基于广域信息的同调机群聚类识别方法.pdfVIP

39 卷 10 期 电 网 技 术 Vol. 39 No. 10 20 15 年10 月 Power System Technology Oct. 2015 文章编号：1000-3673 （20 15）10-2889-05 中图分类号：TM 74 文献标志码：A 学科代码：470 40 基于广域信息的同调机群聚类识别方法 张亚洲，张艳霞，蒙高鹏，赵冉，高瑾 （智能电网教育部重点实验室(天津大学) ，天津市 南开区 300072 ） A Wide Area Information Based Clustering Recognition Method of Coherent Generators ZHANG Yazhou, ZHANG Yanxia, MENG Gaopeng, ZHAO Ran, GAO Jin (Key Laboratory of Smart Grid(Tianjin University), Ministry of Education, Nankai District, Tianjin 300072, China) ABSTRACT: Existing generator coherency cluster grouping 的分群方法相比，轨迹分析更全面，分群结果更具有实际 methods based on WAMS only regard distances between power 意义。 angle curves as clustering index, ignoring local characteristics 关键词：广域测量系统；同调识别；轨迹聚类；结构差异度 and varying trends of power angle tracks, therefore lead to problems of incomprehensive traj ectory analysis. This paper DOI 10.13335/j.1000-3673.pst.2015.10.031 presents a coherent clustering algorithm based on diversity of 0 引言 track structures. The method of extracting traj ectory 随着国民经济的迅速发展，电力系统结构日益 characteristics in communication field is applied to power system. This method takes comprehensive considerations of all 复杂、规模日趋扩大，系统在许多运行状况下表现 kinds of differences, including distance between generator 出复杂的动态特性，这就大大增加了系统安全性分 power angle curves, changing direction, degree of volatility 析的难度。为了快速分析大规模电力系统的动态过 and changing speed of power angle tracks, uses variation 程，目前普遍采用动态等值(extended equal area coefficient method to calculate feature weights, takes structure criterion，EEAC)等方法对系统进行简化以减少计算 difference