Blind source separation of rotating machinery fault signals based on fractional lower order statistics

Vibration signals generated from rotating machinery fault sources are essentially random processes consisting of large amount of fault impulses, which belong to a special class of statistical distribution called α stable distribution. In this paper, a blind source separation method based on the utilization of fractional lower order statistics derived from the α stable distribution is proposed for the separation of rotating machinery fault signals. Simulation experiments based on the application for the separation of bearing fault signals have shown that the proposed approach has great potential for the separation of rotating machinery fault signals.