Jixie chuandong (Jan 2017)
Fault Diagnosis of Rolling Element Bearing based on Angular Domain Empirical Wavelet Transform
Abstract
Considering the difficulty of extracting fault features of rolling element bearings,a fault diagnosis method for rolling element bearings under variable speed conditions based on angular domain empirical wavelet transform（ EWT） is proposed. Firstly,original rolling element bearing vibration signal is re-sampled with even-angular sampling so that the non-stationary signal in time-domain could be transformed into a stationary one in angular-domain. Then,the stationary signal is decomposed adaptively into some empirical mode components by using the empirical wavelet transform（ EWT）. Finally,the envelope spectrum analysis is carried on the empirical mode component that kurtosis value is maximum to extract the order ratio fault features. In order to promote the decomposition efficiency of EWT,the spectrum segmentation of EWT is improved. The rolling element bearing experimental signal analysis results show that the proposed method can effectively inhibit the influence of noise,other interference factors and extract the order ratio fault features accurately,it provides an effective method for rollering bearing fault diagnosis under variable speed.
