A novel method for diagnosing rolling bearing faults based on the frequency spectrum distribution of the modulation signal

Abstract

Abstract Bearing fault diagnosis is required to monitor the running status of rolling bearings, and can greatly reduce the loss caused by rolling bearing faults. It is a very important aspect of prognostic and health management. In this paper, a new method for fault diagnosis, based on an improved fast kurtogram and novel envelope spectrum analysis, is proposed to diagnose rolling bearing faults. In the proposed method, the improved fast kurtogram method is used to select the center frequency and bandwidth of the optimal signal filter which is used to filter the raw bearing vibration signals. Then, the filtered signal is transformed to the frequency domain. Novel envelope spectrum analysis is used to analyze the amplitude distribution of the envelope spectrum waveforms in order to extract more useful features from different zones rather than the whole frequency domain. The extracted features are used to calculate the fitting ratio for diagnosing bearing faults. The proposed method is validated on the fault data of rolling bearings provided by CWRU and QPZZ-II platforms. The experimental results show that the proposed method can efficiently extract features and diagnose rolling bearing faults.