Abstract
The study highlighted the potentiality and effectiveness of Acoustic Emission (AE) signals monitoring along with data clustering analysis as a powerful tool applicable to the rolling elements under examination for incipient damage failure. The AE technique was applied to rollers in contact using rolling contact fatigue test-rig running under constant load and speed for detecting the incipient damage initiation and its damage location. The results demonstrated the successful use of the AE activity monitoring combination with AE source locator and AE data analyzer as a new technique for incipient damage detection. The recorded AE signals from run-to-incipient damage life testing were investigated by unsupervised clustering analysis to examine and to produce numerical validation of the results by separating AE sources data into several classes that reflected the internal structure of the data during contact. A methodology including descriptor selection, methods, procedures for numerical verification and cluster validity criteria were followed.
