Study On Wind Speed Forecasting And Control Scheme For The Wind Power System

The wind power has gained much attention all over the word asrenewable resource and pollution-free energy resource. The controltechnologies of wind turbine generator have rapidly developed. Becauseof the good control effect of the variable speed pitch regulated windturbine controller in different wind speed, more and more researchersfocus on it. In wind power systems, accurate forecasting of wind speed isrecognized as a major contribution to effectively improve the controlresults of the system.In this paper, a further study has been made about the followingaspects of the wind speed forecasting and the wind power generationcontrol on the basis of previous studies:In wind power systems, accurate forecasting of wind speed isrecognized as a major contribution to effectively improve the stability ofthe system, which is of great importance to the control of the wind power generation. In this paper, a wind speed forecasting method based on localAR time series was put forward to improve the accuracy of wind speedestimation. After wavelet decomposition of the data， the data group withhigh similarity to the main data group was selected from the historicaldata，then their contribution degree to the model was determined usingthe euclidean distance to realize the rolling forecasting of wind speed.The results show that the proposed method could significantly improvethe precision of the wind speed forecasting.In wind power systems, high frequency wind may lead to acutechange of the controlling torque, which might lead to damage of themechanical system. In this paper, a modified control of linear quadraticGaussian (LQG) is put forward to realize the dynamic optimization of thesystem and reduce the mechanical stress. By adding adaptive parametersin LQG control and regulating their change rates, this modified methodrealizes the on-line adjustment of the objective function，thus obtainingthe optimal control strategy. While maintaining the similar control effectas the traditional PID control system, it effectively decreases themechanical load of the control system to50%or so, which is beneficialfor prolonging the service life of the system.