The prediction model for electrical power system using an improved hybrid optimization model

In this paper, an improved hybrid optimization model based on grey GM (1,1) model is proposed to develop the prediction model in power systems. To realize more accurate prediction, the regression model is firstly integrated into GM (1,1) through compensation for the residual error series. The improved model is defined as RGM (1,1). Furthermore, Markov chain model is applied to RGM (1,1) to enhance the prediction performance. We call the proposed model as MC-RGM (1,1). Finally, Taylor approximation method is presented MC-RGM (1,1) for achieving the high prediction accuracy. The improved model is defined as T-MC-RGM (1,1). A real case of thermal electric power generation in Japan is used to validate the effectiveness of proposed model. 2012 Elsevier Ltd. All rights reserved.