Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction

Abstract

In addition to providing high-profile successes in computer vision and natural language processing, neural networks also provide an emerging set of techniques for scientific problems. Such data-driven models, however, typically ignore physical insights from the scientific system under consideration. Among other things, a physics-informed model formulation should encode some degree of stability or robustness or well-conditioning (in that a small change of the input will not lead to drastic changes in the output), characteristic of the underlying scientific problem. We investigate whether it is possible to include physics-informed prior knowledge for improving the model quality (e.g., generalization performance, sensitivity to parameter tuning, or robustness in the presence of noisy data). To that extent, we focus on the stability of an equilibrium, one of the most basic properties a dynamic system can have, via the lens of Lyapunov analysis. For the prototypical problem of fluid flow prediction, we show that models preserving Lyapunov stability improve the generalization error and reduce the prediction uncertainty.

References in corpus (9)

• Deep learning for universal linear embeddings of nonlinear dynamics
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• Super-resolution reconstruction of turbulent flows with machine learning
• Enforcing Statistical Constraints in Generative Adversarial Networks for Modeling Chaotic Dynamical Systems
• Deep learning algorithm for data-driven simulation of noisy dynamical system
• Deep Dynamical Modeling and Control of Unsteady Fluid Flows
• Long-time predictive modeling of nonlinear dynamical systems using neural networks
• Physics-informed deep generative models
• From Deep to Physics-Informed Learning of Turbulence: Diagnostics

Cited by in corpus (33)

• Machine learning accelerated computational fluid dynamics
• Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders
• Shallow Neural Networks for Fluid Flow Reconstruction with Limited Sensors
• Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data
• Bridging observation, theory and numerical simulation of the ocean using Machine Learning
• On closures for reduced order models $-$ A spectrum of first-principle to machine-learned avenues
• Assessment of unsteady flow predictions using hybrid deep learning based reduced order models
• Reconstruction of turbulent data with deep generative models for semantic inpainting from TURB-Rot database
• Active Learning in Robotics: A Review of Control Principles
• Long short-term memory embedded nudging schemes for nonlinear data assimilation of geophysical flows
• Attention-Enhanced Neural Network Models for Turbulence Simulation
• Three-dimensional deep learning-based reduced order model for unsteady flow dynamics with variable Reynolds number
• Parameterized Neural Ordinary Differential Equations: Applications to Computational Physics Problems
• Nonlinear proper orthogonal decomposition for convection-dominated flows
• From Fourier to Koopman: Spectral Methods for Long-term Time Series Prediction
• A neural network multigrid solver for the Navier-Stokes equations
• Embedding Hard Physical Constraints in Neural Network Coarse-Graining of 3D Turbulence
• An advanced hybrid deep adversarial autoencoder for parameterized nonlinear fluid flow modelling
• Solving Bayesian Inverse Problems via Variational Autoencoders
• Forecasting Sequential Data using Consistent Koopman Autoencoders
• Continuous-in-Depth Neural Networks
• Mechanisms of a Convolutional Neural Network for Learning Three-dimensional Unsteady Wake Flow
• An unsupervised learning approach to solving heat equations on chip based on Auto Encoder and Image Gradient
• Variational Autoencoders for Learning Nonlinear Dynamics of Physical Systems
• Memory-Efficient Learning of Stable Linear Dynamical Systems for Prediction and Control
• Analyzing Koopman approaches to physics-informed machine learning for long-term sea-surface temperature forecasting
• Validation and parameterization of a novel physics-constrained neural dynamics model applied to turbulent fluid flow
• Noisy Recurrent Neural Networks
• Model order reduction with neural networks: Application to laminar and turbulent flows
• A Differential Geometry Perspective on Orthogonal Recurrent Models
• Convergent Graph Solvers
• Investigation of Nonlinear Model Order Reduction of the Quasigeostrophic Equations through a Physics-Informed Convolutional Autoencoder
• Cluster-and-Conquer: A Framework For Time-Series Forecasting