论文信息 - A Combined Middle-Long Load Forecasting Model Base on Differential Evolution Method

A Combined Middle-Long Load Forecasting Model Base on Differential Evolution Method

Middle-long load forecasting is an important issue for power system’s plan, investment and operation. In this paper, differential evolution algorithm is used to determine the weights of the Combined load forecasting models, which is combined with several single middle-long forecasting models. The experiment results point out that the proposed model’s performance is better than any single forecasting model. It also shows that differential evolution algorithm can choose the weights of combined forecasting model effectually.

Li Li | Jian Jun Wang | Dong Xiao Niu

[1] David Zimbra,et al. Medium term system load forecasting with a dynamic artificial neural network model , 2006 .

[2] R. Storn,et al. Differential Evolution - A simple and efficient adaptive scheme for global optimization over continuous spaces , 2004 .

[3] René Thomsen,et al. A comparative study of differential evolution, particle swarm optimization, and evolutionary algorithms on numerical benchmark problems , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).

[4] R. Buizza,et al. Neural Network Load Forecasting with Weather Ensemble Predictions , 2002, IEEE Power Engineering Review.

[5] Jae Hong Park,et al. Composite modeling for adaptive short-term load forecasting , 1991 .