Very short-term prediction of wind farm power: An advanced hybrid intelligent approach

This paper presents a new hybrid intelligent technique for very short-term wind power forecasting (VSWPF) based on the combination of wavelet transform (WT), similar day (SD) method, and emotional neural networks (ENN), i.e., WT+SD+ENN. The forecasting procedure using the proposed hybrid WT+SD+ENN intelligent model involves the refinement of the forecasted output obtained from the SD method by an application of ENN. The predicting performance of the proposed hybrid model is compared with the benchmark persistence method and other hybrid intelligent models in terms of mean absolute percentage error (MAPE), mean absolute error (MAE) and root mean square error (RMSE).