Mimicking complex dislocation dynamics by interaction networks

Henri Salmenjoki*, Mikko J. Alava, Lasse Laurson

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

1 Citation (Scopus)


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.

Original languageEnglish
Article number275
Pages (from-to)1-6
Number of pages6
JournalEuropean Physical Journal B
Issue number11
Publication statusPublished - 13 Nov 2018
MoE publication typeA1 Journal article-refereed


  • FLOW

Cite this