BEARING DEFECT SIZE ESTIMATION BASED ON SUPPORT VECTOR MACHINE AND ACOUSTIC EMISSION SIGNALS

Abstract

Rolling element bearings are widely used in rotating machinery with wide application in industry. A reliable
condition monitoring method is required to avoid unplanned shutdowns in this category of machines. The major reason for
machine breakdown is the failure of rolling element bearings. In this paper, a technique is proposed for identifying faults in
the rolling element bearings, based on an intelligent classifier technique called support vector machine (SVM). The proposed
technique is focused on outer race faults experienced by the rotating machinery. Experiments are conducted on the defective
bearing at different levels of speed and load conditions and acoustic emission (AE) data is collected for analysis. The SVM is
trained with the experimental data to identify outer race faults using Acoustic emission signals collected in real-time by a
data acquisition system. Results showed that the proposed SVM-based method was effective in identifying different outer race
faults in the Rolling element bearings.