Aiming at the problem that the traditional bearing fault diagnosis method relies heavily on prior knowledge and expert experience in feature extraction in the face of strong noise and non-stationary signal recognition,a CWT-CNN-based centrifugal pump bearing fault identification method was proposed combining traditional signal processing methods with deep learning algorithms.The con-tinuous wavelet transform(CWT)was used to transform the original 1D vibration signal into a 2D time-frequency map with richer fault feature information,and the 2D time-frequency map was then input to the convolution layer to complete the automatic feature extrac-tion,finally fault identification was completed on the SoftMax layer.After the verification of the public bearing data set of Western Re-serve University and the centrifugal pump vibration bearing collection experimental platform built in the laboratory,the fault identifica-tion accuracy of this method can reach more than 90％.
关键词
滚动轴承/连续小波变换/卷积神经网络/故障诊断
Key words
rolling bearing/continuous wavelet transform/convolutional neural network/fault diagnosis
维普
万方数据