Detection of rolling element bearing faults based on kurtogram and correlated kurtosis

Envelope analysis is an effective and widely used method of detecting localized fault in rolling element bearings. However, this method requires selecting the optimum resonance-band to give accurate results. Resonance-band selection is not always easy and practical in the field. Therefore, this paper proposes a method based on kurtogram and correlated kurtosis. Correlated Kurtosis instead of spectral kurtosis is used to filter the raw vibration signal and then used to produce the kurtogram where the parameters are used to select optimal resonance-band for envelope analysis. Several cases of experimental bearing fault signals are used to evaluate the accuracy of the proposed method. The performance of the proposed method is also compared to other previous method, namely envelope spectrum based on spectral kurtosis. The results show that the proposed method is effective and robust in detecting local fault in rolling element bearings under heavy noise.