A Hybrid Physical and Data‐Driven Framework for Temperature Evaluation of Permanent Magnets in PMSMs

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Proposes a hybrid method integrating machine learning with physical modeling for temperature evaluation in PMSMs, suggesting improved reliability.

The proposed framework effectively evaluates temperature in permanent magnet synchronous machines, enhancing operational reliability.
Experimental results demonstrate the robustness and generalization of the hybrid model, with successful temperature assessments.
Combining a physical model with machine learning techniques like graph attention networks enables sophisticated thermal analysis.
This approach lays the groundwork for further studies on temperature-dependent magnet degradation in PMSMs.

Wang et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Proposes a hybrid method integrating machine learning with physical modeling for temperature evaluation in PMSMs, suggesting improved reliability.