摘要
为了提高短期负荷预测精度,提出一种基于改进灰色关联分析(IGRA)和混沌粒子群算法(CMPSO)优化最小二乘支持向量机(LSSVM)参数的短期负荷预测的方法。该预测模型首先在传统的灰色关联分析方法基础上做出改进,定义了综合灰色关联度从而选取相似日;其次,针对标准粒子群算法求解LSSVM参数优化问题时存在的易陷入局部最优的缺陷,引入混沌理论对粒子群算法加以改造,建立CMPSO⁃LSSVM预测模型;最后将该方法应用于某市2018年夏季短期负荷预测,仿真结果表明该方法不仅可以避免算法陷入局部极值,还能提高预测的精准度。
In order to improve the accuracy of short⁃term load prediction,a short⁃term load prediction method which is based on improved grey relational analysis(IGRA)and chaos modified particle swarm optimization(CMPSO)to optimize least square support vector machine(LSSVM)parameters is proposed.The prediction model is improved on the basis of the traditional grey relational analysis method first,and then defines synthetic grey relational degree to select similar days.In order to solve the defect that the standard PSO could easily fall into the local optimal,the chaos theory is introduced to make an improvement and establish the CMPSO⁃LSSVM forecasting model.This method was applied to the short⁃term load prediction of a city in the summer of 2018.The simulation results show that this method can not only protect the algorithm from falling into the local extremum,but also improve the accuracy of load prediction.
作者
解海翔
陈芳芳
刘易
盖佳郇
徐天奇
XIE Haixiang;CHEN Fangfang;LIU Yi;GAI Jiaxun;XU Tianqi(School of Electrical and Information Engineering,Yunnan Minzu University,Kunming 650031,China)
出处
《现代电子技术》
2021年第8期177-181,共5页
Modern Electronics Technique
基金
国家自然科学基金项目资助(61761049)。
关键词
关联分析
相似日选取
特征提取
模型建立
算法改造
短期负荷预测
relational analysis
similar day selection
feature extraction
model establishment
algorithm remoulding
short⁃term load prediction
作者简介
解海翔,男,山东济宁人,硕士研究生,研究方向为人工智能在能源系统中的应用;通讯作者:陈芳芳,女,重庆人,硕士,副教授,研究方向为电气控制、智能电网。
