智能电网中分层网络结构的入侵检测系统研究  

Research on intrusion detection system in hierarchical network architecture of smart grid

作　　者：徐静[1] 姚志垒[2] 徐森[1] 李永忠[3] 吴素芹[1] 

机构地区：[1]盐城工学院信息工程学院,江苏盐城224051 [2]盐城工学院电气工程学院,江苏盐城224051 [3]江苏科技大学计算机科学与工程学院,江苏镇江212003

出　　处：《郑州轻工业学院学报（自然科学版）》2015年第5期88-92,共5页Journal of Zhengzhou University of Light Industry:Natural Science

基　　金：国家自然科学基金项目(51407153)

摘　　要：为了提高智能电网的安全性和可靠性,将入侵检测系统运用到智能电网中.针对智能电网具有网络节点多、信息传输量大、安全要求高等特点,提出一种基于多Agent的入侵检测系统模型,并将该模型部署到智能电网的分层网络结构中,以减少数据传输,节约带宽,解决速度瓶颈问题;针对入侵检测的误检率高、漏检率高等问题,提出适合智能电网的动态克隆选择算法,将所提算法与多Agent技术相结合,构造了具有免疫功能的Agent.利用标准的入侵检测测试数据集——KDD’99数据集——对所提模型和算法进行仿真验证,结果表明:本系统在保证低误检率的同时,提高了智能电网中常见的Dos和Probing类型攻击的检测率.In order to improve the safety and reliability of the smart grid,the intrusion detection system was applied to the smart grid. As smart grid had the characteristics of large number of network nodes,large amount of information transmission and high safety requirements,an intrusion detection model based on multi-agent was established,and was deployed in the hierarchical network structure of smart grid. Therefore,amount of transmission data was reduced,bandwidth was saved and speed bottleneck problem was solved. With the purpose of reducing false positive rate and false negative rate in current intrusion detection system,dynamic clonal selection algorithm for the smart grid environment was proposed. The proposed algorithm could be combined with the multi-agent technology,the immune agents were constructed. The proposed model and algorithm were simulated by KDD＇99 datasets. Simulation results showed that the proposed system had low false positive rate and improved the detection rate of Dos and Probing attack that are common attack in smart grid.

关 键 词：智能电网 入侵检测 多代理 动态克隆选择算法 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]