Title :
Fault diagnosis of gearbox based on matching pursuit
Author :
Feng, Zhi-peng ; Zhang, Jin ; Hao, Ru-jiang ; Zuo, Ming J. ; Chu, Fu-lei
Author_Institution :
Inst. of Vehicular Eng., Univ. of Sci. & Technol. Beijing, Beijing, China
Abstract :
Matching pursuit is effective in matching the characteristic structure of signals and extracting the time-frequency features directly. It is employed to analyze the vibration signals of a gearbox under healthy and faulty statuses. Based on a compound dictionary, the periodic impulses characterizing the vibration of localized damaged gears are extracted in joint time-frequency domain, and the localized gear damage is detected and located. The analysis validates the effectiveness of matching pursuit in detecting and locating localized gear damage.
Keywords :
acoustic signal processing; condition monitoring; fault diagnosis; gears; iterative methods; maintenance engineering; time-frequency analysis; vibrations; condition monitoring; fault diagnosis; gearbox; localized gear damage; matching pursuit; time-frequency domain; time-frequency features; vibration signals; Atomic clocks; Dictionaries; Gears; Matching pursuit algorithms; Time frequency analysis; Vibrations; Wavelet analysis; Fault diagnosis; Gear; Matching pursuit; Time-frequency analysis; Wavelet;
Conference_Titel :
Wavelet Analysis and Pattern Recognition (ICWAPR), 2010 International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-6530-9
DOI :
10.1109/ICWAPR.2010.5576401