Fault diagnosis of rolling bearing based on TKEO-ELM

作者：Ma Jun; Wu Jiande; Wang Xiaodong

来源：UPB Scientific Bulletin, Series D: Mechanical Engineering , 2017, 79(4): 215-226.

• Accuracy of classifications
• Back propagation neural networks
• Classification models
• Energy operators
• Extreme learning machine
• Fault diagnosis method
• Rolling bearings
• Time domain features

Aiming at the issue that the weak vibration signal of early failure in rolling bearing is easily submerged by noise and the time domain feature cannot guarantee the recognition accuracy of the classification model, a fault diagnosis method of rolling bearing based on Teager-Kaiser Energy Operator and Extreme Learning Machine (TKEO-ELM) is proposed. Firstly, the formation mechanism and envelop demodulation will be analyzed. Then, transform the vibration signal from the time domain to Teager-Kaiser domain (TK domain), and extract the TK domain feature to denote local signal characteristics. Finally, establish the classification models of BP neural network, Support Vector Machine and ELM, and compare the performance of classification models using the experimental data of rolling bearing. The experiment results show that the proposed method can ensure the accuracy of classification recognition, decrease the amount of optimized parameters and shorten the optimized time of the model under the same condition.

• 出版日期2017