Wind turbine gearbox fault diagnosis based on time-shift multi-scale attention slope entropy and SCA-SCN
WANG Jiansheng
SHI Kaidong
ZHANG Tingting
WANG Bin
WU Fengjiao
Abstract：[Objective]Planetary gearbox is an important equipment in wind turbine drive train.In order to improve the recognition accuracy of vibration signal of gearboxes under different working conditions,a fault diagnosis method based on time-shift multi-scale attention slope entropy(TSMASE)and improved stochastic configuration network(SCN)was proposed.[Methods]Firstly,to overcome the drawbacks that the parameters of slope entropy need to be adjusted according to prior knowledge,which is difficult to achieve the optimal setting,the attention slope entropy was developed.Secondly,to solve the problem of insufficient coarse granulation of traditional multi-scale entropy,multiple improved multi-scale methods were compared,and the time-shift multi-scale attention slope entropy was developed.The effectiveness,noise resistance and robustness on short time length of the proposed feature extraction method were proved by the simulation platform data.Finally,by inputting the features extracted from TSMASE into the stochastic configuration network model improved by sine cosine algorithm(SCA),the TSMASE-SCA-SCN model was constructed.[Results]The results show that the model has a proposing classification effect,and the diagnostic accuracy reaches 99.80%on gearbox datasets.
Keywords：Attention-based slope entropySine cosine algorithmStochastic configuration networkWind turbineFault diagnosis
Publication Date：2026-05-31
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：13( 63-75 )