Model Reduction with Memory and the Machine Learning of Dynamical Systems

Year:    2019

Communications in Computational Physics, Vol. 25 (2019), Iss. 4 : pp. 947–962

Abstract

The well-known Mori-Zwanzig theory tells us that model reduction leads to memory effect. For a long time, modeling the memory effect accurately and efficiently has been an important but nearly impossible task in developing a good reduced model. In this work, we explore a natural analogy between recurrent neural networks and the Mori-Zwanzig formalism to establish a systematic approach for developing reduced models with memory. Two training models — a direct training model and a dynamically coupled training model — are proposed and compared. We apply these methods to the Kuramoto-Sivashinsky equation and the Navier-Stokes equation. Numerical experiments show that the proposed method can produce reduced model with good performance on both short-term prediction and long-term statistical properties.

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/cicp.OA-2018-0269

Communications in Computational Physics, Vol. 25 (2019), Iss. 4 : pp. 947–962

Published online:    2019-01

AMS Subject Headings:    Global Science Press

Copyright:    COPYRIGHT: © Global Science Press

Pages:    16

Keywords:    Model reduction Mori-Zwanzig recurrent neural networks.

  1. Temporally sparse data assimilation for the small-scale reconstruction of turbulence

    Wang, Yunpeng | Yuan, Zelong | Xie, Chenyue | Wang, Jianchun

    Physics of Fluids, Vol. 34 (2022), Iss. 6

    https://doi.org/10.1063/5.0091391 [Citations: 10]
  2. Artificial neural network approach for turbulence models: A local framework

    Xie, Chenyue | Xiong, Xiangming | Wang, Jianchun

    Physical Review Fluids, Vol. 6 (2021), Iss. 8

    https://doi.org/10.1103/PhysRevFluids.6.084612 [Citations: 15]
  3. The effect of filter anisotropy on the large eddy simulation of turbulence

    Chang, Ning | Yuan, Zelong | Wang, Yunpeng | Wang, Jianchun

    Physics of Fluids, Vol. 35 (2023), Iss. 3

    https://doi.org/10.1063/5.0142643 [Citations: 0]
  4. Deconvolutional artificial-neural-network framework for subfilter-scale models of compressible turbulence

    Yuan, Zelong | Wang, Yunpeng | Xie, Chenyue | Wang, Jianchun

    Acta Mechanica Sinica, Vol. 37 (2021), Iss. 12 P.1773

    https://doi.org/10.1007/s10409-021-01150-7 [Citations: 15]
  5. Artificial neural network-based nonlinear algebraic models for large eddy simulation of turbulence

    Xie, Chenyue | Yuan, Zelong | Wang, Jianchun

    Physics of Fluids, Vol. 32 (2020), Iss. 11

    https://doi.org/10.1063/5.0025138 [Citations: 68]
  6. Memory-based parameterization with differentiable solver: Application to Lorenz ’96

    Bhouri, Mohamed Aziz | Gentine, Pierre

    Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 33 (2023), Iss. 7

    https://doi.org/10.1063/5.0131929 [Citations: 3]
  7. Learning Proper Orthogonal Decomposition of Complex Dynamics Using Heavy-ball Neural ODEs

    Baker, Justin | Cherkaev, Elena | Narayan, Akil | Wang, Bao

    Journal of Scientific Computing, Vol. 95 (2023), Iss. 2

    https://doi.org/10.1007/s10915-023-02176-8 [Citations: 0]
  8. Thermodynamics of Learning Physical Phenomena

    Cueto, Elias | Chinesta, Francisco

    Archives of Computational Methods in Engineering, Vol. 30 (2023), Iss. 8 P.4653

    https://doi.org/10.1007/s11831-023-09954-5 [Citations: 11]
  9. On the structure of time-delay embedding in linear models of non-linear dynamical systems

    Pan, Shaowu | Duraisamy, Karthik

    Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 30 (2020), Iss. 7

    https://doi.org/10.1063/5.0010886 [Citations: 42]
  10. Modeling subgrid-scale forces by spatial artificial neural networks in large eddy simulation of turbulence

    Xie, Chenyue | Wang, Jianchun | E, Weinan

    Physical Review Fluids, Vol. 5 (2020), Iss. 5

    https://doi.org/10.1103/PhysRevFluids.5.054606 [Citations: 82]
  11. Dynamically learning the parameters of a chaotic system using partial observations

    Carlson, Elizabeth | Hudson, Joshua | Larios, Adam | Martinez, Vincent R. | Ng, Eunice | Whitehead, Jared P.

    Discrete and Continuous Dynamical Systems, Vol. 42 (2022), Iss. 8 P.3809

    https://doi.org/10.3934/dcds.2022033 [Citations: 14]
  12. Computational Science – ICCS 2021

    Data-Driven Deep Learning Emulators for Geophysical Forecasting

    Sastry, Varuni Katti | Maulik, Romit | Rao, Vishwas | Lusch, Bethany | Renganathan, S. Ashwin | Kotamarthi, Rao

    2021

    https://doi.org/10.1007/978-3-030-77977-1_35 [Citations: 1]
  13. Overview Frequency Principle/Spectral Bias in Deep Learning

    Xu, Zhi-Qin John | Zhang, Yaoyu | Luo, Tao

    Communications on Applied Mathematics and Computation, Vol. (2024), Iss.

    https://doi.org/10.1007/s42967-024-00398-7 [Citations: 2]
  14. Machine Learning in Modeling and Simulation

    Reduced Order Modeling

    Dar, Zulkeefal | Baiges, Joan | Codina, Ramon

    2023

    https://doi.org/10.1007/978-3-031-36644-4_8 [Citations: 1]
  15. Toward Model Reduction for Power System Transients With Physics-Informed PDE

    Pagnier, Laurent | Fritzsch, Julian | Jacquod, Philippe | Chertkov, Michael

    IEEE Access, Vol. 10 (2022), Iss. P.65118

    https://doi.org/10.1109/ACCESS.2022.3183336 [Citations: 7]
  16. OnsagerNet: Learning stable and interpretable dynamics using a generalized Onsager principle

    Yu, Haijun | Tian, Xinyuan | E, Weinan | Li, Qianxiao

    Physical Review Fluids, Vol. 6 (2021), Iss. 11

    https://doi.org/10.1103/PhysRevFluids.6.114402 [Citations: 25]
  17. Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism

    Wang, Qian | Ripamonti, Nicolò | Hesthaven, Jan S.

    Journal of Computational Physics, Vol. 410 (2020), Iss. P.109402

    https://doi.org/10.1016/j.jcp.2020.109402 [Citations: 81]
  18. A framework for machine learning of model error in dynamical systems

    Levine, Matthew | Stuart, Andrew

    Communications of the American Mathematical Society, Vol. 2 (2022), Iss. 7 P.283

    https://doi.org/10.1090/cams/10 [Citations: 27]
  19. Discovery of interpretable structural model errors by combining Bayesian sparse regression and data assimilation: A chaotic Kuramoto–Sivashinsky test case

    Mojgani, Rambod | Chattopadhyay, Ashesh | Hassanzadeh, Pedram

    Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 32 (2022), Iss. 6

    https://doi.org/10.1063/5.0091282 [Citations: 11]
  20. Artificial neural network-based spatial gradient models for large-eddy simulation of turbulence

    Wang, Yunpeng | Yuan, Zelong | Xie, Chenyue | Wang, Jianchun

    AIP Advances, Vol. 11 (2021), Iss. 5

    https://doi.org/10.1063/5.0053590 [Citations: 30]
  21. Uniformly accurate machine learning-based hydrodynamic models for kinetic equations

    Han, Jiequn | Ma, Chao | Ma, Zheng | E, Weinan

    Proceedings of the National Academy of Sciences, Vol. 116 (2019), Iss. 44 P.21983

    https://doi.org/10.1073/pnas.1909854116 [Citations: 57]
  22. Error bounds of the invariant statistics in machine learning of ergodic Itô diffusions

    Zhang, He | Harlim, John | Li, Xiantao

    Physica D: Nonlinear Phenomena, Vol. 427 (2021), Iss. P.133022

    https://doi.org/10.1016/j.physd.2021.133022 [Citations: 4]
  23. Dynamic iterative approximate deconvolution models for large-eddy simulation of turbulence

    Yuan, Zelong | Wang, Yunpeng | Xie, Chenyue | Wang, Jianchun

    Physics of Fluids, Vol. 33 (2021), Iss. 8

    https://doi.org/10.1063/5.0059643 [Citations: 22]
  24. Data-driven construction of stochastic reduced dynamics encoded with non-Markovian features

    She, Zhiyuan | Ge, Pei | Lei, Huan

    The Journal of Chemical Physics, Vol. 158 (2023), Iss. 3

    https://doi.org/10.1063/5.0130033 [Citations: 11]
  25. Spatial artificial neural network model for subgrid-scale stress and heat flux of compressible turbulence

    Xie, Chenyue | Wang, Jianchun | Li, Hui | Wan, Minping | Chen, Shiyi

    Theoretical and Applied Mechanics Letters, Vol. 10 (2020), Iss. 1 P.27

    https://doi.org/10.1016/j.taml.2020.01.006 [Citations: 24]
  26. The effect of sub-filter scale dynamics in large eddy simulation of turbulence

    Chang, Ning | Yuan, Zelong | Wang, Jianchun

    Physics of Fluids, Vol. 34 (2022), Iss. 9

    https://doi.org/10.1063/5.0098925 [Citations: 6]
  27. Kernel-based prediction of non-Markovian time series

    Gilani, Faheem | Giannakis, Dimitrios | Harlim, John

    Physica D: Nonlinear Phenomena, Vol. 418 (2021), Iss. P.132829

    https://doi.org/10.1016/j.physd.2020.132829 [Citations: 20]
  28. Deconvolutional artificial neural network models for large eddy simulation of turbulence

    Yuan, Zelong | Xie, Chenyue | Wang, Jianchun

    Physics of Fluids, Vol. 32 (2020), Iss. 11

    https://doi.org/10.1063/5.0027146 [Citations: 65]
  29. Extracting Interpretable Physical Parameters from Spatiotemporal Systems Using Unsupervised Learning

    Lu, Peter Y. | Kim, Samuel | Soljačić, Marin

    Physical Review X, Vol. 10 (2020), Iss. 3

    https://doi.org/10.1103/PhysRevX.10.031056 [Citations: 26]
  30. Artificial neural network mixed model for large eddy simulation of compressible isotropic turbulence

    Xie, Chenyue | Wang, Jianchun | Li, Hui | Wan, Minping | Chen, Shiyi

    Physics of Fluids, Vol. 31 (2019), Iss. 8

    https://doi.org/10.1063/1.5110788 [Citations: 71]
  31. Artificial neural network approach to large-eddy simulation of compressible isotropic turbulence

    Xie, Chenyue | Wang, Jianchun | Li, Ke | Ma, Chao

    Physical Review E, Vol. 99 (2019), Iss. 5

    https://doi.org/10.1103/PhysRevE.99.053113 [Citations: 54]
  32. Scientific machine learning for closure models in multiscale problems: a review

    Sanderse, Benjamin | Stinis, Panos | Maulik, Romit | Ahmed, Shady E.

    Foundations of Data Science, Vol. 0 (2024), Iss. 0 P.0

    https://doi.org/10.3934/fods.2024043 [Citations: 2]
  33. Data-driven discovery of quasiperiodically driven dynamics

    Das, Suddhasattwa | Mustavee, Shakib | Agarwal, Shaurya

    Nonlinear Dynamics, Vol. (2024), Iss.

    https://doi.org/10.1007/s11071-024-09970-5 [Citations: 0]
  34. Shock trace prediction by reduced models for a viscous stochastic Burgers equation

    Chen, Nan | Liu, Honghu | Lu, Fei

    Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 32 (2022), Iss. 4

    https://doi.org/10.1063/5.0084955 [Citations: 3]
  35. Spatially multi-scale artificial neural network model for large eddy simulation of compressible isotropic turbulence

    Xie, Chenyue | Wang, Jianchun | Li, Hui | Wan, Minping | Chen, Shiyi

    AIP Advances, Vol. 10 (2020), Iss. 1

    https://doi.org/10.1063/1.5138681 [Citations: 27]
  36. Simultaneous neural network approximation for smooth functions

    Hon, Sean | Yang, Haizhao

    Neural Networks, Vol. 154 (2022), Iss. P.152

    https://doi.org/10.1016/j.neunet.2022.06.040 [Citations: 7]
  37. Analytic continuation of noisy data using Adams Bashforth residual neural network

    Xie, Xuping | Bao, Feng | Maier, Thomas | Webster, Clayton

    Discrete & Continuous Dynamical Systems - S, Vol. 15 (2022), Iss. 4 P.877

    https://doi.org/10.3934/dcdss.2021088 [Citations: 3]
  38. Theoretical tools for understanding the climate crisis from Hasselmann’s programme and beyond

    Lucarini, Valerio | Chekroun, Mickaël D.

    Nature Reviews Physics, Vol. 5 (2023), Iss. 12 P.744

    https://doi.org/10.1038/s42254-023-00650-8 [Citations: 8]
  39. Deep Neural Network Model for Approximating Eigenmodes Localized by a Confining Potential

    Grubišić, Luka | Hajba, Marko | Lacmanović, Domagoj

    Entropy, Vol. 23 (2021), Iss. 1 P.95

    https://doi.org/10.3390/e23010095 [Citations: 9]
  40. A Dynamical Systems Approach to Machine Learning

    Pourmohammad Azizi, S. | Neisy, Abdolsadeh | Ahmad Waloo, Sajad

    International Journal of Computational Methods, Vol. 20 (2023), Iss. 09

    https://doi.org/10.1142/S021987622350007X [Citations: 1]