Conditional probability distribution; Machine learning; Separated flows; Turbulent boundary layer; Wall shear stress model; Wall-modeled large eddy simulation; Conditional probability distributions; Large-eddy simulations; Machine-learning; Stress models; Turbulent boundary layers; Wall shear stress; Wall-shear stress; Chemical Engineering (all); Physics and Astronomy (all); Physical and Theoretical Chemistry
Abstract :
[en] Most wall shear stress models assume the boundary layer to be fully turbulent, at equilibrium, and attached. Under these strong assumptions, that are often not verified in industrial applications, these models predict an averaged behavior. To address the instantaneous and non-equilibrium phenomenon of separation, the mixture density network (MDN), the neural network implementation of a Gaussian Mixture Model, initially deployed for uncertainty prediction, is employed as a wall shear stress model in the context of wall-modeled large eddy simulations (wmLES) of turbulent separated flows. The MDN is trained to estimate the conditional probability p(τw|x), knowing certain entries x, to better predict the instantaneous wall shear stress τw (which is then sampled from the distribution). In this work, an MDN is trained on a turbulent channel flow at the friction Reynolds number Reτ of 1000 and on the two-dimensional periodic hill at the bulk Reynolds number of 10,595. The latter test case is known to feature a massive separation from the hill crest. By construction, the model outputs the probability distribution of the two wall-parallel components of the wall shear stress, conditioned by the model inputs: the instantaneous velocity field, the instantaneous and mean pressure gradients, and the wall curvature. Generalizability is ensured by carefully non-dimensionalizing databases with the kinematic viscosity and wall-model height. The relevance of the MDN model is evaluated a posteriori by performing wmLES using the in-house high-order discontinuous Galerkin (DG) flow solver, named Argo-DG, on a turbulent channel flow at Reτ=2000 and on the same periodic hill flow. The data-driven WSS model significantly improves the prediction of the wall shear stress on both the upper and lower walls of the periodic hill compared to quasi-analytical WSS models.
Safran Tech and computational resources made available on the Tier-1 supercomputer of the F\u00E9d\u00E9ration Wallonie-Bruxelles, infrastructure funded by the Walloon Region under the Grant Agreement No. 1117545.The funding of M. Boxho\u2019s Ph.D. project by Safran Tech is gratefully acknowledged. The present research benefited from computational resources made available on the Tier-1 supercomputer Zenobe of the F\u00E9d\u00E9ration Wallonie-Bruxelles, infrastructure funded by the Walloon Region under the Grant Agreement No. 1117545, and from computational resources made available on Lucia, the new Tier-1 supercomputer of the Walloon Region, infrastructure funded by the Walloon Region under the Grant Agreement No. 1910247.
E. Balaras C. Benocci U. Piomelli Two-layer approximate boundary conditions for large-eddy simulations AIAA J. 34 6 1111 1119 10.2514/3.13200
Benocci, C., Pinelli, A.: The role of the forcing term in the large eddy simulation of equilibrium channel flow. In: International Symposium on Engineering Turbulence Modelling and Measurements. Elsevier (1990). https://api.semanticscholar.org/CorpusID:129356123
S.T. Bose P. Moin A dynamic slip boundary condition for wall-modeled large-eddy simulation Phys. Fluids 26 1 10.1063/1.4849535 015104
S.T. Bose G.I. Park Wall-modeled large-eddy simulation for complex turbulent flows Annu. Rev. Fluid Mech. 50 535 561 3753221 10.1146/annurev-fluid-122316-045241
M. Boxho M. Rasquin T. Toulorge et al. Analysis of space–time correlations to support the development of wall-modeled LES Flow Turbul. Combust. 109 4 1081 1109 10.1007/s10494-022-00365-3
M. Breuer B. Kniazev M. Abel Development of wall models for LES of separated flows using statistical evaluations Comput. Fluids 36 5 817 837 10.1016/j.compfluid.2006.09.001
M. Breuer N. Peller C. Rapp et al. Flow over periodic hills: numerical and experimental study in a wide range of Reynolds numbers Comput. Fluids 38 2 433 457 10.1016/j.compfluid.2008.05.002
S.L. Brunton B.R. Noack P. Koumoutsakos Machine learning for fluid mechanics Annu. Rev. Fluid Mech. 52 477 508 4508944 10.1146/annurev-fluid-010719-060214
Cadieux, F., Sadique, J., Yang, X.I., et al.: Wall-modeled large eddy simulation of laminar and turbulent separation bubble flows. In: 46th AIAA Fluid Dynamics Conference. AIAA, pp. 2016–3189 (2016). https://doi.org/10.2514/6.2016-3189
C. Carton de Wiart K. Hillewaert L. Bricteux et al. Implicit LES of free and wall bounded turbulent flows based on the discontinuous Galerkin/symmetric interior penalty method Int. J. Numer. Methods Fluids 78 6 335 354 3350177 10.1002/fld.4021
C. Carton de Wiart K. Hillewaert L. Bricteux et al. J. Fröhlich H. Kuerten B. Geurts V. Armenio et al. LES using a discontinuous galerkin method: isotropic turbulence, channel flow and periodic hill flow Direct and Large-Eddy Simulation IX ERCOFTAC Series Cham Springer 10.1007/978-3-319-14448-1_13 20
H. Choi P. Moin Grid-point requirements for large eddy simulation: Chapman’s estimates revisited Phys. Fluids 24 1 10.1063/1.3676783 011702
D. Dupuy N. Odier C. Lapeyre Data-driven wall modeling for turbulent separated flows J. Comput. Phys. 487 4582078 10.1016/j.jcp.2023.112173 112173
D. Dupuy N. Odier C. Lapeyre et al. Modeling the wall shear stress in large-eddy simulation using graph neural networks Data-Centric Eng. 4 10.1017/dce.2023.2 e7
K. Duraisamy G. Iaccarino H. Xiao Turbulence modeling in the age of data Annu. Rev. Fluid Mech. 51 357 377 3965623 10.1146/annurev-fluid-010518-040547
Frère, A.: Towards wall-modeled large-eddy simulations of high Reynolds number airfoils using a discontinuous Galerkin method. PhD thesis, UCL - Université Catholique de Louvain (2018)
J. Fröhlich C.P. Mellen W. Rodi et al. Highly resolved large-eddy simulation of separated flow in a channel with streamwise periodic constrictions J. Fluid Mech. 526 19 66 2261429 10.1017/S0022112004002812
Gloerfelt, X., Cinnella, P.: Investigation of the flow dynamics in a channel constricted by periodic hills. In: 45th AIAA Fluid Dynamics Conference. AIAA (2015)
J. Graham K. Kanov X. Yang et al. A web services-accessible database of turbulent channel flow and its use for testing a new integral wall model for les J. Turbul. 17 2 181 215 10.1080/14685248.2015.1088656
S. Heinz A review of hybrid RANS-LES methods for turbulent flows: concepts and applications Prog. Aerosp. Sci. 114 10.1016/j.paerosci.2019.100597 100597
Hillewaert, K.: Development of the discontinuous Galerkin method for high-resolution, large scale CFD and acoustics in industrial geometries. Ph.D. thesis, Ecole polytechnique de Louvain/iMMC (2013)
S. Hoyas J. Jimenez Reynolds number effects on the Reynolds-stress budgets in turbulent channels Phys. Fluids 20 10.1063/1.3005862 101511
G.T. Jamaat Y. Hattori A priori assessment of nonlocal data-driven wall modeling in large eddy simulation Phys. Fluids 35 5 10.1063/5.0146770 055117
B. Krank M. Kronbichler W.A. Wall A multiscale approach to hybrid RANS/LES wall modeling within a high-order discontinuous Galerkin scheme using function enrichment Int. J. Numer. Meth. Fluids 90 2 81 113 3939492 10.1002/fld.4712
J. Larsson S. Kawai J. Bodart et al. Large eddy simulation with modeled wall-stress: recent progress and future directions Mech. Eng. Rev. 3 1 15 00418 10.1299/mer.15-00418
Y.M. Lee J.H. Lee J. Lee Artificial neural network-based wall-modeled large-eddy simulations of turbulent channel and separated boundary layer flows Aerosp. Sci. Technol. 132 10.1016/j.ast.2022.108014 108014
Y. Li E. Perlman M. Wan et al. A public turbulence database cluster and applications to study Lagrangian evolution of velocity increments in turbulence J. Turbul. 9 31 N31 10.1080/14685240802376389
A. Lozano-Durán H.J. Bae Machine learning building-block-flow wall model for large-eddy simulation J. Fluid Mech. 963 A35 4591388 10.1017/jfm.2023.331
I. Marusic J.P. Monty M. Hultmark et al. On the logarithmic region in wall turbulence J. Fluid Mech. 3018521 10.1017/jfm.2012.511
Mellen, C.P., Fröhlich, J., Rodi, W.: Large-eddy simulation of the flow over periodic hills. In: 16th IMACS World Congress, Lausanne, Switzerland (2000)
Perlman, E., Burns, R., Li, Y., et al.: Data exploration of turbulence simulations using a database cluster. Supercomputing SC07, ACM, IEEE (2007)
U. Piomelli Wall-layer models for large-eddy simulations Prog. Aerosp. Sci. 44 6 437 446 10.1016/j.paerosci.2008.06.001
U. Piomelli E. Balaras Wall-layer models for large eddy simulations Annu. Rev. Fluid Mech. 34 349 374 1893771 10.1146/annurev.fluid.34.082901.144919
U. Piomelli P. Moin J.H. Ferziger et al. New approximate boundary conditions for large-eddy simulations of wall-bounded flows Phys. Fluids A 1 1061 1068 10.1063/1.857397
S. Radhakrishnan L.A. Gyamfi A. Miró et al. et al. H. Jagode H. Anzt H. Ltaief et al. et al. A data-driven wall-shear stress model for LES using gradient boosted decision trees High Performance Computing. Lecture Notes in Computer Science Berlin Springer 105 121 10.1007/978-3-030-90539-2_7 12761
M. Rasquin A. Bauer K. Hillewaert Scientific post hoc and in situ visualisation of high-order polynomial solutions from massively parallel simulations Int. J. Comput. Fluid Dyn. 4 22 171 180 3957551 10.1080/10618562.2019.1618453
S. Song J. Eaton Flow structures of a separating, reattaching, and recovering boundary layer for a large range of Reynolds number Exp. Fluids 36 642 653 10.1007/s00348-003-0762-2
Spalart, P.R., et al.: Comments on the feasibility of LES for wings and on a hybrid RANS/LES approach. Advances in DNS/LES: Direct Numerical Simulation and Large Eddy Simulation, pp. 137–148 (1997)
P. Spalart S. Deck M. Shur et al. A new version of detached-eddy simulation, resistant to ambiguous grid densities Theor. Comput. Fluid Dyn. 20 181 195 10.1007/s00162-006-0015-0
P.R. Spalart S. Deck M.L. Shur et al. Approximate wall boundary conditions for large eddy simulations Theor. Comput. Fluid Dyn. 20 3 181 195 10.1007/s00162-006-0015-0
X.I.A. Yang S. Zafar J.X. Wang et al. Predictive large-eddy-simulation wall modeling via physics-informed neural networks Phys. Rev. Fluids 4 10.1103/PhysRevFluids.4.034602 034602
A. Zhang Z.C. Lipton M. Li et al. Dive into Deep Learning Cambridge Cambridge University Press
Z. Zhideng X.I.A. Xiang Z. Fengshun et al. A wall model learned from the periodic hill data and the law of the wall Phys. Fluids 35 5 10.1063/5.0143650 055108
Z. Zhou G. He X. Yang Wall model based on neural networks for LES of turbulent flows over periodic hills Phys. Rev. Fluids 6 10.1103/PhysRevFluids.6.054610 054610
