Bearings Degradation Monitoring Indicator Based on Segmented Hotelling T Square and Piecewise Linear Representation

Condition-based maintenance of rotating mechanics requests efficient bearings degradation monitoring. The accuracy of bearings degradation measure depends largely on degradation indicators. This research aims to extract an indicator which can efficiently characterize the degradation of bearings. To that extent, in this paper we propose a method based on the segmented discarded Hotelling T square (SDHT 2 ) with a piecewise linear representation (PLR) approach. First, we used time domain common features extracted from the bearing vibration signal to roughly describe the bearing degradation. Then, several characteristic values are used to represent the whole historical degradation process by processing the vibration signal based time domain common features through the PLR approach. The SDHT 2 values are used as the characteristic values. The degradation indicator, named (VSDHT 2 ), is a vector where each entry is the discarded Hotelling T square value of a segment. Naturally, the new indicator VSDHT 2 describes the whole degradation process history and also carries the real-time information of bearings degradation. For illustration, a benchmark data set is used in this paper. The results show that the new indicator VSDHT 2 is sensitive and monotonic during the bearings whole lifecycle, which is promising to monitor bearings degradation.