Loading [MathJax]/jax/output/HTML-CSS/config.js
Go to previous page

DL-PDE: Deep-Learning Based Data-Driven Discovery of Partial Differential Equations from Discrete and Noisy Data

DL-PDE: Deep-Learning Based Data-Driven Discovery of Partial Differential Equations from Discrete and Noisy Data

Year:    2021

Author:    Hao Xu, Haibin Chang, Dongxiao Zhang

Communications in Computational Physics, Vol. 29 (2021), Iss. 3 : pp. 698–728

Abstract

In recent years, data-driven methods have been developed to learn dynamical systems and partial differential equations (PDE). The goal of such work is to discover unknown physics and corresponding equations. However, prior to achieving this goal, major challenges remain to be resolved, including learning PDE under noisy data and limited discrete data. To overcome these challenges, in this work, a deep-learning based data-driven method, called DL-PDE, is developed to discover the governing PDEs of underlying physical processes. The DL-PDE method combines deep learning via neural networks and data-driven discovery of PDE via sparse regressions. In the DL-PDE, a neural network is first trained, then a large amount of meta-data is generated, and the required derivatives are calculated by automatic differentiation. Finally, the form of PDE is discovered by sparse regression. The proposed method is tested with physical processes, governed by the diffusion equation, the convection-diffusion equation, the Burgers equation, and the Korteweg-de Vries (KdV) equation, for proof-of-concept and applications in real-world engineering settings. The proposed method achieves satisfactory results when data are noisy and limited.

You do not have full access to this article.

Already a Subscriber? Sign in as an individual or via your institution

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/cicp.OA-2020-0142

Communications in Computational Physics, Vol. 29 (2021), Iss. 3 : pp. 698–728

Published online:    2021-01

AMS Subject Headings:    Global Science Press

Copyright:    COPYRIGHT: © Global Science Press

Pages:    31

Keywords:    Data-driven discovery machine learning deep neural network sparse regression noisy data.

Author Details

Hao Xu Email

Haibin Chang Email

Dongxiao Zhang Email

  1. Physics-informed neural networks for modeling astrophysical shocks

    Moschou, S P | Hicks, E | Parekh, R Y | Mathew, D | Majumdar, S | Vlahakis, N

    Machine Learning: Science and Technology, Vol. 4 (2023), Iss. 3 P.035032

    https://doi.org/10.1088/2632-2153/acf116 [Citations: 4]
  2. Discovering an interpretable mathematical expression for a full wind-turbine wake with artificial intelligence enhanced symbolic regression

    Wang, Ding | Chen, Yuntian | Chen, Shiyi

    Physics of Fluids, Vol. 36 (2024), Iss. 10

    https://doi.org/10.1063/5.0221611 [Citations: 0]
  3. Bayesian deep learning for partial differential equation parameter discovery with sparse and noisy data

    Bonneville, Christophe | Earls, Christopher

    Journal of Computational Physics: X, Vol. 16 (2022), Iss. P.100115

    https://doi.org/10.1016/j.jcpx.2022.100115 [Citations: 1]
  4. DF-ParPINN: parallel PINN based on velocity potential field division and single time slice focus

    Chen, Jingjian | Yuan, Chunxin | Xu, Jiali | Bie, Pengfei | Wei, Zhiqiang

    Frontiers in Marine Science, Vol. 11 (2024), Iss.

    https://doi.org/10.3389/fmars.2024.1309775 [Citations: 0]
  5. Spatial acoustic properties recovery with deep learning

    Liu, Ruixian | Gerstoft, Peter

    The Journal of the Acoustical Society of America, Vol. 155 (2024), Iss. 6 P.3690

    https://doi.org/10.1121/10.0026231 [Citations: 0]
  6. Learning unbounded-domain spatiotemporal differential equations using adaptive spectral methods

    Xia, Mingtao | Li, Xiangting | Shen, Qijing | Chou, Tom

    Journal of Applied Mathematics and Computing, Vol. 70 (2024), Iss. 5 P.4395

    https://doi.org/10.1007/s12190-024-02131-2 [Citations: 1]
  7. SD-PINN: Physics Informed Neural Networks for Spatially Dependent PDES

    Liu, Ruixian | Gerstoft, Peter

    ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), (2023), P.1

    https://doi.org/10.1109/ICASSP49357.2023.10095076 [Citations: 1]
  8. A hybrid modelling approach to model process dynamics by the discovery of a system of partial differential equations

    Raviprakash, Kiran | Huang, Biao | Prasad, Vinay

    Computers & Chemical Engineering, Vol. 164 (2022), Iss. P.107862

    https://doi.org/10.1016/j.compchemeng.2022.107862 [Citations: 7]
  9. Robust discovery of partial differential equations in complex situations

    Xu, Hao | Zhang, Dongxiao

    Physical Review Research, Vol. 3 (2021), Iss. 3

    https://doi.org/10.1103/PhysRevResearch.3.033270 [Citations: 18]
  10. Physics-constrained robust learning of open-form partial differential equations from limited and noisy data

    Du, Mengge | Chen, Yuntian | Nie, Longfeng | Lou, Siyu | Zhang, Dongxiao

    Physics of Fluids, Vol. 36 (2024), Iss. 5

    https://doi.org/10.1063/5.0204187 [Citations: 2]
  11. A Review of Data‐Driven Discovery for Dynamic Systems

    North, Joshua S. | Wikle, Christopher K. | Schliep, Erin M.

    International Statistical Review, Vol. 91 (2023), Iss. 3 P.464

    https://doi.org/10.1111/insr.12554 [Citations: 7]
  12. Compressible Non-Newtonian Fluid Based Road Traffic Flow Equation Solved by Physical-Informed Rational Neural Network

    Yang, Zan | Li, Dan | Nai, Wei | Liu, Lu | Sun, Jingjing | Lv, Xiaowei

    IEEE Access, Vol. 12 (2024), Iss. P.12992

    https://doi.org/10.1109/ACCESS.2024.3356173 [Citations: 1]
  13. DISCOVER: Deep identification of symbolically concise open-form partial differential equations via enhanced reinforcement learning

    Du, Mengge | Chen, Yuntian | Zhang, Dongxiao

    Physical Review Research, Vol. 6 (2024), Iss. 1

    https://doi.org/10.1103/PhysRevResearch.6.013182 [Citations: 4]
  14. A Bayesian Approach for Spatio-Temporal Data-Driven Dynamic Equation Discovery

    North, Joshua S. | Wikle, Christopher K. | Schliep, Erin M.

    Bayesian Analysis, Vol. -1 (2023), Iss. -1

    https://doi.org/10.1214/23-BA1406 [Citations: 2]
  15. The data-driven discovery of partial differential equations by symbolic genetic algorithm

    Sun, Shifei | Tian, Shifang | Wang, Yuduo | Li, Biao

    Nonlinear Dynamics, Vol. 112 (2024), Iss. 22 P.19871

    https://doi.org/10.1007/s11071-024-10093-0 [Citations: 0]
  16. Bayesian modeling of pattern formation from one snapshot of pattern

    Yoshinaga, Natsuhiko | Tokuda, Satoru

    Physical Review E, Vol. 106 (2022), Iss. 6

    https://doi.org/10.1103/PhysRevE.106.065301 [Citations: 4]
  17. Reconstruction of ship propeller wake field based on self-adaptive loss balanced physics-informed neural networks

    Hou, Xianrui | Zhou, Xingyu | Liu, Yi

    Ocean Engineering, Vol. 309 (2024), Iss. P.118341

    https://doi.org/10.1016/j.oceaneng.2024.118341 [Citations: 0]
  18. Learning chaotic systems from noisy data via multi-step optimization and adaptive training

    Zhang, Lei | Tang, Shaoqiang | He, Guowei

    Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 32 (2022), Iss. 12

    https://doi.org/10.1063/5.0114542 [Citations: 2]
  19. PDE-READ: Human-readable partial differential equation discovery using deep learning

    Stephany, Robert | Earls, Christopher

    Neural Networks, Vol. 154 (2022), Iss. P.360

    https://doi.org/10.1016/j.neunet.2022.07.008 [Citations: 17]
  20. Entropy structure informed learning for solving inverse problems of differential equations

    Jiang, Yan | Yang, Wuyue | Zhu, Yi | Hong, Liu

    Chaos, Solitons & Fractals, Vol. 175 (2023), Iss. P.114057

    https://doi.org/10.1016/j.chaos.2023.114057 [Citations: 1]
  21. Deep-learning based discovery of partial differential equations in integral form from sparse and noisy data

    Xu, Hao | Zhang, Dongxiao | Wang, Nanzhe

    Journal of Computational Physics, Vol. 445 (2021), Iss. P.110592

    https://doi.org/10.1016/j.jcp.2021.110592 [Citations: 18]
  22. Numerical solutions of boundary problems in partial differential equations: A deep learning framework with Green's function

    Dai, Yuanjun | Li, Zhi | An, Yiran | Deng, Wanru

    Journal of Computational Physics, Vol. 511 (2024), Iss. P.113121

    https://doi.org/10.1016/j.jcp.2024.113121 [Citations: 0]
  23. Discovery of Partial Differential Equations from Highly Noisy and Sparse Data with Physics-Informed Information Criterion

    Xu, Hao | Zeng, Junsheng | Zhang, Dongxiao

    Research, Vol. 6 (2023), Iss.

    https://doi.org/10.34133/research.0147 [Citations: 14]
  24. Symbolic genetic algorithm for discovering open-form partial differential equations (SGA-PDE)

    Chen, Yuntian | Luo, Yingtao | Liu, Qiang | Xu, Hao | Zhang, Dongxiao

    Physical Review Research, Vol. 4 (2022), Iss. 2

    https://doi.org/10.1103/PhysRevResearch.4.023174 [Citations: 40]