Failure Analysis And Maintenance Optimization Of Large Direct-Drive Wind Turbines

As the installed capacity of wind turbines continues to increase,how to improve O & M efficiency and reduce O & M costs has become the focus of the wind power industry.This paper takes large direct-drive wind turbines as the research object,analyzes the fault tree of direct-drive wind turbines based on fuzzy gray correlation analysis method,analyzes the pitch faults of direct-drive wind turbines based on SCADA data,establishes a diagnostic benchmark,and performs direct-drive wind power Optimizing the maintenance of unit components,guiding the maintenance optimization of direct-drive wind turbines.The main research contents are as follows: The main work content is as follows:(1)Fault tree analysis of direct-drive wind turbine is conducted based on fuzzy gray correlation analysis method.First,the fault tree model of the direct drive wind turbine is established,and the direct drive permanent magnet synchronous wind turbine complete machine,wind wheel system,pitch system,permanent magnet synchronous generator,yaw system,wind measurement system are established respectively.And the fault tree of the nacelle base and tower;for each subsystem,combined with the fault history statistical data,the gray correlation between each minimum cut set and the top event was analyzed for the fault tree of each subsystem by using the fuzzy gray correlation analysis method.Weak link of each subsystem,and put forward corresponding improvement measures.(2)Based on SCADA data,the performance indicators of existing direct-drive wind turbines are evaluated.First,according to the SCADA data of a normal operation of a wind farm in Hunan collected on site,the correlation and distribution between performance indicators are compared.The results show that the three performance indicators show good consistency,but the power generation ratio is more representative,and it can better reflect the power generation performance of wind turbines;then,analyze the information content of the three performance indicators from the perspective of SCADA information sources;the results show that the power generation ratio index is more suitable as a performance monitoring indicator;finally,the SCADA abnormal operation data collected through the field is used to verify the rationality of power generation ratio as a performance monitoring index.(3)Based on the SCADA data,the pitch fault analysis of the direct drive wind turbine is carried out.Based on the SCADA data collected at a wind farm in Hunan,wind speed-power,wind speed-hub speed,wind speed-pitch angle,hub speed-pitch angle,and hub speed-power curve were analyzed,and two more suitable for fault diagnosis were obtained.The theoretical curve of the performance curve;the failure scenarios of the pitch are classified according to the operating status,and finally the fault diagnosis benchmark is established.(4)Optimize the maintenance cycle of the components of the direct-drive wind turbine based on the correlation of the fault.Considering the correlation of failures between components,by introducing effective age thresholds and degradation impact factors,the maintenance cost increase caused by the degradation effects between components is evaluated,and a component maintenance optimization model considering the correlation of failures between components is established,using Method and interior point algorithm to solve the problem;using direct-drive wind turbine components as an example to verify the effectiveness of the model.