Author(s): XU Jing, YAO Zhi-lei, XU Sen, LI Yong-zhong, WU Su-qin, School of Information Engineering, Yancheng Institute of Technology, School of Electrical Engineering, Yancheng Institute of Technology, School of Computer Science and Engineering, Jiangsu University of Science and Technology
Pages: 88-92
Year: 2015 Issue:  Z2
Journal: Journal of Zhengzhou University of Light Industry(Natural Science Edition)
Keyword:  smart grid; intrusion detection; multi-agent; dynamic clonal selection algorithm;
Abstract: 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.