Time-frequency spectrum based on iterative generalized demodulation for gearbox fault diagnosis under nonstationary conditions

• Published in: 2016 IEEE International Instrumentation and Measurement Technology Conference Proceedings pp. 1 - 5
• Main Authors: Xiaowang Chen, Zhipeng Feng, Kangqiang Li
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2016
• Subjects: Demodulation; Fault diagnosis; Frequency modulation; Gears; generalized demodulation; nonstationary; Shafts; Time-frequency analysis; Transforms
• DOI: 10.1109/I2MTC.2016.7520334

Extraction of time-variant characteristic frequency components is an important topic for gearbox fault diagnosis under nonstationary conditions. Time-frequency analysis is capable of revealing time as well as frequency information simultaneously on time-frequency spectrum. However, conventional linear time-frequency analysis methods are subject to Heisenberg uncertainty principle, and bilinear representations suffer from cross term interferences. Hilbert transform is capable of accurately estimating the instantaneous frequency of nonstationary signals, yet it requires the analyzed signal to be mono-component. In this paper an improved version of generalized demodulation, i.e., iterative generalized demodulation is used to decompose multi-component nonstationary signals into monocomponents. Hence time-frequency spectrum based on iterative generalized demodulation, which is of fine time-frequency resolution and free from cross term interferences, is achieved. Both numerical simulated and lab experimental gearbox vibration signals are analyzed, and gear fault is successfully diagnosed.