Using Fractional Order Grey Seasonal Model to Predict the Power Generation in China

Abstract

To accurately predict the amount of power generation, the particle swarm optimization grey season model with fractional order accumulation (PSO-FGSM(1,1) model) is proposed. Seasonal indices are introduced into the new model to enhance its seasonality, and particle swarm algorithm is used to find the optimal order. In order to evaluate the performance of the proposed model, the calculation results of the Holt-Winters model are used for comparison. The experimental results show that the prediction errors of the proposed model and Holt-Winters model are 2.4% and 3.93% respectively. It is proved that the new model has better predictive performance. Finally, the new model is discussed in two specific cases, which further reflect the prediction ability of the proposed model to predict seasonal data. The accurate prediction results can provide reference for the allocation of power resources.