A Neural Network Method for Bearing Fault Diagnosis

To solve the problem of low accuracy of rolling bearing fault diagnosis under complex noise and variable load conditions, this paper proposes a neural network based solution SSRNet. First, the rolling bearing signal is preprocessed by short-time Fourier transform, and the model structure and residual structure of the neural network are adjusted, and LeakyReLU function is integrated into it. The accuracy of rolling bearing fault diagnosis is improved under the condition of complex noise and variable load. At the same time, the data set of Case Western Reserve University is used for experimental verification. In the SNR of - 4dB, the SSRNet model proposed in this paper can achieve 97.11% accuracy and has better performance than the existing methods.