Short-term Wind Speed And Power Forecasting Study Of Wind Farm Based On Time Series And Neural Network

Along with the development of wind power, the proportion of wind power in the grid is increasing too. Because of the intermittence and fluctuation of the wind speed, grid integration of big wind energy will result in some problems, such as the balance between the provide and demand of power system, the power quality and the safety of the power system. So, in order to solve those problems, we will provide the prediction of the wind power advance.If the prediction can reach a high level,which can be used to the power system and improve the competition of the wind energy.This paper introduces the theory and application of time series method and artificial neural network. Firstly, builds the basic model in time series method, in order to obtain the variables largely in forecasting result. Then apply the results of time series method to artificial neural network which can make out the quantity of artificial neural networks input variables. Furthermore, puts forward rolling method to adjust weight factors. The method solves the problem that artificial neural networks weight factors become unsuitable with time passing by. The result shows the building method has certain using value.The most important work in this paper is wind speed and power research which used time series and artificial neural network methods. According contrasting accuracy of wind speed forecasting with wind power forecasting, then points out that the forecasting accuracy exits difference between them owing to the self-regular difference between wind power and wind speed.