J4 ›› 2010, Vol. 48 ›› Issue (02): 251-255.

• 计算机科学 • 上一篇    下一篇

基于遗传算法优化支持向量回归机的网格负载预测模型

唐阔, 胡国圣, 车喜龙, 胡亮   

  1. 吉林大学 计算机科学与技术学院, 长春 130012
  • 收稿日期:2009-03-10 出版日期:2010-03-26 发布日期:2010-03-22
  • 通讯作者: 胡亮 E-mail:hul@jlu.edu.cn

Grid Host Load Prediction Model of Support VectorRegression Optimized by Genetic Algorithm

TANG Kuo, HU Guosheng, CHE Xilong, HU Liang   

  1. College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2009-03-10 Online:2010-03-26 Published:2010-03-22
  • Contact: HU Liang E-mail:hul@jlu.edu.cn

摘要:

提出一种基于遗传算法优化支持向量回归机的模型进行网格负载预测, 使用遗传算法和交叉验证技术解决了支持向量回归机参数难以确定的问题. 标准数据集仿真实验结果表明, 该模型与试验法定参的支持向量回归机和BP神经网络相比具有更优的预测性能.

关键词: 网格负载预测, 支持向量回归, 遗传算法

Abstract:

A support vector regression optimized by genetic algorithm model was developed for grid host load prediction. Genetic algorithm and cross validation technology were applied to solve parameter optimization of support vector regression. Simulation experiments were performed on benchmark data set. Experimental results indicate that the proposed model exhibits better performance than support vector regression model with parameters selected by trialanderror method and the backpropagation neural network.

Key words: grid host load prediction, support vector regression, genetic algorithm

中图分类号: 

  • TP316.4