TR2026-158

Adaptive Observer-Based Detection of Inter-Turn Short-Circuit and Demagnetization Faults in Permanent Magnet Synchronous Motors

    •  Yang, G., Wang, Y., Liu, D., Lin, C., Wang, B., "Adaptive Observer-Based Detection of Inter-Turn Short-Circuit and Demagnetization Faults in Permanent Magnet Synchronous Motors", IEEE Industrial Electronics Society (IECON), October 2026.
      BibTeX TR2026-158 PDF
      • @inproceedings{Yang2026oct2,
      • author = {Yang, Ge and Wang, Yebin and Liu, Dehong and Lin, Chungwei and Wang, Bingnan},
      • title = {{Adaptive Observer-Based Detection of Inter-Turn Short-Circuit and Demagnetization Faults in Permanent Magnet Synchronous Motors}},
      • booktitle = {IEEE Industrial Electronics Society (IECON)},
      • year = 2026,
      • month = oct,
      • url = {https://www.merl.com/publications/TR2026-158}
      • }
  • MERL Contacts:
  • Research Areas:

    Dynamical Systems, Electric Systems

Abstract:

This paper presents an adaptive observer-based method for detecting inter-turn short-circuit (ITSC) and permanent-magnet (PM) demagnetization faults in permanent magnet synchronous motors (PMSMs). The effects of these faults are incorporated into the PMSM dq-frame current model, where the ITSC fault is represented by equivalent fault-current compo- nents and the demagnetization fault is reflected by a variation in the PM flux linkage. An adaptive observer, driven by measured dq-axis voltages, stator currents, and electrical speed, estimates the fault-current components and the PM flux linkage online. Two residuals are then constructed from the estimated fault-current magnitude and the deviation of the estimated flux linkage from its nominal value, enabling discrimination among ITSC, demagneti- zation, and combined-fault scenarios. MATLAB/Simulink studies under load disturbances, demagnetization, ITSC, and sequential combined faults validate the residual behavior and demonstrate robustness against non-fault load variations.