An Indirect Short-Term Wind Power Forecast Approach with Multi-variable Inputs

This paper proposes a novel approach for short-term wind power forecast, where wind speed is predicted and used to forecast wind power through a power curve obtained from historical data. With the help of the empirical mode decomposition (EMD) method, wind speed is decomposed into mean trend and stochastic component. Subsequently, p-step forecast is conducted for the two components separately. The mean trend is forecasted by polynomial regression, while the forecast for the stochastic component is carried out by the least-square support vector machine (LS-SVM) model. Instead of using idealized deterministic power curve, a power curve which is obtained from historical data using polynomial fitting is proposed to derive the relationship between wind speed and wind power. In order to evaluate the performance of the proposed method, simulation studies are carried out using the data obtained from a wind farm located in northern China. The results have demonstrated that the proposed method provides a more accurate and stable forecast compared to the traditional methods.