Local Demagnetization Fault Diagnosis of Permanent Magnet Synchronous Motor Based on CS-LDM

Demagnetization failure of permanent magnet synchronous motor (PMSM) will greatly reduce the running performance of the motor. To solve the problem of PMSM local demagnetization failure, this paper proposes a cost-sensitive large margin distribution machine (CS-LDM) for the PMSM fault diagnosis method. The CS-LDM method is improved by the traditional SVM classification model optimization, which combines the idea of cost-sensitive thinking and margin distribution effectively, at the same time, the misclassification cost is constructed for the classification model, and the margin distribution is optimized by maximizing the margin mean and minimizing the margin variance, and improve the classification accuracy of the model, so as to achieve accurate identification of PMSM local demagnetization faults. The experimental results show that CS-LDM fault diagnosis method proposed in this paper has higher accuracy by comparing several existing diagnostic methods.