AnisoGNN: physics-informed graph neural networks that generalize to anisotropic properties of polycrystals
Submission Track: Papers
Submission Category: AI-Guided Design
Keywords: Graph neural networks, microstructure-property relationships, simulations, polycrystals
Supplementary Material: pdf
TL;DR: We present graph neural networks predicting anisotropic properties of polycrystals in arbitrary directions
Abstract: We present AnisoGNNs -- graph neural networks (GNNs) that generalize predictions of anisotropic properties of polycrystals in arbitrary testing directions without the need in excessive training data. To this end, we develop GNNs with a physics-inspired combination of node attributes and aggregation function. We demonstrate the excellent generalization capabilities of AnisoGNNs in predicting anisotropic elastic and inelastic properties of two alloys.
Digital Discovery Special Issue: Yes
Submission Number: 70
Loading