基于改进混合蛙跳-K均值聚类算法的无功电压控制分区  

Reactive power voltage control partitioning based on modified shuffled frog-leaping-K-means clustering algorithm

作　　者：王联国[1] 韩晓慧[2] 宋磊[2] 

机构地区：[1]甘肃农业大学信息科学技术学院,甘肃兰州730070 [2]甘肃农业大学工学院,甘肃兰州730070

出　　处：《传感器与微系统》2013年第6期18-21,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(61063028)

摘　　要：针对无功电压控制分区的特点,并借鉴聚类算法思想,提出一种基于改进混合蛙跳-K均值聚类算法的无功电压控制分区。依据无功源控制的思想,以电网中的节点为分类对象,每个无功源对其控制作用构成这个对象的特征指标,将系统各节点映射到一个多维空间中,在此基础上通过改进混合蛙跳-K均值聚类算法得出被控节点所在的分区号,最后通过发电机节点归并原则确定发电机节点的分区。应用所述方法对IEEE 39节点系统进行了分析计算,分区结果与其它算法结果进行了比较,验证所提出的算法的准确性和可行性。A reactive power voltage control partitioning based on modified shuffled frog-leaping-K-means clustering algorithm is proposed taking into account the characteristic of reactive power voltage control partitioning.According to idea of control of no power source,nodes in electric networks are regarded as classification of object,control of each no power source on it form characteristic index,all the nodes in the system are mapped into a multi-dimensional space.The modified shuffled frog leaping-K-means clustering algorithm is used to obtain partition number of the controlled nodes.Partition of electric generator nodes is determined by using merge principle.IEEE 39 system is analyzed and calculated using the method,partitioning result is compared with the result of other algorithms to verify the accuracy and feasibility of the proposed algorithm.

关 键 词：聚类分析 混合蛙跳算法 K均值算法 电力系统 无功电压控制分区 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]