Daily Maximum Electric Load Forecasting with RBF Optimized by AFSA in K-Means Clustering Algorithm


Article Preview

Electric load forecasting is an important aspect in the operation of energy market. Many researchers have tried various methods and have achieved considerable results. In this paper, we used Radial Basis Function Neural Network (RBFN) to train data and forecast daily maximum electric load of a costal city in North China. In order to have a better result, we introduced Artificial Fish Swarm Algorithm (AFSA) to optimize RBF and adjust the center of K-means clustering algorithm. Data mining techniques were also employed to select indicators with impact on electric load. By comparing the forecast values and actual data, we arrived at conclusion that RBF optimized by AFSA could produce accurate result in forecasting daily maximum electric load. We also found that climate factors (temperature, humidity and air-pressure) had significant impact on daily maximum electric load.



Key Engineering Materials (Volumes 467-469)

Edited by:

Dehuai Zeng




W. Shen and Y. S. Sun, "Daily Maximum Electric Load Forecasting with RBF Optimized by AFSA in K-Means Clustering Algorithm", Key Engineering Materials, Vols. 467-469, pp. 1225-1230, 2011

Online since:

February 2011





[1] Zhang, Caiqing et al. (2008), BP neural network optimized with PSO algorithm for daily load forecasting,. Proceedings of the International Conference on Information Management, Innovation Management and Industrial Engineering, Vol. 3, pp.82-85.

DOI: https://doi.org/10.1109/iciii.2008.195

[2] Bashir Z.A. (2009), Applying wavelets to short-term load forecasting using PSO-based neural networks, . IEEE Transactions on Power Systems, Vol. 24, No. 1, pp.20-27.

DOI: https://doi.org/10.1109/tpwrs.2008.2008606

[3] Li Wei(2007), Application of an improved ant colony clustering in the power load forecasting, Journal of Beijing Institute of Technology , Vol. 16, pp.110-113.

[4] Sun Wei(2008), Ant colony based feed forward NN short-term load forecasting model with input selection and da clustering, Proceedings - 4th International Conference on Natural Computation, Vol. 2, pp.300-304.

DOI: https://doi.org/10.1109/icnc.2008.432

[5] Wang Ye et al. (2008), "Facial expression recognition using RBF neural network based on improved artificial fish swarm algorithm. IEEE, Proceedings of the 27th Chinese control conference. July 16-18, Kunming, Yunnan, China, Vol. 27, pp.416-419.

DOI: https://doi.org/10.1109/chicc.2008.4604925

[6] Dong-Shan, Gong. (2008).

[7] Xuejun, Chen(2008), Time series forecasting based on novel support vector machine using artificial fish swarm algorithm, Proceedings - 4th International Conference on Natural Computation, Vol. 2, pp.206-211.

DOI: https://doi.org/10.1109/icnc.2008.48

[8] Li, Chuan(2006), Next-day power market clearing price forecasting using artificial fish-swarm based neural network, Lecture Notes in Computer Science, Vol. 3972 , pp.1290-1295.

DOI: https://doi.org/10.1007/11760023_187