The bearing health condition assessment method based on alpha-stable probability distribution model

In this paper, an efficient bearing health condition assessment approach is proposed based on alpha-stable distribution model. The health condition assessment index is constructed with Kullback-Leibler distance (KLD) between two alpha-stable distributions, which are statistical models of vibration signals in normal condition and fault condition. Experimental results based on the run-to-failure bearing test data have shown that KLD index based on alpha-stable distribution model can efficiently assess the health condition of tested bearings. The bearing performance assessment approach has a great potential for industrial applications.