An optimal hybrid model for atomic power generation prediction in Japan

The grey model GM(1,1), which is based on grey system theory, has become a powerful tool for the prediction problems in power systems. However, the prediction accuracy of grey model is unsatisfying when original data set shows great randomness. In this paper, in order to improve the prediction capability of grey model, the exponential smoothing (ES) method is integrated into GM(1,1) through the preprocessing for original data set. We call the proposed model as ESGM(1,1). The Taylor approximation method is then presented to find the optimal coefficient values of ESGM(1,1). The improved model is defined as T-ESGM(1,1). Finally, Markov chain model is applied to T-ESGM(1,1) for achieving the high prediction accuracy. We call the proposed model as MC-T-ESGM(1,1). A real case of atomic power generation in Japan is used to validate the effectiveness of proposed model.