Geometric deep learning for computational mechanics Part I: Anisotropic Hyperelasticity

Abstract

This paper is the first attempt to use geometric deep learning and Sobolev training to incorporate non-Euclidean microstructural data such that anisotropic hyperelastic material machine learning models can be trained in the finite deformation range. While traditional hyperelasticity models often incorporate homogenized measures of microstructural attributes, such as porosity averaged orientation of constitutes, these measures cannot reflect the topological structures of the attributes. We fill this knowledge gap by introducing the concept of weighted graph as a new mean to store topological information, such as the connectivity of anisotropic grains in assembles. Then, by leveraging a graph convolutional deep neural network architecture in the spectral domain, we introduce a mechanism to incorporate these non-Euclidean weighted graph data directly as input for training and for predicting the elastic responses of materials with complex microstructures. To ensure smoothness and prevent non-convexity of the trained stored energy functional, we introduce a Sobolev training technique for neural networks such that stress measure is obtained implicitly from taking directional derivatives of the trained energy functional. By optimizing the neural network to approximate both the energy functional output and the stress measure, we introduce a training procedure the improves efficiency and generalize the learned energy functional for different microstructures. The trained hybrid neural network model is then used to generate new stored energy functional for unseen microstructures in a parametric study to predict the influence of elastic anisotropy on the nucleation and propagation of fracture in the brittle regime.

References in corpus (18)

• Distributed Representations of Words and Phrases and their Compositionality
• Semi-Supervised Classification with Graph Convolutional Networks
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Data-driven computational mechanics
• graph2vec: Learning Distributed Representations of Graphs
• A deep material network for multiscale topology learning and accelerated nonlinear modeling of heterogeneous materials
• Model-Free Data-Driven Inelasticity
• Neural Network with Unbounded Activation Functions is Universal Approximator
• Learning Constitutive Relations from Indirect Observations Using Deep Neural Networks
• Inferring low-dimensional microstructure representations using convolutional neural networks
• Meta-modeling game for deriving theoretical-consistent, micro-structural-based traction-separation laws via deep reinforcement learning
• Machine learning materials physics: Integrable deep neural networks enable scale bridging by learning free energy functions
• Stress-induced anisotropy in granular materials: fabric, stiffness, and permeability
• Machine learning materials physics: Surrogate optimization and multi-fidelity algorithms predict precipitate morphology in an alternative to phase field dynamics
• A Physics-Constrained Data-Driven Approach Based on Locally Convex Reconstruction for Noisy Database
• Sobolev Training for Neural Networks
• ILS-MPM: an implicit level-set-based material point method for frictional particulate contact mechanics of deformable particles

Cited by in corpus (52)

• Polyconvex anisotropic hyperelasticity with neural networks
• Unifying the design space and optimizing linear and nonlinear truss metamaterials by generative modeling
• NN-EUCLID: deep-learning hyperelasticity without stress data
• Neural networks meet hyperelasticity: A guide to enforcing physics
• Multi-Objective Loss Balancing for Physics-Informed Deep Learning
• Deep Learning in Deterministic Computational Mechanics
• A Data-Driven Approach to Full-Field Damage and Failure Pattern Prediction in Microstructure-Dependent Composites using Deep Learning
• FE${}^{\text{ANN}}$ $-$ An efficient data-driven multiscale approach based on physics-constrained neural networks and automated data mining
• Data-driven Tissue Mechanics with Polyconvex Neural Ordinary Differential Equations
• Local approximate Gaussian process regression for data-driven constitutive laws: Development and comparison with neural networks
• Integrated Finite Element Neural Network (I-FENN) for non-local continuum damage mechanics
• A comparative study on different neural network architectures to model inelasticity
• Learning hyperelastic anisotropy from data via a tensor basis neural network
• Finite electro-elasticity with physics-augmented neural networks
• Graph neural networks for simulating crack coalescence and propagation in brittle materials
• MAgNET: A Graph U-Net Architecture for Mesh-Based Simulations
• Data-driven anisotropic finite viscoelasticity using neural ordinary differential equations
• Towards Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses
• On the use of graph neural networks and shape-function-based gradient computation in the deep energy method
• Predicting the Mechanical Properties of Biopolymer Gels Using Neural Networks Trained on Discrete Fiber Network Data
• Geometric deep learning for computational mechanics Part II: Graph embedding for interpretable multiscale plasticity
• LS-DYNA Machine Learning-based Multiscale Method for Nonlinear Modeling of Short Fiber-Reinforced Composites
• Discovering interpretable elastoplasticity models via the neural polynomial method enabled symbolic regressions
• Can KAN CANs? Input-convex Kolmogorov-Arnold Networks (KANs) as hyperelastic constitutive artificial neural networks (CANs)
• What machine learning can do for computational solid mechanics
• Physics informed deep learning for computational elastodynamics without labeled data
• Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics
• Intelligent multiscale simulation based on process-guided composite database
• HyperCAN: Hypernetwork-Driven Deep Parameterized Constitutive Models for Metamaterials
• Equivariant geometric learning for digital rock physics: estimating formation factor and effective permeability tensors from Morse graph
• Cooperative data-driven modeling
• FIP-GNN: Graph neural networks for scalable prediction of grain-level fatigue indicator parameters
• I-FENN with Temporal Convolutional Networks: expediting the load-history analysis of non-local gradient damage propagation
• A graph-based probabilistic geometric deep learning framework with online enforcement of physical constraints to predict the criticality of defects in porous materials
• Similarity Equivariant Graph Neural Networks for Homogenization of Metamaterials
• When invariants matter: The role of I1 and I2 in neural network models of incompressible hyperelasticity
• Recovering Mullins damage hyperelastic behaviour with physics augmented neural networks
• Meshless physics-informed deep learning method for three-dimensional solid mechanics
• N-Adaptive Ritz Method: A Neural Network Enriched Partition of Unity for Boundary Value Problems
• Enforcing exact physics in scientific machine learning: a data-driven exterior calculus on graphs
• How machine learning can help the design and analysis of composite materials and structures?
• Data-Driven Modeling of an Unsaturated Bentonite Buffer Model Test Under High Temperatures Using an Enhanced Axisymmetric Reproducing Kernel Particle Method
• Precise, efficient and flexible modeling of crystallizing elastomers based on physics-augmented neural networks
• COMMET: orders-of-magnitude speed-up in finite element method via batch-vectorized neural constitutive updates
• Conformal Quantile Regression for Neural Probabilistic Constitutive Modeling
• A physics-augmented neural network framework for finite strain incompressible viscoelasticity
• Sobolev training of thermodynamic-informed neural networks for smoothed elasto-plasticity models with level set hardening
• Bayesian-EUCLID: discovering hyperelastic material laws with uncertainties
• Data-driven discovery of interpretable causal relations for deep learning material laws with uncertainty propagation
• Learning Mechanically Driven Emergent Behavior with Message Passing Neural Networks
• Hetero-EUCLID: Interpretable model discovery for heterogeneous hyperelastic materials using stress-unsupervised learning
• Constitutive Priors for Inverse Design