Using independent component analysis scheme for helicopter main gearbox bearing defect identification

Vibration signal analysis is the most common technique for helicopter health condition monitoring. It has been widely employed to detect helicopter gearbox fault and ensure the safe operation. Through the years, vibration signal analysis has a significant contribution to successfully prevent a number of accidents. However, vibration based bearing identification remains a challenge because bearing defects signatures are contaminated by strong background noise. In this paper, the independent component analysis (ICA) scheme was utilized to analyze vibration signals captured from a CS29 Category `A' helicopter main gearbox, where bearing faults were seeded on the second epicyclic stage planetary gears bearing. The ICA scheme could separate the multichannel signals into the mutually independent components. The bearing defect signature can be clearly observed in one of the independent components. The analysis result showed that ICA scheme is a promising method for detecting the bearing fault signatures.