Magnetic microstructure machine learning analysis

TL;DR

This paper employs machine learning, specifically a random forest classifier, combined with micromagnetic modeling to identify key microstructure features influencing magnetization reversal in NdFeB magnets, revealing the impact of grain misorientation and position.

Contribution

It introduces a novel machine learning framework that integrates reduced order modeling to analyze microstructure effects on magnetization reversal in permanent magnets.

Findings

Misorientation and grain position are key factors in magnetization reversal.

Edge regions of the magnet have lower switching fields.

Edge hardening with Dy-diffusion increases coercive fields.

Abstract

We use a machine learning approach to identify the importance of microstructure characteristics in causing magnetization reversal in ideally structured large-grained Nd$_2$Fe$_{14}$B permanent magnets. The embedded Stoner-Wohlfarth method is used as a reduced order model for determining local switching field maps which guide the data-driven learning procedure. The predictor model is a random forest classifier which we validate by comparing with full micromagnetic simulations in the case of small granular test structures. In the course of the machine learning microstructure analysis the most important features explaining magnetization reversal were found to be the misorientation and the position of the grain within the magnet. The lowest switching fields occur near the top and bottom edges of the magnet. While the dependence of the local switching field on the grain orientation is known from theory, the influence of the position of the grain on the local coercive field strength is less obvious. As a direct result of our findings of the machine learning analysis we show that edge hardening via Dy-diffusion leads to higher coercive fields.