Intelligent Diagnosis of Equipment Based on Acoustic Vibration Fusion
A single sensor detection is vulnerable to external interference or its own failure and other factors leading to poor diagnosis of rolling bearing failure has been a difficult problem in the field of intelligent di-agnosis of equipment.Aiming at the above problems,an intelligent diagnosis method based on acoustic and vibration signal fusion is proposed.Firstly,the vibration signal and acoustic signal of the rolling bearing are collected through the sensor configuration.Then,the vibration signal and acoustic signal are decomposed and reconstructed by variational mode decomposition(VMD).Furthermore,the reconstructed acoustic and vibration signals are input into a dual-channel convolutional neural network(DCNN)to achieve fault fea-ture extraction and feature fusion.Finally,the extracted and fused fault features are input into the SoftMax layer of the DCNN network for fault classification modeling.The results show that the proposed fault diag-nosis method based on acoustic-vibration signal fusion can reach 99.3%accuracy compared with the CNN fault diagnosis model based on a single vibration signal,and the fused features are more effective in distin-guishing different fault states of equipment.
NETL
NSTL
万方数据
