Conceptual Approach to Permanent Magnet Synchronous Motor Turn-to-Turn Short Circuit and Uniform Demagnetization Fault Diagnosis

Permanent magnet synchronous motors (PMSMs) play a crucial role in industrial production, and in response to the problem of PMSM turn-to-turn short-circuit and demagnetization faults affecting production safety, this paper proposes a PMSM turn-to-turn short-circuit and demagnetization fault diagnostic method based on a convolutional neural network and bidirectional long and short-term memory neural network (CNN-BiLSTM). Firstly, analyzing the PMSM turn-to-turn short-circuit and demagnetization faults, one takes the PMSM stator current as the fault signal and optimizes the variational modal decomposition (VMD) by using the Gray Wolf Optimization (GWO) algorithm in order to achieve efficient noise reduction processing of the stator current signal and improve the fault feature content in the stator current signal. Finally, the fault diagnostics are classified by using the CNN-BiLSTM, which collects advanced optimization algorithms and deep learning networks. The effectiveness of the method is verified by simulation experiment results. This scheme has high practical value and broad application prospects in the field of PMSM turn-to-turn short circuit and demagnetization fault diagnosis.