Predicting the failure of two-dimensional silica glasses
arXiv:2201.09723 · doi:10.1038/s41467-022-30530-1
Abstract
Being able to predict the failure of materials based on structural information is a fundamental issue with enormous practical and industrial relevance for the monitoring of devices and components. Thanks to recent advances in deep learning, accurate failure predictions are becoming possible even for strongly disordered solids, but the sheer number of parameters used in the process renders a physical interpretation of the results impossible. Here we address this issue and use machine learning methods to predict the failure of simulated two dimensional silica glasses from their initial undeformed structure. We then exploit Gradient-weighted Class Activation Mapping (Grad-CAM) to build attention maps associated with the predictions, and we demonstrate that these maps are amenable to physical interpretation in terms of topological defects and local potential energies. We show that our predictions can be transferred to samples with different shape or size than those used in training, as well as to experimental images. Our strategy illustrates how artificial neural networks trained with numerical simulation results can provide interpretable predictions of the behavior of experimentally measured structures.
arXiv comment41 pages 13 figures
References in corpus (16)
- Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
- Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
- Machine Learning Force Fields
- Machine Learning of Accurate Energy-Conserving Molecular Force Fields
- Data-driven discovery of coordinates and governing equations
- Identifying structural flow defects in disordered solids using machine learning methods
- Novel Two-Dimensional Silicon Dioxide with in-plane Negative Poisson's Ratio
- Statistics and properties of low-frequency vibrational modes in structural glasses
- Predicting plasticity in disordered solids from structural indicators
- Automatic Design of Mechanical Metamaterial Actuators
- Machine learning plastic deformation of crystals
- Predicting orientation-dependent plastic susceptibility from static structure in amorphous solids via deep learning
- Universal low-frequency vibrational modes in silica glasses
- Elementary plastic events in amorphous silica
- Anisotropic Structural Predictor in Glassy Materials
- Probing the transition from dislocation jamming to pinning by machine learning