Hostname: page-component-857557d7f7-8wkb5 Total loading time: 0 Render date: 2025-11-25T22:31:50.565Z Has data issue: false hasContentIssue false
  • English
  • Français

Neural operator-based stochastic forcing for resolvent prediction of space–time turbulence statistics in channel flows

Published online by Cambridge University Press:  25 November 2025

Chutian Wu ,
Xin-Lei Zhang Open the ORCID record for Xin-Lei Zhang [Opens in a new window]  and
Guowei He Open the ORCID record for Guowei He [Opens in a new window]
Chutian Wu
Affiliation:
The State Key Laboratory of Nonlinear Mechanics, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, PR China School of Engineering Sciences, University of Chinese Academy of Sciences, Beijing 100049, PR China
Xin-Lei Zhang*
Affiliation:
The State Key Laboratory of Nonlinear Mechanics, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, PR China School of Engineering Sciences, University of Chinese Academy of Sciences, Beijing 100049, PR China
Guowei He*
Affiliation:
The State Key Laboratory of Nonlinear Mechanics, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, PR China School of Engineering Sciences, University of Chinese Academy of Sciences, Beijing 100049, PR China
*
Corresponding authors: Guowei He, hgw@lnm.imech.ac.cn; Xin-Lei Zhang, zhangxinlei@imech.ac.cn
Corresponding authors: Guowei He, hgw@lnm.imech.ac.cn; Xin-Lei Zhang, zhangxinlei@imech.ac.cn
Get access Rights & Permissions [Opens in a new window]

Abstract

In this work we propose a neural operator-based coloured-in-time forcing model to predict space–time characteristics of large-scale turbulent structures in channel flows. The resolvent-based method has emerged as a powerful tool to capture dominant dynamics and associated spatial structures of turbulent flows. However, the method faces the difficulty in modelling the coloured-in-time nonlinear forcing, which often leads to large predictive discrepancies in the frequency spectra of velocity fluctuations. Although the eddy viscosity has been introduced to enhance the resolvent-based method by partially accounting for the forcing colour, it is still not able to accurately capture the decay rate of the time-correlation function. Also, the uncertainty in the modelled eddy viscosity can significantly limit the predictive reliability of the method. In view of these difficulties, we propose using the neural operator based on the DeepONet architecture to model the stochastic forcing as a function of mean velocity and eddy viscosity. Specifically, the DeepONet-based model is constructed to map an arbitrary eddy-viscosity profile and corresponding mean velocity to stochastic forcing spectra based on the direct numerical simulation data at $Re_\tau =180$. Furthermore, the learned forcing model is integrated with the resolvent operator, which enables predicting the space–time flow statistics based on the eddy viscosity and mean velocity from the Reynolds-averaged Navier–Stokes (RANS) method. Our results show that the proposed forcing model can accurately predict the frequency spectra of velocity in channel flows at different characteristic scales. Moreover, the model remains robust across different RANS-provided eddy viscosities and generalises well to $Re_\tau =550$.

JFM classification

Waves/Free-surface Flows: Channel flow Turbulent Flows: Turbulence modelling Mathematical Foundations: Machine learning

Information

Type
JFM Papers
Information
Journal of Fluid Mechanics , Volume 1024 , 10 December 2025 , A1
Check for updates
Copyright
© The Author(s), 2025. Published by Cambridge University Press

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

References

Ali, N., Mellado, J.P. & Wilczek, M. 2023 A wavenumber–frequency spectrum model for sheared convective atmospheric boundary layer flows. J. Atmos. Sci. 80 (3), 763–776.10.1175/JAS-D-22-0079.1CrossRefGoogle Scholar
Bhatia, R. 1997 Symmetric Norms. Springer New York, pp. 84–111.Google Scholar
Blake, W.K. 2017 Mechanics of Flow-Induced Sound and Vibration, Complex flow-structure interactions, vol. 2. Academic press.Google Scholar
Cai, S., Wang, Z., Lu, L., Zaki, T.A. & Karniadakis, G.E. 2021 DeepM&Mnet: inferring the electroconvection multiphysics fields based on operator approximation by neural networks. J. Comput. Phys. 436, 110296.10.1016/j.jcp.2021.110296CrossRefGoogle Scholar
Cess, R.D. 1958 A survey of the literature on heat transfer in turbulent tube flow. Res. Rep. 8–0529.Google Scholar
Choi, H. & Moin, P. 1990 On the space-time characteristics of wall-pressure fluctuations. Phys. Fluids A: Fluid Dyn. 2 (8), 1450–1460.10.1063/1.857593CrossRefGoogle Scholar
Cossu, C. & Hwang, Y. 2017 Self-sustaining processes at all scales in wall-bounded turbulent shear flows. Philos. Trans. Royal Soc. A: Math. Phys. Engine. Sci. 375(2089), 20160088 10.1098/rsta.2016.0088CrossRefGoogle ScholarPubMed
Cossu, C., Pujals, G. & Depardon, S. 2009 Optimal transient growth and very large-scale structures in turbulent boundary layers. J. Fluid Mech. 619, 79–94.10.1017/S0022112008004370CrossRefGoogle Scholar
Di, L., Clark, P., Lu, L., Meneveau, C., Karniadakis, G.E. & Zaki, T.A. 2023 Neural operator prediction of linear instability waves in high-speed boundary layers. J. Comput. Phys. 474, 111793.Google Scholar
Douglas, D.H. & Peucker, T.K. 2011 Algorithms for the Reduction of the Number of Points Required to Represent a Digitized Line or Its Caricature. Chap. 2, John Wiley & Sons, Ltd, pp. 15–28.Google Scholar
Duraisamy, K., Iaccarino, G. & Xiao, H. 2019 Turbulence modeling in the age of data. Annu. Rev. Fluid Mech. 51 (1), 357–377.10.1146/annurev-fluid-010518-040547CrossRefGoogle Scholar
Ganapathisubramani, B., Hutchins, N., Monty, J.P., Chung, D. & Marusic, I. 2012 Amplitude and frequency modulation in wall turbulence. J. Fluid Mech. 712, 61–91.10.1017/jfm.2012.398CrossRefGoogle Scholar
Glorot, X. & Bengio, Y. 2010 Understanding the difficulty of training deep feedforward neural networks. In Proc.13th Int. Conf. Artif. Intell. Statist. (AISTATS), in Proc. Mach. Learn. Res. (ed. Y. W. Teh & M. Titterington), vol. 9, 249–256. PMLR.Google Scholar
Gupta, V., Madhusudanan, A., Wan, M., Illingworth, S.J. & Juniper, M.P. 2021 Linear-model-based estimation in wall turbulence: improved stochastic forcing and eddy viscosity terms. J. Fluid Mech. 925, A18.10.1017/jfm.2021.671CrossRefGoogle Scholar
Hairer, E., Nørsett, S.P. & Wanner, G. 1993 Multistep methods and general linear methods. In Solving Ordinary Differential Equations I: Nonstiff Problems, pp. 355–474, Springer.Google Scholar
He, G., Jin, G. & Yang, Y. 2017 Space-time correlations and dynamic coupling in turbulent flows. Annu. Rev. Fluid Mech. 49 (1), 51–70.10.1146/annurev-fluid-010816-060309CrossRefGoogle Scholar
He, G., Wang, M. & Lele, S.K. 2004 On the computation of space-time correlations by large-eddy simulation. Phys. Fluids 16 (11), 3859–3867.10.1063/1.1779251CrossRefGoogle Scholar
Holford, J.J., Lee, M. & Hwang, Y. 2023 Optimal white-noise stochastic forcing for linear models of turbulent channel flow. J. Fluid Mech. 961, A32.10.1017/jfm.2023.234CrossRefGoogle Scholar
Holford, J.J., Lee, M. & Hwang, Y. 2024 A data-driven quasi-linear approximation for turbulent channel flow. J. Fluid Mech. 980, A12.10.1017/jfm.2023.1073CrossRefGoogle Scholar
Hwang, Y. 2015 Statistical structure of self-sustaining attached eddies in turbulent channel flow. J. Fluid Mech. 767, 254–289.10.1017/jfm.2015.24CrossRefGoogle Scholar
Hwang, Y. & Cossu, C. 2010 a Amplification of coherent streaks in the turbulent Couette flow: an input–output analysis at low Reynolds number. J. Fluid Mech. 643, 333–348.10.1017/S0022112009992151CrossRefGoogle Scholar
Hwang, Y. & Cossu, C. 2010 b Linear non-normal energy amplification of harmonic and stochastic forcing in the turbulent channel flow. J. Fluid Mech. 664, 51–73.10.1017/S0022112010003629CrossRefGoogle Scholar
Hwang, Y. & Cossu, C. 2010 c Self-sustained process at large scales in turbulent channel flow. Phys. Rev. Lett. 105, 044505.10.1103/PhysRevLett.105.044505CrossRefGoogle Scholar
Jovanović, M. & Bamieh, B. 2001 Modeling flow statistics using the linearized Navier–Stokes equations, In Proceedings of the 40th IEEE Conference on Decision and Control (Cat. No.01CH37228), vol. 5, pp. 4944–4949.Google Scholar
Jovanović, M.R. & Bamieh, B. 2005 Componentwise energy amplification in channel flows. J. Fluid Mech. 534, 145–183.10.1017/S0022112005004295CrossRefGoogle Scholar
Karniadakis, G.E., Israeli, M. & Orszag, S.A. 1991 High-order splitting methods for the incompressible Navier–Stokes equations. J. Comput. Phys. 97 (2), 414–443.10.1016/0021-9991(91)90007-8CrossRefGoogle Scholar
Kovachki, N., Li, Z., Liu, B., Azizzadenesheli, K., Bhattacharya, K., Stuart, A. & Anandkumar, A. 2023 Neural operator: learning maps between function spaces with applications to PDEs. J. Mach. Learn. Res. 24 (89), 1–97.Google Scholar
Del, Á., Juan, C. & Jiménez, J. 2006 Linear energy amplification in turbulent channels. J. Fluid Mech. 559, 205–213.Google Scholar
Del, Á., Juan, C. & Jiménez, J. 2009 Estimation of turbulent convection velocities and corrections to Taylor’s approximation. J. Fluid Mech. 640, 5–26.Google Scholar
Lee, M., Malaya, N. & Moser, R.D. 2013 Petascale direct numerical simulation of turbulent channel flow on up to 786K cores. In SC ’13: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, pp. 1–11.Google Scholar
Lee, M. & Moser, R.D. 2015 Direct numerical simulation of turbulent channel flow up to ${\textit{Re}}_{\tau}\approx 5200$ . J. Fluid Mech. 774, 395–415.10.1017/jfm.2015.268CrossRefGoogle Scholar
Li, Z. & Yang, X. 2024 Resolvent-based motion-to-wake modelling of wind turbine wakes under dynamic rotor motion. J. Fluid Mech. 980, A48.10.1017/jfm.2023.1097CrossRefGoogle Scholar
Lin, C., Maxey, M., Li, Z. & Karniadakis, G.E. 2021 A seamless multiscale operator neural network for inferring bubble dynamics. J. Fluid Mech. 929, A18.10.1017/jfm.2021.866CrossRefGoogle Scholar
Ling, J. & Templeton, J. 2015 Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty. Phys. Fluids 27 (8), 085103.10.1063/1.4927765CrossRefGoogle Scholar
Liu, Q., Sun, Y., Yeh, C.-A., Ukeiley, L.S., Cattafesta III, L.N. & Taira, K. 2021 Unsteady control of supersonic turbulent cavity flow based on resolvent analysis. J. Fluid Mech. 925, A5.10.1017/jfm.2021.652CrossRefGoogle Scholar
Lu, L., Jin, P., Pang, G., Zhang, Z. & Karniadakis, G.N. 2021 Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. Nat. Machine Intell. 3 (3), 218–229.10.1038/s42256-021-00302-5CrossRefGoogle Scholar
Lu, L., Meng, X., Cai, S., Mao, Z., Goswami, S., Zhang, Z. & Karniadakis, G.E. 2022 A comprehensive and fair comparison of two neural operators (with practical extensions) based on FAIR data. Comput. Method. Appl. M. 393, 114778.10.1016/j.cma.2022.114778CrossRefGoogle Scholar
Luhar, M., Sharma, A.S. & McKeon, B.J. 2014 Opposition control within the resolvent analysis framework. J. Fluid Mech. 749, 597–626.10.1017/jfm.2014.209CrossRefGoogle Scholar
Lundberg, S.M. & Lee, S.-I. 2017 A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems, pp. 4765–4774.Google Scholar
Mao, Z., Lu, L., Marxen, O., Zaki, T.A. & Karniadakis, G.E. 2021 DeepM&Mnet for hypersonics: predicting the coupled flow and finite-rate chemistry behind a normal shock using neural-network approximation of operators. J. Comput. Phys. 447, 110698.10.1016/j.jcp.2021.110698CrossRefGoogle Scholar
McKeon, B.J. 2017 The engine behind (wall) turbulence: perspectives on scale interactions. J. Fluid Mech. 817, P1.10.1017/jfm.2017.115CrossRefGoogle Scholar
McKeon, B.J. & Sharma, A.S. 2010 A critical-layer framework for turbulent pipe flow. J. Fluid Mech. 658, 336–382.10.1017/S002211201000176XCrossRefGoogle Scholar
Moarref, R., Sharma, A.S., Tropp, J.A. & McKeon, B.J. 2013 Model-based scaling of the streamwise energy density in high-Reynolds-number turbulent channels. J. Fluid Mech. 734, 275–316.10.1017/jfm.2013.457CrossRefGoogle Scholar
Morra, P., Meneveau, C. & Zaki, T.A. 2024 ML for fast assimilation of wall-pressure measurements from hypersonic flow over a cone. Sci. Rep. 14 (1), 12853.10.1038/s41598-024-63053-4CrossRefGoogle Scholar
Morra, P., Nogueira, P.A.S., Cavalieri, A.V.G. & Henningson, D.S. 2021 The colour of forcing statistics in resolvent analyses of turbulent channel flows. J. Fluid Mech. 907, A24.10.1017/jfm.2020.802CrossRefGoogle Scholar
Morra, P., Semeraro, O., Henningson, D.S. & Cossu, C. 2019 On the relevance of Reynolds stresses in resolvent analyses of turbulent wall-bounded flows. J. Fluid Mech. 867, 969–984.10.1017/jfm.2019.196CrossRefGoogle Scholar
Nogueira, P.A.S., Morra, P., Martini, E., Cavalieri, A.V.G. & Henningson, D.S. 2021 Forcing statistics in resolvent analysis: application in minimal turbulent Couette flow. J. Fluid Mech. 908, A32.10.1017/jfm.2020.918CrossRefGoogle Scholar
Orszag, S.A. 1971 a Accurate solution of the Orr–Sommerfeld stability equation. J. Fluid Mech. 50 (4), 689–703.10.1017/S0022112071002842CrossRefGoogle Scholar
Orszag, S.A. 1971 b On the elimination of aliasing in finite-difference schemes by filtering high-wavenumber components. J. Atmospheric Sci. 28 (6), 1074–1074.10.1175/1520-0469(1971)028<1074:OTEOAI>2.0.CO;22.0.CO;2>CrossRefGoogle Scholar
Park, J. & Choi, H. 2021 Toward neural-network-based large eddy simulation: application to turbulent channel flow. J. Fluid Mech. 914, A16.10.1017/jfm.2020.931CrossRefGoogle Scholar
Pickering, E., Rigas, G., Schmidt, O.T., Sipp, D. & Colonius, T. 2021 a Optimal eddy viscosity for resolvent-based models of coherent structures in turbulent jets. J. Fluid Mech. 917, A29.10.1017/jfm.2021.232CrossRefGoogle Scholar
Pickering, E., Towne, A., Jordan, P. & Colonius, T. 2021 b Resolvent-based modeling of turbulent jet noise. J. Acoustical Soc. Am. 150 (4), 2421–2433.10.1121/10.0006453CrossRefGoogle ScholarPubMed
Pope, S.B. 2000 Turbulent Flows. Cambridge University Press.Google Scholar
Ran, W., Zare, A. & Jovanović, M.R. 2021 Model-based design of riblets for turbulent drag reduction. J. Fluid Mech. 906, A7.10.1017/jfm.2020.722CrossRefGoogle Scholar
Reynolds, W.C. & Hussain, A.K.M.F. 1972 The mechanics of an organized wave in turbulent shear flow. Part 3. Theoretical models and comparisons with experiments. J. Fluid Mech. 54 (2), 263–288.10.1017/S0022112072000679CrossRefGoogle Scholar
Reynolds, W.C. & Tiederman, W.G. 1967 Stability of turbulent channel flow, with application to Malkus’s theory. J. Fluid Mech. 27 (2), 253–272.10.1017/S0022112067000308CrossRefGoogle Scholar
Rosenberg, K. & McKeon, B.J. 2019 Efficient representation of exact coherent states of the Navier–Stokes equations using resolvent analysis. Fluid Dyn. Res. 51 (1), 011401.10.1088/1873-7005/aab1abCrossRefGoogle Scholar
Russo, S. & Luchini, P. 2016 The linear response of turbulent flow to a volume force: comparison between eddy-viscosity model and DNS. J. Fluid Mech. 790, 104–127.10.1017/jfm.2016.4CrossRefGoogle Scholar
von, S., Jakob, G.R., Schmidt, O.T., Jordan, P. & Oberleithner, K. 2024 On the role of eddy viscosity in resolvent analysis of turbulent jets. J. Fluid Mech. 1000, A51.Google Scholar
Schmid, P.J., Henningson, D.S. & Jankowski, D.F. 2002 Stability and transition in shear flows. Appl. Math. Sci. Appl. Mech. Rev 142–55 (3), B57–B59.10.1115/1.1470687CrossRefGoogle Scholar
Smits, A.J., McKeon, B.J. & Marusic, I. 2011 High-Reynolds number wall turbulence. Annu. Rev. Fluid Mech. 43 (2011), 353–375.10.1146/annurev-fluid-122109-160753CrossRefGoogle Scholar
Spalart, P. & Allmaras, S. 1992 A one-equation turbulence model for aerodynamic flows, In 30th aerospace sciences meeting and exhibit, p. 439.Google Scholar
Symon, S., Madhusudanan, A., Illingworth, S.J. & Marusic, I. 2023 Use of eddy viscosity in resolvent analysis of turbulent channel flow. Phys. Rev. Fluids 8, 064601.10.1103/PhysRevFluids.8.064601CrossRefGoogle Scholar
Symon, S., Rosenberg, K., Dawson, S.T.M. & McKeon, B.J. 2018 Non-normality and classification of amplification mechanisms in stability and resolvent analysis. Phys. Rev. Fluids 3 (5), 053902.10.1103/PhysRevFluids.3.053902CrossRefGoogle Scholar
Taassob, A., Kumar, A., Gitushi, K.M., Ranade, R. & Echekki, T. 2024 A PINN-DeepONet framework for extracting turbulent combustion closure from multiscalar measurements. Comput. Method. Appl. Mechan. Engine. 429, 117163.10.1016/j.cma.2024.117163CrossRefGoogle Scholar
Taira, K. 2017 Modal analysis of fluid flows: an overview. AIAA J. 55 (12), 4013–4041.10.2514/1.J056060CrossRefGoogle Scholar
Towne, A., Lozano-Durán, A. & Yang, X. 2020 Resolvent-based estimation of space-time flow statistics. J. Fluid Mech. 883, A17.10.1017/jfm.2019.854CrossRefGoogle Scholar
Wang, Y., Wu, T. & He, G. 2025 The significant contribution of stochastic forcing to nonlinear energy transfer in resolvent analysis. Theoretical Appl. Mechan. Lett. 15 (1), 100521.10.1016/j.taml.2024.100521CrossRefGoogle Scholar
Wilcox, D.C. 1998 Turbulence Modeling for CFD. DCW industries.Google Scholar
Wilcox, D.C. 2008 Formulation of the $k$ - $\omega$ turbulence model revisited. AIAA J. 46 (11), 2823–2838.10.2514/1.36541CrossRefGoogle Scholar
Williams, J.E.F. & Kempton, A.J. 1978 The noise from the large-scale structure of a jet. J. Fluid Mech. 84 (4), 673–694.10.1017/S0022112078000415CrossRefGoogle Scholar
Willoughby, R.A. 1973 Numerical initial value problems in ordinary differential equations (C. William Gear). SIAM Rev. 15 (3), 676–678.10.1137/1015088CrossRefGoogle Scholar
Wu, C., Wang, S., Zhang, X.-L. & He, G. 2023 Explainability analysis of neural network-based turbulence modeling for transonic axial compressor rotor flows. Aerosp. Sci. Technol. 141, 108542.10.1016/j.ast.2023.108542CrossRefGoogle Scholar
Wu, C., Zhang, X.-L. & He, G. 2025 a Code for learning resolvent neural operator. Available at https://github.com/XinleiZhang/RNO.Google Scholar
Wu, C., Zhang, X.-L., Xu, D. & He, G. 2025 b A framework for learning symbolic turbulence models from indirect observation data via neural networks and feature importance analysis. J. Comput. Phys. 537, 114068.10.1016/j.jcp.2025.114068CrossRefGoogle Scholar
Wu, T. & He, G. 2020 Local modulated wave model for the reconstruction of space-time energy spectra in turbulent flows. J. Fluid Mech. 886, A11.10.1017/jfm.2019.1044CrossRefGoogle Scholar
Wu, T. & He, G. 2021 Stochastic dynamical model for space-time energy spectra in turbulent shear flows. Phys. Rev. Fluids 6, 054602.10.1103/PhysRevFluids.6.054602CrossRefGoogle Scholar
Wu, T. & He, G. 2023 Composition of resolvents enhanced by random sweeping for large-scale structures in turbulent channel flows. J. Fluid Mech. 956, A31.10.1017/jfm.2023.39CrossRefGoogle Scholar
Yeh, C.-A. & Taira, K. 2019 Resolvent-analysis-based design of airfoil separation control. J. Fluid Mech. 867, 572–610.10.1017/jfm.2019.163CrossRefGoogle Scholar
Yin, M., Zhang, E., Yu, Y. & Karniadakis, G.E. 2022 Interfacing finite elements with deep neural operators for fast multiscale modeling of mechanics problems. Comput. Method. Appl. Mechan. Engine. 402, 115027.10.1016/j.cma.2022.115027CrossRefGoogle ScholarPubMed
Ying, A., Chen, X., Li, Z. & Fu, L. 2024 Optimisation and modelling of eddy viscosity in the resolvent analysis of turbulent channel flows. J. Fluid Mech. 1001, A20.10.1017/jfm.2024.1099CrossRefGoogle Scholar
Ying, A., Liang, T., Li, Z. & Fu, L. 2023 A resolvent-based prediction framework for incompressible turbulent channel flow with limited measurements. J. Fluid Mech. 976, A31.10.1017/jfm.2023.867CrossRefGoogle Scholar
Zare, A., Jovanović, M.R. & Georgiou, T.T. 2017 Colour of turbulence. J. Fluid Mech. 812, 636–680.10.1017/jfm.2016.682CrossRefGoogle Scholar
Zhang, X.-L., Xiao, H., Luo, X. & He, G. 2022 Ensemble Kalman method for learning turbulence models from indirect observation data. J. Fluid Mech. 949, A26.10.1017/jfm.2022.744CrossRefGoogle Scholar
Zhou, Z., He, G., Wang, S. & Jin, G. 2019 Subgrid-scale model for large-eddy simulation of isotropic turbulent flows using an artificial neural network. Comput. Fluids 195, 104319.10.1016/j.compfluid.2019.104319CrossRefGoogle Scholar