Abstract
Based on insufficient fault data of permanent magnet motor, it is difficult to identify and analyze the fault of permanent magnet motor. Aiming at the problem of insufficient fault data and difficult fault diagnosis of a permanent magnet motor, a fault identification method is designed to improve the stability and safety of a permanent magnet motor. Firstly, for the small sample problem of stator current fault of permanent magnet motor, DCGAN is used to generate virtual fault data and expand the training sample. Secondly, RCCNN model is used for feature extraction and classification of stator current data to realize fault diagnosis. Finally, the original data and expanded data are used to verify RCCNN, which proves that the expanded data can effectively improve the fault diagnosis ability of RCCNN. The experimental results show that by using DCGAN to expand the training samples, fault diagnosis accuracy of RCCNN permanent magnet motor is improved by 13.1%, which verifies the accuracy of the model.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have