Fault Diagnosis Of Gearbox Based On Doubly Fed Induction Generator

With the development of new energy technologies,the proportion of wind power in the power grid has increased significantly,and the connection between wind farms and the power grid has become increasingly close.After wind power is connected to the grid,it will cause the main problems such as the failure of gear box and the sharp rise of input and operation cost.Artificial intelligence technology is now widely used to simulate and imitate the human brain thinking of a technical discipline.The application of this technology in wind power grid can process some abstract and nonlinear structural information,such as the fault diagnosis problem of wind power gearbox and the parameter optimization problem involved in fault diagnosis.Based on the analysis of doubly-fed fan of interconnection and using wavelet packet and fuzzy neural network for wind turbine gearbox fault conducted a preliminary analysis,and then take advantage of the improved particle swarm optimization fuzzy c-means clustering algorithm for wind turbine group classification of a fault of the unit is very good,finally aiming at the shortcomings of the fuzzy neural network,is proposed based on improved fuzzy kernel clustering algorithm of particle swarm optimization for wind turbine gearboxes make a more accurate diagnosis of fault classification.First of all,the common fault types of wind generating sets are introduced,the fault diagnosis technology used to analyze wind generating sets and the application of artificial intelligence technology in fault diagnosis of wind generating sets are introduced.Introduces the typical composition of doubly-fed wind power generator set principle and basic structure,and the wind power grid after running fault and high incidence of gearbox fault has carried on the simple introduction and analysis of the main,by using the wavelet packet fuzzy neural network method for wind turbine gearbox fault vibration signals are analyzed in a preliminary understanding.For the grid size of wind turbine is more and more big,the traditionalrepair plan big workload,low efficiency,so the improved particle swarm optimization fuzzy c-means clustering algorithm for large-scale wind turbine in the fault set a reasonable classification accurately,so as to greatly reduce the wind generating set operations staff workload,improve the efficiency of operation of maintenance.The fuzzy neural network has some defects when it is applied to the fault diagnosis of wind turbines,and the existing fault diagnosis scheme cannot make accurate and reasonable fault diagnosis results after the unknown fault occurs.Therefore,an improved fuzzy kernel clustering algorithm for particle swarm optimization is proposed and applied to the fault diagnosis of the gearbox of a wind turbine generator.By analyzing the vibration data of the gearbox collected by the actual wind farm,it is verified that the fault diagnosis method of the gearbox in this paper can not only accurately and rapidly identify the known fault,but also classify the fault in the case of unknown fault.