Demagnetization Fault Diagnosis of Permanent Magnet Synchronous Motors Based on Decoupling Observer

ABSTRACT This paper addresses the demagnetization fault diagnosis problem of interior permanent magnet synchronous motors (IPMSM) by proposing an online observation and fault classification method based on system fault decoupling transformation. By constructing a state coordinate transformation matrix, faults and disturbances are separated into different subsystems, achieving simultaneous estimation of flux linkage attenuation and offset angle. Based on the observed faults, a demagnetization fault classification method using improved beluga whale optimization (IBWO) algorithm and support vector machine (SVM) is proposed, where IBWO is utilized to optimize key parameters of SVM offline. A comprehensive fault experimental platform was established, simulating five different levels of demagnetization faults by replacing magnets. The IBWO‐SVM system was trained offline using experimental data. Finally, the effectiveness of the proposed method was validated through simulations and experiments. Results demonstrate that the proposed scheme achieves classification performance comparable to the IBWO‐LSTM system while consuming significantly lower computational resources.