A Novel Blind Deconvolution Algorithm with Improved Fusion Index for Bearing Fault Diagnosis
In order to solve the problem that the existing blind deconvolution algorithms are vulnerable to random pulses,a new approach is proposed.In this method,the time domain and frequency domain features are integrated into a composite index named Envelope spectral kurtosis-envelope Gini(ESKEG)index.This new index is more sensitive to periodic pulses and less susceptible to random pulses.On this basis,a new blind deconvolution algorithm based on the maximum ESKEG is proposed,which uses the particle swarm optimization(PSO)algorithm to solve the filter coefficients.By comparing its results with the simulated vibration signals and experimental simulation signals,the correctness and efficiency of this algorithm is verified.It is demonstrated that the proposed PSO-ESKEG algorithm outperforms other blind deconvolution algorithms in overcoming the influence of random pulse signals when the prior knowledge of the fault is unknown.
万方数据
维普
