A Convolutional Neural Network Model-Based Approach for Multi-Fault Diagnosis of Asynchronous Motors

In view of the low accuracy of fault diagnosis for asynchronous motors, this paper constructs an stacked fault diagnosis model based on stacked autoencoder and convolutional neural network. The model uses stacked autoencoder to reduce the dimensionality and noise of the asynchronous motor fault data, and inputs the processed data into the convolutional neural network for fault diagnosis. By removing the pooling layer in the traditional convolutional neural network and using LeakyRelu activation function, the robustness and generalization of the diagnosis model are further improved, and the accurate diagnosis of fault types is realized. The experimental results show that the proposed method can effectively improve the accuracy of multi-fault diagnosis for asynchronous motors, and has excellent performance.