Similarity Equivariant Graph Neural Networks for Homogenization of Metamaterials

Abstract

Soft, porous mechanical metamaterials exhibit pattern transformations that may have important applications in soft robotics, sound reduction and biomedicine. To design these innovative materials, it is important to be able to simulate them accurately and quickly, in order to tune their mechanical properties. Since conventional simulations using the finite element method entail a high computational cost, in this article we aim to develop a machine learning-based approach that scales favorably to serve as a surrogate model. To ensure that the model is also able to handle various microstructures, including those not encountered during training, we include the microstructure as part of the network input. Therefore, we introduce a graph neural network that predicts global quantities (energy, stress stiffness) as well as the pattern transformations that occur (the kinematics). To make our model as accurate and data-efficient as possible, various symmetries are incorporated into the model. The starting point is an E(n)-equivariant graph neural network (which respects translation, rotation and reflection) that has periodic boundary conditions (i.e., it is in-/equivariant with respect to the choice of RVE), is scale in-/equivariant, can simulate large deformations, and can predict scalars, vectors as well as second and fourth order tensors (specifically energy, stress and stiffness). The incorporation of scale equivariance makes the model equivariant with respect to the similarities group, of which the Euclidean group E(n) is a subgroup. We show that this network is more accurate and data-efficient than graph neural networks with fewer symmetries. To create an efficient graph representation of the finite element discretization, we use only the internal geometrical hole boundaries from the finite element mesh to achieve a better speed-up and scaling with the mesh size.

60 pages, 22 figures. Published in CMAME (Computer Methods in Applied Mechanics and Engineering)

References in corpus (12)

• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• Mechanical Metamaterials with Negative Compressibility Transitions
• SchNet: A continuous-filter convolutional neural network for modeling quantum interactions
• A new family of Constitutive Artificial Neural Networks towards automated model discovery
• Equivariant message passing for the prediction of tensorial properties and molecular spectra
• Neural networks meet hyperelasticity: A guide to enforcing physics
• A Newton Solver for Micromorphic Computational Homogenization Enabling Multiscale Buckling Analysis of Pattern-Transforming Metamaterials
• On the Expressive Power of Geometric Graph Neural Networks
• Extended Micromorphic Computational Homogenization for Mechanical Metamaterials Exhibiting Multiple Geometric Pattern Transformations
• Learning constitutive models from microstructural simulations via a non-intrusive reduced basis method
• Learning constitutive models from microstructural simulations via a non-intrusive reduced basis method: Extension to geometrical parameterizations
• Learning the nonlinear dynamics of soft mechanical metamaterials with graph networks