Research On Fault Diagnosis Of Wind Turbine Based On Operational Data

With the rapid development of the wind power industry,the operation and maintenance of wind farms has become a concern of wind farms and many scholars.The SCADA system records the real-time operating status information of key components of wind turbines.It is an effective way to reduce the operation and maintenance costs by implementing data mining and predictive analysis to realize online monitoring and fault diagnosis of wind turbine operating conditions.First of all,this paper introduces the structure of wind turbines.Then,the operating principle,common faults and causes of each subsystem of the fan are illustrated.On this basis,the function,composition and monitoring data of the wind turbine SCADA system are introduced.Aiming at the large amount of data in the existing SCADA system,but lacking the limitations of mining and utilizing effective information,the significance of this paper is pointed out.Secondly,for the gearbox rolling bearing problem with high failure rate,the Nonlinear State Estimate Technology(NSET)method based on Mahalanobis distance optimization is used to predict the running condition of the gearbox rolling bearing.On this basis,the Variational Mode Decomposition(VMD)algorithm is used as the fault diagnosis method for the gearbox rolling bearing.The principle of the VMD method and the method for determining the number of modal decompositions are introduced.The gearbox rolling bearing fault data of Case Western Reserve University is selected for analysis.The analysis results show that the VMD method can effectively identify the fault of the gearbox rolling bearing.Finally,the SCADA data of the real gearbox rolling bearing fault case of a wind farm is selected for analysis.The analysis results are consistent with the test report results of the wind farm.This indicates that the NSET modeling method based on Mahalanobis distance optimization combined with the sliding window analysis method can achieve the effect of detecting and warning the running condition of the rolling bearing of the wind turbine gearbox,and the fault warning method based on the variational mode decomposition method can be used to achieve an effective diagnosis of gearbox rolling bearing failures.