Exploring the hysteresis properties of nanocrystalline permanent magnets using deep learning

TL;DR

This paper presents a data-driven deep learning approach using variational autoencoders and neural networks to efficiently predict the hysteresis properties of nanocrystalline permanent magnets from microstructure data.

Contribution

It introduces a novel method combining model order reduction and neural networks to encode microstructure and predict magnetic hysteresis properties, enabling structure generation and property estimation.

Findings

Latent space effectively encodes magnet microstructure.

Neural network accurately predicts demagnetization curves.

Interpolation in latent space generates new magnet structures with predictable properties.

Abstract

We demonstrate the use of model order reduction and neural networks for estimating the hysteresis properties of nanocrystalline permanent magnets from microstructure. With a data-driven approach, we learn the demagnetization curve from data-sets created by grain growth and micromagnetic simulations. We show that the granular structure of a magnet can be encoded within a low-dimensional latent space. Latent codes are constructed using a variational autoencoder. The mapping of structure code to hysteresis properties is a multi-target regression problem. We apply deep neural network and use parameter sharing, in order to predict anchor points along the demagnetization curves from the magnet's structure code. The method is applied to study the magnetic properties of nanocrystalline permanent magnets. We show how new grain structures can be generated by interpolation between two points in the latent space and how the magnetic properties of the resulting magnets can be predicted.