Failure libration model of bearing and its fault diagnosis based on EMD

Aiming at the problem that the bearings of antenna drive organ on vessel easily occur failure, failure libration model of bearing and its fault diagnosis based on EMD are researched. Failure libration models of the inner circle, outer circle and roll body of the bearing are discussed respectively. Based on the models, fault diagnosis method of bearing based on EMD is presented. First, original libration signal is decomposed using EMD. Decibel value of Hilbert envelope spectrum for each IMF is calculated. Then, Decibel value of Hilbert envelope spectrum at characteristic frequency of bearing is extracted for each IMF, and the maximal decibel value is consequently obtained. Lastly, the state of the bearing is identified by shock pulse method. Experimental results show that the method can diagnose the location and degree of bearing fault.