Fault Feature Signal Extraction Method of Rolling Bearing Based on GA-VMD

To solve the problem of difficulty in bearing fault extraction, a VMD parameter optimization method with the envelope entropy and kurtosis as the comprehensive objective function is established in this paper. The genetic algorithm is used to compute the minimum value of the integrated objective function to obtain the optimal number of modal decompositions and the value of the penalty factor. The VMD method optimized by GA is used to obtain the IMFs of the simulated and measured signals, and the sensitive component of IMFs is selected based on the method with the maximum value of correlation kurtosis and then analyzed by Hilbert envelope spectrum. The results show that the VMD method based on genetic algorithm optimization can effectively extract the fault feature signals.