SHORT-TERM WIND POWER FORECASTING AND PREDOMINANT WIND DIRECTION USING SVM KERNEL FUNCTION

In general, forecasting of wind power is categorized into two types: (i) short-term forecasting and (ii) long-term forecasting. Short-term forecasting of wind energy affects grid reliability and market-based ancillary service costs whereas long-term forecasting offers a scheme about a particular site location. To predict the wind speed and the units wind power generation there are few attributes to be considered, one among them is wind direction. This direction helps in analyzing the power generation at the earliest. In this paper we use kernel function and SVM classifier to account the data without any data loss and to find the direction of wind which will be helpful in increasing the power production. Hence, the proposed method helps in increasing the efficiency and accuracy of the prediction algorithm