Fault Early Diagnosis of Rolling Element Bearings Combining Wavelet Filtering and Degree of Cyclostationarity Analysis  

作　　者：ZHOU Fu-chang（周福昌） CHEN Jin（陈进） HE Jun（何俊） BI Guo （毕果） LI Fu-cai （李富才） ZHANG Gui-cai （张桂材） 

机构地区：[1]State Key Lab. of Vibration, Shock and Noise, Shanghai Jiaotong Univ. , Shanghai 200240, China

出　　处：《Journal of Shanghai Jiaotong university(Science)》2005年第4期446-448,455,共4页上海交通大学学报（英文版）

基　　金：National Natural Science Foundation ofChina(No.50175068) and the Key Project sup-ported by National Natural Science Foundationof China(No.50335030)

摘　　要：The vibration signals of rolling element bearing are produced by a combination of periodic and random processes due to the machine’s rotation cycle and interaction with the real world. The combination of such components can give rise to signals, which have periodically time-varying ensemble statistical and are best considered as cyclostationary. When the early fault occurs, the background noise is very heavy, it is difficult to disclose the latent periodic components successfully using cyclostationary analysis alone. In this paper the degree of cyclostationarity is combined with wavelet filtering for detection of rolling element bearing early faults. Using the proposed entropy minimization rule. The parameters of the wavelet filter are optimized. This method is shown to be effective in detecting rolling element bearing early fault when cyclostationary analysis by itself fails.The vibration signals of rolling element bearing are produced by a combination of periodic and random processes due to the machine＇s rotation cycle and interaction with the real world. The combination of such components can give rise to signals, which have periodically time-varying ensemble statistical and are best considered as cyclostationary. When the early fault occurs, the background noise is very heavy, it is difficult to disclose the la- tent periodic components successfully using cyclostationary analysis alone. In this paper the degree of cyclostationarity is combined with wavelet filtering for detection of rolling element bearing early faults. Using the proposed entropy minimization rule. The parameters of the wavelet filter are optimized. This method is shown to be effective in detecting rolling element bearing early fault when cyclostationary analysis by itself fails.

关 键 词：CYCLOSTATIONARY degree of cyclostationarity fault diagnosis rolling element bearings 

分 类 号：TH13[机械工程—机械制造及自动化]