Compound fault detection in gearbox based on time synchronous resample and adaptive variational mode decomposition

Compound fault detection of gearboxes is an ambitious matter considering its interconnection and complication. An innovative means for compound fault detection based on time synchronous resample (TSR) and adaptive variational mode decomposition (AVMD) is put forward in this work. TSR used in the method can enhance fault signals of synchronous shaft gears by eliminating signal components independent of synchronous shaft. Therefore, the TSR is used to separate the synchronous shaft signal corresponding to the gear fault from the raw compound fault signal. Then a series of mode components are obtained by decomposing the synchronous shaft signals of all faults by AVMD. The variational mode decomposition (VMD) can overcome the mode aliasing problem of empirical mode decomposition (EMD), but the decomposition effect of VMD is affected by its parameter setting. Thus, the paper proposes an AVMD algorithm based on whale optimization algorithm (WOA). In the AVMD, the WOA is used to optimizes the parameters of the VMD. After AVMD decomposition, the correlated kurtosis of the mode components obtained by AVMD decomposition is calculated. Then the mode components with the maximum correlated kurtosis are selected to carry out envelope analysis. Finally, the compound fault feature can be found from the envelope spectrum to get the diagnosis results. In order to test the validity of the proposed method, a compound fault experiment is implemented in a gearbox. Through the analysis of the experimental data, it is proved that the method shows a good performance in the compound fault detection of gearbox.