一种不良数据辨识系统的研究与实现  

Research and Realization for a Bad-data Identification System

作　　者：付金光[1] 蒋德珑[1] 王克文[1] 孙栗[1] 

机构地区：[1]郑州大学电气工程学院,河南郑州450001

出　　处：《低压电器》2011年第12期13-18,共6页Low Voltage Apparatus

基　　金：国家自然科学基金(50177028);郑州大学研究生科学研究基金(08YKYA018)

摘　　要：采用VC++与Fortran 6.5语言混合编程,开发了一种基于模糊聚类分析的电力系统不良数据辨识系统。在理论研究的基础上,将基于模糊等价关系和基于模糊等价划分两种模糊聚类方法进行有效综合。通过编程,分别实现了基于等价关系法、基于模糊ISODATA法和基于聚类综合法的不良数据辨识系统,并对比分析了三种模糊聚类方法的特点。算例分析表明,该系统能够快速、准确地辨识出不良数据,并能有效克服残差污染及残差淹没现象,同时具有辨识方法选取灵活、软件界面友好、计算速度快等特点,有良好的应用前景。With VC＋ ＋ and Fortran 6.5,a bad-data identification system for power systems had been developed based on fuzzy clustering analysis.By the analytic comparison between the fuzzy equivalence and fuzzy ISODATA methods,two methods were synthesized to identify the bad-data in power system real-time data.The bad-data identification system realized separately based on fuzzy equivalence relation,fuzzy equivalence relation and integrated fuzzy clustering to identify bad-data.Meanwhile,the features of the three cluster methods had been analyzed by comparison.The analysis of example revealed that the system not only could identify the bad data instantly and accurately,but also could avoid the residual contamination and residual submerge,with the characteristics of flexible choice for the identification approach,friendly interface,and fast speed of calculation,etc.Therefore,it would have a great application prospect in the future.

关 键 词：不良数据辨识 聚类分析 等价关系 模糊ISODATA法 

分 类 号：TM712[电气工程—电力系统及自动化]