Research On Pitch Angle Fault Diagnosis Technology Of Large Wind Turbine Generator System

With the optimization of global energy structure,the proportion of clean energy represented by wind has increased rapidly,the number and capacity of installed wind turbines have increased rapidly.However,the development speed of wind turbine fault diagnosis technology is far behind the development speed of the industry.Pitch system is the subsystem with the highest failure rate of wind turbine,once the system fails,it will seriously endanger the safe,reliable and economic operation of the unit.In this thesis,the pitch angle fault in the wind turbine pitch system is studied,aiming to explore a new method that can use the existing data of SCADA system to accurately and quickly identify the pitch angle fault,and develop an online monitoring and diagnosis system of pitch angle fault.The main contents and achievements of this thesis are as follows:Firstly,by analyzing the structure and operation characteristics of the wind turbine and its pitch system,it is found that the dynamic relationship between these parameters reflects the operation state of the wind turbine.In this thesis,the common fault types of the wind turbine pitch system are selected as the research object,and the pitch fault information of the wind turbine subsystem is further classified.Second,by comparing the average method,the least square method and the non-parametric kernel density estimation method,the non-parametric kernel density estimation method is selected for SCADA raw data processing.By comparing the Relief series of feature parameter extraction algorithms,the Relief-F algorithm which is most suitable for pitch angle state feature parameter extraction is selected,by using Relief-F algorithm to extract fault characteristic parameters,five kinds(nine)characteristic parameters of pitch angle state are obtained.Third,an improved PCA-KNN fusion algorithm for pitch angle fault is proposed.By compared with KNN algorithm,PCA-KNN algorithm and BP neural network,the results show that the method has obvious advantages in improving the accuracy of fault diagnosis results.Finally,based on the above research and by using LabVIEW software development platform,a fault diagnosis system of wind turbine pitch angle is designed and developed,which integrates data display and analysis,condition monitoring,fault diagnosis and other functions.Based on the actual wind farm SCADA data,the accuracy of the system is verified.The results show that the pitch angle fault diagnosis software system can accurately identify the operation status of the pitch system,which has a certain engineering application value.In this thesis,aiming at the pitch angle fault of wind turbine,the fault characteristic analysis,data preprocessing,feature parameter extraction,fault identification method and fault diagnosis system establishment are carried out,which improves the efficiency and accuracy of pitch angle fault identification,and has well theoretical and engineering application value.