Research On Converter Fault Analysis Of Doubly Fed Wind Turbine Generator

With the progress of science and technology as well as the preparation technology of wind turbine becoming more and more perfect,the capacity of wind farm is expanding constantly.However,after the wind power equipment works for a certain period of time,the maintenance and repair of its running fault has become a problem to be solved in engineering technology.At present,the wind power equipment in the doubly-fed wind turbine has been used widely.Comparing with the generator and other mechanical components in the wind turbine,the stability of wind power converters is relatively poor.During the operation,the power devices in the converter are always affected by the temperature and its surrounding environment.Under some circumstances,this may cause some problems of the operation of the converter,which is easy to lead to severe damage to the wind turbine,or even machine halt.Therefore,it is very important to study and analyze the fault diagnosis of doubly fed wind power converter to ensure the stability of the operation of grid connected wind power equipment.First of all,this essay based on the theory of doubly-fed wind power system builds the mathematic model of special double fed motor and double PWM converter,and on the basis of this model and the strategy of controlling every parts of the wind turbine generator,a Doubly-fed wind turbine grid simulation platform is built by using the efficient software called Matlab/Simulink.The final simulation result shows that the structures of the doubly-fed wind generator system can realize stable operation under normal circumstances and can simulate the open-circuit fault accurately when the problems of equipment operating occur,which provides a good foundation for further research on the problems of the switch of converter in the future.Secondly,in the research of the fault feature extraction of converter,it puts forward a kind of method based on fault characteristics of wavelet transform.For the FFT transform is a reflection of the frequency characteristics of fault signals of the time,but not suitable for analyzing non-stationary signals and the problems of local abnormal datum of time domain.By using wavelet transform,the three-phase fault current signal can be decomposed,regarding the low frequency energy of three-phase fault current value acquired by wavelet decomposition as the running state of the feature vector,which lays a good foundation for the research of fault recognition in the next chapter.The simulation result shows that the method mentioned in this essay has good separability of normal and abnormal state,so it can be very good for the rotor side converter open circuit fault classification.Finally,aiming at the problems that the training time of neural network is long and that the design of network is complex,a new method based on wavelet and rough set is proposed,which can diagnose the open problems of Doubly-fed wind rotor side converter.On the basis of the characteristics of the open-circuit fault of the rotor side converter by wavelet transform,the feature vectors extracted by the wavelet transform are used to diagnose the open-circuit fault of the converter by the discretization and attributes reduction of rough set.The simulation result shows that the method based on wavelet and the open-circuit fault diagnosis of doubly fed wind turbine rotor side converter of rough set can not only improve the rapidity of fault diagnosis but also improve the accuracy of classification and identification.Therefore,it does good to improve the efficiency of diagnosis and has a bright future.