Fractional-order discrete grey models for China's electricity consumption forecasting

This paper presents a fractional-order discrete grey model to accurately forecast China's power consumption. The first order accumulated generating operation is replaced with the r-order accumulated generating operation to bring the priority of new information. A fractional-order discrete GM(r)(q, 1) model is established by constructing the q order difference equation to increase the fitting precision. The optimal orders of the proposed fractional-order discrete grey model are determined by the particle swarm optimization algorithm. The calculation formula of the parameters in the proposed grey model is given by using the least squares method. The results of China's total power consumption forecasting is offered.