Fault Diagnosis of Permanent Magnet Synchronous Motor Based on Improved Probabilistic Neural Network

In view of the fact that it is difficult to extract the fault characteristics of current signals of permanent magnet demagnetization and winding turn-to-turn short circuit faults and the accuracy of fault diagnosis is not high. In this paper, the fault classification method of permanent magnet synchronous motor is based on the combination of variational modal decomposition and FCM-PNN. The eigenmode function IMF is obtained by variational mode decomposition, and then the energy value of each IMF is calculated as the eigenvector. In order to overcome the problem that the complexity of the model layer is too high when the training set of PNN is too large, Fuzzy C-means clustering is used to optimize the structure of the model layer of probabilistic neural network. Finally, the current feature vector is input into the improved probabilistic neural network to obtain the classification of motor faults. The finite element simulation is used for experimental verification, and the experimental results show that the fault diagnosis accuracy of this method reaches 95%, which can effectively and accurately classify motor faults.