Study On Spatial Dispersion Of Wind Speed In Wind Farm And Its Influence On Day-ahead Wind Power Forecast

Wind energy as a clean renewable energy,more and more attention of the world.But the uncertainty of wind power makes the grid connected to the wind power negatively affect the stable operation of the power system.It is necessary to accurately predict wind power,so it is of great significance to reduce the error of wind power prediction.This thesis takes wind farm as the research object,identifies the abnormal data of wind turbine,analyzes the spatial dispersion of wind speed in large scale wind farm and its influence on the prediction error of power before day,and puts forward the equivalent average wind speed model of wind farm.In view of the large amount of abnormal data usually appeared in the data collected by wind farm,a cloud segment optimal entropy algorithm for identifying abnormal data of wind turbine is proposed.The historical operation data of wind farm,especially wind speed and power data,are the important data base for wind power prediction.A large number of abnormal data collected from wind farm are composed of abnormal data of each wind turbine,so it is necessary to identify the abnormal data of wind turbine.The algorithm is based on the entropy of cloud model to identify the abnormal data set of generating unit and separate the abnormal data.The results show that the algorithm is universal for every wind turbine with different distribution of abnormal data,and it can effectively identify abnormal data.It lays a data base for the following research of wind farm prediction before the day.It is an important method for reducing the uncertainty of wind power generator to predict the wind power.The prediction accuracy of the whole wind power which is guided by the wind speed prediction would be affected by the dispersion characteristics of wind farm.In this thesis,the mechanism of the wind power prediction error which is caused by the dispersion of wind speed space and the dispersion of wind speed-power transfer characteristics under the accurate prediction of spatial average wind speed.The errors in the evaluation of the dispersion characteristics of wind farms are defined.The relationship between the errors is revealed.It is showed by the example that the wind power prediction errors is mainly caused by the wind speed-power transfer characteristics.It would be varied in the error values with the different seasons and the different scale of the wind farm.The prediction error of full-field wind power,which is guided by wind speed prediction,is so high that it is difficult to meet the requirements of engineering application.The wind speed model of wind turbine considering wind shear and tower shadow effect and the equivalent average wind speed model of wind farm considering the spatial dispersion of wind speed in windfarm are proposed.Based on the equivalent average wind speed model of wind farm,the prediction method of wind power before day is obtained,and the prediction results are compared with that of the traditional spatial average wind speed prediction method.The results show that the method proposed in this thesis can improve the prediction accuracy obviously,and the model is simple and feasible,it can provide an important reference for improving the prediction accuracy of wind power and the optimal operation of wind farm.