Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems
- 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
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.
- Research Organization:
- Univ. of Texas, Austin, TX (United States); New York Univ. (NYU), NY (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC); US Air Force Office of Scientific Research (AFOSR)
- Grant/Contract Number:
- SC0019303; SC0019334; FA9550-17-1-0195; FA9550-15-1-0038; FA9550-18-1-0023
- OSTI ID:
- 1803677
- Alternate ID(s):
- OSTI ID: 1603700
- Journal Information:
- Physica. D, Nonlinear Phenomena, Vol. 406, Issue C; ISSN 0167-2789
- Publisher:
- ElsevierCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Web of Science
Similar Records
Data-driven reduced-order models via regularised Operator Inference for a single-injector combustion process
Operator inference for non-intrusive model reduction with quadratic manifolds