Local demagnetization fault diagnosis of permanent magnet synchronous motor based on half-period back EMF residual

Accurate quantitative diagnosis of permanent magnet demagnetization fault is essential for prevention of catastrophic motor fault and formulation of system operation and maintenance strategies. In order to realize the demagnetization permanent magnet location and degree evaluation, a local demagnetization fault diagnosis method was proposed based on half-period back electromotive force (EMF) residual. In the method, a search coil was installed in the stator slot to extract its EMF residual signal and partition it; the partitioned EMF residual peak value was used as a characteristic quantity for demagnetization fault type identification, which in turn determines the healthy conditions of two adjacent permanent magnets and the demagnetization degree, using the signal of one rotation period to determine the status of all permanent magnets. The proposed method is less invasive, less computationally intensive and highly accurate. The installation and detection mechanism of search coil was introduced, and the mathematical model of search coil half-period back EMF residual was established under demagnetization fault of permanent magnet synchronous motor (PMSM). Then, an algorithm was proposed to extract the EMF residual considering the branch circulating current, which effectively improves accuracy of the search coil EMF residual extraction. Finally, the mapping relationship between the EMF residual peak value and the location and demagnetization degree of demagnetization permanent magnet was established, and an algorithm for location and degree evaluation of demagnetization permanent magnet was proposed to achieve accurate quantitative diagnosis of local demagnetization fault in PMSM. Simulation and experimental results verify correctness and accuracy of the proposed method. © 2023 Editorial Department of Electric Machines and Control. All rights reserved.