Bi-spectrum analysis based bearing fault diagnosis

Abstract

A novel method for detection and diagnosis the bearing inner and outer race fault according to bi-spectrum analysis technique is presented. The bi-spectrum analysis is widely recognized as an effective technique for machinery fault diagnosis using vibration signals since it can be used to eliminate the effect of strong noise. This advantage makes it very suitable for extracting useful features from the noisy mechanical vibration signals in gearbox transmission system. The experimental results show that the bi-spectrum analysis technique can effectively extract the transients from strong noise signals and diagnose the fault of bearing.