Rolling Bearing Fault Diagnosis Based on Deep Residual Network
In view of the difficulties in extracting eigenvalues and poor generalization of common deep learning-based rolling bearing fault diagnosis methods under varying working conditions,a rolling bearing fault diagnosis model based on continuous wavelet transform and deep residual network was proposed.The collected original vibration signal is transformed into a two-dimensional time-frequency graph by continuous wavelet transform(CWT).Group Convolution has the advantage of reducing the complexity of models,and it replaces the standard convolution layer of traditional residual networks.The anti-jamming capability of the model is improved by adding the Dropout layer and using the global average pooling layer to avoid overfitting.The experimental results show that the recognition accuracy of this model reaches 98.33％.The fault diagnosis accuracy of rolling bearings under variable working conditions is more than 90％.
万方数据
维普
