs =0.36.
    Figure 5. Relaxation time, T, as a function of shear rate; see also figure 4, Fs =0.36.
    Figure 6. Experimental and modelled results for compression test on AI-A356 alloy at two temperatures.
    Figure 6. Experimental and modelled results for compression test on AI-A356 alloy at two temperatures.
    Table 1. Calculated parameters for the breakdown in compression tests [20].
    Table 1. Calculated parameters for the breakdown in compression tests [20].
    Figure 7. Drop-forge results from Yurko and Flemings [7].
    Figure 7. Drop-forge results from Yurko and Flemings [7].
    Figure 8. Prediction of FLOW-3D®.
    Figure 8. Prediction of FLOW-3D®.

    Conclusions

    y에서 전단 된 반고체 슬러리의 틱소 트로픽 분해에 대한 속도 방정식은 다음과 같은 형식으로 제안됩니다. T = l / (a ​​+ uym), 여기서 T는 급속 분해 또는 유사 정상 상태 구조에 대한 특성 시간이며, 밴드 m은 상수입니다. 이 관계는 제한된 범위의 전단 속도에서 Sn-Pb 합금의 전단 속도 점프에 의해 실험적으로 확인되었습니다.

    이 파괴율 방정식은 AI-Si 합금의 반고체 슬러그에 대한 빠른 압축 테스트에서 실험적으로 얻은 힘-변위 곡선을 시뮬레이션하기 위해 FLOW-3D® (버전 8.2 : FlowScience Inc.)에 도입되었습니다. 담금 시간과 다른 압축 속도에서. 이 분석의 결과는 모든 경우에 요 변성 거동이 관련되어 있음을 나타내지만, 5 분 동안 담근 후 (산업 관행에서와 같이) 구조가 크게 분해되었으며 초기에는 낮은 전단 속도 영역에서 흐름이 뉴턴에 가깝습니다.

    파괴율은 100 S-I 이상의 전단율에서 극적으로 증가하는 것으로 가정 됩니다. 이 예측은 높은 전단 속도에서 더 세심한 작업에 의해 테스트되어야 하지만 평균 전단 속도가 1300 sol까지 생성된 드롭 단조 실험에 의해 뒷받침되는 것으로 보입니다 [7].

    References

    [I] T.Y Liu, H.Y. Atkinson, PJ. Ward, D.H. Kirkwood: Metall.Mater.TransA, 34A (2003), 409/17.
    [2] A. Zavaliangos and A. Lawley: J. Mater. Eng. Perfonn., 4 (1995),40/47.
    [3] M.R. Barkhudarov, e.L. Bronisz, e.w. Hirt: ProcAth Int. Conf. onSemi-solid Processing of Alloys and Composites,1996, Sheffield,p.llO.
    [4] W.R.Loue, M.Suery, J.L.Querbes: Proc.2ndInt.Conf.on Semi-solidProcessing of Alloys and Composites,1992, Cambridge MA , pp266-75.
    [5] P.Kapranos, D.H.Kirkwood, M.R. Barkhudarov: Proc.5th Int. Conf.on Semi-solid Processing of Alloys and Composites, Golden, Colorado,1998. pp.II-19.
    [6] T.Y. Liu, H.Y. Atkinson, P. Kapranos, D.H. Kirkwood, S.G. Hogg:Metall. Mater. Trans A, 34A (2003), 1545/54.
    [7] J.A. Yurko and M.e. Flemings: Metall. Mater. Trans A, 33A (2002),2737/46.
    [8] M. Modigell and J. Koke: Mechanics of Time Dependent Materials, 3(1999), 15/30.
    [9] Y. Laxmanan and M.e. Flemings: Metall. Trans. A, IIA( 1980),1927/36.
    [IO]A.R.A Mclelland, N.G. Henderson, H.Y. Atkinson, D.H. Kirkwood:Mater. Sci. Eng., A232 (1997), 110/18.
    [II] H.A. Barnes: 1. Non-Newtonian Fluid Mech., 81 (1999),133n8.
    [12]A.N. Alexandrou, E. Due , Y. Entov: 1. Non-Newtonian Fluid Mech.,96 (2001), 383/403.
    [13]C.L. Martin, P. Kumar and S. Brown: Acta Mat. Mater., 42 (1994),3603/14.
    [14]C. Quaak, L. Katgennan and W.H. Kool: Proc. 4th Conf. on Semi-solid Processing of Alloys and Composites, 1996, Sheffield, pp.35/39.
    [15]D.C-H. Cheng: Int. Journal Cosmetic Science, 9 (1987), pp.151/91.
    [16]An Introduction to Rheology: H.A. Barnes, J.F. Hutton and K Walters,Elsevier, Amsterdam, 1989.
    [17]A.M. de Figueredo, A. Kato and M.e. Flemings: Proc.6th Int. Conf.on Semi-solid Processing of Alloys and Composites, 2000, Turin,477/82.
    [18]1.y’ Chen and Z. Fan: Mater. Sci. Tech., 18 (2002), 237/42.
    [19]Z. Fan: Int. Mater. Rev., 47 (2002), No.2, 49/85.
    [20]D.H. Kirkwood and P.J. Ward: Proc. 8th Int. Conf. on Semi-solid Processing of Alloys and Composites, 2004, Cyprus. To be published.

    Mixing Tank with FLOW-3D

    CFD Stirs Up Mixing 일반

    CFD (전산 유체 역학) 전문가가 필요하고 때로는 실행하는데 몇 주가 걸리는 믹싱 시뮬레이션의 시대는 오래 전입니다. 컴퓨팅 및 관련 기술의 엄청난 도약에 힘 입어 Ansys, Comsol 및 Flow Science와 같은 회사는 엔지니어의 데스크톱에 사용하기 쉬운 믹싱 시뮬레이션을 제공하고 있습니다.

    “병렬화 및 고성능 컴퓨팅의 발전과 템플릿화는 비전문 화학 엔지니어에게 정확한 CFD 시뮬레이션을 제공했습니다.”라고 펜실베이니아  피츠버그에있는 Ansys Inc.의 수석 제품 마케팅 관리자인 Bill Kulp는 말합니다 .

    흐름 개선을위한 실용적인 지침이 필요하십니까? 다운로드 화학 처리의 eHandbook을 지금 흐름 도전 싸우는 방법!

    예를 들어, 회사는 휴스턴에있는 Nalco Champion과 함께 프로젝트를 시작했습니다. 이 프로젝트는 시뮬레이션 전문가가 아닌 화학 엔지니어에게 Ansys Fluent 및 ACT (분석 제어 기술) 템플릿 기반 시뮬레이션 앱에 대한 액세스 권한을 부여합니다. 새로운 화학 물질을위한 프로세스를 빠르고 효율적으로 확장합니다.

    Giving Mixing Its Due

    “화학 산업은 CFD와 같은 계산 도구를 사용하여 많은 것을 얻을 수 있지만 혼합 프로세스는 단순하다고 가정하기 때문에 간과되는 경우가 있습니다. 그러나 최신 수치 기법을 사용하여 우수한 성능을 달성하는 흥미로운 방법이 많이 있습니다.”라고 Flow Science Inc. , Santa Fe, NM의 CFD 엔지니어인 Ioannis Karampelas는 말합니다 .

    이러한 많은 기술이 회사의 Flow-3D Multiphysics 모델링 소프트웨어 패키지와 전용 포스트 프로세서 시각화 도구 인 FlowSight에 포함되어 있습니다.

    “모든 상업용 CFD 패키지는 어떤 형태의 시각화 도구와 번들로 제공되지만 FlowSight는 매우 강력하고 사용하기 쉽고 이해하기 쉽게 설계되었습니다. 예를 들어, 프로세스를 재 설계하려는 엔지니어는 다양한 설계 변경의 효과를 평가하기 위해 매우 직관적인 시각화 도구가 필요합니다.”라고 그는 설명합니다.

    이 접근 방식은 실험 측정을 얻기 어려운 공정 (예 : 쉽게 측정 할 수없는 매개 변수 및 독성 물질의 존재로 인해 본질적으로 위험한 공정)을 더 잘 이해하고 최적화하는데 특히 효과적입니다.

    동일한 접근 방식은 또한 믹서 관련 장비 공급 업체가 고객 요구에 맞게 제품을보다 정확하게 개발하고 맞춤화하는 데 도움이되었습니다. “이는 불필요한 프로토 타이핑 비용이나 잠재적 인 과도한 엔지니어링을 방지합니다. 두 가지 모두 일부 공급 업체의 문제였습니다.”라고 Karampelas는 말합니다.

    CFD 기술 자체는 계속해서 발전하고 있습니다. 예를 들어, 수치 알고리즘의 관점에서 볼 때 구형 입자의 상호 작용이 열 전달을 적절하게 모델링하는 데 중요한 다양한 문제에 대해 이산 요소 모델링을 쉽게 적용 할 수있는 반면, LES 난류 모델은 난류 흐름 패턴을 정확하게 시뮬레이션하는 데 이상적입니다.

    컴퓨팅 리소스에 대한 비용과 수요에도 불구하고 Karampelas는 난류 모델의 전체 제품군을 제공 할 수있는 것이 중요하다고 생각합니다. 특히 LES는 이미 대부분의 학계와 일부 산업 (예 : 전력 공학)에서 선택하는 방법이기 때문입니다. .

    그럼에도 불구하고 CFD의 사용이 제한적이거나 비실용적 일 수있는 경우는 확실히 있습니다. 여기에는 나노 입자에서 벌크 유체 증발을 모델링하는 것과 같이 관심의 규모가 다른 규모에 따라 달라질 수있는 문제와 중요한 물리적 현상이 아직 알려지지 않았거나 제대로 이해되지 않았거나 아마도 매우 복잡한 문제 (예 : 모델링)가 포함됩니다. 음 펨바 효과”라고 Karampelas는 경고합니다.

    반면에 더욱 강력한 하드웨어와 업데이트 된 수치 알고리즘의 출현은 CFD 소프트웨어를 사용하여 과다한 설계 및 최적화 문제를 해결하기위한 최적의 접근 방식이 될 것이라고 그는 믿습니다.

    “복잡한 열교환 시스템 및 새로운 혼합 기술과 같이 점점 더 복잡한 공정을 모델링 할 수있는 능력은 가까운 장래에 가능할 수있는 일을 간단히 보여줍니다. 수치적 방법 사용의 주요 이점은 설계자가 상상력에 의해서만 제한되어 소규모 믹서에서 대규모 반응기 및 증류 컬럼에 이르기까지 다양한 화학 플랜트 공정을 최적화 할 수있는 길을 열어 준다는 것입니다. 실험적 또는 경험적 접근 방식은 항상 관련성이 있지만 CFD가 미래의 엔지니어를위한 선택 도구가 될 것이라고 확신합니다.”라고 그는 결론을 내립니다.


    Ottewell2
    Seán Ottewell은 Chemical Processing의 편집장입니다. sottewell@putman.net으로 이메일을 보낼 수 있습니다 .

    기사 원문 : https://www.chemicalprocessing.com/articles/2017/cfd-stirs-up-mixing/

    Fig. 3. Nylon 11 impact sequence onto a preheated substrate

    Impact Modeling of Thermally Sprayed Polymer Particles

    Ivosevic, M., Cairncross, R. A., Knight, R., Philadelphia / USA

    열 스프레이는 전통적으로 금속, 카바이드 및 세라믹 코팅을 증착하는 데 사용되어 왔지만 최근에는 HVOF (High Velocity Oxy-Fuel) 열 스프레이 공정의 높은 운동 에너지로 인해 용융 점도가 높은 폴리머의 무용제 처리도 가능하다는 사실이 밝혀졌습니다. , 유해한 휘발성 유기 용매가 필요하지 않습니다. 이 작업의 주된 목표는 지식 기반을 개발하고 HVOF 연소 스프레이 공정에 의해 분사되는 폴리머 입자의 충격 거동에 대한 질적 이해를 개선하는 것이 었습니다. 고분자 입자의 HVOF 분사 중 입자 가속, 가열 및 충격 변형의 수치 모델이 개발되었습니다. Volume-of-Fluid (VoF) 전산 유체 역학 패키지 인 Flow3D®는 입자가 강철 기판과 충돌하는 동안 유체 역학 및 열 전달을 모델링하는 데 사용되었습니다. 입자 가속 및 열 전달 모델을 사용하여 예측 된 방사형 온도 프로파일은 저온, 고점도 코어 및 고온, 저점도 표면을 가진 폴리머 입자를 시뮬레이션하기 위해 온도 의존 점도 모델과 함께 Flow3D®의 초기 조건으로 사용되었습니다. 이 접근법은 얇은 디스크 내에서 크고 거의 반구형 인 코어를 나타내는 변형 된 입자를 예측했으며 광학 현미경을 사용하여 만든 열 스프레이 스 플랫의 실험 관찰과 일치했습니다.

    폴리머 증착에 열 분무 공정을 사용하는 주요 이점은 다음과 같습니다. (i) 휘발성 유기 화합물 (VOCs)을 사용하지 않는 무용제 코팅; (ii) 거의 모든 환경 조건에서 큰 물체를 코팅 할 수있는 능력; (iii) 용융 점도가 높은 폴리머 코팅을 적용하는 능력; 및 (iv) 일반적으로 정전기 분말 코팅 및 용제 기반 페인트에 필요한 오븐 건조 또는 경화와 같은 증착 후 처리없이 “즉시 사용 가능한”코팅을 생산할 수있는 능력. 이러한 공정에 비해 주요 단점은 다음과 같습니다. (i) 낮은 증착 효율, (ii) 낮은 품질의 표면 마감 및 (iii) 높은 공정 복잡성 (종종 폴리머 용융 및 분해 온도에 의해 정의되는 좁은 공정 창). 폴리머 증착에 세 가지 열 스프레이 공정이 사용 된 것으로 알려졌습니다 [1].

    • 기존의 화염 분사.
    • HVOF 연소 스프레이.
    • 플라즈마 스프레이.

    HVOF 및 플라즈마 스프레이 공정에 의해 분사되는 폴리머의 수는 제한되어 있으며 HVOF 및 플라즈마 스프레이 폴리머 코팅의 상업적 응용은 아직 개발 단계에 있습니다 [1]. 폴리머의 HVOF 스프레이는 화염 스프레이 [최대 ~ 100m / s]에 비해 상당히 높은 입자 속도 [최대 1,000m / s]로 인해 주로 주목을 받았습니다. 이는 특히 고 분자량 폴리머 및 높은 (> 5 vol. %) 세라믹 강화 함량을 갖는 폴리머 / 세라믹 복합재를 포함하여 용융 점도가 높은 코팅의 증착에있어 중요한 이점입니다.

    Fig. 1. Nylon 11 splats deposited onto a room temperature glass slide.
    Fig. 1. Nylon 11 splats deposited onto a room temperature glass slide.
    Fig. 2. Nylon 11 splats deposited onto a preheated glass slide (200 °C).
    Fig. 2. Nylon 11 splats deposited onto a preheated glass slide (200 °C).
    Fig. 3. Nylon 11 impact sequence onto a preheated substrate
    Fig. 3. Nylon 11 impact sequence onto a preheated substrate, (I) partially melted particle before impact, (II) “fried-egg” shaped splat, (III) post-deposition flow of a fully molten droplet, (IV) droplet shrinkage during cooling.
    Fig. 5. Predicted velocities of Nylon 11 particles in an HVOF jet (total O2 + H2 gas flow rate of 1.86 g/s at Φ = 0.83).
    Fig. 5. Predicted velocities of Nylon 11 particles in an HVOF jet (total O2 + H2 gas flow rate of 1.86 g/s at Φ = 0.83).
    Fig. 7. Simulated deformation of a Nylon 11 droplet with a radial temperature gradient and temperaturedependent viscosity during impact.
    Fig. 7. Simulated deformation of a Nylon 11 droplet with a radial temperature gradient and temperaturedependent viscosity during impact.
    Figure 2. Ink fraction contours for mesh 1 through 4 (left to right) at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs.

    Coupled CFD-Response Surface Method (RSM) Methodology for Optimizing Jettability Operating Conditions

    분사성 작동 조건을 최적화하기 위한 결합된 CFD-Response Surface Method(RSM)

    Nuno Couto 1, Valter Silva 1,2,* , João Cardoso 2, Leo M. González-Gutiérrez 3 and Antonio Souto-Iglesias 41
    INEGI-FEUP, Faculty of Engineering, Porto University, 4200-465 Porto, Portugal;
    nunodiniscouto@hotmail.com
    2 VALORIZA, Polytechnic Institute of Portalegre, 7300-110 Portalegre, Portugal; jps.cardoso@ipportalegre.pt
    3 CEHINAV, DMFPA, ETSIN, Universidad Politécnica de Madrid, 28040 Madrid, Spain; leo.gonzalez@upm.es
    4 CEHINAV, DACSON, ETSIN, Universidad Politécnica de Madrid, 28040 Madrid, Spain;
    antonio.souto@upm.es

    • Correspondence: valter.silva@ipportalegre.pt; Tel.: +351-245-301-592

    소개

    물방울 생성에 대한 이해는 여러 산업 응용 분야에서 매우 중요합니다 [ 1 ]. 잉크젯 프린팅 프로세스는 일반적으로 10 ~ 100 μm [ 1 ] 범위의 독특하고 작은 액적 크기를 특징으로 하며 연속적 또는 충동적 흐름을 사용하여 얻을 수 있습니다 (마지막 방식은 주문형 드롭 (DoD)이라고도 함). 잉크젯).

    여러 장점 덕분에 DoD 방법은 산업 환경에서 상당한 수용을 얻고 있습니다 [ 2 ].DoD는 복잡한 프로세스이며 유체 속성, 노즐 형상 및 구동 파형 [ 1 , 3 ]의 세 가지 주요 범주로 분류되는 여러 매개 변수에 따라 달라집니다 .그러나 길이와 시간 척도가 모두 마이크로 오더 [ 4 ] 이기 때문에 실험을하기가 어렵습니다 .

    결과적으로 실험 설정은 항상 비용이 많이 들고 복잡하며 CFD (전산 유체 역학)와 같은 고급 수치 접근 방식이 엄격한 요구 사항입니다 [ 5 , 6 ]. VOF (volume-of-fluid) 접근 방식은 액체 분해 및 액적 생성에 대한 다상 공정을 시뮬레이션하기위한 적절한 대안으로 밝혀졌으며 과거 연구에서 그대로 사용되었습니다 [ 7 , 8], 인쇄 프로세스의 맥락에서 전자는 여전히 현재 연구의 주제입니다. 

    또한 VOF 체계를 사용하면 단일 운동량 방정식 세트를 해결하고 도메인 전체에 걸쳐 각 유체의 체적 분율을 추적하여 명확하게 정의된 인터페이스로 둘 이상의 혼합 불가능한 유체를 효과적으로 시뮬레이션 할 수 있습니다. Feng [ 9 ]는 VOF 접근 방식을 사용하여 일시적인 유체 인터페이스 변형 및 중단을 효과적으로 추적하는 패키지 FLOW-3D를 사용하여 낙하 배출 중 복잡한 유체 역학 프로세스를 시뮬레이션하는 선구자 작업 중 하나를 수행했습니다.

    주요 목표는 볼륨 및 속도와 같은 민감한 변수를 더 잘 이해하면서 장치 개발에서 일반적인 설계 규칙을 구현하는 것이 었습니다. 이러한 종류의 공정과 관련된 주요 질문 중 하나는 안정적인 액적 형성을 위한 작동 범위의 정의입니다.

    Fromm [ 10 ]은 Reynolds 수와 Weber 수의 제곱근 비율이 2보다 작으면 안정적인 방울을 생성 할 수 없다는 것을 확인했습니다. 이 무차원 값은 나중에 Z 번호로 알려졌으며 분사 가능성 범위 [ 11 ]를 정의합니다 . 문헌에서 분사 가능성을 위한 Z 간격은 1 ~ 10 [ 12 ], 4 ~ 14 [ 13 ] 또는 0.67 ~ 50 [ 14]을 찾을 수 있습니다. 

    이것은 Z 값 만으로는 분사 가능성 조건을 나타낼 수 없음을 분명히 의미합니다. 실제로, 다른 속성을 가진 유체는 다른 인쇄 품질을 나타내면서 동일한 Z 값을 나타낼 수 있습니다. 액적 생성 공정과 해당 분사 성은 주로 전체 공정 품질에 큰 영향을 미치는 매개 변수 세트에 의해 결정됩니다. 

    토대 메커니즘을 더 잘 이해하려면 확장 된 작동 조건 및 매개 변수 세트를 고려하여 여러 실험 또는 수치 실행을 수행해야 합니다. DoE (design-of-experiment) 접근 방식과 같은 체계적인 접근 방식이 없으면 이것은 달성하기 매우 어려운 작업이 될 수 있습니다. 최적화 문제를 해결하기 위해 반응 표면 방법을 사용하여 처음으로 체계화된 접근 방식이 개발된 Box and Wilson [ 15 ] 의 선구자 기사 이후 ,이 입증된 방법론은 많은 화학 및 산업 공정[ 16 ] 및 기타 관련 학계에 성공적으로 적용되었습니다.

    예를 들어 Silva와 Rouboa [ 17 ]는 직접 메탄올 연료 전지의 출력 밀도에 영향을 미치는 관련 매개 변수를 식별하기 위해 반응 표면 방법론 (RSM)을 사용했습니다. 많은 실제 산업 응용 분야에서 실험 연구는 작동 매개 변수를 조절하기 어렵 기 때문에 제한적이지만 주로 설정을 개발하거나 실험을 실행하는 데 드는 비용이 높기 때문입니다. 

    따라서 솔루션은 주요 시스템 응답을 시뮬레이션하고 예측할 수 있는 효과적인 수학적 모델의 개발에 의존합니다. DoE와 같은 최적화 방법론을 수치 모델과 결합하면 비용이 많이 들고 시간이 많이 걸리는 실험을 피하고 다양한 입력 조합을 사용하여 최적의 조건을 얻을 수 있습니다 [ 16 ]. 

    실바와 루 보아 [ 18] CFD 프레임 워크 하에서 개발 된 2D Eulerian-Eulerian 바이오 매스 가스화 모델에서 얻은 결과를 RSM과 결합하여 다양한 응용 분야에서 합성 가스를 생성하기 위한 최적의 작동 조건을 찾습니다. 

    저자는 입력 요인으로 인한 최상의 응답과 최소한의 변동을 모두 보장하는 작동 조건을 찾을 수 있었습니다. Frawley et al. [ 19 ] CFD 및 DoE 기술 (특히 RSM)을 결합하여 파이프의 팔꿈치에서 고체 입자 침식에 대한 다양한 주요 요인의 영향을 조사하여 침식 예측 모델을 개발할 수 있습니다.우리가 아는 한, DoD 잉크젯 프로세스의 개선 및 더 나은 이해에 적용되는 DoE 접근법 (실험적으로 또는 모든 종류의 수치 모델과 결합)을 구현하는 연구는 없습니다. 선도 기업이 이러한 접근 방식을 적용 할 가능성이 있지만 관련 결과는 민감할 수 있으므로 더 넓은 커뮤니티에서 사용할 수 없습니다. 이 사실은 DoD 잉크젯 공정에서 액적 생성에 대한 여러 매개 변수의 영향을 평가하기 위한 이러한 종류의 연구로서 현재 논문의 영향을 증가 시킬 수 있습니다.

    CFD 프레임 워크 내에서 VOF 접근 방식을 사용하여 여러 컴퓨터 실험의 설계를 개발하고 RSM을 분석 도구로 사용했습니다. 충분한 수치 정확도와 수용 가능한 시간 계산 시뮬레이션의 균형을 맞추기 위해 메쉬 수렴 연구가 수행되었습니다. 설계 목적을 위해 점도, 표면 장력, 입구 속도 및 노즐 직경이 입력 요인으로 선택되었습니다. 응답은 break-up 시간과 break-up 길이였습니다.

    Figure 1. Schematic of the computational domain
    Figure 1. Schematic of the computational domain
    Figure 2. Ink fraction contours for mesh 1 through 4 (left to right) at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs.
    Figure 2. Ink fraction contours for mesh 1 through 4 (left to right) at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs.
    Figure 3. Comparison between surface tensions at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs
    Figure 3. Comparison between surface tensions at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs
    Figure 4. Comparison between viscosity values at the following four time steps: (a) 6 μs, (b) 12 μs, (c) 18 μs, and (d) 24 μs.
    Figure 4. Comparison between viscosity values at the following four time steps: (a) 6 μs, (b) 12 μs, (c) 18 μs, and (d) 24 μs.
    Figure 5. Comparison between different nozzle diameters at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs
    Figure 5. Comparison between different nozzle diameters at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs
    Figure 6. Comparison between different inlet velocities at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs
    Figure 6. Comparison between different inlet velocities at the following four time steps: (a) 6 µs, (b) 12 µs, (c) 18 µs, and (d) 24 µs
    Figure 8. Contour response plots for break-up time as a function of (a) surface tension and viscosity, (b) nozzle diameter and viscosity, (c) inlet velocity and viscosity, (d) nozzle diameter and surface tension, (e) inlet velocity and surface tension, and (f) inlet velocity and nozzle diameter.
    Figure 8. Contour response plots for break-up time as a function of (a) surface tension and viscosity, (b) nozzle diameter and viscosity, (c) inlet velocity and viscosity, (d) nozzle diameter and surface tension, (e) inlet velocity and surface tension, and (f) inlet velocity and nozzle diameter.
    Figure 12. Break-up length as a function of the We–Ca space (obtained from the 25 runs).
    Figure 12. Break-up length as a function of the We–Ca space (obtained from the 25 runs).

    References

    1. Hutchings, I.M.; Martin, G.D. Inkjet Technology for Digital Fabrication; John Wiley & Sons Ltd.: Hoboken, NJ,
      USA, 2013.
    2. Waasdorp, R.; Heuvel, O.; Versluis, F.; Hajee, B.; GhatKesar, M. Acessing individual 75-micron diameter
      nozzles of a desktop inkjet printer to dispense picoliter droplets on demand. RSC Adv. 2018, 8, 14765.
    3. Zhang, H.; Wang, J.; Lu, G. Numerical investigation of the influence of companion drops on drop-ondemand ink jetting. Appl. Phys. Eng. 2012, 13, 584–595.
    4. Dong, H.; Carr, W. An experimental study of drop-on-demand drop formation. Phys. Fluids 2006, 18,
      072102.
    5. Patel, M.; Pericleous, K.; Cross, M. Numerical Modelling of Circulating Fluidized beds. Int. J. Comput.
    6. Fluid Dyn. 1993, 1, 161–176. [CrossRef]
    7. Zhao, X.; Glenn, C.; Xiao, Z.; Zhang, S. CFD development for macro particle simulations. Int. J. Comput.
    8. Fluid Dyn. 2014, 28, 232–249. [CrossRef]
    9. Hasan, M.N.; Chandy, A.; Choi, J.W. Numerical analysis of post-impact droplet deformation for direct-print.
    10. Eng. Appl. Comput. Fluid Mech. 2015, 9, 543–555. [CrossRef]
    11. Ghafouri-Azar, R.; Mostaghimi, J.; Chandra, S. Numerical study of impact and solidification of a droplet
    12. over a deposited frozen splat. Int. J. Comput. Fluid Dyn. 2004, 18, 133–138. [CrossRef]
    13. Feng, J. A General Fluid Dynamic Analysis of Drop Ejection in Drop-on-Demand Ink Jet Devices. J. Imaging
    14. Sci. Technol. 2002, 46, 398–408.
    15. Fromm, J. Numerical Calculation of the Fluid Dynamics of Drop-on-Demand Jets. IBM J. Res. Dev. 1984, 28,
    16. 322–333. [CrossRef]
    17. Nallan, H.; Sadie, J.; Kitsomboonloha, R.; Volkman, S.; Subramanian, V. Systematic Design of Jettable
    18. Nanoparticle-Based Inkjet Inks: Rheology, Acoustics and Jettability. Langmuir 2014, 30, 13470–13477.
    19. [CrossRef] [PubMed]
    20. Reis, N.; Derby, B. Ink Jet Deposition of Ceramic Suspensions: Modelling and Experiments of Droplet Formation;
    21. Chapter in MRS Online Proceeding Library Archive; Cambridge University Press: Cambridge, UK, 2000;
    22. Volume 624, pp. 117–122.
    23. Jang, D.; Kim, D.; Moon, J. Influence of Fluid Physical Properties on Ink-Jet Printability. Langmuir 2009, 25,
    24. 2629–2635. [CrossRef] [PubMed]
    25. Tai, J.; Gan, H.Y.; Liang, Y.N.; Lok, B.K. Control of Droplet Formation in Inkjet Printing Using Ohnesorge
    26. Number Category: Materials and Processes. In Proceedings of the 10th Electronics Packaging Technology
    27. Conference, EPTC, Singapore, 9–12 December 2008; pp. 761–766.
    28. Box, G.; Wilson, K. On the Experimental Attainment of Optimum Conditions. J. R. Stat. Soc. Ser. B 1951, 13,
    29. 1–45.
    30. Silva, V.; Rouboa, A. Optimizing the gasification operating conditions of forest residues by coupling a
    31. two-stage equilibrium model with a response surface methodology. Fuel Process. Technol. 2014, 122, 163–169.
    32. [CrossRef]
    33. Silva, V.; Rouboa, A. Optimizing the DMFC Operating Conditions using a Response Surface Method.
    34. Appl. Math. Comput. 2012, 218, 6733–6743. [CrossRef]
    35. Silva, V.; Rouboa, A. Combining a 2-D multiphase CFD model with a Response Surface Methodology to
    36. optimize the gasification of Portuguese biomasses. Energy Convers. Manag. 2015, 99, 28–40. [CrossRef]
    37. Frawley, P.; Corish, J.; Niven, A.; Geron, M. Combination of CFD and DOE to analyse solid particle erosion
    38. in elbows. Int. J. Comput. Fluid Dyn. 2009, 23, 411–426. [CrossRef]
    39. Morrison, N.F.; Harlen, O.G. Viscoelasticity in inkjet printing. Rheol. Acta 2010, 49, 619–632. [CrossRef]
    40. ANSYS Inc. ANSYS Fluent Tutorial Guide; Release 15.0; ANSYS Inc.: Canonsburg, PA, USA, November 2013.
    41. ANSYS Inc. ANSYS Fluent Theory Guide; Release 17.0; ANSYS Inc.: Canonsburg, PA, USA, January 2016.
    42. Dinsenmeyer, R.; Fourmigué, J.F.; Caney, N.; Marty, P. Volume of fluid approach of boiling flows in
    43. concentrated solar plants. Int. J. Heat Fluid Flow 2017, 65, 177–191. [CrossRef]
    44. Das, S.; Weerasiri, L.D.; Yang, W. Influence of surface tension on bubble nucleation, formation and onset of
    45. sliding. Colloids Surf. A Physicochem. Eng. Asp. 2017, 516, 23–31. [CrossRef]
    46. Du, W.; Zhang, J.; Lu, P.; Xu, J.; Wei, W.; He, G.; Zhang, L. Advanced understanding of local wetting
    47. behaviour in gas-liquid-solid packed beds using CFD with a volume of fluid (VOF) method. Chem. Eng. Sci.
    48. 2017, 170, 378–392. [CrossRef]
    49. Shrestha, S.; Chou, K. A build surface study of Powder-Bed electron beam additive manufacturing by
    50. 3D thermo-fluid simulation and white-light interferometry. Int. J. Mach. Tools Manuf. 2017, 121, 37–49.
    51. [CrossRef]
    52. Zhong, Y.; Fang, H.; Ma, Q.; Dong, X. Analysis of droplet stability after ejection from an inkjet nozzle. J. Fluid
    53. Mech. 2018, 845, 378–391. [CrossRef]
    54. Zhang, X. Dynamics of drop formation in viscous flows. Chem. Eng. Sci. 1999, 54, 1759–1774. [CrossRef]
    55. Calvert, P. Inkjet printing for materials and devices. Chem. Mater. 2001, 13, 3299–3305. [CrossRef]
    56. Kim, C.S.; Park, S.; Sim, W.; Kim, Y.; Yoo, Y. Modelling and characterization of an industrial inkjet head for
    57. micro-patterning on printed circuit boards. Comput. Fluids 2009, 38, 602–612. [CrossRef]
    58. ChemEngineering 2018, 2, 51 19 of 19
    59. Wang, P. Numerical Analysis of Droplet Formation and Transport of a Highly Viscous Liquid. Master’s Thesis,
    60. University of Kentucky, Lexington, KY, USA, 2014.
    61. Zhang, Z.; Xiong, R.; Corr, D.; Huang, Y. Study of Impingement Types and Printing Quality during Laser
    62. Printing of Viscoelastic Alginate Solutions. Langmuir 2016, 32, 3004–3014. [CrossRef] [PubMed]
    63. Derby, B. Inkjet Printing Ceramics: From Drops to Solid. J. Eur. Ceram. Soc. 2011, 31, 2543–2550. [CrossRef]
    64. Kim, E.; Baek, J. Numerical Study on the Effects of Non Dimensional Parameters on Drop-on-Demand
    65. Droplet Formation Dynamics and Printability Range in the up-Scaled Model. Phys. Fluids 2012, 24, 082103.
    66. [CrossRef]
    FIG. 2. Sequence of images showing capillary-driven neck evolution and droplet formation for low-viscosity fluids

    Computational analysis of self-similar capillary-driven thinning and pinch-off dynamics during dripping using the volume-of-fluid method

    낙하 형성 및 분리는 표면 장력 구동 흐름으로 인해 가늘어지는 유체 목의 형성을 포함하여 큰 위상 변화를 수반하며, 목의 pinch-off에서 Laplace pressure와 같은 속성은 유한한 시간 특이성을 나타냅니다. 드롭 형성 중에 발생하는 큰 위상 변형과 비선형성을 정확하게 시뮬레이션하는 것은 일반적으로 pinch-off 순간에 가까운 작은 특징을 해결하기 위해서는 고해상도 및 정확도가 필요하기 때문에 수치 시뮬레이션이 계산적으로 요구됩니다.

    필요한 질량 및 계산 시간을 보존하고 인터페이스를 추적하는 데 내재된 이점에도 불구하고, 초기 실무자들이 물 점도가 10배 이상인 유체에 대한 수렴 문제를 보고했기 때문에 낙하 형성 연구에 VOF(Volume-of-fluid) 방법을 활용하는 연구는 거의 없습니다.

    이 기여에서, 우리는 FLOW-3D에 구현된 VOF 방법을 사용하여 물 점도보다 4배 더 높은 점도 값을 포함하여 뉴턴 유체에 대한 드리프트의 원형 자유 표면 흐름을 시뮬레이션합니다. 우리는 이 연구의 일부로 수행된 실험에 대해 시뮬레이션된 목 모양, 목 진화 속도 및 헤어짐 길이를 벤치마킹합니다.

    핀치오프 역학은 관성, 점성 및 모세관 응력의 복잡한 상호 작용에 의해 결정되며, 여기서 실험과 시뮬레이션 모두에서 대조되는 자기 유사 스케일링 법칙은 종종 역학에 대해 설명합니다. 우리는 시뮬레이션된 반지름 진화 프로파일이 축 대칭 흐름에 대한 뉴턴 유체에 대해 실험적으로 관찰되고 이론적으로 예측되는 핀치오프 역학과 일치함을 보여준다. 또한, 우리는 가는 목 안에서 법칙, 속도 및 변형 필드의 스케일링에 대한 사전 요인을 결정하고, 우리는 실험과 비교할 수 있는 중단 시간과 길이뿐만 아니라 사전 요인을 VOF 방법을 사용하여 시뮬레이션할 수 있음을 보여줍니다.

    experimental setup, as shown schematically in Fig. 1(a), includes a dispensing system
    experimental setup, as shown schematically in Fig. 1(a), includes a dispensing system
     A numerical simulation of drop formation from a cylindrical nozzle at a constant flow rate is performed. (c) Graphical representation of the VOF approach
    A numerical simulation of drop formation from a cylindrical nozzle at a constant flow rate is performed. (c) Graphical representation of the VOF approach
    FIG. 2. Sequence of images showing capillary-driven neck evolution and droplet formation for low-viscosity fluids
    FIG. 2. Sequence of images showing capillary-driven neck evolution and droplet formation for low-viscosity fluids. (a) A sequence of simulated images of water (0 wt. % glycerol) shows neck formation and subsequent thinning and pinch-off dynamics including the formation of the satellite drop. (b) A sequence of images shows neck radius evolution and drop detachment for the low viscosity fluid composed of 50 wt. % glycerol in water. The time step between images is 500 µs, and the scale bar represents a length of 1 mm for the two cases shown. The color bar shows the velocity field in units of cm/s. The addition of glycerol seems to exercise a relatively minor influence on pinch-off dynamics despite a five-fold increase in viscosity.
    FIG. 3. Computed evolution of the minimum radius of the water neck during the drop formation and detachment process
    FIG. 3. Computed evolution of the minimum radius of the water neck during the drop formation and detachment process. The instantaneous neck radius of water and the inertio-capillary fit are shown. The inset shows a self-similar nature of neck thinning dynamics close to a pinch-off moment. The characteristic cone angle of 18.1◦ as predicted by Day et al.50 and visualized in experiments52 is captured well using the VOF method.
    FIG. 5. Glycerol thinning image sequence and break-up length visualization for three cases
    FIG. 5. Glycerol thinning image sequence and break-up length visualization for three cases. (a) Glycerol thinning is shown through a sequence of snapshots with a time step ∆t = 5 ms and reveals quite different dynamics compared to previously seen for low viscosity fluids. The length of a filament changes significantly when the glycerol content increases above 70 wt. %. (b) Final lengths of the simulated liquid filaments before pinch-off for three cases of glycerol + water mixtures (0 wt. %, 70 wt. %, and 100 wt. %).
    FIG. 8. Comparison of experiments and simulations for the case of a drop formation for 80 wt. % glycerol and water mixture
    FIG. 8. Comparison of experiments and simulations for the case of a drop formation for 80 wt. % glycerol and water mixture. (a) A set of images obtained from experiments (upper row) and simulations (bottom row) with a time step of 1 ms show good agreement. The simulated drop profiles shown in the bottom row are colored by the velocity magnitude [ranging from 0 (dark blue) to 100 cm/s (red) and colored online], and velocity vectors are shown in the images. (b) Radius evolution with time of liquid filament formed during the drop formation process is shown on a log-log plot for the two cases.

    레이놀즈 수

    본 자료는 국내 사용자들의 편의를 위해 원문 번역을 해서 제공하기 때문에 일부 오역이 있을 수 있어서 원문과 함께 수록합니다. 자료를 이용하실 때 참고하시기 바랍니다.

    Reynolds Number

    레이놀즈 수

    주어진 수치 방법에 의해 정확하게 계산 될 수 있는 유동에 대해서 가장 높고, 가장 낮은 레이놀즈 수 무엇입니까? 이 질문은 다양한 답과 그리고 가장 기술적인 문제들로서 주어진 답을 포함하는 가정들로부터 다양한 답을 가지고 있습니다.

    본 목적을 위해, 레이놀즈 수는 R = R LU / ν로 정의되며, 여기서 L과 U는 유동 특성 길이 및 스케일이고, ν는 유체의 동점도(kinematic viscosity )입니다. 즉 물체의 관성이 점성에 비해서 얼마나 큰가를 나타내는 척도로 이 레이놀즈 수가 작을수록 층류(유체의 유선이 유지되면서 흐르는 유동)가, 클수록 난류가 형성된다. 무 차원 레이놀즈 수가 점성의 관성 효과의 측정을 중요성을 상기시킵니다. 높은 레이놀즈 수에서의 흐름은 정성적으로 다른 행동을 나타내고, 난류 될 수 있습니다.

    일반적으로 고려해야 할 가장 중요한 한계는 높은 레이놀즈 수입니다. 이것은 층류가 난류로의 분해 또는 경계층이 표면에서 분리되는 위치에 따라 달라지는 몸체의 양력 및 항력을 예측하는 데 계산이 사용될 수 있는 한계입니다. 유동에 대한 점성 응력의 상대적 효과를 정확하게 시뮬레이션 하는 것이 중요한 이러한 또는 다른 유형의 유동 프로세스에서는 계산에서 어떤 수준의 정확도를 기대할 수 있는지에 대한 아이디어를 갖는 것이 유용합니다.

    일반적으로 고려해야 할 가장 중요한 한계는 높은 레이놀즈 수입니다. 이것은 층류에서 난류로 붕괴되는 것을 예측하곤 하는 계산의 한계치이며, 유동의 경계층이 그 표면에서부터 박리되는 곳에서의 물체의 양력과 항력을 예측하는 한계치이기도 합니다. 유동의 다양한 유형에서 유동의 점성 응력의 상대적 효과를 정확하게 시뮬레이션하는 것은 중요하며, 계산상 예측되는 정확도의 수준에 대한 어떤 아이디어를 확보하는 것 또한 매우 유용할 것입니다.

    높은 레이놀즈 수 제한 – 물리적 인수

    흐름을 정확하게 표현하는데 필요한 계산 요구 사항 (즉, 해상도)을 추정하기 위해 간단한 물리적 인수를 사용할 수 있습니다. 이 주장은 흐름 영역이 작은 요소로 세분화 될 때 요소 내의 모든 흐름량이 천천히 변한다는 가정을 기반으로 합니다. 이 가정은 각 요소의 평균 수량 값이 요소 내의 실제 값에 대한 좋은 근사치라는 의미를 전달합니다.

    요소 내에서 느리게 변하는 속도를 가지려면 요소 크기의 척도에서 흐름의 레이놀즈 수가 작아야 합니다 (예 : 1 차 Rd = dx · du / ν ≤ 1.0). 이 표현에서 dx와 du는 요소의 길이와 속도 스케일입니다. 이 물리적 요구 사항, 요소의 흐름의 부드러움 (즉, 낮은 레이놀즈 수, 이 척도의 층류 흐름)은 정확한 수치 분해능에 필요한 요소의 크기를 정의하는데 사용될 수 있습니다.

    위의 부등식은 L = Ndx 및 U = Ndu 관계에 의해 거시적 레이놀즈 수로 변환 될 수 있으며, 이는 R ≤ N 2로 이어집니다 . 즉, 개별 요소의 규모에 대한 부드러운 흐름의 물리적 정확도 요구 사항은 정확도로 계산할 수 있는 최대 레이놀즈 넘버원이 NN 2 정도라는 것을 의미합니다. 여기서 N은 특성을 해결하는 데 사용되는 요소의 수입니다. 길이 L.

    대표적인 응용에서 N은 종종10 내지 20의 범위에 있는 수로서 매우 큰 수 아닙니다. 그리고 이는 단지 약400 의 정확한 계산을 위해 최대 레이놀즈 수로 변환합니다. 이 결과에 대해 해석을 달기 전에 정확한 레이놀즈 수 계산을 위한 추정을 위해서 다른 접근 방법을 시도하는 유익합니다.

    High Reynolds Number Limit – A Numerical Argument

    수치 근사에 의해서 계산 도입된 viscous-like smoothing의 양은 truncation error로부터 평가 될 수 있습니다. 알다시피 아이디어는 요소 크기 (그리고 적정한 시간 간격 크기) 멱함수을 미분 근사하는 테일러 급수 전개를 하는 것입니다. 물론, 일관성 있는 근사는 원래 근사환 된 편미분 방정식의 가장 낮은 차수를 이용하는 것입니다.

    다음으로 높은 차수는 보통 확산 (즉, 2차 차수 공간 미분형태) 항입니다. 점성 계수와 더불어 이러한 항의 계수 비교는 점성 효과를 더 정확하게 계산 할 수 없을 때의 추정치를 제공합니다.

    1차 수치 근사 (예를 들어 대류에 대한donor cell 또는upwind technique )에 대해서 정확도를 위해서 1보다 적어야만 하는 항들의 비는 다음의 판별식을 유도하게 됩니다( R ≤ 2N.) 그리고 2차 수치 근사의 결과, R ≤ N얻어지고 물리적인 인자(Physical Argument)로부터 같은 결과가 얻어 집니다.

    이러한 관계의 우변을 곱하는 작은 숫자 요소가 사용되며, 이는 사용 된 특정 수치 근사에 따라 달라 지지만 N에 대한 기본 종속성은 변경되지 않습니다. 모든 2 차 방법이 1 차 방법보다 분명히 훨씬 낫지 만 결과는 고무적이지 않습니다. 정확하게 계산할 수 있는 최대 레이놀즈 수는 N을 늘리지 않는 한 매우 제한적인 것으로 보입니다. 이는 매우 큰 그리드를 처리한다는 의미입니다.

    하이 레이놀즈 수에 대한 일반적인 의견

    이러한 평가들은 첫 발생 시에는 실망스런 부분도 있으나 종종 완화되는 상황으로 전개됩니다. 무엇보다도 중용한 것은 대부분의 문제들은 점성 응력에 대한 정확한 처리를 요구하지 않습니다. 이러한 문제에 대해서 높은 레이놀즈 수의 상한은 점성 효과가 중요하지 않다는 것을 의도한 의미를 갖습니다.

    어떤 유동이 난류에 의해 운동량 혼합이 이루워진 fully turbulent 되기 위해 충분히 높은 레이놀즈 수를 가질 때, 종종 잘 분류될 수 있는 scale을 가진 영역 내에서 100 미만의 유효한 레이놀즈 수의 평균 유동으로 진행되곤 합니다. 물론, 이것은 난류를 기술할 수 있는 적당한 난류 모델이 사용되고 있다는 것을 가정합니다.

    마지막으로 점성 효과의 정확한 정보에 따라 일부 유동 특성을 가질 필요가 있을 때 인위적인 의미의 효과를 유도하는 것이 가능 할 수 있습니다. 예를 들어, 풍동 trip wire는 종종 레이놀즈 수 상사성( similarity )의 부족을 고려하여 trigger 유동의 박리에 사용되곤 합니다. 비슷한 처리가 풍동의 수치 시뮬레이션에 추가 될 수 있습니다.

    결론은 CFD 방법을 사용하여 높은 레이놀즈 수 흐름을 계산하는 데 사용할 수 있지만 수치해석상의 전산 오차가 물리적인 효과를 압도 할 수 있는 상황에 대한 경고는 해당 난류 모델에 달려있다고 말할 수 있습니다.

    낮은 레이놀즈 수 제한

    낮은 레이놀즈 수에서 한계는 정밀도의 한계가 아니라 계산을 완료하는데 필요한 계산 시간을 기준으로 한계입니다.  점성 응력 항에 explicit 수치 근사를 사용하면 숫자의 안정성을 유지하기 위해 시간 단계의 크기에 한계가 있습니다.  이 한계는 본질적으로 점성으로 인한 운동량의 변화는 하나의 시간 단계에서 대략 1 개의 요소를 넘어 전파하는 것은 아니라는 것을 보여줍니다.  단순한 2 차원의 경우에는 이 한계는 νdt ≤ dx2/4입니다.

    이것은 T = Mdt 및 TU = L이라는 대응을 작성하여 레이놀즈 수를 포함하는 식으로 변형 할 수 있습니다.  즉, 흐름의 특성 시간은 속도 U의 유체가 거리 L을 이동하는 시간이며, 시간 T를 분해 시간 단계의 수는 M입니다.  이러한 관계식에 의해 안정된 조건은 M = 4N2/R 입니다.

    이 결과에서 중요한 것은 M이 R에 반비례하여 증가하는 것입니다.  레이놀즈 수가 매우 작은 흐름의 경우 explicit 수치 법에는 매우 많은 시간 단계가 필요할 수 있으며,이 숫자는 해상도의 상승에 따라 급속히 증가하고 있습니다.  낮은 레이놀즈 수의 한계를 가장 효과적으로 제거하는 방법은 implicit 수치 법을 사용하여 점성 응력을 평가하는 것입니다.


    Reynolds Number

    What are the highest and lowest Reynolds number flows that can be accurately computed by a given numerical method? This question has a variety of answers, and, as with most technical issues, the variety of answers arises from the assumptions involved in giving the answer.

    For present purposes, the Reynolds number R is defined as R=LU/ν, where L and U are characteristic length and velocity scales for a flow, and ν is the kinematic viscosity of the fluid. It will be recalled that the non dimensional Reynolds number is a measure of the importance of inertia to viscosity effects. At high Reynolds numbers a flow may become turbulent, exhibiting qualitatively different behavior.

    Generally, the most important limit to consider is that of high Reynolds numbers. This is the limit where computations might be used to predict the breakdown of a laminar flow into turbulence, or the lift and drag of a body that is dependent on where boundary layers separate from its surface. In these or other types of flow processes in which it is critical to correctly simulate the relative effect of viscous stresses on the flow, it is useful to have some idea of what level of accuracy can be expected in a computation.

    The reason that a Reynolds number limitation exists in computational fluid dynamics CFD) is that the computational stability of most CFD methods relies on some type of numerical smoothing or homogenizing within the computational elements. Since viscosity is a physical mechanism for smoothing flow variations, there can be a problem differentiating between numerical and physical smoothing. This is especially important when critical Reynolds number situations are encountered, because they require an especially accurate estimate of viscous stresses.

    High Reynolds Number Limit – A Physical Argument

    A simple physical argument can be used to estimate the computational requirements (i.e., resolution) needed to achieve an accurate representation of a flow. The argument is based on the assumption that when a flow region is subdivided into small elements all flow quantities within an element are slowly varying. This assumption carries the implication that the average values of quantities in each element are good approximations for the actual values within the element.

    To have a slowly varying velocity within an element, the Reynolds number of the flow on scales of the element size must be small, say of order one, Rd=dx·du/ν ≤ 1.0. In this expression dx and du are length and velocity scales characteristic of the element. This physical requirement, the smoothness of the flow in elements (i.e., a low Reynolds number, laminar flow on this scale), may be used to define the size of elements needed for an accurate numerical resolution.

    The above inequality can be converted to a macroscopic Reynolds number by the relations, L=Ndx and U=Ndu, which leads to R ≤ N2. In other words, the physical accuracy requirement of a smooth flow on the scale of individual elements implies that the maximum Reynolds number one can expect to compute with accuracy is on the order of NN2 where N is the number of elements used to resolve a characteristic length L.

    In typical applications, N is often in the range of 10 to 20, which translates to a maximum Reynolds number for accurate computations of only about 400, not a very large number! Before commenting on this result it is instructive to try a different approach for estimating the limit for accurate Reynolds number computations.

    High Reynolds Number Limit – A Numerical Argument

    The amount of viscous-like smoothing introduced into a computation by numerical approximations can be estimated from truncation errors. The idea is to do a Taylor Series expansion on the difference approximations in powers of the element size (and time-step size if that is appropriate). Of course, a consistent approximation should have as its lowest order terms the partial differential equation that was originally being approximated.

    At the next higher order there are usually terms that have the character of a diffusion (i.e., second-order space derivatives). A comparison of the coefficients of these terms with the coefficient of viscosity gives an estimate of when viscous effects would no longer be computed accurately.

    For a first-order numerical approximation (e.g., a donor cell or upwind technique for advection) the ratio of terms, which must be less than one for accuracy, leads to the criteria R ≤ 2N. With a second-order approximation the result is R ≤ N2, the same result obtained from the “Physical Argument.”

    There are small numerical factors multiplying the right-hand sides of these relations, which depend on the specific numerical approximations used, but the basic dependencies on N remain unchanged. Any second-order method is clearly much better than a first-order method, but the results are not encouraging. The maximum Reynolds number that can be computed accurately appears to be quite limited, unless one is willing to increase N, which means dealing with extremely large grids.

    General Comments on High Reynolds Numbers

    These estimates are discouraging when first encountered, but there are frequently mitigating circumstances. Foremost is the realization that most problems do not require an accurate treatment of viscous stresses. For these problems the high Reynolds number limit has the intended meaning that viscous effects are not important.

    When flows have a high enough Reynolds number to be fully turbulent the momentum mixing induced by the turbulence often leads to a mean flow with an effective Reynolds number that is less than 100, well within the range of resolvable scales. Of course, this assumes that a suitable turbulence model is available to describe the turbulence.

    Finally, when it is necessary to have some flow property that depends on an accurate knowledge of viscous effects, it may be possible to induce that effect by artificial means. For example, in wind tunnels trip wires are sometimes used to trigger flow separations to account for a lack of Reynolds number similarity. A similar treatment can be added to a numerical simulation of a wind tunnel.

    The bottom line is, CFD methods can be used to compute high Reynolds number flows, but it is up to the modeler to be alert for situations where numerical errors could overshadow physical effects.

    Low Reynolds Number Limit

    At low Reynolds numbers the limit is not one of accuracy but a limit based on the computational time necessary to complete a computation. When explicit numerical approximations are used for viscous stress terms there is a limit on the size of the time step to maintain numerical stability. That limit is essentially a statement that momentum changes caused by viscosity do not propagate more than about one element in one time step. In a simple two-dimensional case this limit is νdt ≤ dx2/4.

    This can be transformed into an expression involving the Reynolds number by making the correspondences: T=Mdt and TU=L. That is, the characteristic time for a flow is the time for fluid at velocity U to move a distance L, and the number of time steps resolving time T is M. With these relations the stability condition is then, M = 4N2/R.

    The importance of this result is that M increases inversely with R. For very low Reynolds number flows, explicit numerical methods may require a very large number of time steps, and this number increases rapidly with an increase in resolution. The low Reynolds number limit is best eliminated by employing an implicit numerical method for evaluating viscous stresses.

    FLOW-3D 튜토리얼 V12

    FLOW-3D 튜토리얼 V12

    빠른 시작

    이 튜토리얼 매뉴얼은 FLOW-3D 처음 사용하는 사용자에게 그래픽 사용자 인터페이스(GUI)의 주요 구성 요소를 쉽게 익히도록 하고, 다양한 시뮬레이션의 설정 및 실행 방법을 안내하기 위한 것입니다.

    이 매뉴얼에 있는 실습과정은 FLOW-3D의 기본 사항을 다루기 위한 것입니다. 이 매뉴얼에서 제시하는 문제는 다양한 주제를 설명하고, 발생할 수 있는 많은 질문을 해결하기 위해 선정되었습니다. 이 매뉴얼의 실습과정은 FLOW-3D실행하는 컴퓨터에 앉아 사용하는 것이 가장 좋습니다.

    CFD 사용 철학에 대한 간단한 섹션 다음에는 중요 파일과 시뮬레이션 파일을 실행하는 방법이 소개되어 있습니다. 이 소개 섹션 다음에는 모델 설정, 시뮬레이션 실행 및 포스트 프로세스, Simulation Manager 탐색 방법에 대한 설명이 있습니다. 이러한 각 단계에 대한 자세한 내용은 모델 설정, 컴퓨팅 결과 및 후처리 장에서 확인할 수 있습니다.

    1.CFD 사용에 대한 철학

    CFD (Computational Fluid Dynamics)는 유체 흐름(질량, 운동량 및 에너지 보존)에 대한 지배 방정식의 컴퓨터 솔루션입니다. 지정된 지배방정식은 이론 장에 설명된 Numerical방법을 사용하여 이산화되고 계산됩니다.

    CFD 소프트웨어를 사용하는 것은 여러 면에서 실험을 설정하는 것과 유사합니다. 실제 상황을 시뮬레이션하기 위해 실험을 올바르게 설정하지 않으면, 그 결과는 실제 상황을 반영하지 않습니다. 같은 방법으로 수치 모델이 실제 상황을 정확하게 나타내지 않으면, 그 결과는 실제 상황을 반영하지 않습니다. 사용자는 어떤 것이 중요한지, 어떻게 표현해야 하는지를 결정해야 합니다. 시작하기 전에 다음과 같은 질문을 하는 것이 중요합니다.

    • CFD 계산에서 무엇을 알고 싶습니까?
    • 중요한 현상을 포착하기 위해 규모와 Mesh는 어떻게 설계되어야 하는가?
    • 실제 물리적 상황을 가장 잘 나타내는 경계 조건은 무엇입니까?
    • 어떤 종류의 유체를 사용해야합니까?
    • 이 문제에 어떤 유체 특성이 중요합니까?
    • 다른 어떤 물리적 현상이 중요합니까?
    • 초기 유체 상태는 어떻게 됩니까?
    • 어떤 단위 시스템을 사용해야합니까?

    모델링 되는 문제가 실제 상황을 가능한 한 유사하게 나타내는지 확인하는 것이 중요합니다. 사용자는 복잡한 시뮬레이션 작업을 해결 가능한 부분으로 나누는 것이 좋습니다.

    복잡한 물리 효과를 추가하기 전에, 간단하고 쉽게 이해할 수 있는 근사값으로 점차적으로 시작하여 프로세스 진행하십시오. 간단한 손 계산(베르누이 방정식, 에너지 균형, 파동
    전파, 경계층 성장 등)은 물리 및 매개 변수를 선택하는데 도움이 되고, 결과와 비교할 수 있는 점검항목을 제공합니다.

    CFD의 장단점을 이해하면 분석을 진행하는데 도움이 될 수 있습니다. CFD는 다음과 같은 경우 탁월한 분석 옵션입니다.

    • 기하 구조, 물리학 또는 필요한 상세 수준으로 인해 표준 엔지니어링 계산이 유용하지 않은 경우가 많습니다.
    • 실제 실험은 비용이 많이 소요됩니다.
    • 실험에서 수집할 수 있는 것보다 유체흐름에 대한 자세한 정보가 필요한 경우 유용합니다.
    • 위험하거나 적대적인 조건, 확장이 잘되지 않는 프로세스 등으로 인해 정확한 실험 측정을 하기가 어려운 경우
    • 복잡한 흐름 정보에 대한 커뮤니케이션

    CFD는 다음과 같은 경우에 덜 효과적입니다.

    • 솔루션이 계산 리소스가 매우 많이 소요되거나, 도메인 크기를 줄이기 위한 가정 또는 해결되지 않은 물리적 현상을 설명하기 위한 반 임계 모델이 필요한 경우
    • CFD 시뮬레이션에 대한 입력이 되는 중요한 물리적 현상이 알려지지 않은 경우
    • 물리적 현상이 잘 이해되지 않거나 매우 복잡한 경우

    CFD를 사용할 때 명심해야 할 몇 가지 중요한 참고 사항이 있습니다.

    • CFD는 규정된 초기 및 경계 조건에 따라 지정된 지배 방정식의 수치해석 솔루션입니다. 따라서 모델 설정, 즉 어떤 방정식을 풀어야 하는지, 재료 특성, 초기 조건 및 경계 조건이, 가능한 한 물리적 상황과 최대한 일치해야 합니다.
    • 방정식의 수치 해는 일반적으로 어떤 종류의 근사치를 필요로 합니다. 물리적 모델에 대한 가정과 해결방법을 검토한 후 사용하는 것이 좋습니다.
    • 디지털 컴퓨터는 숫자가 유한 정밀도로 이진수로 표시되는 방식으로 인해 반올림 오류가 발생합니다. 이는 문제를 악화시키기 때문에 매우 근소한 숫자의 차이를 계산해야 하는 상황을 피하십시오. 이러한 상황의 예는 시뮬레이션 도메인이 원점에서 멀리 떨어져 있을 때입니다.

     

    2.중요한 파일

    FLOW-3D 시뮬레이션과 관련된 많은 파일이 있습니다. 가장 중요한 것들이 아래에 설명되어 있습니다. 모든 prepin.* 파일의 명칭에서 prepin는 파일 형식을 의미하며, 별표시* 위치는 시뮬레이션 이름을 의미합니다. ( : prepin.example_simulation.)

    • ·prepin.*: 시뮬레이션용 입력 파일입니다. 시뮬레이션 설정을 설명하는 모든 입력 변수가 포함되어 있습니다.
    • ·prpgrf.*: 이것은 전 처리기 출력 파일입니다. 여기에는 계산된 초기 조건이 포함되며 시뮬레이션을 실행하기 전에 설정을 확인하는 데 사용될 수 있습니다.
    • ·flsgrf.*: 솔버 출력 파일입니다. 시뮬레이션의 최종 결과가 포함됩니다.
    • ·prperr.*, report.*, prpout.*: 이 파일들은 Preprocessor Diagnostic Files.
    • ·hd3err.*, hd3msg.*, hd3out.*: 이 파일들은 Solver Diagnostic Files.

    모든 시뮬레이션 파일은 단일 폴더에 함께 유지하므로, 설명이 될 수 있는 시뮬레이션 이름을 사용하는 것이 좋습니다. 그러나 매우 긴 파일 이름은 운영 체제에 따라 문제가 될 수 있습니다.

    노트

    • 시뮬레이션 이름이 inp(즉, 입력 파일이 있다면 prepin.inp) 출력 및 진단 파일은 모두 .dat이름을 갖습니다. 예: flsgrf.dat.
    • 모든 입력 파일은 네트워크 위치의 컴퓨터 대신 로컬 디렉토리에 저장하는 것이 좋습니다. 이것은 솔버가 더 빠르게 실행되고 GUI의 응답 속도가 빨라지며 실행중인 시뮬레이션을 방해하는 네트워크 문제 가능성을 제거합니다.

    3.시뮬레이션 관리자

    FLOW-3D 시뮬레이션 관리자의 탭은 주로 시뮬레이션을 실행할 수 있도록 시뮬레이션 환경을 구성하고 실행 시뮬레이션에 대한 상태 정보를 표시하는데 사용됩니다.

    작업 공간 (Workspaces)

    작업 공간(Workspaces)Simulation Manager의 필수 부분이며 파일을 FLOW-3D에서 처리하는 방식입니다. 기본적으로 시뮬레이션을 포함하고 구성하는 폴더입니다. 몇 가지 예를 들면 시뮬레이션과 또 다른 작업 공간인 검증 사례를 포함하도록 할 수 있습니다:

    포트폴리오의 작업 공간

    새로운 작업 공간 만들기

    튜토리얼에서는 작성하려는 시뮬레이션을 포함할 작업 공간(Workspaces)을 작성하십시오.

    1.File -> New workspace 이동

    2.작업 공간 이름으로 Tutorial를 입력하십시오.

    3.기본 위치는 현재 사용자의 홈 디렉토리에 있습니다. 다른 곳에서 찾을 수 있지만 기본 위치가 우리의 목적에 적합합니다.

    4.하위 디렉토리를 사용하여 작업 공간 이름 만들기 확인란을 선택합니다. 이렇게 하면 파일 시스템에서 작업 공간에 대한 새로운 하위 디렉토리가 만들어져 시뮬레이션 파일을 훨씬 쉽게 구성할 수 있습니다.

    새로운 작업 공간 만들기

    5.확인을 눌러 새 작업 공간을 작성하십시오. 이제 포트폴리오에 표시됩니다.

    새로운 작업 공간 만들기

    작업 공간 닫기

    포트폴리오를 정리하고 탐색하기 쉽도록 필요 없는 작업공간을 닫는 것이 편리합니다. 작업 공간을 닫으면 포트폴리오에서 해당 작업 공간만 제거됩니다. 그러나, 컴퓨터에서 작업 공간을 삭제하지는 않습니다.

    작업 공간을 닫으려면

    1.기존 작업 공간을 마우스 오른쪽 버튼으로 클릭하고 작업 Close Workspace 선택하십시오. 또는 포트폴리오에서 작업 공간을 선택 (왼쪽 클릭) 하고 Delete 키를 누를 수 있습니다.

    2.작업 공간을 닫을 것인지 묻는 메세지가 표시됩니다. 예를 선택하십시오.

    3.포트폴리오는 더 이상 닫힌 작업 공간을 포함하지 않습니다.

    기존 작업 공간 열기

    오래된 작업 공간을 열어야 할 때가 있을 것입니다. 예를 들어, 새 프로젝트에 유사한 시뮬레이션을 작성하기 전에 기존 시뮬레이션의 설정을 검토할 수 있습니다. 기존 작업 공간을 열려면

    1.File -> Open Workspace를 선택하십시오

    2.작업 공간 파일이 있는 디렉토리를 찾으십시오. Tutorial.FLOW-3D_Workspace.

    작업 공간 열기

    3.작업 공간을 로드 하려면 OK누르십시오.

    작업 공간에서 시뮬레이션 작업

    작업 공간을 사용하는 방법을 알았으니, 여기에 시뮬레이션을 추가해 봅시다.

    Example를 추가하십시오

    작업 공간에 작업 시뮬레이션을 추가하는 가장 간단한 방법은 포함된 예제 시뮬레이션 중 하나를 추가하는 것입니다. FLOW-3D의 다양한 기능을 사용하는
    방법을 보여주기 위해 설계된 간단하고 빠른 시뮬레이션입니다. 기존 작업 공간에 예제를 추가하려면 다음을 수행하십시오.

    1.포트폴리오에서 원하는 작업 공간을 강조 표시하십시오

    2.File -> Add example 선택하십시오. 또는 작업공간을 마우스 오른쪽 버튼으로 클릭하고 예제 추가선택할 수 있습니다.

    3.예제 대화 상자에서 예제를 선택하고 열기를 누르십시오. 자연 대류(Natural Convection) 예제를 선택했습니다.

    시뮬레이션 예제 추가

    4.새 시뮬레이션 대화 상자가 열립니다.

    5.디렉토리가 작업 공간 위치에 있는지 확인하는 것이 좋으므로 기본 시뮬레이션 이름과 위치를 잘 확인하는 것이 좋습니다. FLOW-3D는 모든 시뮬레이션 파일을 이 작업 공간 디렉토리의 별도 하위 디렉토리에 배치하여 파일 구성을 쉽게 만들어 줍니다.

    6.시뮬레이션을 위한 단위 시스템을 선택하십시오. 표준 단위 시스템이 권장되지만 각 단위를 독립적으로 선택하기 위해 사용자 지정 단위 시스템을 선택할 수 있습니다.

    7.확인을 눌러 새 시뮬레이션을 작업 공간에 추가하십시오.

    작업 공간에서의 시뮬레이션

    작업 공간에서 시뮬레이션 제거

    작업 공간에서 시뮬레이션을 제거해야 하는 경우가 있습니다 (이는 작업 공간에서 시뮬레이션을 제거만 하며, 컴퓨터에서 시뮬레이션을 삭제하지는 않습니다). 작업 공간에서 시뮬레이션을 제거하려면 다음을 수행하십시오.

    1.작업 공간에서 기존 시뮬레이션을 마우스 오른쪽 버튼으로 클릭하고 (이 경우 이전 섹션에서 추가 한 예제 사용) 시뮬레이션 제거를 선택하십시오. 또는 작업 공간에서 시뮬레이션을 선택 (왼쪽 클릭)하고 Delete 키를 누를 수 있습니다.

    2.작업 공간에는 더 이상 시뮬레이션이 포함되지 않습니다.

    모든 작업 공간 및 디스크에서 시뮬레이션 삭제

    작업 공간에서 시뮬레이션을 제거하는 것 외에도 디스크에서 모든 시뮬레이션 파일을 삭제해야 할 수도 있습니다. 작업 공간에서 시뮬레이션을 제거하고 디스크에서 시뮬레이션
    파일을 삭제하려면 다음을 수행하십시오.

    1.작업 공간에서 기존 시뮬레이션을 마우스 오른쪽 단추로 클릭하고 (이 경우 이전 섹션에서 추가 한 예제 사용) 모든 작업 공간 및 디스크에서 시뮬레이션
    삭제를
    선택하십시오.

    2.시뮬레이션 디렉토리에서 삭제할 파일을 선택할 수 있는 창이 나타납니다. 삭제할 파일을 선택한 다음 확인을 눌러 해당 파일을 삭제하거나 취소를 눌러 작업을 중단하십시오.

    3.OK를 선택한 경우 선택한 작업 공간은 더 이상 시뮬레이션을 포함하지 않습니다. 선택한 작업 공간의 모든 시뮬레이션 파일은 디렉토리에서 삭제됩니다.

    경고

    이 작업은 취소할 수 없으므로 계속하기 확인 후 파일을 삭제해야 합니다.

    작업 공간에 기존 시뮬레이션 추가

    기존 시뮬레이션을 작업 공간에 추가하려면 다음을 수행하십시오.

    1.열린 작업 공간을 마우스 오른쪽 버튼으로 클릭하고 기존 시뮬레이션 추가 선택합니다. 작업 공간을 선택한 다음 File->Add Existing Simulation 을 선택할 수도 있습니다.

    2.prepin.*파일 위치로 이동하여 열기를 선택하십시오.

    작업 공간에 기존 시뮬레이션 추가

    3.시뮬레이션이 이제 작업 공간에 나타납니다.

    작업 공간에 새로운 시뮬레이션 추가

    대부분의 경우 기존 시뮬레이션을 사용하는 대신 새 시뮬레이션을 작성하게 됩니다. 작업 공간에 새로운 시뮬레이션을 추가하려면:

    1.기존 작업 공간을 마우스 오른쪽 버튼으로 클릭하고 새 시뮬레이션 추가 선택하십시오.

    2.시뮬레이션 이름을 입력하라는 message가 표시됩니다. 이 예제에서는 heat transfer example 불러오십시오.

    3.그런 다음 드롭다운 목록을 사용하여 시뮬레이션을 위한 단위 시스템을 결정합니다. 사용 가능한 옵션은 질량, 길이, 시간, 전기요금
    각각 g, cm, s, coul기준의 Kg, m, s, CGS입니다. 또한 엔지니어링 단위도 사용할 수 있으며, slug, ft, s의 기초 단위가 있지만, 전기
    충전을 위한 단위는 없습니다. 이러한 옵션 중 어느 것도 해당되지 않는 경우, 질량, 길이, 시간 및 전기요금에 대한 기준 등을 사용자 정의하여 사용자 지정 단위 시스템을 사용할 수 있습니다.

    4.온도 단위는 드롭다운 목록을 사용하여 지정해야 합니다. 사용 가능한 옵션은 SI CGS 단위의 경우 Celsius
    Kelvin, 엔지니어링 단위의 경우 Fahrenheit Rankine입니다. Custom units(사용자 정의 단위) 옵션을 선택한 경우, 사용 가능한 온도 단위는 질량
    및 길이에 대해 선택한 기본 단위에 따라 변경됩니다.

    노트

    새 시뮬레이션의 시뮬레이션 단위는 신중하게 선택하십시오. 일단 설정하면 단위를 변경할 수 없습니다.

    5.이 시뮬레이션에 사용된 템플릿이 기본 템플릿이 됩니다. 템플릿은 포함된 설정을 새 시뮬레이션에 적용하는 저장된 값 세트입니다. 다른 템플릿을 사용해야하는 경우
    찾아보기 아이콘 (
    browse_icon_v12)