A Novel Adaptive Fault Diagnosis Method for Wind Power Gearbox

In the noisy environment, fault characteristics of composite faults wind power gearbox are coupled with each other, which makes extraction features more difficult. order to extract faults, a new diagnosis method for is proposed in this paper, namely modified Savitzky Golay Laplacian Gaussian filter (MSGloG). The can not only solve defects that scale parameters Modified (MloG) adaptive, but also overcome problems smoothing effect too much affected by noise. Firstly, determining Laplace model filter, and using least square convolution process improve signal-to-noise ratio vibration signal. Secondly, marginal envelope spectrum entropy index measure complex characteristics. Finally, chaotic grey wolf optimization algorithm proposed, uses spectral as fitness function, purpose make MSGloG noise reduction adaptive. extracted bearing outer ring rolling elements successfully.