EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A physics-inspired memory-augmented deep learning framework for magnetic core loss prediction

Haifang Cong, Siyu Chen, Yang Yang, Tianyun Luan iD, Chao Yang

PLOS One · 2026

DOI: 10.1371/journal.pone.0339490 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Haifang Cong und 4 weiteren Autor:innen. Beteiligte Einrichtung: Changchun University of Science and Technology, Changchun, China. Hauptthemen laut MEDLINE: Deep Learning, Magnetics, Physics. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Development Program of Jilin Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate prediction of magnetic core loss is a key challenge for improving the efficiency and reliability of power electronic systems. Traditional empirical models such as the Steinmetz equation are only applicable to sinusoidal steady-state conditions and struggle with the complex non-sinusoidal waveforms and variable operating conditions in modern power electronics. While existing deep learning methods have shown improvements, they still face fundamental limitations in handling the nonlinear mismatch between B(t) and H(t) waveforms, coupling of multi-scale loss mechanisms, and generalization under extreme operating conditions. This paper proposes an Enhanced Memory Augmented Mamba (EMA-Mamba) model that achieves breakthrough progress in magnetic core loss prediction. It utilizes a state-space memory augmentation mechanism that stores and retrieves typical magnetization patterns through a trainable external memory matrix, endowing the model with a capability similar to the “magnetic memory” of magnetic materials, effectively solving the gradient vanishing problem in long sequence modeling. Combined with an attention-guided intelligent feature selection mechanism, it adaptively identifies critical turning points in hysteresis curves through a Top-K strategy, fundamentally solving the temporal mismatch problem between B(t) and H(t) waveforms. Finally, through a physics-constrained multi-objective optimization framework, it achieves decoupled modeling of hysteresis loss, eddy current loss, and residual loss through loss function combination, overcoming the optimization difficulties caused by data spanning six orders of magnitude. Experiments on the MagNet dataset containing 10 materials and over 150,000 data points show that EMA-Mamba achieves an average prediction error of 4.50% and a coefficient of determination of 99.9947%, reducing error by 34.2% compared to state-of-the-art baseline methods, with a 36.2% reduction in 95th percentile error under extreme conditions. The model demonstrates excellent temperature robustness and cross-material generalization capability, providing a reliable theoretical tool for intelligent design and optimization of magnetic components.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningMagneticsPhysicsAlgorithmsModels, Theoretical

Autor:innen und Institutionen

  • Changchun University of Science and Technology, Changchun, China.
  • Changchun Sci-Tech University, Changchun, China.
  • Beijing Aerospace Times Laser Navigation Technology Co., Ltd, Beijing, China.

ORCID-Profile: Tianyun Luan

Förderung

  • Science and Technology Development Program of Jilin Province

Zitieren

APA

Cong, H., Chen, S., Yang, Y., Luan, T., & Yang, C. (2026). A physics-inspired memory-augmented deep learning framework for magnetic core loss prediction. PLOS One. https://doi.org/10.1371/journal.pone.0339490

Vancouver

Cong H, Chen S, Yang Y, Luan T, Yang C. A physics-inspired memory-augmented deep learning framework for magnetic core loss prediction. PLOS One. 2026. doi:10.1371/journal.pone.0339490

BibTeX
@article{Cong2026,
  author = {Cong, Haifang and Chen, Siyu and Yang, Yang and Luan, Tianyun and Yang, Chao},
  title = {A physics-inspired memory-augmented deep learning framework for magnetic core loss prediction},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0339490}
}
Kostenlos starten →