Deep learning and multi-level featurization of graph representations of microstructural data

DOI10.1007/S00466-023-02300-3MaRDI QIDQ6159319zbMATH OpenOpenAlexFDO

Authors Reese Jones, Cosmin Safta, Ari Frankel

Publication date 1 June 2023

Published in Computational Mechanics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2210.00854

zbMATH Keywords

graph convolutional neural networkinelasticitymicrostructure responsereduced graph

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Artificial neural networks and deep learning (68T07) Micromechanics of solids (74M25) Numerical and other methods in solid mechanics (74S99)

Abstract: Many material response functions depend strongly on microstructure, such as inhomogeneities in phase or orientation. Homogenization presents the task of predicting the mean response of a sample of the microstructure to external loading for use in subgrid models and structure-property explorations. Although many microstructural fields have obvious segmentations, learning directly from the graph induced by the segmentation can be difficult because this representation does not encode all the information of the full field. We develop a means of deep learning of hidden features on the reduced graph given the native discretization and a segmentation of the initial input field. The features are associated with regions represented as nodes on the reduced graph. This reduced representation is then the basis for the subsequent multi-level/scale graph convolutional network model. There are a number of advantages of reducing the graph before fully processing with convolutional layers it, such as interpretable features and efficiency on large meshes. We demonstrate the performance of the proposed network relative to convolutional neural networks operating directly on the native discretization of the data using three physical exemplars.

Recommendations

• A deep material network for multiscale topology learning and accelerated nonlinear modeling of heterogeneous materials
• Efficient multiscale modeling of heterogeneous materials using deep neural networks
• Machine learning materials physics: multi-resolution neural networks learn the free energy and nonlinear elastic response of evolving microstructures
• De-homogenization using convolutional neural networks
• Statistical characterization and reconstruction of heterogeneous microstructures using deep neural network

Cites work

• A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs
• Computational crystal plasticity
• Geometric deep learning for computational mechanics. I: Anisotropic hyperelasticity
• scientific article; zbMATH DE number 1217993 (Why is no real title available?)
• scientific article; zbMATH DE number 3064911 (Why is no real title available?)
• scientific article; zbMATH DE number 3069258 (Why is no real title available?)
• MgNet: a unified framework of multigrid and convolutional neural network
• Shape recognition using FFT preprocessing and neural network

Cited in

(3)

• Uncertainty quantification of graph convolution neural network models of evolving processes
• Equivariant graph convolutional neural networks for the representation of homogenized anisotropic microstructural mechanical response
• Graph-enhanced deep material network: multiscale materials modeling with microstructural informatics

This page was built for publication: Deep learning and multi-level featurization of graph representations of microstructural data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6159319)