Design Of State Monitoring System For Doubly-fed Wind Power Generator

With the shortage of energy resources at domestic and overseas,some countries were starting to pay more and more attention to the utilization of wind energy,at the same time,there were more and more countries had step their efforts to develop wind energy.Also the total filling content of wind turbine is bigger and bigger.In the development and utilization of wind energy resource contained Huarui wind power and Jinfeng Technology Company and so on.However,the wind turbine generator were generally installed in areas where abundant wind resources,such as the western mountainous areas and China's southeast coastal areas,where environment was extremely hostile,and the wind turbine worked in this environment long-time,affected by exposure to extreme weather,such as lightning,thunder,snow and so on,it would cause some hidden trouble,if the hidden trouble cannot be effectively to predicted and diagnosed in time,which would cause economic losses and negative influence.Therefore,the fault diagnosis for wind turbine has been developing continuously.This paper analyzed the possibility of electrical failure of wind turbine,a set of state monitoring system for doubly fed wind turbine was designed.The whole system was divided into hardware part and software part,the hardware part includes the enhanced STC15F2K60S2 single-chip microcomputer,synchronous sampling chip AD7606,voltage sensor,current sensor,temperature sensor,CAN bus interface,LCD,PCF8563 and so on,and hardware circuit of each part was designed.It could acquire the voltage,current data and temperature of stator and rotor for wind turbine generator in low-power consumption.In the second,the system software for the system was also designed.Software development environment was Keil ?Vision5,and each part of software process had been designed.The voltage,current and temperature acquired data by AD7606 were processed on STC15F2K60S2,if there were occurred hidden trouble,would be alarmed.For wind turbine worked in hash environments,the acquired signals contained noise in ordinary,which to reduce the influence of noise interference on fault diagnosis,an adaptive threshold wavelet analysis method was selected,which could remove the noise from the noisy signal to improve the accuracy of state monitoring.In the end,state monitoring system for doubly-fed wind power generator could be used to monitor the voltage,current and temperature of the stator and rotor of the wind power generator,when the wind turbine faults occurred,it could real-time monitoring and maintain stability of power system and reduce the loss of wind power in effective.