Computational fluid dynamics; Input features; Multivariate random forest; Turbulent bluff body flow; Variable importance prediction; Bluff body flows; Computational costs; Computational fluid; Fluid-dynamics; Random forests; Variable importances; Chemical Engineering (all); Physics and Astronomy (all); Physical and Theoretical Chemistry
Abstract :
[en] High-fidelity computational fluid dynamics (CFD) is widely used to understand turbulence and guide engineering design. While effective in predicting complex flow phenomena, CFD simulations at high Reynolds numbers require fine grids, resulting in prohibitive computational costs for parametric studies. To address this, we proposed a framework that uses machine learning (ML) to predict fine-grid results from coarse-grid simulations in a previous work. Coarsening the grid increases grid-induced error and affects turbulence prediction, necessitating a data-driven surrogate model to predict and correct these errors. A Random Forest (RF) regression was used to construct the surrogate model. The proposed framework was tested using a turbulent flow configuration consisting of an enclosed duct with a triangular bluff body acting as a blockage to the incoming flow. The chosen input features (IFs) were shown to be critical in predicting the turbulent flow field. In the current paper, we introduce further enhancements to the framework to allow it to be more robust in its prediction and application. These extensions also serve to reduce the computational cost of the approach without compromising on the accuracy. The proposed extensions include (i) adoption of Multivariate Random Forest (MRF) to replace the RF approach; (ii) identification and reduction of the IFs required for training and prediction using Variable IMportance Prediction (VIMP); (iii) predictions of flow field with changes in the bluff body configurations. The present paper aims to investigate the capability of the proposed extensions within the framework. We show that (i) the MRF allows for the accurate prediction of multiple outputs within one training instance but with a reduced computational cost relative to the RF approach. (ii) the impact of the IFs on the training can be understood via VIMP, and applying the MRF model with reduced IFs selected through VIMP does not cause any detriment to the accuracy of the prediction (iii) the extended framework trained with different bluff body configurations could be robustly applied to predict the flow field in an unseen configuration that is different from those trained. The predictive capability of the approach with these proposed extensions is demonstrated.
M.W. Ahmad M. Mourshed Y. Rezgui Trees vs neurons: comparison between random forest and ANN for high-resolution prediction of building energy consumption Energy Build 147 77 89
A.G. Balasubramanian L. Guastoni P. Schlatter H. Azizpour R. Vinuesa Predicting the wall-shear stress and wall pressure through convolutional neural networks Int. J. Heat Fluid Flow 103 109200
Bardina, J., Ferziger, J. H., Reynolds, W. C.: Improved subgrid-scale models for large-eddy simulation, AIAA Paper No. 80–1357 (1980)
A.D. Beck D.G. Flad C.D. Munz Deep neural networks for data-driven LES closure models J. Comput. Phys. 398 1 23 4000072
S.T. Bose P. Moin D. You Grid-independent large-eddy simulation using explicit filtering Phys. Fluids 22 1 6
L. Breiman Random forests Mach. Learn. 45 5 32
T. Craft B. Launder K. Suga Development and application of a cubic eddy-viscosity model of turbulence Int. J. Heat Fluid Flow 17 108 115
Cremades, A., Hoyas, S., Quintero, P., Lellep, M., Linkmann, M., Vinuesa, R.: Explaining wall-bounded turbulence through deep learning, arXiv:2302.01250 (2023)
J.S. Evans M.A. Murphy Z.A. Holden S.A. Cushman C. Drew Y. Wiersma F. Huettmann Modeling species distribution and change using Random Forest Predictive Species and Habitat Modeling in Landscape Ecology New York, NY Springer
Ferziger, J. H., Perić, M.: Computational methods for fluid dynamics. 3rd ed., Springer, Berlin. (2002)
N. Gourdain F. Sicot F. Duchaine L. Gicquel Large eddy simulation of flows in industrial compressors: a path from 2015 to 2035 Phil. Trans. R. Soc. A. 372 20130323
L. Grinsztajn E. Oyallon G. Varoquaux Why do tree-based models still outperform deep learning on typical tabular data? Adv. Neural Inf. Process. Syst. 35 1 14
T. Han D. Jiang Q. Zhao L. Wang K. Yin Comparison of random forest, artificial neural networks and support vector machine for intelligent diagnosis of rotating machinery Trans. Ins. Meas. Control 40 2681 2693
J.C. Hunt A.M. Savill J.C. Vassilicos Chapter 8: guidelines and criteria for the use of turbulence models in complex flows Predictions of Turbulent Flows
Ishwaran, H., Kogalur, U. B.: Fast unified random forests for survival, regression, and classification (RF-SRC), R package version 3.2.0. (2023)
Ishwaran, H., Lu, M., Kogalur, U. B.: randomForestSRC: variable importance (VIMP) with subsampling inference vignette (2021) http://randomforestsrc.org/articles/vimp.html.
R.I. Issa Solution of the implicitly discretised fluid flow equations by operator-splitting J. Comput. Phys. 62 40 65 825890
N. Jarrin S. Benhamadouche D. Laurence R. Prosser A synthetic-eddy-method for generating inflow conditions for large-eddy simulations Int. J. Heat Fluid Flow 27 585 593
R. Johnson The Handbook of Fluid Dynamics CRC Press
M.L. Kaandorp R.P. Dwight Data-driven modelling of the Reynolds stress tensor using random forest with invariance Comp. Fluids 202 1 16 4078889
A. Kiener S. Langer P. Bekemeyer Data-driven correction of coarse grid CFD simulations Comp. Fluids 264 1 14
M. Klein An attempt to assess the quality of large eddy simulations in the context of implicit filtering Flow Turb. Combust. 75 131 147
C.Y. Lee S. Cant A grid-induced and physics-informed machine learning CFD framework for turbulent flows Flow Turb. Combust. 112 407 442
M. Lesieur O. Métais New trends in large-eddy simulations of turbulence Annu. Rev. Fluid Mech. 28 45 82 1371162
J. Ling R. Jones J. Templeton Machine learning strategies for systems with invariance properties J. Comput. Phys. 318 22 35 3503986
J. Ling A. Kurzawski J. Templeton Reynolds averaged turbulence modelling using deep neural networks with embedded invariance J Fluid Mech 807 155 166 3569308
J. Ling J. Templeton Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty Phys. Fluids 27 1 8
R. Maulik O. San A. Rasheed P. Vedula Subgrid modelling for two-dimensional turbulence using neural networks J Fluid Mech 858 122 144 3871385
M. Milano P. Koumoutsakos Neural network modeling for near wall turbulent flow J. Comput. Phys. 182 1 26
P. Moin K. Mahesh Direct numerical simulation: a tool in turbulence research Annu. Rev. Fluid Mech. 30 539 578 1609753
E.J. Parish K. Duraisamy A paradigm for data-driven predictive modeling using field inversion and machine learning J. Comput. Phys. 305 758 774 3429605
F. Pedregosa G. Varoquaux A. Gramfort V. Michel B. Thirion O. Grisel M. Blondel P. Prettenhofer R. Weiss V. Dubourg J. Vanderplas A. Passos D. Cournapeau M. Brucher M. Perrot E. Duchesnay Scikit-learn: machine learning in Python J. Mach. Learn. Res. 12 2825 2830 2854348
H. Pitsch Large-eddy simulation of turbulent combustion Annu. Rev. Fluid Mech. 38 453 482 2206982
S.B. Pope Turbulent Flows Cambridge University Press
Pope, S. B.: Ten questions concerning the large-eddy simulation of turbulent flows. New J. Phys. 6 (2004)
Probst, P, Wright, M. N, Boulesteix, A. –L.: Hyperparameters and tuning strategies for random forest. WIREs Data Min. Knowl. Discov. 9, e1301 (2019)
P. Sagaut Large Eddy Simulation for Incompressible Flows Berlin, Heidelberg Springer
F. Sarghini G. De Felice S. Santini Neural networks based subgrid scale modeling in large eddy simulations Comp. Fluids 32 97 108
M. Segal Y. Xiao Multivariate random forests WIREs Data Min. Knowl. Discov. 1 80 87
Sjunnesson, A., Olovsson, S., Sjoblom, S.: Validation rig - a tool for flame studies. In: ISABE Conference, Nottingham, UK (1991a)
Sjunnesson, A, Olovsson, S, Max, E.: Measurements of velocities and turbulence in a bluff body stabilized flame. In: Fourth International Conference on Laser Anemometry - Advances and Application, ASME Cleveland, US (1991b)
J. Smagorinsky General circulation experiments with the primitive equations: I. The basic experiment Mon. Weather Rev. 91 99 164
A. Spencer R. Rivlin Isotropic integrity bases for vectors and second-order tensors Arch. Ration. Mech. Anal. 9 45 63 141038
Tracey, B., Duraisamy, K., Alonso, J.: A machine learning strategy to assist turbulence model development, in: 53rd AIAA Aerospace Sciences Meeting, Kissimmee, Florida, AIAA 2015–1287 (2013)
E.R. Van Driest On turbulent flow near a wall J. Aeronau. Sci. 23 1007 1011
P.S. Volpiani Are random forests better suited than neural networks to augment RANS turbulence models? Int. J. Heat Fluid Flow 107 109348
A. Vollant G. Balarac C. Corre Subgrid-scale scalar flux modelling based on optimal estimation theory and machine-learning procedures J. Turb. 18 854 878 3677958
J.X. Wang J.L. Wu H. Xiao Physics-informed machine learning approach for reconstructing Reynolds stress modeling discrepancies based on DNS data Phys. Rev. Fluids 2 034603
H.G. Weller G.R. Tabor H. Jasak C. Fureby A tensorial approach to computational continuum mechanics using object-oriented techniques J. Comput. Phys. 12 620 631
Y.T. Wu F. Porté-Agel Large-eddy simulation of wind-turbine wakes: evaluation of turbine parametrisations Boud.-Layer Meteorol. 138 345 366
J.L. Wu H. Xiao E. Paterson Physics-informed machine learning approach for augmenting turbulence models: a comprehensive framework Phys. Rev. Fluids 3 1 42
J. Wu X.Y. Chen H. Zhang L.D. Xiong H. Lei S.H. Deng Hyperparameter optimization for machine learning models based on bayesian optimization J. Elec. Sci. Tech. 17 26 40
