Rolling Bearing Fault Diagnosis Based on Improved VMD Binary Image
Author:佚名 Author Unit: 1. School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan 411201, China; 2 .Hunan Sine Electronic Technology Company Limited, Xiangtan 411100, China.
Abstract:Because it is easy to feature signal loss during signal feature extraction by the Variational Mode Decomposition (VMD), we have proposed a method of rolling bearing fault diagnosis based on the improved VMD algorithm of binary diagram and then combined with the LeNet-5 network. It can convert the vibration signal into a binary image with distinct characteristics, and the LeNet-5 network can identify the fault. Experimental results show that the recognition accuracy of this proposed model can achieve 100%, which exceeds 97.5% of the traditional support vector machine (SVM) classiffer and 97.61% of empirical mode decomposition (EMD) binarized graph +
convolutional neural network (CNN). In the anti-noise ability test, the recognition accuracy of the model is still as high as 100%, with good noise immunity and stability.