Deep Learning-Based Eddy Viscosity Modeling for Improved RANS Simulations of Wind Pressures on Bluff Bodies

Aly, A. M.

doi:10.47176/jafm.17.12.2770

Deep Learning-Based Eddy Viscosity Modeling for Improved RANS Simulations of Wind Pressures on Bluff Bodies

Document Type : Regular Article

Author

A. M. Aly

Windstorm Impact, Science and Engineering (WISE) Research Lab, Louisiana State University, 3230 H Patrick F. Taylor Hall, Baton Rouge, LA 70803, USA

10.47176/jafm.17.12.2770

Abstract

Accurate prediction of wind pressures on buildings is crucial for designing safe and efficient structures. Existing computational methods, like Reynolds-averaged Navier-Stokes (RANS) simulations, often fail to predict pressures accurately in separation zones. This study proposes a novel deep-learning methodology to enhance the accuracy and performance of eddy viscosity modeling within RANS turbulence closures, particularly improving predictions for bluff body aerodynamics. A deep learning model, trained on large eddy simulation (LES) data for various bluff body geometries, including a flat-roof building and forward/backward facing steps, was used to adjust eddy viscosity in RANS equations. The results show that incorporating the machine learning-predicted eddy viscosity significantly improves agreement with LES results and experimental data, particularly in the separation bubble and shear layer. The deep learning model employed a neural network architecture with four hidden layers, 32 neurons, and tanh activation functions, trained using the Adam optimizer with a learning rate of 0.001. The training data consisted of LES simulations for forward/backward facing steps with width-to-height ratios ranging from 0.2 to 6. The study reveals that the machine learning model achieves a balance in eddy viscosity that delays flow reattachment, leading to more accurate pressure and velocity predictions than traditional turbulence closures like k-ω SST and k-ε. A sensitivity analysis demonstrated the pivotal role of eddy viscosity in governing flow separation, reattachment, and pressure distributions. Additionally, the investigation underscores the disparity in eddy viscosity values between RANS and LES models, highlighting the need for enhanced turbulence modeling. The findings presented in this paper offer substantive insights that can inform the advancement of more dependable computational methodologies tailored for engineering applications, encompassing wind load considerations for structural design and the intricate dynamics of unsteady aerodynamic phenomena.

Keywords

Main Subjects

Computational Fluid Dynamics

References

Ai, Z. T., & Mak, C. M. (2015). Large-eddy simulation of flow and dispersion around an isolated building: analysis of influencing factors. Computers and Fluids, 118, 89–100. https://doi.org/10.1016/j.compfluid.2015.06.006

Akhlaghi, M., Asadbeigi, M., & Ghafoorian, F. (2023). Novel CFD and DMST dual method parametric study and optimization of a darrieus vertical axis wind turbine. Journal of Applied Fluid Mechanics, 17(1), 205–218. https://doi.org/10.47176/jafm.17.1.1985

Aly, A., & Dougherty, E. (2021). Bridge pier geometry effects on local scour potential: A comparative study. Ocean Engineering, https://doi.org/10.1016/j.oceaneng.2021.109326

Aly, A. M. (2014). Atmospheric boundary-layer simulation for the built environment: Past, present and future. Building and Environment, 75, 206–221. https://doi.org/10.1016/j.buildenv.2014.02.004

Aly, A. M. (2016). On the evaluation of wind loads on solar panels: The scale issue. Solar Energy, 135, 423–434. https://doi.org/10.1016/j.solener.2016.06.018

Aly, A. M., & Bitsuamlak, G. (2013). Aerodynamics of ground-mounted solar panels: Test model scale effects. Journal of Wind Engineering and Industrial Aerodynamics, 123, 250–260. https://doi.org/10.1016/j.jweia.2013.07.007

Aly, A. M., & Clarke, J. (2023). Wind design of solar panels for resilient and green communities: CFD with machine learning. Sustainable Cities and Society, 94, 104529. https://doi.org/10.1016/j.scs.2023.104529

Aly, A. M., & Dragomirescu, E. (2018). Wind engineering for natural hazards: modeling, simulation, and mitigation of windstorm impact on critical infrastructure. American Society of Civil Engineers. https://doi.org/10.1061/9780784415153

Aly, A. M., & Gol-Zaroudi, H. (2020). Peak pressures on low rise buildings: CFD with LES versus full scale and wind tunnel measurements. Wind and Structures, An International Journal, 30(1), 99–117. https://doi.org/10.12989/was.2020.30.1.099

Aly, A. M., Khaled, M. F., & Clancy, R. (2022). Large-scale open-jet testing: a new frontier in structural wind engineering. Engineering Structures, 266, 114567. https://doi.org/10.1016/j.engstruct.2022.114567

Banari, A., Hertel, D., Schlink, U., Hampel, U., & Lecrivain, G. (2023). Simulation of particle resuspension by wind in an urban system. Environmental Fluid Mechanics, 23(1), 41–63. https://doi.org/10.1007/s10652-022-09905-x

Bao, K., Zhang, X., Peng, W., & Yao, W. (2023). Deep learning method for super-resolution reconstruction of the spatio-temporal flow field. Advances in Aerodynamics. https://doi.org/10.1186/s42774-023-00148-y

Chen, S., Qiu, J., Yang, H., Yuan, W., & Gao, Z. (2023). Deep learning for inverse design of low-boom supersonic configurations. Advances in Aerodynamics. https://doi.org/10.1186/s42774-023-00145-1

Dickison, M., Ghaleeh, M., Milady, S., Wen, L. T., & Al Qubeissi, M. (2020). Investigation into the aerodynamic performance of a concept sports car. Journal of Applied Fluid Mechanics, 13(2), 583–601. https://doi.org/10.29252/jafm.13.02.30179

Duraisamy, K., Iaccarino, G., & Xiao, H. (2019). Turbulence modeling in the age of data. In Annual Review of Fluid Mechanics. https://doi.org/10.1146/annurev-fluid-010518-040547

ElDegwy, A., Elsayed, A., & Darwish, M. (2022). Aerodynamics of ancient egyptian obelisks and their structural response to boundary layer wind. Environmental Fluid Mechanics, 22(5), 1035–1053. https://doi.org/10.1007/s10652-022-09877-y

Ferreira, A. D., Thiis, T., Freire, N. A., & Ferreira, A. M. C. (2019). A wind tunnel and numerical study on the surface friction distribution on a flat roof with solar panels. Environmental Fluid Mechanics, 19, 601–617. http://dx.doi.org/10.1007%2Fs10652-018-9641-5

Fukami, K., Fukagata, K., & Taira, K. (2019). Super-resolution reconstruction of turbulent flows with machine learning. Journal of Fluid Mechanics, 870, 106–120. https://doi.org/10.1017/jfm.2019.238

Gargallo-Peiró, A., Avila, M., Owen, H., Prieto-Godino, L., & Folch, A. (2018). Mesh generation, sizing and convergence for onshore and offshore wind farm Atmospheric Boundary Layer flow simulation with actuator discs. Journal of Computational Physics. https://doi.org/10.1016/j.jcp.2018.08.031

Geneva, N., & Zabaras, N. (2019). Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networks. Journal of Computational Physics, 383, 125–147. https://doi.org/10.1016/j.jcp.2019.01.021

Gnatowska, R. (2019). Wind-induced pressure loads on buildings in tandem arrangement in urban environment. Environmental Fluid Mechanics, 19(3), 699–718. https://doi.org/10.1007/s10652-018-9646-0

Guichard, R. (2019). Assessment of an improved Random Flow Generation method to predict unsteady wind pressures on an isolated building using Large-Eddy Simulation. Journal of Wind Engineering and Industrial Aerodynamics, 189, 304–313. https://doi.org/10.1016/j.jweia.2019.04.006

Jones, W. P., & Launder, B. E. (1973). The calculation of low-Reynolds-number phenomena with a two-equation model of turbulence. International Journal of Heat and Mass Transfer, 16(6), 119–1130. https://doi.org/10.1016/0017-9310(73)90125-7

Khaled, M., Aly, A., & Elshaer, A. (2021). Computational efficiency of CFD modeling for building engineering: An empty domain study. Journal of Building Engineering. https://doi.org/10.1016/j.jobe.2021.102792

Khaled, M. F., & Aly, A. M. (2022). Assessing aerodynamic loads on low-rise buildings considering Reynolds number and turbulence effects: a review. Advances in Aerodynamics, 4(1), 1–33. https://doi.org/10.1186/s42774-022-00114-0

Khaled, M. F., & Aly, A. M. (2023). Augmenting external surface pressures’ predictions on isolated low-rise buildings using CFD simulations. Wind and Structures, An International Journal, 37(4), 255–274. https://doi.org/10.12989/was.2023.37.4.255

Kochkov, D., Smith, J. A., Alieva, A., Wang, Q., Brenner, M. P., & Hoyer, S. (2021). Machine learning–accelerated computational fluid dynamics. Proceedings of the National Academy of Sciences of the United States of America. https://doi.org/10.1073/pnas.2101784118

Kundu, P. K., Cohen, I. M., & Dowling, D. R. (2012). Fluid Mechanics. Academic Press. https://doi.org/10.1016/B978-0-12-382100-3.10008-3

Le, H., Moin, P., & Kim, J. (1997). Direct numerical simulation of turbulent flow over a backward-facing step. Journal of Fluid Mechanics, 330, 349–374. https://doi.org/https://doi.org/10.1017/S0022112096003941

Li, T., Zhang, J., Rashidi, M. M., & Yu, M. (2019). On the reynolds-averaged navier-stokes modelling of the flow around a simplified train in crosswinds. Journal of Applied Fluid Mechanics, 12(2), 551–563. https://doi.org/10.29252/jafm.12.02.28958

Lim, H. C., Castro, I. P., & Hoxey, R. P. (2007). Bluff bodies in deep turbulent boundary layers: Reynolds-number issues. Journal of Fluid Mechanics, 571, 97–118. https://doi.org/10.1017/S0022112006003223

Liu, L., & Stevens, R. J. A. M. (2020). Wall modeled immersed boundary method for high Reynolds number flow over complex terrain. Computers and Fluids. https://doi.org/10.1016/j.compfluid.2020.104604

Liu, W., Fang, J., Rolfo, S., Moulinec, C., & Emerson, D. R. (2021). An iterative machine-learning framework for RANS turbulence modeling. International Journal of Heat and Fluid Flow, 90, 108822. https://doi.org/10.1016/j.ijheatfluidflow.2021.108822

Liu, W., Fang, J., Rolfo, S., Moulinec, C., & Emerson, D. R. (2023). On the improvement of the extrapolation capability of an iterative machine-learning based RANS Framework. Computers & Fluids, 105864. https://doi.org/10.1016/j.compfluid.2023.105864

Ma, K., & Lai, H. (2016). Comparison of five two-equation turbulence models for calculation of flow in 90 curved rectangular ducts. Journal of Applied Fluid Mechanics, 9(6), 2917–2931. https://doi.org/10.29252/jafm.09.06.25568

Mahboub, A., Bouzit, M., & Ghenaim, A. (2022). Effects of different shaped cavities and bumps on flow structure and wing performance. Journal of Applied Fluid Mechanics, 15(6), 1649–1660. https://doi.org/10.47176/jafm.15.06.1108

Majchrzak, M., Marciniak-Lukasiak, K., & Lukasiak, P. (2023). A Survey on the application of machine learning in turbulent flow simulations. Energies, 16(4), 1755. https://doi.org/10.3390/en16041755

Malaspinas, O., & Sagaut, P. (2014). Wall model for large-eddy simulation based on the lattice Boltzmann method. Journal of Computational Physics. https://doi.org/10.1016/j.jcp.2014.06.020

Maulik, R., Sharma, H., Patel, S., Lusch, B., & Jennings, E. (2021). Deploying deep learning in OpenFOAM with TensorFlow. AIAA Scitech 2021 Forum (p. 1485). https://doi.org/10.2514/6.2021-1485

Menter, F. R. (1992). Improved two-equation turbulence models for aerodynamic flows. NASA Technical Manual 103975.

Mitchell, T. M. (1997). Machine Learning. McGraw-Hill.

Orszag, S. A. (1970). Analytical theories of turbulence. Journal of Fluid Mechanics, 41(2), 363–386. https://doi.org/10.1017/S0022112070000642

Pathak, J., Mustafa, M., Kashinath, K., Motheau, E., Kurth, T., & Day, M. (2020). Using machine learning to augment coarse-grid computational fluid dynamics simulations. 1–8. https://doi.org/arXiv:2010.00072

Richards, P. J., & Hoxey, R. P. (2002). Unsteady flow on the sides of a 6m cube. Journal of Wind Engineering and Industrial Aerodynamics, 90(12–15), 1855–1866. https://doi.org/10.1016/S0167-6105(02)00293-3

Richards, P. J., & Hoxey, R. P. (2012). Pressures on a cubic building-Part 2: Quasi-steady and other processes. Journal of Wind Engineering and Industrial Aerodynamics, 102, 87–96. https://doi.org/10.1016/j.jweia.2011.11.003

Samuel, A. L. (1959). Some studies in machine learning using the game of checkers. IBM Journal of Research and Development, 3(3), 210–229. https://doi.org/10.1147/rd.33.0210

Samuel, A. L. (1962). Artificial Intelligence: A frontier of automation. The ANNALS of the American Academy of Political and Social Science, 340(1), 10–20. https://doi.org/10.1177/000271626234000103

San, B., Zhao, Y., & Qiu, Y. (2019). Numerical simulation and optimization study of surface pressure and flow field around a triangular prism behind a porous fence. Environmental Fluid Mechanics, 19, 969–987. https://link.springer.com/article/10.1007/s10652-019-09695-9

Sanhueza, R. D., Smit, S. H. H. J., Peeters, J. W. R., & Pecnik, R. (2023). Machine learning for RANS turbulence modeling of variable property flows. Computers & Fluids, 255, 105835. https://doi.org/10.1016/j.compfluid.2023.105835

Sharma, A., Mittal, H., & Gairola, A. (2019). Wind-induced forces and flow field of aerodynamically modified buildings. Environmental Fluid Mechanics, 19, 1599–1623. http://dx.doi.org/10.1007%2Fs10652-019-09687-9

Singh, A. P., Medida, S., & Duraisamy, K. (2017). Machine-learning-augmented predictive modeling of turbulent separated flows over airfoils. AIAA Journal. https://doi.org/10.2514/1.J055595

Slotnick, J., Khodadoust, A., Alonso, J., & Darmofal, D. (2014). CFD vision 2030 study: A path to revolutionary computational aerosciences. NNASA/CR-2014-218178.

Smolarkiewicz, P. K., Sharman, R., Weil, J., Perry, S. G., Heist, D., & Bowker, G. (2007). Building resolving large-eddy simulations and comparison with wind tunnel experiments. Journal of Computational Physics. https://doi.org/10.1016/j.jcp.2007.08.005

Thuerey, N., Weissenow, K., Prantl, L., & Hu, X. (2020). Deep learning methods for reynolds-averaged navier-stokes simulations of airfoil flows. AIAA Journal, 58(1), 25–36. https://doi.org/10.48550/arXiv.1810.08217

Ti, Z., Deng, X. W., & Yang, H. (2020). Wake modeling of wind turbines using machine learning. Applied Energy. https://doi.org/10.1016/j.apenergy.2019.114025

TienPhuc, D., ZhengQi, G., & Zhen, C. (2015). Numerical simulation of the flow field around generic formula one. Journal of Applied Fluid Mechanics, 9(1), 443–450. https://doi.org/10.18869/acadpub.jafm.68.224.24260

Wadi Al-Fatlawi, A., Hashemi, J., Hossain, S., & El Haj Assad, M. (2024). Applying machine learning in CFD to study the impact of thermal characteristics on the aerodynamic characteristics of an airfoil. Journal of Applied Fluid Mechanics, 17(4), 742–755. https://doi.org/10.47176/jafm.17.4.2276

Xingjun, H., Yufei, L., Keyuan, S., Peng, G., & Jingyu, W. (2023). Suppressing methods of the pressure fluctuation in open jet wind tunnels. Journal of Applied Fluid Mechanics, 16(10), 1901–1915. https://doi.org/10.47176/jafm.16.10.1889

Yan, B. W., & Li, Q. S. (2017). Detached-eddy and large-eddy simulations of wind effects on a high-rise structure. Computers and Fluids, 150, 74–83. https://doi.org/10.1016/j.compfluid.2017.02.009

Zan, B. W., Han, Z. H., Xu, C. Z., Liu, M. Q., & Wang, W. Z. (2022). High-dimensional aerodynamic data modeling using a machine learning method based on a convolutional neural network. Advances in Aerodynamics. https://doi.org/10.1186/s42774-022-00128-8

Zhang, J., Gao, G., Liu, T., & Li, Z. (2017). Shape optimization of a kind of earth embankment type windbreak wall along the Lanzhou-Xinjiang railway. Journal of Applied Fluid Mechanics, 10(4), 1189–1200. https://doi.org/10.18869/acadpub.jafm.73.241.27353

Zhang, Z. Q., Liu, H. L., Liu, Z., Zhang, Z., Cheng, G. G., Wang, X. D., & Ding, J. N. (2019). Anisotropic interfacial properties between monolayered black phosphorus and water. Applied Surface Science. https://doi.org/10.1016/j.apsusc.2019.01.037

Zhao, W., Jiang, L., Yao, S., & Chen, W. (2021). Data-driven nonlinear constitutive relations for rarefied flow computations. Advances in Aerodynamics. https://doi.org/10.1186/s42774-021-00085-8

Zhao, Y., Akolekar, H. D., Weatheritt, J., Michelassi, V., & Sandberg, R. D. (2020). RANS turbulence model development using CFD-driven machine learning. Journal of Computational Physics. https://doi.org/10.1016/j.jcp.2020.109413

Zhou, L. (2017). Numerical modelling of scour in steady flows [Université de Lyon]. https://tel.archives-ouvertes.fr/tel-01598600

Zhu, L., Zhang, W., & Tu, G. (2022). Generalization enhancement of artificial neural network for turbulence closure by feature selection. Advances in Aerodynamics. https://doi.org/10.1186/s42774-021-00088-5