Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems
Abstract
In this work, we present Lift & Learn, a physics-informed method for learning low-dimensional models for large-scale dynamical systems. The method exploits knowledge of a system’s governing equations to identify a coordinate transformation in which the system dynamics have quadratic structure. This transformation is called a lifting map because it often adds auxiliary variables to the system state. The lifting map is applied to data obtained by evaluating a model for the original nonlinear system. This lifted data is projected onto its leading principal components, and low-dimensional linear and quadratic matrix operators are fit to the lifted reduced data using a least-squares operator inference procedure. Analysis of our method shows that the Lift & Learn models are able to capture the system physics in the lifted coordinates at least as accurately as traditional intrusive model reduction approaches. This preservation of system physics makes the Lift & Learn models robust to changes in inputs. Numerical experiments on the FitzHugh–Nagumo neuron activation model and the compressible Euler equations demonstrate the generalizability of our model.
- Authors:
-
- Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
- Univ. of California, San Diego, CA (United States)
- New York Univ. (NYU), NY (United States). Courant Institute of Mathematical Sciences
- Univ. of Texas, Austin, TX (United States). Oden Institute for Computational Engineering and Sciences
- Publication Date:
- Research Org.:
- Univ. of Texas, Austin, TX (United States); New York Univ. (NYU), NY (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC); US Air Force Office of Scientific Research (AFOSR)
- OSTI Identifier:
- 1803677
- Alternate Identifier(s):
- OSTI ID: 1603700
- Grant/Contract Number:
- SC0019303; SC0019334; FA9550-17-1-0195; FA9550-15-1-0038; FA9550-18-1-0023
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Physica. D, Nonlinear Phenomena
- Additional Journal Information:
- Journal Volume: 406; Journal Issue: C; Journal ID: ISSN 0167-2789
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; Data-driven model reduction; Scientific machine learning; Dynamical systems; Partial differential equations; Lifting map
Citation Formats
Qian, Elizabeth, Kramer, Boris, Peherstorfer, Benjamin, and Willcox, Karen. Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems. United States: N. p., 2020.
Web. doi:10.1016/j.physd.2020.132401.
Qian, Elizabeth, Kramer, Boris, Peherstorfer, Benjamin, & Willcox, Karen. Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems. United States. https://doi.org/10.1016/j.physd.2020.132401
Qian, Elizabeth, Kramer, Boris, Peherstorfer, Benjamin, and Willcox, Karen. Wed .
"Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems". United States. https://doi.org/10.1016/j.physd.2020.132401. https://www.osti.gov/servlets/purl/1803677.
@article{osti_1803677,
title = {Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems},
author = {Qian, Elizabeth and Kramer, Boris and Peherstorfer, Benjamin and Willcox, Karen},
abstractNote = {In this work, we present Lift & Learn, a physics-informed method for learning low-dimensional models for large-scale dynamical systems. The method exploits knowledge of a system’s governing equations to identify a coordinate transformation in which the system dynamics have quadratic structure. This transformation is called a lifting map because it often adds auxiliary variables to the system state. The lifting map is applied to data obtained by evaluating a model for the original nonlinear system. This lifted data is projected onto its leading principal components, and low-dimensional linear and quadratic matrix operators are fit to the lifted reduced data using a least-squares operator inference procedure. Analysis of our method shows that the Lift & Learn models are able to capture the system physics in the lifted coordinates at least as accurately as traditional intrusive model reduction approaches. This preservation of system physics makes the Lift & Learn models robust to changes in inputs. Numerical experiments on the FitzHugh–Nagumo neuron activation model and the compressible Euler equations demonstrate the generalizability of our model.},
doi = {10.1016/j.physd.2020.132401},
journal = {Physica. D, Nonlinear Phenomena},
number = C,
volume = 406,
place = {United States},
year = {Wed Feb 19 00:00:00 EST 2020},
month = {Wed Feb 19 00:00:00 EST 2020}
}
Web of Science
Works referenced in this record:
Dynamic mode decomposition of numerical and experimental data
journal, July 2010
- Schmid, Peter J.
- Journal of Fluid Mechanics, Vol. 656
A Data–Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition
journal, June 2015
- Williams, Matthew O.; Kevrekidis, Ioannis G.; Rowley, Clarence W.
- Journal of Nonlinear Science, Vol. 25, Issue 6
Aerodynamic Data Reconstruction and Inverse Design Using Proper Orthogonal Decomposition
journal, August 2004
- Bui-Thanh, T.; Damodaran, M.; Willcox, K.
- AIAA Journal, Vol. 42, Issue 8
Computability of global solutions to factorable nonconvex programs: Part I — Convex underestimating problems
journal, December 1976
- McCormick, Garth P.
- Mathematical Programming, Vol. 10, Issue 1
A new finite element formulation for computational fluid dynamics: I. Symmetric forms of the compressible Euler and Navier-Stokes equations and the second law of thermodynamics
journal, February 1986
- Hughes, T. J. R.; Franca, L. P.; Mallet, M.
- Computer Methods in Applied Mechanics and Engineering, Vol. 54, Issue 2
Data-driven operator inference for nonintrusive projection-based model reduction
journal, July 2016
- Peherstorfer, Benjamin; Willcox, Karen
- Computer Methods in Applied Mechanics and Engineering, Vol. 306
Impulses and Physiological States in Theoretical Models of Nerve Membrane
journal, July 1961
- FitzHugh, Richard
- Biophysical Journal, Vol. 1, Issue 6
Extracting Sparse High-Dimensional Dynamics from Limited Data
journal, January 2018
- Schaeffer, Hayden; Tran, Giang; Ward, Rachel
- SIAM Journal on Applied Mathematics, Vol. 78, Issue 6
Nonintrusive reduced-order modeling of parametrized time-dependent partial differential equations
journal, February 2013
- Audouze, Christophe; De Vuyst, Florian; Nair, Prasanth B.
- Numerical Methods for Partial Differential Equations, Vol. 29, Issue 5
Discovering governing equations from data by sparse identification of nonlinear dynamical systems
journal, March 2016
- Brunton, Steven L.; Proctor, Joshua L.; Kutz, J. Nathan
- Proceedings of the National Academy of Sciences, Vol. 113, Issue 15
Two-Sided Projection Methods for Nonlinear Model Order Reduction
journal, January 2015
- Benner, Peter; Breiten, Tobias
- SIAM Journal on Scientific Computing, Vol. 37, Issue 2
Data-driven discovery of partial differential equations
journal, April 2017
- Rudy, Samuel H.; Brunton, Steven L.; Proctor, Joshua L.
- Science Advances, Vol. 3, Issue 4
An Active Pulse Transmission Line Simulating Nerve Axon
journal, October 1962
- Nagumo, J.; Arimoto, S.; Yoshizawa, S.
- Proceedings of the IRE, Vol. 50, Issue 10
Data-driven model order reduction of quadratic-bilinear systems: The Loewner framework for QB systems
journal, July 2018
- Gosea, Ion Victor; Antoulas, Athanasios C.
- Numerical Linear Algebra with Applications, Vol. 25, Issue 6
Projection-based model reduction: Formulations for physics-based machine learning
journal, January 2019
- Swischuk, Renee; Mainini, Laura; Peherstorfer, Benjamin
- Computers & Fluids, Vol. 179
Dynamical Systems of Continuous Spectra
journal, March 1932
- Koopman, B. O.; Neumann, J. v.
- Proceedings of the National Academy of Sciences, Vol. 18, Issue 3
The Proper Orthogonal Decomposition in the Analysis of Turbulent Flows
journal, January 1993
- Berkooz, G.; Holmes, P.; Lumley, J. L.
- Annual Review of Fluid Mechanics, Vol. 25, Issue 1
Spectral analysis of nonlinear flows
journal, November 2009
- Rowley, Clarence W.; MeziĆ, Igor; Bagheri, Shervin
- Journal of Fluid Mechanics, Vol. 641
QLMOR: A Projection-Based Nonlinear Model Order Reduction Approach Using Quadratic-Linear Representation of Nonlinear Systems
journal, September 2011
- Gu, Chenjie
- IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Vol. 30, Issue 9
Extended dynamic mode decomposition with dictionary learning: A data-driven adaptive spectral decomposition of the Koopman operator
journal, October 2017
- Li, Qianxiao; Dietrich, Felix; Bollt, Erik M.
- Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 27, Issue 10
Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem
journal, May 2019
- Wang, Qian; Hesthaven, Jan S.; Ray, Deep
- Journal of Computational Physics, Vol. 384
Turbulence and the dynamics of coherent structures. I. Coherent structures
journal, January 1987
- Sirovich, Lawrence
- Quarterly of Applied Mathematics, Vol. 45, Issue 3
Fusing wind-tunnel measurements and CFD data using constrained gappy proper orthogonal decomposition
journal, March 2019
- Mifsud, Michael; Vendl, Alexander; Hansen, Lars-Uwe
- Aerospace Science and Technology, Vol. 86
Minimal subspace rotation on the Stiefel manifold for stabilization and enhancement of projection-based reduced order models for the compressible Navier–Stokes equations
journal, September 2016
- Balajewicz, Maciej; Tezaur, Irina; Dowell, Earl
- Journal of Computational Physics, Vol. 321
Data to decisions: Real-time structural assessment from sparse measurements affected by uncertainty
journal, April 2017
- Mainini, Laura; Willcox, Karen E.
- Computers & Structures, Vol. 182
Kernel Methods for the Approximation of Nonlinear Systems
journal, January 2017
- Bouvrie, Jake; Hamzi, Boumediene
- SIAM Journal on Control and Optimization, Vol. 55, Issue 4
Nonlinear Model Order Reduction via Lifting Transformations and Proper Orthogonal Decomposition
journal, June 2019
- Kramer, Boris; Willcox, Karen E.
- AIAA Journal, Vol. 57, Issue 6
Nonlinear Model Reduction via Discrete Empirical Interpolation
journal, January 2010
- Chaturantabut, Saifon; Sorensen, Danny C.
- SIAM Journal on Scientific Computing, Vol. 32, Issue 5
Data-driven model order reduction of quadratic-bilinear systems: The Loewner framework for QB systems
journal, July 2018
- Gosea, Ion Victor; Antoulas, Athanasios C.
- Numerical Linear Algebra with Applications, Vol. 25, Issue 6
Stochastic resonance in neuron models
journal, January 1993
- Longtin, Andr�
- Journal of Statistical Physics, Vol. 70, Issue 1-2
Fusing wind-tunnel measurements and CFD data using constrained gappy proper orthogonal decomposition
journal, March 2019
- Mifsud, Michael; Vendl, Alexander; Hansen, Lars-Uwe
- Aerospace Science and Technology, Vol. 86
Projection-based model reduction: Formulations for physics-based machine learning
journal, January 2019
- Swischuk, Renee; Mainini, Laura; Peherstorfer, Benjamin
- Computers & Fluids, Vol. 179
Data to decisions: Real-time structural assessment from sparse measurements affected by uncertainty
journal, April 2017
- Mainini, Laura; Willcox, Karen E.
- Computers & Structures, Vol. 182
Dynamical Systems of Continuous Spectra
journal, March 1932
- Koopman, B. O.; Neumann, J. v.
- Proceedings of the National Academy of Sciences, Vol. 18, Issue 3
Turbulence and the dynamics of coherent structures. III. Dynamics and scaling
journal, January 1987
- Sirovich, Lawrence
- Quarterly of Applied Mathematics, Vol. 45, Issue 3
Nonlinear Model Reduction via Discrete Empirical Interpolation
journal, January 2010
- Chaturantabut, Saifon; Sorensen, Danny C.
- SIAM Journal on Scientific Computing, Vol. 32, Issue 5