Abstract: Under variable speed conditions, the fault diagnosis of rolling bearings in heavy-duty equipment such as cranes and wind turbine was not accurate enough. Aiming at this problem, a fault diagnosis method based on robust local mean decomposition (RLMD) and wavelet-based synchro extracting transform (WSET) were proposed. Based on vibration signal analysis, the fault characteristic frequency of rolling bearing under variable speed was studied, and the formulas of RLMD and WSET were deduced. RLMD was used to decompose the vibration signal of rolling bearing, the best component was selected according to the principle of maximum cross-correlation coefficient, then WSET was performed on the selected components to obtain the time-frequency representation of energy concentration. The fault characteristic curve was extracted from the time-frequency plane and compared with the theoretical fault characteristic frequency curve, so as to perform fault diagnosis under variable speed. The numerical simulation experiment of multi-component signal was studied and designed, and the fault diagnosis of variable speed rolling bearing in bearing gear fault comprehensive test-bench was carried out. The research results show that the proposed method can eliminate the influence of noise and obtain accurate fault characteristic frequency curve, which is effective in variable speed rolling bearing fault diagnosis.

Key words:  rolling bearing; fault diagnosis; variable speed conditions; robust local mean decomposition (RLMD); wavelet-based synchro extracting transform (WSET)