Mimicking complex dislocation dynamics by interaction networks
arXiv:1809.03793 · doi:10.1140/epjb/e2018-90419-7
Abstract
Two-dimensional discrete dislocation models exhibit complex dynamics in relaxation and under external loading. This is manifested both in the time-dependent velocities of individual dislocations and in the ensemble response, the strain rate. Here we study how well this complexity may be reproduced using so-called Interaction Networks, an Artificial Intelligence method for learning the dynamics of complex interacting systems. We test how to learn such networks using creep data, and show results on reproducing individual and collective dislocation velocities. The quality of reproducing the interaction kernel is discussed.
References in corpus (9)
- Machine learning phases of matter
- Learning phase transitions by confusion
- Intermittent dislocation flow in viscoplastic deformation
- Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
- A learning scheme to predict atomic forces and accelerate materials simulations
- Mutual Information, Neural Networks and the Renormalization Group
- Avalanches and Plastic Flow in Crystal Plasticity: An Overview
- Depinning transition of dislocation assemblies: pileup and low-angle grain boundary
- Fluctuations and scaling in creep deformation