Physics-Informed Active Learning With Simultaneous Weak-Form Latent Space Dynamics Identification

He, Xiaolong; Tran, April; Bortz, David M.; Choi, Youngsoo

doi:10.1002/nme.7634

Physics-Informed Active Learning With Simultaneous Weak-Form Latent Space Dynamics Identification

Journal Article · Sun Dec 08 23:00:00 EST 2024 · International Journal for Numerical Methods in Engineering

DOI:https://doi.org/10.1002/nme.7634· OSTI ID:2533635

^[1]; Tran, April ^[2]; ^[2]; ^[3]

ANSYS Inc., Livermore, CA (United States)
Univ. of Colorado, Boulder, CO (United States)
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)

The parametric greedy latent space dynamics identification (gLaSDI) framework has demonstrated promising potential for accurate and efficient modeling of high-dimensional nonlinear physical systems. However, it remains challenging to handle noisy data. Here, to enhance robustness against noise, we incorporate the weak-form estimation of nonlinear dynamics (WENDy) into gLaSDI. In the proposed weak-form gLaSDI (WgLaSDI) framework, an autoencoder and WENDy are trained simultaneously to discover intrinsic nonlinear latent-space dynamics of high-dimensional data. Compared with the standard sparse identification of nonlinear dynamics (SINDy) employed in gLaSDI, WENDy enables variance reduction and robust latent space discovery, therefore leading to more accurate and efficient reduced-order modeling. Furthermore, the greedy physics-informed active learning in WgLaSDI enables adaptive sampling of optimal training data on the fly for enhanced modeling accuracy. The effectiveness of the proposed framework is demonstrated by modeling various nonlinear dynamical problems, including viscous and inviscid Burgers' equations, time-dependent radial advection, and the Vlasov equation for plasma physics. With data that contains 5%–10% Gaussian white noise, WgLaSDI outperforms gLaSDI by orders of magnitude, achieving 1%–7% relative errors. Compared with the high-fidelity models, WgLaSDI achieves 121 to 1779x speed-up.

View Accepted Manuscript (DOE)

Research Organization:: Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)

Sponsoring Organization:: USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: AC52-07NA27344; SC0023164; SC0023346

OSTI ID:: 2533635

Report Number(s):: LLNL--JRNL-865859; 1100236

Journal Information:: International Journal for Numerical Methods in Engineering, Journal Name: International Journal for Numerical Methods in Engineering Journal Issue: 1 Vol. 126; ISSN 0029-5981

Publisher:: WileyCopyright Statement

Country of Publication:: United States

Language:: English

References (48)

Model order reduction for meshfree solution of Poisson singularity problems Chen, Jiun‐Shyan; Marodon, Camille; Hu, Hsin‐Yun International Journal for Numerical Methods in Engineering, Vol. 102, Issue 5 https://doi.org/10.1002/nme.4743	journal	August 2014
Improving the efficiency of large scale topology optimization through on-the-fly reduced order model construction: EFFICIENT TOPOLOGY OPTIMIZATION THROUGH ON-THE-FLY ROM CONSTRUCTION Gogu, Christian International Journal for Numerical Methods in Engineering, Vol. 101, Issue 4 https://doi.org/10.1002/nme.4797	journal	October 2014
Variational multiscale proper orthogonal decomposition: Navier-stokes equations: VARIATIONAL MULTISCALE POD Iliescu, Traian; Wang, Zhu Numerical Methods for Partial Differential Equations, Vol. 30, Issue 2 https://doi.org/10.1002/num.21835	journal	December 2013
Efficient Implementation of Weighted ENO Schemes Jiang, Guang-Shan; Shu, Chi-Wang Journal of Computational Physics, Vol. 126, Issue 1 https://doi.org/10.1006/jcph.1996.0130	journal	June 1996
A decomposed subspace reduction for fracture mechanics based on the meshfree integrated singular basis function method He, Qizhi; Chen, Jiun-Shyan; Marodon, Camille Computational Mechanics, Vol. 63, Issue 3 https://doi.org/10.1007/s00466-018-1611-8	journal	August 2018
Direct Estimation of Parameters in ODE Models Using WENDy: Weak-Form Estimation of Nonlinear Dynamics Bortz, David M.; Messenger, Daniel A.; Dukic, Vanja Bulletin of Mathematical Biology, Vol. 85, Issue 11 https://doi.org/10.1007/s11538-023-01208-6	journal	October 2023
Non-intrusive reduced order modeling of natural convection in porous media using convolutional autoencoders: Comparison with linear subspace techniques Kadeethum, T.; Ballarin, F.; Choi, Y. Advances in Water Resources, Vol. 160 https://doi.org/10.1016/j.advwatres.2021.104098	journal	February 2022
Continuous conditional generative adversarial networks for data-driven solutions of poroelasticity with heterogeneous material properties Kadeethum, T.; O’Malley, D.; Choi, Y. Computers & Geosciences, Vol. 167 https://doi.org/10.1016/j.cageo.2022.105212	journal	October 2022
MFEM: A modular finite element methods library Anderson, Robert; Andrej, Julian; Barker, Andrew Computers & Mathematics with Applications, Vol. 81 https://doi.org/10.1016/j.camwa.2020.06.009	journal	January 2021
Data-driven operator inference for nonintrusive projection-based model reduction Peherstorfer, Benjamin; Willcox, Karen Computer Methods in Applied Mechanics and Engineering, Vol. 306 https://doi.org/10.1016/j.cma.2016.03.025	journal	July 2016
Operator inference for non-intrusive model reduction of systems with non-polynomial nonlinear terms Benner, Peter; Goyal, Pawan; Kramer, Boris Computer Methods in Applied Mechanics and Engineering, Vol. 372 https://doi.org/10.1016/j.cma.2020.113433	journal	December 2020
Component-wise reduced order model lattice-type structure design McBane, Sean; Choi, Youngsoo Computer Methods in Applied Mechanics and Engineering, Vol. 381 https://doi.org/10.1016/j.cma.2021.113813	journal	August 2021
Reduced order models for Lagrangian hydrodynamics Copeland, Dylan Matthew; Cheung, Siu Wun; Huynh, Kevin Computer Methods in Applied Mechanics and Engineering, Vol. 388 https://doi.org/10.1016/j.cma.2021.114259	journal	January 2022
LaSDI: Parametric Latent Space Dynamics Identification Fries, William D.; He, Xiaolong; Choi, Youngsoo Computer Methods in Applied Mechanics and Engineering, Vol. 399 https://doi.org/10.1016/j.cma.2022.115436	journal	September 2022
Stress-constrained topology optimization of lattice-like structures using component-wise reduced order models McBane, Sean; Choi, Youngsoo; Willcox, Karen Computer Methods in Applied Mechanics and Engineering, Vol. 400 https://doi.org/10.1016/j.cma.2022.115525	journal	October 2022
GPLaSDI: Gaussian Process-based interpretable Latent Space Dynamics Identification through deep autoencoder Bonneville, Christophe; Choi, Youngsoo; Ghosh, Debojyoti Computer Methods in Applied Mechanics and Engineering, Vol. 418 https://doi.org/10.1016/j.cma.2023.116535	journal	January 2024
A fast and accurate domain decomposition nonlinear manifold reduced order model Diaz, Alejandro N.; Choi, Youngsoo; Heinkenschloss, Matthias Computer Methods in Applied Mechanics and Engineering, Vol. 425 https://doi.org/10.1016/j.cma.2024.116943	journal	May 2024
Weak-form latent space dynamics identification Tran, April; He, Xiaolong; Messenger, Daniel A. Computer Methods in Applied Mechanics and Engineering, Vol. 427 https://doi.org/10.1016/j.cma.2024.116998	journal	July 2024
Train small, model big: Scalable physics simulators via reduced order modeling and domain decomposition Chung, Seung Whan; Choi, Youngsoo; Roy, Pratanu Computer Methods in Applied Mechanics and Engineering, Vol. 427 https://doi.org/10.1016/j.cma.2024.117041	journal	July 2024
tLaSDI: Thermodynamics-informed latent space dynamics identification Park, Jun Sur Richard; Cheung, Siu Wun; Choi, Youngsoo Computer Methods in Applied Mechanics and Engineering, Vol. 429 https://doi.org/10.1016/j.cma.2024.117144	journal	September 2024
Finite volume POD-Galerkin stabilised reduced order methods for the parametrised incompressible Navier–Stokes equations Stabile, Giovanni; Rozza, Gianluigi Computers & Fluids, Vol. 173 https://doi.org/10.1016/j.compfluid.2018.01.035	journal	September 2018
Gradient-based constrained optimization using a database of linear reduced-order models Choi, Youngsoo; Boncoraglio, Gabriele; Anderson, Spenser Journal of Computational Physics, Vol. 423 https://doi.org/10.1016/j.jcp.2020.109787	journal	December 2020
Weak SINDy for partial differential equations Messenger, Daniel A.; Bortz, David M. Journal of Computational Physics, Vol. 443 https://doi.org/10.1016/j.jcp.2021.110525	journal	October 2021
A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder Kim, Youngkyu; Choi, Youngsoo; Widemann, David Journal of Computational Physics, Vol. 451 https://doi.org/10.1016/j.jcp.2021.110841	journal	February 2022
Local Lagrangian reduced-order modeling for the Rayleigh-Taylor instability by solution manifold decomposition Cheung, Siu Wun; Choi, Youngsoo; Copeland, Dylan Matthew Journal of Computational Physics, Vol. 472 https://doi.org/10.1016/j.jcp.2022.111655	journal	January 2023
Predicting solar wind streams from the inner-heliosphere to Earth via shifted operator inference Issan, Opal; Kramer, Boris Journal of Computational Physics, Vol. 473 https://doi.org/10.1016/j.jcp.2022.111689	journal	January 2023
gLaSDI: Parametric physics-informed greedy latent space dynamics identification He, Xiaolong; Choi, Youngsoo; Fries, William D. Journal of Computational Physics, Vol. 489 https://doi.org/10.1016/j.jcp.2023.112267	journal	September 2023
Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems Qian, Elizabeth; Kramer, Boris; Peherstorfer, Benjamin Physica D: Nonlinear Phenomena, Vol. 406 https://doi.org/10.1016/j.physd.2020.132401	journal	May 2020
Learning mean-field equations from particle data using WSINDy Messenger, Daniel A.; Bortz, David M. Physica D: Nonlinear Phenomena, Vol. 439 https://doi.org/10.1016/j.physd.2022.133406	journal	November 2022
Computational Physics Thijssen, Jos Cambridge University Press https://doi.org/10.1017/CBO9781139171397	book	January 2007
Deep learning in fluid dynamics Kutz, J. Nathan Journal of Fluid Mechanics, Vol. 814 https://doi.org/10.1017/jfm.2016.803	journal	January 2017
The rise of computational biology Noble, Denis Nature Reviews Molecular Cell Biology, Vol. 3, Issue 6 https://doi.org/10.1038/nrm810	journal	June 2002
Reduced order modeling for flow and transport problems with Barlow Twins self-supervised learning Kadeethum, Teeratorn; Ballarin, Francesco; O’Malley, Daniel Scientific Reports, Vol. 12, Issue 1 https://doi.org/10.1038/s41598-022-24545-3	journal	November 2022
A framework for data-driven solution and parameter estimation of PDEs using conditional generative adversarial networks Kadeethum, Teeratorn; O’Malley, Daniel; Fuhg, Jan Niklas Nature Computational Science, Vol. 1, Issue 12 https://doi.org/10.1038/s43588-021-00171-3	journal	December 2021
Discovering governing equations from data by sparse identification of nonlinear dynamical systems Brunton, Steven L.; Proctor, Joshua L.; Kutz, J. Nathan Proceedings of the National Academy of Sciences, Vol. 113, Issue 15 https://doi.org/10.1073/pnas.1517384113	journal	March 2016
Data-driven discovery of coordinates and governing equations Champion, Kathleen; Lusch, Bethany; Kutz, J. Nathan Proceedings of the National Academy of Sciences, Vol. 116, Issue 45 https://doi.org/10.1073/pnas.1906995116	journal	October 2019
Learning anisotropic interaction rules from individual trajectories in a heterogeneous cellular population Messenger, Daniel A.; Wheeler, Graycen E.; Liu, Xuedong Journal of The Royal Society Interface, Vol. 19, Issue 195 https://doi.org/10.1098/rsif.2022.0412	journal	October 2022
Weak-form inference for hybrid dynamical systems in ecology Messenger, Daniel; Dwyer, Greg; Dukic, Vanja Journal of The Royal Society Interface, Vol. 21, Issue 221 https://doi.org/10.1098/rsif.2024.0376	journal	December 2024
CaloGAN: Simulating 3D high energy particle showers in multilayer electromagnetic calorimeters with generative adversarial networks Paganini, Michela; de Oliveira, Luke; Nachman, Benjamin Physical Review D, Vol. 97, Issue 1 https://doi.org/10.1103/PhysRevD.97.014021	journal	January 2018
A Schur method for balanced-truncation model reduction Safonov, M. G.; Chiang, R. Y. IEEE Transactions on Automatic Control, Vol. 34, Issue 7 https://doi.org/10.1109/9.29399	journal	July 1989
Simulation of a Car crash using ANSYS Muhammad, Aisha; Shanono, Ibrahim Haruna 2019 15th International Conference on Electronics, Computer and Computation (ICECCO) https://doi.org/10.1109/ICECCO48375.2019.9043275	conference	December 2019
Deep Fluids: A Generative Network for Parameterized Fluid Simulations Kim, Byungsoo; Azevedo, Vinicius C.; Thuerey, Nils Computer Graphics Forum, Vol. 38, Issue 2 https://doi.org/10.1111/cgf.13619	journal	May 2019
Reducing the Dimensionality of Data with Neural Networks Hinton, G. E. Science, Vol. 313, Issue 5786 https://doi.org/10.1126/science.1127647	journal	July 2006
Distilling Free-Form Natural Laws from Experimental Data Schmidt, Michael; Lipson, Hod Science, Vol. 324, Issue 5923 https://doi.org/10.1126/science.1165893	journal	April 2009
Weak SINDy: Galerkin-Based Data-Driven Model Selection Messenger, Daniel A.; Bortz, David M. Multiscale Modeling & Simulation, Vol. 19, Issue 3 https://doi.org/10.1137/20M1343166	journal	January 2021
S-OPT: A Points Selection Algorithm for Hyper-Reduction in Reduced Order Models Lauzon, Jessica T.; Cheung, Siu Wun; Shin, Yeonjong SIAM Journal on Scientific Computing, Vol. 46, Issue 4 https://doi.org/10.1137/22M1484018	journal	July 2024
The Proper Orthogonal Decomposition in the Analysis of Turbulent Flows Berkooz, G.; Holmes, P.; Lumley, J. L. Annual Review of Fluid Mechanics, Vol. 25, Issue 1 https://doi.org/10.1146/annurev.fl.25.010193.002543	journal	January 1993
Advanced Modeling and Simulation of Vehicle Active Aerodynamic Safety Kurec, Krzysztof; Remer, Michał; Broniszewski, Jakub Journal of Advanced Transportation, Vol. 2019 https://doi.org/10.1155/2019/7308590	journal	February 2019

Similar Records

gLaSDI: Parametric physics-informed greedy latent space dynamics identification

Journal Article · Mon Jun 05 00:00:00 EDT 2023 · Journal of Computational Physics · OSTI ID:1999976

Weak-Form Latent Space Dynamics Identification

Software · Thu Jul 04 20:00:00 EDT 2024 · OSTI ID:code-142351

Related Subjects

97 MATHEMATICS AND COMPUTING
autoencoder
data-driven modeling
latent space dynamics learning
physics-informed active learning
reduced-order modeling
weak form

Physics-Informed Active Learning With Simultaneous Weak-Form Latent Space Dynamics Identification

Citation Formats

References (48)

Similar Records

Related Subjects