Research On The Pitch Control Method Of Wind Turbine Generator Based On RBF Neural Network

Energy is an important material basis for the existence and development of human society,with the constant development and progress of human society,the demand for energy continues to grow,and energy structure is also constantly changing.In recent years,the development and utilization of renewable energy is being paid more and more attention,and wind power generation is one of the cheapest and most promising green energy.In the wind power generation technology,variable speed constant frequency wind power technology has become one of the main directions of research and development so far,and how to improve the utilization of wind energy,obtain the maximum wind energy tracking,improve system performance and ensure stable wind output power of the system are the research hotspot and focus in the field.The wind power generation system is a complicated,nonlinear,multi-variable and coupling system.In order to improve the operating performance of the wind turbine and ensure the stability of the operation,this thesis studies the control method of the pitch system of the wind turbine with the RBF neural network algorithm.This thesis begins with a brief analysis of the variable speed constant frequency wind turbine pitch control,it elaborates on the main pitch control process and as the wind turbine operating region for a detailed analysis.On this basis,this thesis establish the mathematical model of each module for the wind turbine,and set up the corresponding simulation module.In view of the problems of the current pitch system,the thesis presents the method of variable pitch control based on RBF neural network.In this thesis,the author will combine RBF neural network with traditional PID control method to design the RBF-PID adaptive variable pitch control system,through the variable pitch system it can ensure that when the wind speed is higher than the rated wind speed,achieve wind energy capture limiting function by the pitch adjustment function,thereby ensuring the stability of the wind turbine output power.During the RBF-PID variable pitch system performance validation process,we found that Pitch implementing agencies exist dead zones,and it would bring certain risks on the precise adjustment of the Pitch Systems and security and stability operation of the wind turbine.In order to solve this problem,this thesis on the basis of designing the RBF-PID variable oar distance system.On the basis of the design of RBF-PID pitch system,this thesis design the dead-zone compensation specifically for variable pitch actuator,through the dead-zone compensator it will generate a pitch angle compensation value to compensate for the dead zone brought in to ensure the accuracy of pitch system regulation of the RBF-PID pitch systems designed by this thesis.In order to verify the beneficial effects of the dead zone compensator of the RBF-PID variable pitch control system and variable pitch actuator in this thesis,the author built simulation platform based on MATLAB/Simulink,through simulation analysis,it is proved that the study in this thesis based on RBF neural network variable propeller pitch control method has the advantages of fast response speed,high accuracy,good stability,overshoot,robustness and so on.