Abstract:

Aiming at the problem of poor performance of AC relay hum fault feature extraction under large background noise,a fault feature extraction method based on the parameter optimization variational modal decomposition (VMD) is proposed.First,the salp swarm algorithm is used to optimize the parameters of VMD,and the obtained optimal value is used to perform VMD decomposition of the fault signal to obtain a number of intrinsic mode functions (IMF).Then the decomposed IMF components is filtered according to the kurtosis criterion,the IMF component with the largest kurtosis is selected as the effective component,and the Hilbert envelope demodulation analysis is performed to extract the feature of the hum fault.Finally,the analysis results of the simulation signals and experimental datas show that the proposed method is feasible and effective in extracting the hum fault characteristics of AC relays and has higher application value compared to the empirical mode decomposition (EMD) method.

Key words: fault feature extraction, hum, variational modal decomposition(VMD), salp swarm algorithm(SSA), kurtosis criterion