Spectral entropy prior-guided deep feature fusion architecture for magnetic core loss

TL;DR

This paper introduces SEPI-TFPNet, a hybrid deep learning model guided by spectral entropy priors for magnetic core loss prediction, enhancing accuracy and interpretability over traditional and purely data-driven models.

Contribution

It proposes a novel hybrid model combining empirical spectral entropy-based priors with deep learning for improved magnetic core loss modeling.

Findings

Achieves higher accuracy than 21 challenge models

Demonstrates robustness across various magnetic materials

Outperforms recent advanced methods in 2024-2025

Abstract

Accurate core loss modeling is critical for the design of high-efficiency power electronic systems. Traditional core loss modeling methods have limitations in prediction accuracy. To advance this field, the IEEE Power Electronics Society launched the MagNet Challenge in 2023, the first international competition focused on data-driven power electronics design methods, aiming to uncover complex loss patterns in magnetic components through a data-driven paradigm. Although purely data-driven models demonstrate strong fitting performance, their interpretability and cross-distribution generalization capabilities remain limited. To address these issues, this paper proposes a hybrid model, SEPI-TFPNet, which integrates empirical models with deep learning. The physical-prior submodule employs a spectral entropy discrimination mechanism to select the most suitable empirical model under different excitation waveforms. The data-driven submodule incorporates convolutional neural networks, multi-head attention mechanisms, and bidirectional long short-term memory networks to extract flux-density time-series features. An adaptive feature fusion module is introduced to improve multimodal feature interaction and integration. Using the MagNet dataset containing various magnetic materials, this paper evaluates the proposed method and compares it with 21 representative models from the 2023 challenge and three advanced methods from 2024-2025. The results show that the proposed method achieves improved modeling accuracy and robustness.