Fault diagnosis of motor bearing based on improved convolution neural network based on VMD

To improve the diagnostic accuracy of motor bearing fault diagnosis under complex conditions, an ensemble approach based on variable mode decomposition (VMD) and convolutional neural network(CNN) is proposed. Firstly, after the motor bearing fault signal are collected, the data of the motor bearing is denoised by VMD, and the data of the denoised data is extracted and diagnosed by CNN. The experiment shows that the accuracy of the VMD-CNN for motor bearing fault diagnosis is better than CNN.