Thermodynamics-informed graph neural networks

Abstract

In this paper we present a deep learning method to predict the temporal evolution of dissipative dynamic systems. We propose using both geometric and thermodynamic inductive biases to improve accuracy and generalization of the resulting integration scheme. The first is achieved with Graph Neural Networks, which induces a non-Euclidean geometrical prior with permutation invariant node and edge update functions. The second bias is forced by learning the GENERIC structure of the problem, an extension of the Hamiltonian formalism, to model more general non-conservative dynamics. Several examples are provided in both Eulerian and Lagrangian description in the context of fluid and solid mechanics respectively, achieving relative mean errors of less than 3% in all the tested examples. Two ablation studies are provided based on recent works in both physics-informed and geometric deep learning.

14 pages, 11 figures

References in corpus (4)

• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Machine learning materials physics: Integrable deep neural networks enable scale bridging by learning free energy functions
• Port-Hamiltonian Neural Networks for Learning Explicit Time-Dependent Dynamical Systems
• ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations

Cited by in corpus (8)

• Gibbs-Duhem-Informed Neural Networks for Binary Activity Coefficient Prediction
• Thermodynamics-informed neural networks for physically realistic mixed reality
• Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems
• Physics-Aware Neural Implicit Solvers for multiscale, parametric PDEs with applications in heterogeneous media
• Reversible and irreversible bracket-based dynamics for deep graph neural networks
• Critical Sampling for Robust Evolution Operator Learning of Unknown Dynamical Systems
• Latent Dynamics Graph Convolutional Networks for model order reduction of parameterized time-dependent PDEs
• Enhancing material behavior discovery using embedding-oriented Physically-Guided Neural Networks with Internal Variables