Research on the combined predicting model of short-term load in smart grid

Abstract

As an important basis of power grid planning and dispatching, short-term load predicting model with high accuracy is very important to ensure the efficient and reliable operation of power grid. In this paper, the influencing factors of the short-term load of smart grid are determined by the method of grey correlation analysis. BP, RBF and Elman neural network construct the single prediction model of the short-term load of smart grid. The single prediction model is weighted by GA genetic algorithm, and the combined prediction model of the short-term load of smart grid is constructed and verified by an example. The results show that the error of the combined prediction model can be kept at about 0.4%, which has higher prediction accuracy and stability.