本文已被：浏览 1190次 下载 586次
投稿时间：2014-09-24
投稿时间：2014-09-24
中文摘要: 讨论了基于支持向量机的电力系统负荷预测模型建模方法.通过对模型结构的分析,提出了最小二乘支持向量机算法学习参数的选取方法.结合粒子群优化算法,给出了粒子群优化对最小二乘支持向量机系数优化选择的方法.采用某省的经济、人口、天气和电价等实证数据对几种预测方法进行比较分析,算例结果表明,所提出的方法可以加快计算速度,并有效提高预测精度.
Abstract:Loading forecasting algorithms based on Support Vector Machine(SVM) are discussed in combination with least square method of its merits on reducing unsatisfactory of conventional parameters. Furthermore, Particle Swarm Optimization(PSO) is used to refine the weighted coefficient for the Least Square SVMin order to improve the computational efficiency and precision. The comparative analysis with empirical data of a province proves the intended advantages of the proposed method.
keywords: load forecast LS-SVM PSO regression
文章编号：20150313 中图分类号： 文献标志码：
基金项目:上海绿色能源并网工程技术研究中心资助项目(13DZ2251900)
| 作者 | 单位 | |
| 潘雪涛 | 上海电力学院电气工程学院 | panxuetao@shiep.edu.cn |
| Author Name | Affiliation | |
| PAN Xuetao | School of Electrical Engineering, Shanghai University of Electric Power, Shanghai 200090, China | panxuetao@shiep.edu.cn |
引用文本：
潘雪涛.电力系统中长期负荷预测改进算法分析[J].上海电力大学学报,2015,31(3):255-257,261.
PAN Xuetao.Improved Medium and Long-term Load Forecast Algorithm and Application of Power System[J].Journal of Shanghai University of Electric Power,2015,31(3):255-257,261.
潘雪涛.电力系统中长期负荷预测改进算法分析[J].上海电力大学学报,2015,31(3):255-257,261.
PAN Xuetao.Improved Medium and Long-term Load Forecast Algorithm and Application of Power System[J].Journal of Shanghai University of Electric Power,2015,31(3):255-257,261.