A combined model for short-term load forecasting based on neural network and grey wolf optimization

Short-term load forecasting (STLF) plays a considerable role in the operation of power system. Many methods were proposed for STLF, but all the single methods have their own weakness. In order to take full advantages of individual models and avoid their defects, this paper proposed a new combined model applied to STLF. This combined model was composed of the extreme learning machine (ELM), radial basis function neural network (RBFNN), the least squares support vector machine (LSSVM) and the grey wolf optimization (GWO) algorithm. The GWO was used to determine the weight coefficients of the prediction results of ELM, RBFNN and LSSVM. And the weighted results of the three individual method were as the final forecasting results of the combined model. Finally, the good performance of this proposed combined model is demonstrated by the case study.