Jixie qiangdu (Jan 2022)
APPLICATION OF IMPROVED EXPERIENCE WAVELET TRANSFORM IN FAULT DIAGNOSIS OF WIND TURBINE GEARBOX
Abstract
Aiming at the problems that the minor fault features of wind turbine gearbox bearings are easily submerged and difficult to extract under the background of strong noise, a continuous average spectral negative entropy（CASN） is proposed to improve the empirical wavelet transform（EWT）. The CASNEWT method is used to decompose the fault signal of the wind turbine gearbox bearing, and then the obtained components are filtered and reconstructed by the spectrum negentropy criterion, and the reconstructed signal is analyzed by envelope analysis to accurately extract the fault characteristics. Finally, a feature vector set is formed and input to the support vector machine for fault diagnosis. The results show that the CASNEWT method retains the advantages of the EWT algorithm, which can effectively avoid the modal aliasing effect and the end effect, while greatly improving the robustness of the EWT decomposition algorithm to noise, removes the noise and retains the original signal characteristic information. Accurately extract the characteristic frequency of the fault to improve the accuracy of fault identification.
