FEATURE EXTRACTION OF MICRO-IMPULSE FAULT USING NOVEL RESONANCE DEMODULATION BASED ON THE TIME-FREQUENCY ANALYSIS
Rolling bearing damage impulse fault signal is very weak when the fault is in early stage and very difficult to extract fault feature. Novel resonance demodulation based on the time-frequency analysis is put forward in order to purify information and extract fault feature for the micro-impulse vibration modulation signal. In this method, original signals are firstly decomposed and reconstructed by using the second generation wavelet. Then the highfrequency signal from the decomposition and restruction of original signal is analyzed by the EMD, from which several Intrinsic Mode Functions (IMFs) are obtained. Then the high-frequency components of IMFs are analyzed by the Hilbert envelop demodulation, by which the modulate fault feature is extracted exactly. Simulation and experiment analysis results indicate the validity and the practicability of the method proposed here. Research results of engineering example indicate that the method could extract the early fault feature of scathing rolling bear.
Feature extraction Resonance demodulation Time-frequency analysis
Cui Ling-li Yang Zijing Gao Lixin Zhang Jianyu Xu Yonggang
College of Mechanical Engineering and Applied Electronics Technology, Beijing University of Technology, The Key Laboratory of Advanced Manufacturing Technology, Beijing, 100124, China.
国际会议
北京
英文
385-389
2008-10-27（万方平台首次上网日期，不代表论文的发表时间）