Learning constitutive relations from indirect observations using deep neural networks

DOI10.1016/j.jcp.2020.109491zbMath1437.65192arXiv1905.12530OpenAlexW3022973855MaRDI QIDQ781968

Kailai Xu, Eric Darve, Daniel Z. Huang, Charbel Farhat

Publication date: 21 July 2020

Published in: Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1905.12530

zbMATH Keywords

finite element methodneural networkshomogenizationuncertainty quantification

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Laplace operator, Helmholtz equation (reduced wave equation), Poisson equation (35J05) Numerical differentiation (65D25)

Related Items

Image inversion and uncertainty quantification for constitutive laws of pattern formation ⋮ Efficient derivative-free Bayesian inference for large-scale inverse problems ⋮ Hybrid FEM-NN models: combining artificial neural networks with the finite element method ⋮ Physics constrained learning for data-driven inverse modeling from sparse observations ⋮ Learning generative neural networks with physics knowledge ⋮ Multiscale modeling of inelastic materials with thermodynamics-based artificial neural networks (TANN) ⋮ Bayesian-EUCLID: discovering hyperelastic material laws with uncertainties ⋮ Iterated Kalman methodology for inverse problems ⋮ A deep learning energy method for hyperelasticity and viscoelasticity ⋮ Probabilistic partition of unity networks for high‐dimensional regression problems ⋮ Hybrid analysis and modeling, eclecticism, and multifidelity computing toward digital twin revolution ⋮ A mechanics‐informed artificial neural network approach in data‐driven constitutive modeling ⋮ Molecular dynamics inferred transfer learning models for finite‐strain hyperelasticity of monoclinic crystals: Sobolev training and validations against physical constraints ⋮ Data-driven elasto-(visco)-plasticity involving hidden state variables ⋮ Bayesian calibration for large‐scale fluid structure interaction problems under embedded/immersed boundary framework ⋮ Uncertainty quantification in scientific machine learning: methods, metrics, and comparisons ⋮ Addressing discontinuous root-finding for subsequent differentiability in machine learning, inverse problems, and control ⋮ Transfer learning based physics-informed neural networks for solving inverse problems in engineering structures under different loading scenarios ⋮ Automated discovery of generalized standard material models with EUCLID ⋮ DPK: Deep Neural Network Approximation of the First Piola-Kirchhoff Stress ⋮ Improved Analysis of PINNs: Alleviate the CoD for Compositional Solutions ⋮ Physics-integrated neural differentiable (PiNDiff) model for composites manufacturing ⋮ An equivariant neural operator for developing nonlocal tensorial constitutive models ⋮ Spectral operator learning for parametric PDEs without data reliance ⋮ A mechanics-informed deep learning framework for data-driven nonlinear viscoelasticity ⋮ Unsupervised discovery of interpretable hyperelastic constitutive laws ⋮ Learning constitutive models from microstructural simulations via a non-intrusive reduced basis method ⋮ Learning nonlocal constitutive models with neural networks ⋮ Solving inverse problems in stochastic models using deep neural networks and adversarial training ⋮ Learning viscoelasticity models from indirect data using deep neural networks ⋮ Geometric deep learning for computational mechanics. I: Anisotropic hyperelasticity ⋮ Sobolev training of thermodynamic-informed neural networks for interpretable elasto-plasticity models with level set hardening ⋮ Frame-independent vector-cloud neural network for nonlocal constitutive modeling on arbitrary grids ⋮ Material Modeling via Thermodynamics-Based Artificial Neural Networks

Uses Software

• LBFGS-B

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Cross-covariance functions for multivariate geostatistics
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• Identification of elastic parameters by displacement field measurement
• Machine learning strategies for systems with invariance properties
• Finite element analysis of V-ribbed belts using neural network based hyperelastic material model
• Determination of the size of the representative volume element for random composites: Statistical and numerical approach.
• \(FE^2\) multiscale approach for modelling the elastoviscoplastic behaviour of long fibre SiC/Ti composite materials
• Modeling uncertainty in flow simulations via generalized polynomial chaos.
• Error estimates and condition numbers for radial basis function interpolation
• An algorithm for selecting a good value for the parameter \(c\) in radial basis function interpolation
• Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
• A framework for data-driven analysis of materials under uncertainty: countering the curse of dimensionality
• Data driven computing with noisy material data sets
• A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Data-driven computational mechanics
• Bayesian Calibration of Computer Models
• Active Subspace Methods in Theory and Practice: Applications to Kriging Surfaces
• Computational homogenization of nonlinear elastic materials using neural networks
• Analysis of Composite Materials—A Survey
• Toward realization of computational homogenization in practice
• Laplacian smoothing and Delaunay triangulations
• Two algorithms for constructing a Delaunay triangulation
• Finite element computation of large rubber membrane deformations
• Line search algorithms with guaranteed sufficient decrease
• Barycentric Lagrange Interpolation
• The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
• A Limited Memory Algorithm for Bound Constrained Optimization
• Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
• Advanced Lectures on Machine Learning
• Spectral Methods
• Genetic evolution of nonlinear material constitutive models