Condition Monitoring of Rolling Bearings using Statistical Linguistic Analysis

Abstract

Defective rolling bearings generally provoke a demonstration of nonstationary and nonlinear properties. As a result, condition monitoring of a rolling bearing seems challenging due to difficulties in fault feature extraction. This study introduces statistical linguistic analysis (SLA) to investigate rolling bearing vibration data. By SLA, original vibration data are allowed to be distilled into a rank index sequence, which preserves fundamental dynamics hidden in the original data. Afterwards, a correlation coefficient is defined for detecting a change of conditions of rolling bearings. Consequently, this study develops a novel method for condition monitoring or rolling bearings using SLA. Moreover, the feasibility of the proposed method is assessed by using a set of full-lifecycle vibration data from a realistic rolling bearing. The results showed that the proposed method has the capability of detecting a change of running conditions of rolling bearings.