入侵检测技术在智能配电系统中的应用研究  

Application Research of Intrusion Detection Technology in Intelligent Distribution System

作　　者：黄沙里 郭其一[1] 柳悦 黄世泽 屠旭慰 HUANG Shali;GUO Qiyi;LIU Yue;HUANG Shize;TU Xuwei(College of Electronics and Information Engineering,Tongji University,Shanghai 201804,China;Shanghai Key Laboratory of Structural Durability and System Safety for Rail Transit,Shanghai 201804,China;Zhejiang Zhongkai Technology Co.,Ltd.,Wenzhou 325604,China)

机构地区：[1]同济大学电子与信息工程学院,上海201804 [2]上海市轨道交通结构耐久与系统安全重点实验室,上海201804 [3]浙江中凯科技股份有限公司,浙江温州325604

出　　处：《电器与能效管理技术》2020年第4期23-29,共7页Electrical & Energy Management Technology

基　　金：国家自然科学基金(61703308)。

摘　　要：近年来智能配电系统在数据化、网络化、智能化方向发展迅速。随着网络空间安全形势的日益复杂,智能配电系统的网络安全问题正面临巨大的挑战。如何检测针对智能配电系统的入侵行为是一个亟待解决的关键技术问题。以控制与保护开关为终端设备的智能配电系统为研究对象,提出了系统的入侵检测关键算法,引用了Relief、MCFS、mRMR 3种特征选择算法分别进行数据降维,运用K近邻算法作为入侵检测系统的分类器,并对比了检测结果。实验结果表明,提出的入侵检测算法能够有效检测智能配电系统遭受的非法攻击。In recent years,the intelligent power distribution systems have developed rapidly in the direction of data,networking and intelligence.With the increasing complexity of the cyberspace security situation,the network security of intelligent power distribution systems is facing enormous challenges.How to detect intrusion behavior against intelligent power distribution systems is a key technical issue that needs to be solved urgently.The intelligent power distribution system with control and protection switch as the terminal equipment was taken as the research object.The key algorithm of system intrusion detection was proposed.Three feature selection algorithms of Relief,MCFS and mRMR,were used to reduce the data respectively.The K nearest neighbor(KNN)algorithm was used as classifiers of the intrusion detection system,and the detection results were compared.The experiment results show that the intrusion detection algorithm proposed can effectively detect illegal attacks on intelligent power distribution systems.

关 键 词：智能配电系统 入侵检测 特征选择算法 K近邻算法 

分 类 号：TM76[电气工程—电力系统及自动化]