基于一种改进蚁群算法的动态无功优化  

Dynamic reactive power optimization based on a modified ant colony algorithm

作　　者：王韶[1] 张煜成[1] 周鑫 李颖[1] 

机构地区：[1]输配电装备及系统安全与新技术国家重点实验室(重庆大学电气工程学院),重庆400030 [2]云南电力试验研究院(集团)有限公司电力研究院,云南昆明650217

出　　处：《电力系统保护与控制》2012年第17期100-104,109,共6页Power System Protection and Control

基　　金：输配电装备及系统安全与新技术国家重点实验室自主研究项目(2007DA10512709212);国家'111'计划项目(B08036)

摘　　要：为满足并联电容器开关或有载调压变压器分接头等离散控制设备最大允许动作次数的限制,根据负荷变化情况提出采用全天各时段的离散控制设备延缓动作时间来避免不必要的调节。运用层次聚类法和引入种群多样性思想,提出一种求解动态无功优化的改进蚁群算法。该算法用各时段的离散控制设备动作次数修改初始信息素;在每个时段优化过程中,对系统各节点灵敏度进行聚类分析后确定信息素变化量,并用离散控制设备的动作次数对其修改。通过对IEEE14节点系统进行的仿真计算验证了所提方法的可行性和有效性。Due to the limits of maximum allowable operation numbers of discrete control device （ONDCD） such as the capacitor switches and on-load tap changers, this paper proposes a way of using delay operation time of discrete control device to avoid the unnecessary operation according to the changes of the load. A modified ant colony algorithm combined with Hierarchical Clustering Algorithm and population diversity is proposed to solve dynamic reactive power optimization. The initial pheromone is modified according to ONDCD among time segments. In optimization process of each time segment, the pheromone variation is determined by means of the cluster analysis of the sensitivity of each node and corrected by ONDCD. The simulation results of the IEEE 14-bus system verify the feasibility and effectiveness of the proposed method.

关 键 词：动态无功优化 蚁群算法 离散控制设备 延缓动作时间 层次聚类法 种群多样性 

分 类 号：TM715[电气工程—电力系统及自动化]