Fault diagnosis of rolling bearing based on PPCA and 1.5-dimensional energy spectrum

Aiming at the problem that the nonlinear and non-stationary fault characteristics extraction of the rolling bearing under the strong background noise, the fault diagnosis method is proposed by the combination of PPCA (Probabilistic Principal Component Analysis) and 1.5-dimensional Teager energy spectrum. Firstly, the PPCA of signal is carried out to reduce the dimension of signal, then the signal is reconstructed and its principal fault characteristic component is constructed, and the strong background noise is removed. Then the 1.5-dimensional energy spectrum of the reconstructed signal is analyzed to obtain the characteristic spectrum information of the bearing fault. The proposed method is adopted to analyze the simulative and experimental data of rolling bearing, the results show that compared with EEMD (Ensemble Empirical Mode Decomposition) spectrum envelope, the method combining the PPCA with 1.5-dimensional energy spectrum has certain advantages in high-order frequency extraction of rolling bearing fault. © 2018, Electric Power Automation Equipment Press. All right reserved.