Bearing Performance Degradation Assessment based on A Combination of Multi-Scale Entropy and K-medoids Clustering

Abstract

Bearing performance degradation assessment (PDA) underlies the residual useful life prediction and maintenance decision-making. In bearing fault diagnosis and PDA, a commonly used method is clustering analysis, among which K-medoids clustering is not susceptible to extreme data. The amalgamation of K-medoids clustering and Degree of Membership (DOM) is expected to give a health indicator with a determined value range for PDA. In this study a diagnostic model for bearing PDA based on Multi-scale entropy (MSE) and K-medoids clustering is proposed. Extracting the multi-scale entropy of the rolling bearing vibration signal to construct the K-medoids clustering model. The tested data are input into the model to obtain the Degree of Membership as the index to evaluate the current bearing operating state. Resultant health indicator is used to depict bearing health condition with the help of an adaptive threshold. Results on artificially induced faults and bearing run-to-failure data demonstrate the proposed method is able to track the progress of bearing faults and detect them at incipient stage.