Fault Diagnosis of Three-phase Full-bridge Rectifier Circuit based on Deep Neural Network

摘要

This paper is concerned with the diagnostic issue of the open-circuit fault for the three-phase full-bridge rectifier based on an autoencoder-based deep neural network (AE-DNN). Firstly, the preliminary of the AE-DNN is briefly introduced. Then, the fault diagnosis model is presented for the open-circuit fault of the three-phase full-bridge rectifier. Finally, the superiority and effectiveness of the AE-DNN based fault diagnostic system is verified by several simulation experiments. It is concluded that the AE-DNN based fault diagnostic system can automatically extract and learn the features from the raw fault data and perform more robust ability to the noisy signals.