A Fault Diagnosis Method Based on NTFES-FCCT for Variable Working Condition Bearing Signals

The vibration signals of rolling bearings under the influence of strong background noise and large fluctuation of rotational speed often show the characteristics of strong nonstationarity and spectral dispersion. These characteristics will bring about problems such as the failure of traditional filtering methods, large errors, and low efficiency of order analysis methods. A fault diagnosis method of normalized time-frequency entropy spectrum (NTFES) combined with a fault characteristic coefficient template (FCCT) for variable operating conditions is proposed in this work. First, the time-frequency distribution matrix is obtained by the short-time Fourier transform (STFT) of the vibration signal, and the entropy of the spectrum on each time window is calculated in turn and normalized to get the NTFES. Second, the instantaneous frequency change curve (IFCC) is extracted from the NTFES using the multiple time-frequency curve extraction (MTFCE). Then, the rotational frequency change curve (RFCC) is obtained by fitting the calculation to the simultaneously acquired rotational speed pulse signal. The instantaneous fault characterization change line (IFCCL) is obtained by calculating the ratio of IFCC to RFCC. Finally, the standard fault characterization coefficients are calculated according to the relevant parameters of rolling bearings, and the error analysis of IFCCL and standard fault characteristic change line (SFCCL) in different modes is carried out to realize the identification of bearing fault types under variable speed conditions. The simulation analysis and two real cases verify the effectiveness of the proposed method.