Variational Onsager Neural Networks (VONNs): A thermodynamics-based variational learning strategy for non-equilibrium PDEs

Abstract

We propose a thermodynamics-based learning strategy for non-equilibrium evolution equations based on Onsager's variational principle, which allows to write such PDEs in terms of two potentials: the free energy and the dissipation potential. Specifically, these two potentials are learned from spatio-temporal measurements of macroscopic observables via proposed neural network architectures that strongly enforce the satisfaction of the second law of thermodynamics. The method is applied to three distinct physical processes aimed at highlighting the robustness and versatility of the proposed approach. These include (i) the phase transformation of a coiled-coil protein, characterized by a non-convex free-energy density; (ii) the one-dimensional dynamic response of a three-dimensional viscoelastic solid, which leverages the variational formulation as a tool for obtaining reduced order models; and (iii) linear and nonlinear diffusion models, characterized by a lack of uniqueness of the free energy and dissipation potentials. These illustrative examples showcase the possibility of learning partial differential equations through their variational action density (i.e., a function instead), by leveraging the thermodynamic structure intrinsic to mechanical and multiphysics problems.

References in corpus (21)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decomposition
• Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks
• Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
• Odd elasticity
• PDE-Net: Learning PDEs from Data
• Machine learning materials physics: Integrable deep neural networks enable scale bridging by learning free energy functions
• Deep learning of thermodynamics-aware reduced-order models from data
• Structure-preserving neural networks
• Variational system identification of the partial differential equations governing pattern-forming physics: Inference under varying fidelity and noise
• Coarse-scale PDEs from fine-scale observations via machine learning
• GFINNs: GENERIC Formalism Informed Neural Networks for Deterministic and Stochastic Dynamical Systems
• A generalization of Onsager's reciprocity relations to gradient flows with nonlinear mobility
• Variational Modelling: Energies, gradient flows, and large deviations
• Input Convex Neural Networks for Building MPC
• Scale bridging materials physics: Active learning workflows and integrable deep neural networks for free energy function representations in alloys
• Fluctuation symmetry leads to GENERIC equations with non-quadratic dissipation
• A framework of nonequilibrium statistical mechanics. II. Coarse-graining
• Computing diffusivities from particle models out of equilibrium
• Learning the GENERIC evolution

Cited by in corpus (16)

• NN-EUCLID: deep-learning hyperelasticity without stress data
• Neural networks meet hyperelasticity: A guide to enforcing physics
• Automated discovery of generalized standard material models with EUCLID
• A comparative study on different neural network architectures to model inelasticity
• Modular machine learning-based elastoplasticity: generalization in the context of limited data
• Automated identification of linear viscoelastic constitutive laws with EUCLID
• Reduced and All-at-Once Approaches for Model Calibration and Discovery in Computational Solid Mechanics
• Physics-augmented neural networks for constitutive modeling of hyperelastic geometrically exact beams
• A Complement to Neural Networks for Anisotropic Inelasticity at Finite Strains
• Carbon Nanotubes as a Basis of Metamaterials and Nanostructures: Crafting via Design Optimization
• A computational framework for nanotrusses: input convex neural networks approach
• Precise, efficient and flexible modeling of crystallizing elastomers based on physics-augmented neural networks
• Deep learning-based computational method for soft matter dynamics: Deep Onsager-Machlup method
• A physics-augmented neural network framework for finite strain incompressible viscoelasticity
• Learning inelastic constitutive models from stress-strain data under hard thermodynamic constraints
• Hetero-EUCLID: Interpretable model discovery for heterogeneous hyperelastic materials using stress-unsupervised learning