Deep learning of thermodynamics-aware reduced-order models from data

Abstract

We present an algorithm to learn the relevant latent variables of a large-scale discretized physical system and predict its time evolution using thermodynamically-consistent deep neural networks. Our method relies on sparse autoencoders, which reduce the dimensionality of the full order model to a set of sparse latent variables with no prior knowledge of the coded space dimensionality. Then, a second neural network is trained to learn the metriplectic structure of those reduced physical variables and predict its time evolution with a so-called structure-preserving neural network. This data-based integrator is guaranteed to conserve the total energy of the system and the entropy inequality, and can be applied to both conservative and dissipative systems. The integrated paths can then be decoded to the original full-dimensional manifold and be compared to the ground truth solution. This method is tested with two examples applied to fluid and solid mechanics.

17 pages, 7 figures

References in corpus (6)

• Deep Learning in Neural Networks: An Overview
• Machine learning materials physics: Integrable deep neural networks enable scale bridging by learning free energy functions
• Structure-preserving neural networks
• Learning Differential Equations that are Easy to Solve
• Symplectic Neural Networks in Taylor Series Form for Hamiltonian Systems
• Learning Poisson systems and trajectories of autonomous systems via Poisson neural networks

Cited by in corpus (16)

• Multiscale modeling of inelastic materials with Thermodynamics-based Artificial Neural Networks (TANN)
• On physics-informed data-driven isotropic and anisotropic constitutive models through probabilistic machine learning and space-filling sampling
• Variational Onsager Neural Networks (VONNs): A thermodynamics-based variational learning strategy for non-equilibrium PDEs
• GFINNs: GENERIC Formalism Informed Neural Networks for Deterministic and Stochastic Dynamical Systems
• Evolution TANN and the identification of internal variables and evolution equations in solid mechanics
• Thermodynamics-informed graph neural networks
• Parameterized Neural Ordinary Differential Equations: Applications to Computational Physics Problems
• Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems
• PGD-based advanced nonlinear multiparametric regressions for constructing metamodels at the scarce-data limit
• On the data-driven description of lattice materials mechanics
• The mixed deep energy method for resolving concentration features in finite strain hyperelasticity
• Thermodynamics-informed super-resolution of scarce temporal dynamics data
• Uncertainty quantification in a mechanical submodel driven by a Wasserstein-GAN
• A surrogate model for topology optimisation of elastic structures via parametric autoencoders
• A deep learning energy method for hyperelasticity and viscoelasticity
• Learning inelastic constitutive models from stress-strain data under hard thermodynamic constraints