Fault Diagnosis of Rotating Machine Based on Denoising Source Separation
Wangqun Deng · Noise and Vibration Control · 2013
Vibration signal of rotating machines is an important information source for fault identification and diagnosis.The denoising source separation(DSS) technique which separates the mixed signals by using statistical characteristics was applied in the fault diagnosis of the rotating machines.The basic theory of DSS and its tangential denoising function were studied.Then the analogous signals were separated.The results show that the performance index and the correlation coefficients of DSS are better than those of blind source separation.Applying the DSS method to the fault diagnosis of a gas turbine,the unbalance and pseudo resonance phenomenon of the rotor were diagnosed through the measured fault signals.It shows that the DSS method is efficient for fault diagnosis of rotating machines.This work provides new ideas and methods for condition monitoring and fault diagnosis of rotating machines.