RHINO-MAG: Recursive H-Field Inference based on Observed Magnetic Flux under Dynamic Excitation

TL;DR

This paper introduces RHINO-MAG, a GRU-based model for accurate, efficient H-field prediction in ferrite materials under dynamic excitation, winning the MagNet Challenge 2025.

Contribution

The study demonstrates that a lightweight GRU model outperforms physics-inspired models in transient magnetic field prediction tasks.

Findings

A 325-parameter GRU model achieves 8.02% sequence error.

The model attains 1.07% normalized energy error across five materials.

It won first place in the MagNet Challenge 2025 performance category.

Abstract

Driven by the MagNet Challenge 2025 (MC2), increased research interest is directed towards modeling transient magnetic fields within ferrite material. An accurate time-resolved and temperature-aware H-field prediction is essential for optimizing magnetic components in applications with quasi-stationary / non-stationary excitation waveforms. Within the scope of this investigation, a selection of model structures with varying degrees of physically motivated structure are compared. Based on a Pareto investigation, a rather black-box gated recurrent unit (GRU) model structure with a graceful initialization setup is found to offer the most attractive model size vs. model accuracy trade-off, while the physics-inspired models performed worse. For a GRU-based model with only 325 parameters, a sequence relative error of 8.02 % and a normalized energy relative error of 1.07 % averaged across five different materials are achieved on unseen test data. With this excellent parameter efficiency, the proposed model won the first place in the performance category of the MC2.