Research On Fault Diagnosis Method Of The Wind Turbine System

In recent years,with the continuous increase of wind power installed capacity in China,the cumulative installed capacity of wind power in China has ranked first in the world.During the operation of wind turbines,a large number of wind power accidents have caused widespread concern.According to statistics,the downtime caused by the failure of the drive system is the longest,and it causes huge economic losses.As an important part of the wind turbine drive system,the rolling bearing has real-time monitoring and analysis of its operating state,which is of great significance to the operation and maintenance and fault diagnosis of the entire power generation system of wind turbine.This topic takes the rolling bearing in the wind turbine drive system as the research object.The main research contents are as follows:Firstly,Chapter 2 researches the basic structure of the wind turbine rolling bearing,and then systematically analyzed the fault form of the rolling bearing,the vibration mechanism of the rolling bearing,the fault characteristic frequency and the natural frequency of each component of the rolling bearing.Aiming at the failure of wind turbine rolling bearing faults without historical fault data,Chapter 3 presents fault diagnosis method for wind turbine rolling bearing based on improved EEMD-Hilbert envelope demodulation.On the basis of EEMD,Pearson correlation coefficient is introduced to eliminate the pseudo component and interference component generated in EEMD decomposition.The improved signal analysis method combining EEMD and Hilbert envelope demodulation is applied to the fault diagnosis of rolling bearings.Taking the experimental data of rolling bearing published by the Western Reserve University as an example,the simulation experiment is carried out.The simulation experiment proves the simplicity of the method and the feasibility and rapidity of fault diagnosis of the inner and outer rings of the rolling bearing.In order to further improve the accuracy of wind turbine rolling bearing fault diagnosis,in the background of wind turbine supervisory control and data acquisition(SCADA),the fault diagnosis method of wind turbine rolling bearing based on multi-feature fusion and boosting decision tree is proposed.Taking the experimental data of rolling bearing test published by the Western Reserve University as an example,the simulation result shows that the multi-feature fusion and boosting algorithm fault diagnosis algorithm can extract features with high discrimination and independence,and superior to support vector machine(SVM),k-nearest neighbor(KNN)and artificial neural network(ANN)in fault diagnosis rate of the rolling bearing.Considering that the traditional feature extraction method can not fully exploit the potential information in the rolling bearings vibration signal,a fault diagnosis method based BiGRU neural network for rolling bearings is proposed.the fault diagnosis simulation experiment was carried out by using BiGRU network model,and the proposed algorithm has the ability of adaptive feature extraction and the feasibility of processing time series problems.The algorithm is fast and accurate in the fault diagnosis of rolling bearings.SVM,KNN,ensemble learning and other algorithms are used to compare the experimental results of rolling bearing fault diagnosis.