Han Yongjun

Application of the combined model in short-term wind power forecasting

Short-term wind power forecasting plays a major role in wind energy plant operations and the integration of wind power into traditional grid systems. It is the purpose of the present paper to provide a combined model, which is composed of the FFNN (Feed-Forward Neural Network) and LS-SVR (Least Squares Support Vector Regression) model, pertaining to the short-term wind power prediction. In this proposed approach, the FFNN and LS-SVR model can offer wind power predictions using inputs processed by PCA (Principal Component Analysis) respectively and the combined forecasting method is employed to obtain the new forecasting result. Additionally, in order to optimize the LS-SVR model, CSA (coupled simulated annealing) can assist the LS-SVR model achieve optimal performance. Our results indicate that the suggested combined model improves short-term wind power forecasts in comparison with the single models in the combined model and the traditional model.