Physics-informed machine learning and stray field computation with application to micromagnetic energy minimization

TL;DR

This paper introduces a physics-informed neural network approach for 3D static micromagnetic equations, enabling efficient energy minimization and stray field computation without traditional meshing, validated on standard problems.

Contribution

It presents a novel mesh-free, unsupervised neural network method for micromagnetic energy minimization and stray field calculation, reducing computational complexity and enabling flexible parameter handling.

Findings

Successfully minimized micromagnetic energy using PINNs.

Efficient stray field computation via linear least squares.

Validated approach on standard micromagnetic problems.

Abstract

We study the full 3d static micromagnetic equations via a physics-informed neural network (PINN) ansatz for the continuous magnetization configuration. PINNs are inherently mesh-free and unsupervised learning models. In our approach we can learn to minimize the total Gibbs free energy with additional conditional parameters, such as the exchange length, by a single low-parametric neural network model. In contrast, traditional numerical methods would require the computation and storage of a large number of solutions to interpolate the continuous spectrum of quasi-optimal magnetization configurations. We also consider the important and computationally expensive stray field problem via PINNs, where we use a basically linear learning ansatz, called extreme learning machines (ELM) within a splitting method for the scalar potential. This reduces the stray field training to a linear least squares problem with precomputable solution operator. We validate the stray field method by means of numerical example comparisons from literature and illustrate the full micromagnetic approach via the NIST $\mu$MAG standard problem $\#3$.