Axle box bearings are one of the fundamental components of train, and their failure is the most common cause of machine breakdown. Fault diagnosis can effectively ensure the safe operation of axle bearings. However, the fault diagnosis technology of train axle bearings shares the following issues: (1) In the complex operating environment of the train, the fault features of raw signal are covered by the complex noises. (2) The most fault diagnosis methods need users to interpret the result based on their professional knowledge and visual analysis. To address the above issues, this paper combines the Gini coefficient with the 1/3-binary tree to form the Ginigram, which is used to extract fault component from the raw signal. Then, squared envelope spectrum (SES) of filtered signal is used to obtain a new health indicator (HI), so as to identify the bearing fault location and degree. Finally, by taking the statistical threshold of historical data of healthy bearings, an autonomous method is proposed for bearing fault diagnosis. The experimental results of train axle bearings show that (1) Ginigram is more robust to extract the weak fault signals, compared with classical Kurtogram. (2) HI is more accurate in identifying the fault location and degree, compared with the classical indicator.
Autonomous Fault Identification Method of Train Axle Bearings Based on Ginigram and Squared Envelope Spectrum
Lect. Notes Electrical Eng.
International Conference on Electrical and Information Technologies for Rail Transportation ; 2021 October 21, 2021 - October 23, 2021
Proceedings of the 5th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2021 ; Chapter : 58 ; 516-523
2022-02-23
8 pages
Article/Chapter (Book)
Electronic Resource
English
British Library Conference Proceedings | 2022
|