摘要
配电网网架优化是一个多目标综合优化问题,粒子群算法因其易实现、收敛速度快等特点逐渐成为电力系统优化领域研究热点之一.针对粒子群算法易陷于局部最优问题,提出一种基于聚类策略的改进粒子群算法,动态地将粒子聚类为三种级别的粒子并对应采用不同的学习模型更新速度,增强了粒子群体多样性和全局搜索能力.通过算例仿真验证了算法在配电网网架优化问题上的可行性.
Distribution network optimization is a multi-objective comprehensive problem.Meanwhile,particle swarm optimization(PSO)has become a hotspot in the field of power system optimization due to its easy implementation and fast convergence.To avoid trapping in local optimality,a modified PSO particle swarm optimization algorithm based on clustering strategy is proposed in this paper.Wherein,particles are dynamically clustered into three levels.Correspondingly,different learning models are used to update speed,thus enhancing the particle diversity and global search capabilities.The superiority of the proposed algorithm in distribution network optimization is verified by numerical simulation.
作者
王姝
张海龙
王恩荣
Wang Shu;Zhang Hailong;Wang Enrong(School of NARI Electrical and Automation,Nanjing Normal University,Nanjing 210023,China)
出处
《南京师范大学学报（工程技术版）》
CAS
2020年第1期15-19,共5页
Journal of Nanjing Normal University(Engineering and Technology Edition)
基金
南京师范大学企业合作项目(KJZX17015)
关键词
配电网
网架优化
聚类分层
粒子群
distribution network
grid optimization
cluster stratification
particle swarm optimization
作者简介
通讯作者:张海龙,博士,副教授,研究方向:电工理论与新技术.E-mail:61204@njnu.edu.cn。
