基于改进蚁群算法优化参数的LSSVM短期负荷预测

来源期刊:中南大学学报(自然科学版)2011年第11期

论文作者:龙文 梁昔明 龙祖强 李朝辉

文章页码:3408 - 3414

关键词:最小二乘支持向量机;蚁群优化算法;参数优化;短期负荷预测

Key words:least square support vector machine (LSSVM); ant colony optimization (ACO) algorithm; parameter optimization; short-term load forecasting

摘    要:提出一种自动优选最小二乘支持向量机(LSSVM)模型参数的改进蚁群(MACO)算法。该算法将LSSVM模型的参数作为蚂蚁的位置向量,然后采用动态随机抽取的方法来确定目标个体引导蚁群进行全局搜索,同时在最优蚂蚁邻域内进行小步长局部搜索,找到模型的最优参数,得到基于MACO算法优化的LSSVM(MACO-LSSVM)预测模型。将优化后的LSSVM模型应用于短期电力负荷预测问题,选择湖南某地区日期为2009-08-01至2009-08-30各小时点的数据进行分析,对2009-08-31该日24 h的负荷进行预测,并与BP神经网络和SVM模型进行比较。研究结果表明: 本文方法得到的均方根相对误差为1.71%,比用BP神经网络和SVM模型得到的均方根相对误差分别低1.61%和1.05%。

Abstract: An optimization method based on the modified ant colony optimization (MACO) algorithm was used to select the two parameters of least square support vector machine (LSSVM) model. In this method, the parameters of LSSVM model were considered the position vector of ants. Target individuals which lead the ant colony to do global rapid search were determined by dynamic and stochastic extraction, and the optimal ant of this generation searched in small step nearly. The optimal parameter value was obtained by MACO and modified ant colony optimization-least square support vector machine (MACO-LSSVM) forecasting model was obtained. The proposed model is applied to the short-term electrical power load forecasting problem. Every hour’s load from 2009-08-01 to 2009-08-30 of area in Hunan province was taken as the sample data to be analyzed. The results indicate that the root-mean-square relative error of the proposed method is only 1.71%, which is less than those of BP and SVM model by 1.61% and 1.05%, respectively.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号