Chendi Zhang1 , Yuncheng Xu1,2, Marwan A Hassan3 , Mengzhen Xu1 , Pukang He1 1State Key Laboratory of Hydroscience and Engineering, Tsinghua University, Beijing, 100084, China. 2 College of Water Resources and Civil Engineering, China Agricultural University, Beijing, 100081, China. 5 3Department of Geography, University of British Columbia, 1984 West Mall, Vancouver BC, V6T1Z2, Canada. Correspondence to: Chendi Zhang (chendinorthwest@163.com) and Mengzhen Xu (mzxu@mail.tsinghua.edu.cn)
Abstract
스텝 풀 시스템은 계류의 일반적인 기반이며 전 세계의 하천 복원 프로젝트에 활용되었습니다. 스텝 풀 장치는 스텝 풀 기능의 형태학적 진화 및 안정성과 밀접하게 상호 작용하는 것으로 보고된 매우 균일하지 않은 수력 특성을 나타냅니다.
그러나 스텝 풀 형태에 대한 3차원 수리학의 자세한 정보는 측정의 어려움으로 인해 부족했습니다. 이러한 지식 격차를 메우기 위해 SfM(Structure from Motion) 및 CFD(Computational Fluid Dynamics) 기술을 기반으로 하이브리드 모델을 구축했습니다. 이 모델은 CFD 시뮬레이션을 위한 입력으로 6가지 유속의 자연석으로 만든 인공 스텝 풀 장치가 있는 침대 표면의 3D 재구성을 사용했습니다.
하이브리드 모델은 스텝 풀 장치에 대한 3D 흐름 구조의 고해상도 시각화를 제공하는 데 성공했습니다. 결과는 계단 아래의 흐름 영역의 분할, 즉 수면에서의 통합 점프, 침대 근처의 줄무늬 후류 및 그 사이의 고속 제트를 보여줍니다.
수영장에서 난류 에너지의 매우 불균일한 분포가 밝혀졌으며 비슷한 용량을 가진 두 개의 에너지 소산기가 수영장에 공존하는 것으로 나타났습니다. 흐름 증가에 따른 풀 세굴 개발은 점프 및 후류 와류의 확장으로 이어지지만 이러한 증가는 스텝 풀 실패에 대한 임계 조건에 가까운 높은 흐름에서 점프에 대해 멈춥니다.
음의 경사면에서 발달된 곡물 20 클러스터와 같은 미세 지반은 국부 수력학에 상당한 영향을 주지만 이러한 영향은 수영장 바닥에서 억제됩니다. 스텝 스톤의 항력은 가장 높은 흐름이 사용되기 전에 배출과 함께 증가하는 반면 양력은 더 큰 크기와 더 넓은 범위를 갖습니다. 우리의 결과는 계단 풀 형태의 복잡한 흐름 특성을 조사할 때 물리적 및 수치적 모델링을 결합한 하이브리드 모델 접근 방식의 가능성과 큰 잠재력을 강조합니다.
Step-pool systems are common bedforms in mountain streams and have been utilized in river restoration projects around the world. Step-pool units exhibit highly non-uniform hydraulic characteristics which have been reported to closely 10 interact with the morphological evolution and stability of step-pool features. However, detailed information of the threedimensional hydraulics for step-pool morphology has been scarce due to the difficulty of measurement. To fill in this knowledge gap, we established a hybrid model based on the technologies of Structure from Motion (SfM) and computational fluid dynamics (CFD). The model used 3D reconstructions of bed surfaces with an artificial step-pool unit built by natural stones at six flow rates as inputs for CFD simulations. The hybrid model succeeded in providing high-resolution visualization 15 of 3D flow structures for the step-pool unit. The results illustrate the segmentation of flow regimes below the step, i.e., the integral jump at the water surface, streaky wake vortexes near the bed, and high-speed jets in between. The highly non-uniform distribution of turbulence energy in the pool has been revealed and two energy dissipaters with comparable capacity are found to co-exist in the pool. Pool scour development under flow increase leads to the expansion of the jump and wake vortexes but this increase stops for the jump at high flows close to the critical condition for step-pool failure. The micro-bedforms as grain 20 clusters developed on the negative slope affect the local hydraulics significantly but this influence is suppressed at pool bottom. The drag forces on the step stones increase with discharge before the highest flow is used while the lift force has a larger magnitude and wider varying range. Our results highlight the feasibility and great potential of the hybrid model approach combining physical and numerical modeling in investigating the complex flow characteristics of step-pool morphology.
Figure 1: Workflow of the hybrid modeling. SfM-MVS refers to the technology of Structure from Motion with Multi View Stereo.
DSM is short for digital surface model. RNG-VOF is short for Renormalized Group (RNG) k-ε turbulence model coupled with
Volume of Fluid method.Figure 2: Flume experiment settings in Zhang et al., (2020): (a) the artificially built-up step-pool model using natural stones, with
stone number labelled; (b) the unsteady hydrograph of the run of CIFR (continually-increasing-flow-rate) T2 used in this study.Figure 3: Setup of the CFD model: (a) three-dimensional digital surface model (DSM) of the step-pool unit by structure from motion
with multi view stereo (SfM-MVS) method as the input to the 3D computational fluid dynamics (CFD) modeling; (b) extruded bed
160 surface model connected to the extra downstream component (in purple blue) and rectangular columns to fill leaks (in green), with
the boundary conditions shown on mesh planes; (c) recognized geometry with mesh grids of two mesh blocks shown where MS is
short for mesh size; (d) sampling volumes to capture the flow forces acting on each step stone at X, Y, and Z directions; and (e) an
example for the simulated 3D flow over the step-pool unit colored by velocity magnitude at the discharge of 49.9 L/s. The
abbreviations for boundary conditions in (b) are: V for specified velocity; C for continuative; P for specific pressure; and W for wall
165 condition. The contraction section in Figure (e) refers to the edge between the jet and jump at water surface.Figure 4: Distribution of time-averaged velocity magnitude (VM_mean) and vectors in three longitudinal sections. The section at Y = 0 cm goes across the keystone while the other two (Y = -18 and 13.5 cm) are located at the step stones beside the keystone with lower top elevations. Q refers to the discharge at the inlet of the computational domain. The spacing for X, Y, and Z axes are all 10 cm in the plots.Figure 5: Distribution of time-averaged flow velocity at five cross sections which are set according to the reference section (x0). The
reference cross section x0 is located at the downstream end of the keystone (KS). The five sections are located at 18 cm and 6 cm
upstream of the reference section (x0-18 and x0-6), and 2 cm, 15 cm and 40 cm downstream of the reference section (x0+2, x0+15,
x0+40). The spacing for X, Y, and Z axes are all 10 cm in the plots.Figure 6: Distribution of the time-averaged turbulence kinetic energy (TKE) at the five cross sections same with Figure 3.Figure 7: Boxplots for the distributions of the mass-averaged flow kinetic energy (KE, panels a-f), turbulence kinetic energy (TKE,
panels g-l), and turbulent dissipation (εT, panels m-r) in the pool for all the six tested discharges (the plots at the same discharge are
in the same row). The mass-averaged values were calculated every 2 cm in the streamwise direction. The flow direction is from left
to right in all the plots. The general locations of the contraction section for all the flow rates are marked by the dashed lines, except
for Q = 5 L/s when the jump is located too close to the step. The longitudinal distance taken up by negative slope in the pool for the
inspected range is shown by shaded area in each plot.Figure 8: Instantaneous flow structures extracted using the Q-criterion (Qcriterion=1200) and colored by the magnitude of flow velocity.Figure 9: Time-averaged dynamic pressure (DP_mean) on the bed surface in the step-pool model under the two highest discharges,
with the step numbers marked. The negative values in the plots result from the setting of standard atmospheric pressure = 0 Pa,
whose absolute value is 1.013×105 Pa.Figure 10: Time-averaged shear stress (SS_mean) on bed surface in the step-pool model, with the step numbers marked. The
standard atmospheric pressure is set as 0 Pa.Figure 11: Variation of fluid force components and magnitude of resultant flow force acting on step stones with flow rate. The stone
4 is the keystone. Stone numbers are consistent with those in Fig. 9-10. The upper limit of the sampling volumes for flow force
calculation is higher than water surface while the lower limit is set at 3 cm lower than the keystone crest.Figure 12: Variation of drag (CD) and lift (CL) coefficient of the step stones along with flow rate. Stone numbers are consistent with
those in Fig. 8-9. KS is short for keystone. The negative values of CD correspond to the drag forces towards the upstream while the
negative values of CL correspond to lift forces pointing downwards.Figure 13: Longitudinal distributions of section-averaged and -integral turbulent kinetic energy (TKE) for the jump and wake
vortexes at the largest three discharges. The flow direction is from left to right in all the plots. The general locations of the contraction
sections under the three flow rates are marked by dashed lines in figures (d) to (f).Figure A1: Water surface profiles of the simulations with different mesh sizes at the discharge of 43.6 L/s at the longitudinal sections
at: (a) Y = 24.5 cm (left boundary); (b) Y = 0.3 cm (middle section); (c) Y = -24.5 cm (right boundary). MS is short for mesh size.
The flow direction is from left to right in each plot.Figure A2: Contours of velocity magnitude in the longitudinal section at Y = 0 cm at different mesh sizes (MSs) under the flow
condition with the discharge of 43.6 L/s: (a) 0.50 cm; (b) 0.375 cm; (c) 0.30 cm; (d) 0.27 cm; (e) 0.25 cm; (f) 0.24 cm. The flow direction
is from left to right.Figure A3: Measurements of water surfaces (orange lines) used in model verification: (a) water surface profiles from both sides of
the flume; (b) upstream edge of the jump regime from top view. KS refers to keystone in figure (b).Figure A15. Figure (a) shows the locations of the cross sections and target coarse grains at Q = 49.9 L/s. Figures (b) to (e) show the
distribution of velocity magnitude (VM_mean) in the four chosen cross sections: (a) x0+8.0; (b) x0+14.0; (c) x0+21.5; (d) x0+42.5.
G1 to G6 refer to 6 protruding grains in the micro-bedforms in the pool.Figure A16. The distribution of turbulent kinetic energy (TKE) in the same cross sections as in figure S15: (a) x0+8.0; (b) x0+14.0;
(c) x0+21.5; (d) x0+42.5.
References
720 Aberle, J. and Smart, G. M: The influence of roughness structure on flow resistance on steep slopes, J. Hydraul. Res., 41(3), 259-269, https://doi.org/10.1080/00221680309499971, 2003. Abrahams, A. D., Li, G., and Atkinson, J. F.: Step-pool streams: Adjustment to maximum flow resistance. Water Resour. Res., 31(10), 2593-2602, https://doi.org/10.1029/95WR01957, 1995. Adrian, R. J.: Twenty years of particle image velocimetry. Exp. Fluids, 39(2), 159-169, https://doi.org/10.1007/s00348-005- 725 0991-7 2005. Chanson, H.: Hydraulic design of stepped spillways and downstream energy dissipators. Dam Eng., 11(4), 205-242, 2001. Chartrand, S. M., Jellinek, M., Whiting, P. J., and Stamm, J.: Geometric scaling of step-pools in mountain streams: Observations and implications, Geomorphology, 129(1-2), 141-151, https://doi.org/10.1016/j.geomorph.2011.01.020, 2011. 730 Chen, Y., DiBiase, R. A., McCarroll, N., and Liu, X.: Quantifying flow resistance in mountain streams using computational fluid dynamics modeling over structure‐from‐motion photogrammetry‐derived microtopography, Earth Surf. Proc. Land., 44(10), 1973-1987, https://doi.org/10.1002/esp.4624, 2019. Church, M. and Zimmermann, A.: Form and stability of step‐pool channels: Research progress, Water Resour. Res., 43(3), W03415, https://doi.org/10.1029/2006WR005037, 2007. 735 Cignoni, P., Callieri, M., Corsini, M., Dellepiane, M., Ganovelli, F., and Ranzuglia, G.: Meshlab: an open-source mesh processing tool, in: Eurographics Italian chapter conference, Salerno, Italy, 2-4 July 2008, 129-136, 2008.
Comiti, F., Andreoli, A., and Lenzi, M. A.: Morphological effects of local scouring in step-pool streams, Earth Surf. Proc. Land., 30(12), 1567-1581, https://doi.org/10.1002/esp.1217, 2005. Comiti, F., Cadol, D., and Wohl, E.: Flow regimes, bed morphology, and flow resistance in self‐formed step-pool 740 channels, Water Resour. Res., 45(4), 546-550, https://doi.org/10.1029/2008WR007259, 2009. Dudunake, T., Tonina, D., Reeder, W. J., and Monsalve, A.: Local and reach‐scale hyporheic flow response from boulder ‐ induced geomorphic changes, Water Resour. Res., 56, e2020WR027719, https://doi.org/10.1029/2020WR027719, 2020. Flow Science.: Flow-3D Version 11.2 User Manual, Flow Science, Inc., Los Alamos, 2016. Gibson, S., Heath, R., Abraham, D., and Schoellhamer, D.: Visualization and analysis of temporal trends of sand infiltration 745 into a gravel bed, Water Resour. Res., 47(12), W12601, https://doi.org/10.1029/2011WR010486, 2011. Hassan, M. A., Tonina, D., Beckie, R. D., and Kinnear, M.: The effects of discharge and slope on hyporheic flow in step‐pool morphologies, Hydrol. Process., 29(3), 419-433, https://doi.org/10.1002/hyp.10155, 2015. Hirt, C. W. and Nichols, B. D.: Volume of Fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys., 39, 201-225, https://doi.org/10.1016/0021-9991(81)90145-5, 1981. 750 Javernick L., Brasington J., and Caruso B.: Modeling the topography of shallow braided rivers using structure-from-motion photogrammetry, Geomorphology, 213(4), 166-182, https://doi.org/10.1016/j.geomorph.2014.01.006, 2014. Lai, Y. G., Smith, D. L., Bandrowski, D. J., Xu, Y., Woodley, C. M., and Schnell, K.: Development of a CFD model and procedure for flows through in-stream structures, J. Appl. Water Eng. Res., 1-15, https://doi.org/10.1080/23249676.2021.1964388, 2021. 755 Lenzi, M. A.: Step-pool evolution in the Rio Cordon, northeastern Italy, Earth Surf. Proc. Land., 26(9), 991-1008, https://doi.org/10.1002/esp.239, 2001. Lenzi, M. A.: Stream bed stabilization using boulder check dams that mimic step-pool morphology features in Northern Italy, Geomorphology, 45(3-4), 243-260, https://doi.org/10.1016/S0169-555X(01)00157-X, 2002. Lenzi, M. A., Marion, A., and Comiti, F.: Local scouring at grade‐control structures in alluvial mountain rivers, Water Resour. 760 Res., 39(7), 1176, https://doi:10.1029/2002WR001815, 2003. Li, W., Wang Z., Li, Z., Zhang, C., and Lv, L.: Study on hydraulic characteristics of step-pool system, Adv. Water Sci., 25(3), 374-382, https://doi.org/10.14042/j.cnki.32.1309.2014.03.012, 2014. (In Chinese with English abstract) Maas, H. G., Gruen, A., and Papantoniou, D.: Particle tracking velocimetry in three-dimensional flows, Exp. Fluids, 15(2), 133-146. https://doi.org/10.1007/BF00223406, 1993.
765 Montgomery, D. R. and Buffington, J. M.: Channel-reach morphology in mountain drainage basins, Geol. Soc. Am. Bul., 109(5), 596-611, https://doi.org/10.1130/0016-7606(1997)109<0596:CRMIMD>2.3.CO;2, 1997. Morgan J. A., Brogan D. J., and Nelson P. A.: Application of structure-from-motion photogrammetry in laboratory flumes, Geomorphology, 276(1), 125-143, https://doi.org/10.1016/j.geomorph.2016.10.021, 2017. Recking, A., Leduc, P., Liébault, F., and Church, M.: A field investigation of the influence of sediment supply on step-pool 770 morphology and stability. Geomorphology, 139, 53-66, https://doi.org/10.1016/j.geomorph.2011.09.024, 2012. Roth, M. S., Jähnel, C., Stamm, J., and Schneider, L. K.: Turbulent eddy identification of a meander and vertical-slot fishways in numerical models applying the IPOS-framework, J. Ecohydraulics, 1-20, https://doi.org/10.1080/24705357.2020.1869916, 2020. Saletti, M. and Hassan, M. A.: Width variations control the development of grain structuring in steep step‐pool dominated 775 streams: insight from flume experiments, Earth Surf. Proc. Land., 45(6), 1430-1440, https://doi.org/10.1002/esp.4815, 2020. Smith, D. P., Kortman, S. R., Caudillo, A. M., Kwan‐Davis, R. L., Wandke, J. J., Klein, J. W., Gennaro, M. C. S., Bogdan, M. A., and Vannerus, P. A.: Controls on large boulder mobility in an ‘auto-naturalized’ constructed step-pool river: San Clemente Reroute and Dam Removal Project, Carmel River, California, USA, Earth Surf. Proc. Land., 45(9), 1990-2003, 780 https://doi.org/10.1002/esp.4860, 2020. Thappeta, S. K., Bhallamudi, S. M., Fiener, P., and Narasimhan, B.: Resistance in Steep Open Channels due to Randomly Distributed Macroroughness Elements at Large Froude Numbers, J. Hydraul. Eng., 22(12), 04017052, https://doi.org/10.1061/(ASCE)HE.1943-5584.0001587, 2017. Thappeta, S. K., Bhallamudi, S. M., Chandra, V., Fiener, P., and Baki, A. B. M.: Energy loss in steep open channels with step785 pools, Water, 13(1), 72, https://doi.org/10.3390/w13010072, 2021. Turowski, J. M., Yager, E. M., Badoux, A., Rickenmann, D., and Molnar, P.: The impact of exceptional events on erosion, bedload transport and channel stability in a step-pool channel, Earth Surf. Proc. Land., 34(12), 1661-1673, https://doi.org/10.1002/esp.1855, 2009. Vallé, B. L. and Pasternack, G. B.: Air concentrations of submerged and unsubmerged hydraulic jumps in a bedrock step‐pool 790 channel, J. Geophys. Res.-Earth, 111(F3), F03016. https://doi:10.1029/2004JF000140, 2006. Waldon, M. G.: Estimation of average stream velocity, J. Hydraul. Eng., 130(11), 1119-1122. https://doi.org/10.1061/(ASCE)0733-9429(2004)130:11(1119), 2004. Wang, Z., Melching, C., Duan, X., and Yu, G.: Ecological and hydraulic studies of step-pool systems, J. Hydraul. Eng., 135(9), 705-717, https://doi.org/10.1061/(ASCE)0733-9429(2009)135:9(705), 2009
795 Wang, Z., Qi, L., and Wang, X.: A prototype experiment of debris flow control with energy dissipation structures, Nat. Hazards, 60(3), 971-989, https://doi.org/10.1007/s11069-011-9878-5, 2012. Weichert, R. B.: Bed Morphology and Stability in Steep Open Channels, Ph.D. Dissertation, No. 16316. ETH Zurich, Switzerland, 247pp., 2005. Wilcox, A. C., Wohl, E. E., Comiti, F., and Mao, L.: Hydraulics, morphology, and energy dissipation in an alpine step‐pool 800 channel, Water Resour. Res., 47(7), W07514, https://doi.org/10.1029/2010WR010192, 2011. Wohl, E. E. and Thompson, D. M.: Velocity characteristics along a small step–pool channel. Earth Surf. Proc. Land., 25(4), 353-367, https://doi.org/10.1002/(SICI)1096-9837(200004)25:4<353::AID-ESP59>3.0.CO;2-5, 2000. Wu, S. and Rajaratnam, N.: Impinging jet and surface flow regimes at drop. J. Hydraul. Res., 36(1), 69-74, https://doi.org/10.1080/00221689809498378, 1998. 805 Xu, Y. and Liu, X.: 3D computational modeling of stream flow resistance due to large woody debris, in: Proceedings of the 8th International Conference on Fluvial Hydraulics, St. Louis, USA, 11-14, Jul, 2346-2353, 2016. Xu, Y. and Liu, X.: Effects of different in-stream structure representations in computational fluid dynamics models—Taking engineered log jams (ELJ) as an example, Water, 9(2), 110, https://doi.org/10.3390/w9020110, 2017. Zeng, Y. X., Ismail, H., and Liu, X.: Flow Decomposition Method Based on Computational Fluid Dynamics for Rock Weir 810 Head-Discharge Relationship. J. Irrig. Drain. Eng., 147(8), 04021030, https://doi.org/10.1061/(ASCE)IR.1943- 4774.0001584, 2021. Zhang, C., Wang, Z., and Li, Z.: A physically-based model of individual step-pool stability in mountain streams, in: Proceedings of the 13th International Symposium on River Sedimentation, Stuttgart, Germany, 801-809, 2016. Zhang, C., Xu, M., Hassan, M. A., Chartrand, S. M., and Wang, Z.: Experimental study on the stability and failure of individual 815 step-pool, Geomorphology, 311, 51-62, https://doi.org/10.1016/j.geomorph.2018.03.023, 2018. Zhang, C., Xu, M., Hassan, M. A., Chartrand, S. M., Wang, Z., and Ma, Z.: Experiment on morphological and hydraulic adjustments of step‐pool unit to flow increase, Earth Surf. Proc. Land., 45(2), 280-294, https://doi.org/10.1002/esp.4722, 2020. Zimmermann A., E.: Flow resistance in steep streams: An experimental study, Water Resour. Res., 46, W09536, 820 https://doi.org/10.1029/2009WR007913, 2010. Zimmermann A. E., Salleti M., Zhang C., Hassan M. A.: Step-pool Channel Features, in: Treatise on Geomorphology (2nd Edition), vol. 9, Fluvial Geomorphology, edited by: Shroder, J. (Editor in Chief), Wohl, E. (Ed.), Elsevier, Amsterdam, Netherlands, https://doi.org/10.1016/B978-0-12-818234-5.00004-3, 2020.
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.
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.
(1)
(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:
(3)where CDIS1, CDIS2, and CDIS3 are dimensionless parameters and the user can modify them. The diffusion of dissipation, Diff ɛ, is
(4)where u, v 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
(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:
(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
(7)
(8)
(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.
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).
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.
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)
d0
q (m3s−1)
dc/h (–)
1
50
18.6
0.06
0.045
0.1
2.6
2
50
18.6
0.06
0.082
0.235
4.6
3
50
30.0
0.06
0.045
0.1
2.6
4
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).
Figure3 The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.
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.
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.
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.
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 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.
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.
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.
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.
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.
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.
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.
FLOW-3 D v12.0 온라인 교육 과정은 미국 FSI에서 제공되는 컨텐츠로 FLOW-3D 사용자(구매/임차 및 기술지원 계약이 되어 있는 고객)에게 제공되는 교육 리소스입니다. 이 온라인 교육 과정은 FLOW-3D 기본 모델 사용법 전반에 대한 온라인 주문형 비디오를 제공합니다.
각 과정에서는 사용자가 스스로 시뮬레이션을 설정할 수 있도록 예제와 설명을 제공합니다. 모든 신규 FLOW–3D사용자는 프로젝트별 시뮬레이션 작업을 시작하기 전에 기본 과정을 완료하는 것이 좋습니다.
또한 기존 사용자는 FLOW–3D v12.0모델 설정 프로세스에서 사용할 수 있는 향상된 기능과 새로운 기능에 대해 배우고 기본 모델 설정 항목에 대한 리프레시로 배우는 데 유용한 새로운 교육 시리즈를 찾게 될 것입니다. 과정 비디오는 특정 주제 및 세그먼트를 쉽게 찾을 수 있도록 구성되어 있고, 즐겨 찾기에 추가될 수 있으며, 언제든지 참조할 수 있는 유용한 리소스를 제공합니다.
본 교육 과정은 미국 본사 정책에 따라 유지보수 계약이 체결된 고객 ID를 통해 미국의 Users Site 에서 제공됩니다.
FLOW-3D Training Modules
FLOW-3D GUI
Introduction to FLOW-3D graphical user interface
Simulation Manager Tab
Portfolio
Running Simulations and the Queue
Runtime Diagnostics: Text Output
Runtime Diagnostics: Plots
Runtime Controls
FLOW-3D File Structure Review the important files that are created when running simulations in FLOW-3D. Access the simulation files through a link on the Simulation Manager Tab. Identify the important setup and solver outputs files
모델 설정 탭
Introduction to the Model Setup TabIntroduction to the Model Setup Tab including an orientation to its layout and how to access model inputs though the dock widgets on the process toolbar. Options for customizing the layout of the process toolbar are also reviewed.
Navigating the 3D ViewportLearn the basic controls for navigating the 3D viewport. This includes mouse controls, toolbar shortcuts, saving views, and moving the pivot point.
Other Menu/Toolbar Navigation Options
Working with Dock Widget Inputs
Model DependenciesRecognize and understand dock widget input dependencies.
전역 설정
Global Dock Widget Overview
Pressure Type
Finish Time
Finish Options: Additional Finish Condition
Finish Options: Active Simulation ControlDefine a logical condition to stop the simulation using active simulation control.
Restart OptionsHow to manually define the Restart options to continue running a previously completed simulation.
Version OptionsDefine the Version options to specify the solver version and the number of processors used when starting a new simulation run.
물리 모델
Physics Dock Widget OverviewDescription of the available options in the Physics dock widget
Interface Tracking, Number of Fluids and Flow ModeBackground information on interface tracking methods and defining the number of fluids. Description of the Volume of Fluid (VOF) method for simulation of complex free surfaces, and how this affects the selection of the number of fluids. Examples are presented for one fluid and two fluid simulations.
Activating Physics ModelsDemonstration for how to activate physics models and how to limit the display of inactive physics models using the physics model filter.
유체 속성
Fluids Dock Widget OverviewIntroduction to the Fluids dock widget and how to define properties for fluids in the simulation.
Defining Fluid Properties ManuallyExample for how to manually define fluid properties.
Defining Fluid Properties from the Materials DatabaseExample for how to load fluid properties from the fluids database.
Managing the Materials Database How to add and edit entries in the materials database.
지오메트리
Introduction
Component and Subcomponent Overview
Creating Subcomponents: Overview
Creating Subcomponents: STL
Creating Subcomponents: Primitives Manually
Creating Subcomponents: Primitives Interactively
Creating Subcomponents: Raster
Subcomponent Types
Subcomponent Order
Component Order
Component and Subcomponent Properties
Transformations
Meshing
Meshing Introduction
Coordinate Systems
FAVOR™
Meshing Basics: Meshing Overview
Meshing Basics: Creating Mesh Blocks
Meshing Basics: Domain Extents
Meshing Basics: Global Controls
Meshing Basics: Local Controls
Reviewing Mesh Quality: FAVORize
Reviewing Mesh Quality: Preprocessing
Multi-block Meshing
Conforming Mesh Blocks
Meshing Best Practices
Boundary Conditions
Introduction Introductory comments regarding how boundary conditions are applied and other considerations when defining BCs.
Boundaries Dock Widget Overview
Velocity
Volume Flow Rate
Wall
Symmetry
Grid Overlay
Pressure
Continuative
Outflow Description and example setup of the Outflow BC type.
Initial Conditions
Introduction Discussion of how the initial conditions and can affect simulation results and run times.
Options for Defining ICs Example: Global Settings Example: Fluid Regions
Example: Function Coefficients Description and example for defining spatially varying fluid properties with user defined functions.
Example: Pointers Description and example for defining an initial condition by filling contiguous cells with the Pointer object.
Output Options
Output Dock Widget Overview
Spatial Data
Spatial Data: Restart Data
Spatial Data: Selected Data
History Data
Diagnostics: Short Print Data
Diagnostics: Long Print Data
Example Setup
Batch Post-processing
Batch Mode: Context File
Batch Mode: Manual
Batch Mode: Generate Reports
Baffles
Introduction An introduction to the available options for creating and defining baffle objects. Creating Baffle Objects Limitations Force Outputs Porosity Scalar Reset Options Summary A summary of the important options for creating baffles and defining properties.
Measurement Devices
History Probes History probes are point measurement devices and are used to record solver output at a specific location. Examples are provided for how to create these objects interactively and by defining a coordinate value.
Flux Surfaces Flux surfaces are a special type of baffle object with a fixed porosity of 1, and are used to calculate flux quantities. Examples are provided for how to create flux surfaces and the types of data available from their output.
Sampling volumes Sampling volumes are are three-dimensional data collection regions. Examples are provided for how to create sampling volumes and the types of data available from their output.
W&E Exercise: Ogee Weir
This exercise demonstrates the steps to setup a basic free surface or open channel flow simulation in FLOW-3D. It is intended to be a simple and fast running simulation that demonstrates the key setup steps that can be applied to a wide range of other common open channel flow applications. In this exercise, we will simulate flow over an ogee weir to predict the discharge capacity. Simulation results can be validated against discharge rating curves obtained from physical model measurements (USBR, 1996). Special attention is given to the common types of boundary conditions used in open channel flow simulations and how to select them during the model setup. We also provide examples for common post-processing tasks using both FLOW-3D and FlowSight.
FLOW-3D/MP v6.1 은 FLOW-3D v11.1 솔버에 기초하여 물리 모델, 특징 및 그래픽 사용자 인터페이스가 동일합니다. FLOW-3D v11.1의 새로운 기능은 아래 파란색으로 표시되어 있으며 FLOW-3D/MP v6.1 에서 사용할 수 있습니다. 새로운 개발 기능에 대한 자세한 설명은 FLOW-3D v11.1에서 새로운 기능을 참조하십시오.
Meshing & Geometry
Structured finite difference/control volume meshes for fluid and thermal solutions
Finite element meshes in Cartesian and cylindrical coordinates for structural analysis
Multi-Block gridding with nested, linked, partially overlapping and conforming mesh blocks
Fractional areas/volumes (FAVOR™) for efficient & accurate geometry definition
Mesh quality checking
Basic Solids Modeler
Import CAD data
Import/export finite element meshes via Exodus-II file format
Grid & geometry independence
Cartesian or cylindrical coordinates
Flow Type Options
Internal, external & free-surface flows
3D, 2D & 1D problems
Transient flows
Inviscid, viscous laminar & turbulent flows
Hybrid shallow water/3D flows
Non-inertial reference frame motion
Multiple scalar species
Two-phase flows
Heat transfer with phase change
Saturated & unsaturated porous media
Physical Modeling Options
Fluid structure interaction
Thermally-induced stresses
Plastic deformation of solids
Granular flow
Moisture drying
Solid solute dissolution
Sediment transport and scour
Cavitation (potential, passive tracking, active tracking)
Phase change (liquid-vapor, liquid-solid)
Surface tension
Thermocapillary effects
Wall adhesion
Wall roughness
Vapor & gas bubbles
Solidification & melting
Mass/momentum/energy sources
Shear, density & temperature-dependent viscosity
Thixotropic viscosity
Visco-elastic-plastic fluids
Elastic membranes & walls
Evaporation residue
Electro-mechanical effects
Dielectric phenomena
Electro-osmosis
Electrostatic particles
Joule heating
Air entrainment
Molecular & turbulent diffusion
Temperature-dependent material properties
Spray cooling
Flow Definition Options
General boundary conditions
Symmetry
Rigid and flexible walls
Continuative
Periodic
Specified pressure
Specified velocity
Outflow
Grid overlay
Hydrostatic pressure
Volume flow rate
Non-linear periodic and solitary surface waves
Rating curve and natural hydraulics
Wave absorbing layer
Restart from previous simulation
Continuation of a simulation
Overlay boundary conditions
Change mesh and modeling options
Change model parameters
Thermal Modeling Options
Natural convection
Forced convection
Conduction in fluid & solid
Fluid-solid heat transfer
Distributed energy sources/sinks in fluids and solids
Radiation
Viscous heating
Orthotropic thermal conductivity
Thermally-induced stresses
Turbulence Models
RNG model
Two-equation k-epsilon model
Two-equation k-omega model
Large eddy simulation
Metal Casting Models
Thermal stress & deformations
Iron solidification
Sand core blowing
Sand core drying
Permeable molds
Solidification & melting
Solidification shrinkage with interdendritic feeding
Micro & macro porosity
Binary alloy segregation
Thermal die cycling
Surface oxide defects
Cavitation potential
Lost-foam casting
Semi-solid material
Core gas generation
Back pressure & vents
Shot sleeves
PQ2 diagram
Squeeze pins
Filters
Air entrainment
Temperature-dependent material properties
Cooling channels
Fluid/wall contact time
Numerical Modeling Options
TruVOF Volume-of-Fluid (VOF) method for fluid interfaces
First and second order advection
Sharp and diffuse interface tracking
Implicit & explicit numerical methods
GMRES, point and line relaxation pressure solvers
User-defined variables, subroutines & output
Utilities for runtime interaction during execution
Fluid Modeling Options
One incompressible fluid – confined or with free surfaces
Two incompressible fluids – miscible or with sharp interfaces
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.
Model, the Two-Equation Renormalization-Group (RNG) Model and large Eddy Simulation Model (
it solves an additional transport equation:
, 
, 
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
: 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:
: the turbulent length scale.
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.
