Support Vector Machines Models Optimized by Particles Swarm for Day-Ahead Price Forecasting Based on Self-Organizing Map

摘要：

In a competitive power market, hourly electricity price varies greatly. It is hard to forecast all the hourly prices with only one model. Neither will it get excellent results with 24 different models to forecast the 24 hourly prices, for there are always not sufficient data to train these models, especially the models for summer electricity price data. This paper applies SOM (Self-Organizing Map) neural network to cluster the data automatically for modeling. SVM (Support Vector Machines) Models for regression are built on the data clustered respectively. Parameters of the SVM models are chosen by PSO (Particle Swarm Optimization) automatically to avoid the arbitrary of the decision by the tester. A case studied shows that the hybrid models have good performance in forecasting the summer hourly price.

作者: Dongxiao Niu Da Liu Mian Xing Yuanyuan Li

作者单位: School of Business Administration North China Electric Power University Beijing, China 102206 School of Mathematics & Physics North China Electric Power University (Baoding) Baoding, China 07100

会议类型: 国际会议

会议名称: Fourth International Conference on Impulsive and Hybrid Dynamical Systems(ICIHDS 2007)(第四届国际脉冲和混合动力系统学术会议)

会议地点: 南宁

会议语种:英文

在线出版日期: 2007-07-20（万方平台首次上网日期，不代表论文的发表时间）