摘要
高渗透率随机性、间歇性分布式电源的大量接入,给传统配电系统的安全、经济和可靠运行带来了一系列的问题,使得传统的负荷预测方法已不再适用.针对这一问题,提出了利用混沌优化粒子群最小二乘支持向量机(PSO-LSSVM)的算法实现对短期电力系统负荷的精确预测.利用粒子群(PSO)算法的全局搜索能力和混沌算法随机、遍历的特性,使其分别克服选参时的盲目性和寻优时粒子群(PSO)算法易出现早熟而陷入局部最优的缺点.最后在Matlab2014a软件平台上验证了混沌优化PSO-LSSVM算法的有效性和收敛性.
Massive connection of random and intermittent distributive power sources with high permeability bring a series of problems to safe,economic,and reliable operation of traditional power distribution system making the traditional load forecasting method applicative no longer.Aimed at this problem,it is proposed that the chaotic optimization particle swarm least square support vector machine(PSO-LSSVM)algorithm should be used for accurate forecast of short-term power system load.The global searching ability of particle swarm optimization(PSO)and the randomness and wide-experience feature of chaos algorithm are employed for overcoming the blindness of parameter selection and premature and local optima due to the particle swarm optimization(PSO)algorithm.The effectiveness and convergence of the chaotic optimization of PSO-LSSVM is verified on Matlab2014a software platform.
作者
郝晓弘
刘鹏娟
汪宁渤
HAO Xiao-hong;LIU Peng-juan;WANG Ning-bo(College of Electrical and Information Engineering,Lanzhou Univ. of Tech.,Lanzhou 730050,China;Gansu Electric Power Research Institute,Lanzhou 730070,China)
出处
《兰州理工大学学报》
CAS
北大核心
2019年第1期85-90,共6页
Journal of Lanzhou University of Technology
基金
国家自然科学基金(61540033)
高比例风光电集中接入弱电网的暂态稳定控制关键技术研究项目(SGTYHT/15-JS-191)
关键词
智能配电网
负荷预测
短期
混沌算法
粒子群算法
最小二乘支持向量机
intelligent power distribution network
load forecasting
short-term
chaos algorithm
particle swarm optimization algorithm
least square support vector machine
