Predicting the mechanical response of oligocrystals with deep learning

Abstract

In this work we employ data-driven homogenization approaches to predict the particular mechanical evolution of polycrystalline aggregates with tens of individual crystals. In these oligocrystals the differences in stress response due to microstructural variation is pronounced. Shell-like structures produced by metal-based additive manufacturing and the like make the prediction of the behavior of oligocrystals technologically relevant. The predictions of traditional homogenization theories based on grain volumes are not sensitive to variations in local grain neighborhoods. Direct simulation of the local response with crystal plasticity finite element methods is more detailed, but the computations are expensive. To represent the stress-strain response of a polycrystalline sample given its initial grain texture and morphology we have designed a novel neural network that incorporates a convolution component to observe and reduce the information in the crystal texture field and a recursive component to represent the causal nature of the history information. This model exhibits accuracy on par with crystal plasticity simulations at minimal computational cost per prediction. * rjones@sandia.gov arXiv:1901.10669v2 [cond-mat.mes-hall]

Feed-Forward Neural Networks for Failure Mechanics Problems

“…[49] introduced a graph convolutional deep neural network, incorporating the non-Euclidean weighted graph data to predict the elastic response of materials with complex microstructures. For recent works on CNNs, we refer to [50][51][52], and the citations therein.…”

SO(3)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials

“…Within structural dynamics and the computation of response statistics, Koeppe et al [34] proposed the inclusion of RNN to speed up Monte Carlo method by replacing the iterative calculation procedure used to evaluate the nonlinear, inelastic, hysteresis structural behavior. In connection with employing machine learning in the context of mechanical behavior of polycrystalline aggregates, Frankel et al [35] recently developed a new approach that incorporate the pre-deformed grain morphology and orientation in the sense of image data and using convolutional neural network. Meanwhile, Huang et al [27] introduced uncertainty quantification to neural network models designed for elasticity constitutive laws and the corresponding boundary value problems that employs neural network constitutive laws.…”

One for all: Universal material model based on minimal state-space neural networks

“…S1), thereby adding memory units to the network. As a result, applications of LSTMs and GRUs to material modeling (9)(10)(11)(12)(13)(14)(15) involve far more state variables than typical physics-based models.…”