Research on vmd-fastica in gearbox fault diagnosis

In order to make up for the disadvantage that ICA can only be applied to the blind source separation problem when the number of observation sources is not less than the number of signal sources, this paper uses VMD decomposition method to reconstruct the signal and the introduced virtual noise channel as the input matrix of ICA, so as to effectively solve the underdetermined problem of single channel ICA.

The implementation steps of VMD FastICA are as follows:

Step 1: collect the vibration signal of the gearbox and set the parameters in the VMD program, in which the initial mode number k = 5, the penalty factor α = 2000 and the bandwidth τ = 0;

Step 2, the denoised signal of MCKD is loaded into the VMD program for decomposition;

Step 3: observe the center frequency of each IMF component. If the center frequency is similar, then k = K + 1, otherwise k = k-1, continue with step 2;

Step 4, the maximum frequency range of the original signal and each IMF component is calculated by fast spectral kurtosis, and the correlation coefficient of each IMF component is calculated;

Step 5, the two channels obtained in step 4 are used as FastICA inputs, which are unmixed to obtain the signal after joint noise reduction;

Step 6, the signal envelope after joint noise reduction is demodulated to judge the fault.

spacer