改进新息图法在不良数据检测与辨识中的应用  

Application of Improved Innovation Graph Method in Detection and Identification of Bad Data

作　　者：钟建伟 刘佳芳 倪俊 吕静 ZHONG Jianwei;LIU Jiafang;NI Jun;LV Jing(School of Information Engineering,Hubei University for Nationalities,Enshi 445000,China;Enshi Power Supply Company,State Grid Hubei Electric Power Co.,Ltd.,Enshi 445000,China)

机构地区：[1]湖北民族学院信息工程学院,恩施445000 [2]国网湖北省电力有限公司恩施供电公司,恩施445000

出　　处：《电力系统及其自动化学报》2018年第9期83-88,共6页Proceedings of the CSU-EPSA

基　　金：国家自然科学基金资助项目(61263030;61463014)

摘　　要：针对连支测量值为不良数据时,传统新息图法检测与辨识过程的无连续性问题,提出基于蚁群算法的改进新息图法。该方法根据配电网的网孔对配电网的支路进行编码,建立独立回路矩阵,通过对连支测量值的自动判断,利用简化的蚁群算法生成新树,对配电网络中的不良数据进行高效地检测与辨识。采用IEEE14节点配电系统对此方法进行验证,并与新息图法进行对比分析,仿真结果证明改进新息图法在时间上的高效性和在检测与辨识上的准确性。When the measurement values on link branches are bad data,there is no continuity in the detection or identi-fication using the traditional innovation graph method.To solve this problem,an improved innovation graph method based on ant colony algorithm is proposed.According to the mesh of distribution network,the link branches are coded,and an independent circuit matrix is established;in addition,through the automatic judgement of measurement values on link branches,a new tree is built using the simplified ant colony algorithm to efficiently detect and identify bad data in distribution network.The presented method is verified with an IEEE 14-bus distribution system as a numerical exam-ple,and it is further compared with the traditional innovation graph method.Simulation results show that the improved innovation graph method has a higher time efficiency,and higher detection and identification accuracies.

关 键 词：不良数据 新息向量 新息图法 蚁群算法 检测 辨识 

分 类 号：TM732[电气工程—电力系统及自动化]