Rolling Bearing Fault Diagnosis Based on Model Migration

A rolling bearing fault diagnosis method based on deep transfer learning was proposed to solve the problems of low efficiency of rolling bearing fault classification under variable working conditions, complex model and traditional machine learning that could not adapt to weak calculation and less label. Firstly, the preprocessed data is used as the input layer of the one-dimensional convolutional neural network, and the learning rate multi-step attenuation strategy is used to train the model and construct the optimal model. Secondly, the optimal model is used to complete the rolling bearing fault classification in the target domain. Finally, compared with the ResNet model and TCA algorithm, the experimental results show that the proposed method has higher fault diagnosis accuracy than the ResNet model and TCA method, and is an effective method for automatic fault feature extraction and classification recognition.