Analysis Of Rotor Imbalanced Characteristics Of Wind Turbine And Research On Fault Diagnosis By Reverse Reasoning

In response to global climate change,the wind power industry,as the most important part of new energy power generation,has developed rapidly in recent decades.At present,wind turbines continue to develop towards large-scale development.The single-unit power generation capacity has reached more than 10 MW,and the volume of the unit has also become larger.Wind turbines are the power source of wind turbines.The diameter of the wind turbines of largecapacity turbines has exceeded 100 meters.Due to the harsh operating environment of wind turbines,blades may become unbalanced,such as icing on the blades.Disposal,it may cause unit damage or economic loss.Therefore,by carrying out characteristic analysis and fault diagnosis research on the imbalance of the wind turbine generator,and discovering the fault in time,the safe and stable operation of the wind turbine generator can be guaranteed.This article first builds a 3MW wind turbine simulation model based on the GH Bladed simulation software based on wind turbine aerodynamic theory,transmission structure and vibration form,and sets the simulation unit model according to the type of wind turbine imbalance fault,including the rotor mass imbalance 10%.Eight kinds of operating conditions,collecting wind turbine simulation signal data.Then take the wind turbine blade load signal and the tower top load signal as examples to analyze in detail the time domain characteristics of the unit signal under different working conditions of the wind turbine imbalance.It can be seen that the impact of the wind turbine aerodynamic imbalance fault on the wind turbine is in the time domain characteristics of the unit signal The performance is more obvious than the mass imbalance fault,and the time domain characteristics of the unit signal when the wind wheel coupling imbalance fault is similar to the aerodynamic imbalance fault,but the impact on the unit is more serious than the latter;take the engine room vibration signal as For example,based on the fast Fourier transform and wavelet decomposition algorithm,the frequency domain characteristics of the unit signal under various working conditions are analyzed in detail,and it can be seen that the amplitude of the unit frequency domain signal will be changed when the wind turbine quality,aerodynamic,or coupling unbalance fault occurs.Obvious increase or decrease,the fault characteristics are obvious,and the same kind of imbalanced faults can be distinguished in detail through the detailed frequency domain signal after decomposition.The time domain and frequency domain characteristics of the signals of each part of the other ten kinds of units are given.Through the analysis of the time domain and frequency domain characteristics of various unit signals,it can be obtained that the impact of wind turbine imbalance fault on each part of the unit is different,and the time domain or frequency domain characteristics of the corresponding unit signal of each part are also different.Therefore,the unit signal with obvious time-domain or frequencydomain characteristics can be selected as the basis for the diagnosis of wind turbine imbalance faults.Finally,based on Particle Swarm Optimization-Support Vector Machine(PSO-SVM)to build a wind turbine imbalance fault diagnosis algorithm,select the unit signal with obvious signal characteristics under the fault conditions for example to extract the fault feature vector,As a training sample to train and classify the fault diagnosis algorithm,obtain the detection sample corresponding to the fault type,and verify the accuracy of the fault diagnosis at the same time;finally use the corresponding signal under the approximate imbalance condition with the fault type as the diagnosis sample to verify the fault diagnosis algorithm The ability of reverse reasoning fault diagnosis.