Non-intrusive model reduction of large-scale, nonlinear dynamical systems using deep learning

Abstract

Projection-based model reduction has become a popular approach to reduce the cost associated with integrating large-scale dynamical systems so they can be used in many-query settings such as optimization and uncertainty quantification. For nonlinear systems, significant cost reduction is only possible with an additional layer of approximation to reduce the computational bottleneck of evaluating the projected nonlinear terms. Prevailing methods to approximate the nonlinear terms are code intrusive, potentially requiring years of development time to integrate into an existing codebase, and have been known to lack parametric robustness. This work develops a non-intrusive method to efficiently and accurately approximate the expensive nonlinear terms that arise in reduced nonlinear dynamical system using deep neural networks. The neural network is trained using only the simulation data used to construct the reduced basis and evaluations of the nonlinear terms at these snapshots. Once trained, the neural network-based reduced-order model only requires forward and backward propagation through the network to evaluate the nonlinear term and its derivative, which are used to integrate the reduced dynamical system at a new parameter configuration. We provide two numerical experiments---the dynamical systems result from the semi-discretization of parametrized, nonlinear, hyperbolic partial differential equations---that show, in addition to non-intrusivity, the proposed approach provides more stable and accurate approximations to each dynamical system across a large number of training and testing points than the popular empirical interpolation method.

arXiv comment18 pages, 7 figures

References in corpus (18)

• Adam: A Method for Stochastic Optimization
• The GNAT method for nonlinear model reduction: effective implementation and application to computational fluid dynamics and turbulent flows
• A Physics Informed Machine Learning Approach for Reconstructing Reynolds Stress Modeling Discrepancies Based on DNS Data
• Data-assisted reduced-order modeling of extreme events in complex dynamical systems
• A deep learning enabler for non-intrusive reduced order modeling of fluid flows
• An artificial neural network framework for reduced order modeling of transient flows
• Construction of Reduced Order Models for Fluid Flows Using Deep Feedforward Neural Networks
• A Deep Learning based Approach to Reduced Order Modeling for Turbulent Flow Control using LSTM Neural Networks
• Time-series learning of latent-space dynamics for reduced-order model closure
• Data-driven Discovery of Closure Models
• Physically-Constrained Data-Driven, Filtered Reduced Order Modeling of Fluid Flows
• DR-RNN: A deep residual recurrent neural network for model reduction
• Data driven approximation of parametrized PDEs by Reduced Basis and Neural Networks
• Machine learning action parameters in lattice quantum chromodynamics
• An Adjoint Method for a High-Order Discretization of Deforming Domain Conservation Laws for Optimization of Flow Problems
• Deep-learning-based reduced-order modeling for subsurface flow simulation
• A Fully Discrete Adjoint Method for Optimization of Flow Problems on Deforming Domains with Time-Periodicity Constraints
• Using recurrent neural networks for nonlinear component computation in advection-dominated reduced-order models

Cited by in corpus (16)

• Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels
• Transformers for Modeling Physical Systems
• On closures for reduced order models $-$ A spectrum of first-principle to machine-learned avenues
• The Random Feature Model for Input-Output Maps between Banach Spaces
• Operator inference for non-intrusive model reduction with quadratic manifolds
• Deep learning-based surrogate model for 3-D patient-specific computational fluid dynamics
• Multi-resolution partial differential equations preserved learning framework for spatiotemporal dynamics
• Gravitational-wave surrogate models powered by artificial neural networks: The ANN-Sur for waveform generation
• A Bi-fidelity Surrogate Modeling Approach for Uncertainty Propagation in Three-Dimensional Hemodynamic Simulations
• Lagrangian operator inference enhanced with structure-preserving machine learning for nonintrusive model reduction of mechanical systems
• Operator Learning Using Random Features: A Tool for Scientific Computing
• Non-intrusive reduced order modeling of poroelasticity of heterogeneous media based on a discontinuous Galerkin approximation
• A comparison of dynamical fluctuations of biased diffusion and run-and-tumble dynamics in one dimension
• CD-ROM: Complemented Deep-Reduced Order Model
• Physics-Constrained Bayesian Neural Network for Fluid Flow Reconstruction with Sparse and Noisy Data
• Energy-Preserving Reduced Operator Inference for Efficient Design and Control