Motor Fault Diagnosis Method Based on Deep Learning
Zhen Xu, Dong Yu, Yi Hu · 2022 11th International Conference of Information and Communication Technology (ICTech)) · 2022
Motors occupies an important position in the manufacturing industry.Aiming at the problem of low accuracy of motor fault recognition,the temporal convolutional neural network(TCN) and attention mechanism are introduced. The fault diagnosis method based on TCN, by adding the SE-block module, improves the performance of the model at the cost of minimal complexity.Comparative experiments show that compared with the GRU model, LSTM model and the original TCN model,this method can better extract fault features in motor fault diagnosis, with higher accuracy and better generalization ability.