Research On Fault Diagnosis System Of Wind Turbine Generateor Gearbox Base On The Wavelet Packet

Gearbox is an important component of the wind turbine,and it is also the component with the highest failure rate.Its operation state directly affects the overall performance of the wind turbine.Therefore,developing the fault diagnosis of wind turbine gearbox has important practical significance to improve the utilization rate of wind energy and reduce the maintenance cost of equipment.In this paper,the gearbox of doubly fed asynchronous wind turbine generator is taken as the object of study,and the fault diagnosis is studied.Firstly,the main internal structure and energy conversion process of the wind turbine are introduced,and the fault type and vibration principle of the gear and bearing in the gear box are analyzed emphatically.Secondly,the wavelet packet transform is used to denoise the signal,wavelet packet decomposition and reconstruction,and the fault feature is extracted by combining the energy spectrum analysis method to form the fault feature vector.Finally,the feature vectors are divided into training set and learning set as input of support vector machines to identify faults.In this paper,the parameter selection of support vector machines(SVM)is still dependent on empirical selection,and particle swarm optimization(PSO)and genetic algorithm(GA)are proposed to optimize the parameters of svm.Bearing vibration data by Case Western Reserve University test bench and the actual wind farm,fault diagnosis approach based on support vector machine wavelet packet and the standard support vector machine and after parameter optimization,simulation analysis and examples,and do comparative analysis.Simulation results show that the method can effectively extract fault features,identify faults accurately,and improve classification accuracy while saving classification time.Therefore,this method has certain practicability,popularization and accuracy.