Short-term wind speed forecasting simulation research based on ARIMA-LSSVM combination method

A high accurate wind speed forecasting can effectively reduce or avoid the adverse effect of wind farm on power grid, meanwhile enhances the competitive ability of wind power in electricity market. In this paper, a short-term wind speed forecasting method based on auto-regressive integrated moving average (ARIMA) and least square support vector machine (LS-SVM) is proposed. The weights are calculated by the two methods, equal weight average method and covariance optimization combination forecast. Research results show that the forecast accuracy from different methods is diverse one another; even though a method can offer high forecast accuracy in total, at individual point the forecast error of this method may be larger, while combination forecasting model can avoid larger forecast error in each point, so it is favorable to improve forecast accuracy.