Learning to fail: Predicting fracture evolution in brittle material models using recurrent graph convolutional neural networks

Abstract

We propose a machine learning approach to address a key challenge in materials science: predicting how fractures propagate in brittle materials under stress, and how these materials ultimately fail. Our methods use deep learning and train on simulation data from high-fidelity models, emulating the results of these models while avoiding the overwhelming computational demands associated with running a statistically significant sample of simulations. We employ a graph convolutional network that recognizes features of the fracturing material and a recurrent neural network that models the evolution of these features, along with a novel form of data augmentation that compensates for the modest size of our training data. We simultaneously generate predictions for qualitatively distinct material properties. Results on fracture damage and length are within 3% of their simulated values, and results on time to material failure, which is notoriously difficult to predict even with high-fidelity models, are within approximately 15% of simulated values. Once trained, our neural networks generate predictions within seconds, rather than the hours needed to run a single simulation.

References in corpus (7)

• Semi-Supervised Classification with Graph Convolutional Networks
• An overview of gradient descent optimization algorithms
• Multilayer Networks
• Deep Convolutional Networks on Graph-Structured Data
• Dynamic Graph Convolutional Networks
• Why does deep and cheap learning work so well?
• Machine learning for graph-based representations of three-dimensional discrete fracture networks

Cited by in corpus (9)

• A Data-Driven Approach to Full-Field Damage and Failure Pattern Prediction in Microstructure-Dependent Composites using Deep Learning
• Novel DeepONet architecture to predict stresses in elastoplastic structures with variable complex geometries and loads
• Averaging local structure to predict the dynamic propensity in supercooled liquids
• On the use of graph neural networks and shape-function-based gradient computation in the deep energy method
• I-FENN with Temporal Convolutional Networks: expediting the load-history analysis of non-local gradient damage propagation
• Accelerating High-Strain Continuum-Scale Brittle Fracture Simulations with Machine Learning
• StressNet: Deep Learning to Predict Stress With Fracture Propagation in Brittle Materials
• Exploring the potential of transfer learning for metamodels of heterogeneous material deformation
• Visualizing key features in X-ray images of epoxy resins for improved material classification using singular value decomposition of deep learning features