支持向量机在电力负荷预测中的应用研究  

Research on Power Load Forecasting Base on Support Vector Machines

作　　者：蒋喆[1] 

机构地区：[1]沈阳汽车工业学院,辽宁沈阳110015

出　　处：《计算机仿真》2010年第8期282-285,共4页Computer Simulation

基　　金：辽宁省"十一五"教育科学规划项目(辽教函[2006]8号);沈阳市总工会科技计划(2009SR023427)

摘　　要：研究电力系统负荷预测问题,针对电力负荷过程存在非线性技术,为提高预测精度,保证安全供电,改变传统方法,提出改进支持向量机的预测性能,更精确地预测电力负荷,提出粒子群算法优化支持向量机(PSO-SVM)的电力负荷预测方法。PSO-SVM用粒子群算法优化支持向量机参数,减少了对支持向量机参数选择的盲目性,获得较优的支持向量机预测模型。并以贵州省为例在2008.7-2009.7电力负荷数据进行测试和分析,并进行仿真。实验结果表明,在电力负荷预测中,PSO-SVM比SVM和BPNN有着更高的预测精度,测试表明PSO-SVM方法用于电力负荷预测是有效可行的。Research on power load forecasting base on support vector machines is an important problem.In order to improve the forecasting performance of support vector machines and to forecast the power load more accurately,a new method for exchange rate time series forecasting was proposed,in which particle swarm optimization is used to determine free parameters of support vector machines.PSO is an intelligent swarm optimization method,which derives from the research for behavior of bird flocking.The method not only has strong global search capability,but also is very easy to implement.Thus,particle swarm optimization is suitable to determine the parameters of support vector machine.The power load data from 2008.7 to 2009.7 of Guizhou are used to testify and analyze the performance of the proposed model.The result shows that SVM based on particle swarm has both fast training speed and small number of errors.The forecast precision has also been significantly improved,thus proving the validity of this model for power load forecasting.

关 键 词：粒子群算法 支持向量机 电力负荷预测 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论] TP301[自动化与计算机技术—计算机科学与技术]