The Transient Area Power Control Of Wind Turbine Based On Self-studying Algorithm

In order to optimize the control strategy of transient area,this thesis has built a BP neural network model which can predict the operational parameter s of the turbine after a short term based on the real-time data.And propose a kind of self-studying algorithm based on that model to optimize the original PID control.The conclusions are as follows.First,the wind turbine's parameters of the next 0.5 seconds can be accurately predicted by the use of BP neural network.Second,the transient area control of wind turbine can be optimized by the self-studying algorithm presented in this thesis.Third,the self-studying algorithm based on only one fixed BP neural network has a certain applicability to different wind regime.The last,to retrain the BP neural network used in the self-studying algorithm can improve accuracy of the prediction and the optimization.