A hybrid EMD-SVM based short-term wind power forecasting model

This paper proposes a wind power forecasting model based on the empirical mode decomposition (EMD) and the support vector machine (SVM). In this model, the EMD is used to decompose wind power sequence into several intrinsic mode functions (IMF) and a residual component. Then, the SVM is used to train each component for the optimal parameters and kernel function. Finally, sum the prediction results of each component to obtain the wind power prediction values. Compared with the traditional forecasting methods, the hybrid EMD-SVM forecasting method can effectively reduce the root mean square error and the relative error, improve the forecasting accuracy and track the change of wind power.