Font Size: a A A

Research On Network Attack Prediction And Defense Method In Electricity CPS

Posted on:2019-10-12Degree:MasterType:Thesis
Country:ChinaCandidate:W H LiFull Text:PDF
GTID:2382330572995088Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
The cyber physical system integrates the physical environment of the computer system,communication network,and power system through 3C technology to form a multidimensional,heterogeneous and complex system that integrates real-time sensing,dynamic control,and information services.As the dependence of the physical system and the information system of the power CPS is getting higher and higher,the entire system gradually becomes inclusive and open from isolation and closure.With a large number of remote terminals accessing the information platform,the environment has gradually become more complex,making the system more and more security threats,and the types of attacks that appear to be diversified,toward multi-step,cross-domain,and synergies.The development of power CPS security research is also increasingly important.In smart grid CPS security research,attack prediction can infer subsequent attacks based on the attacks that have occurred,and help reduce the damage caused by network attacks.For defenses that may occur,defensive resource allocation can improve defense effectiveness.This paper deeply studies the domestic and international smart grid network security technologies,and proposes attack graph-based attack prediction and game-based defense resource allocation methods in power CPS.The main research work of this article is summarized as follows:(1)In the context of a multi-step cross-domain attack on a power information physical system,this paper proposes a power-based CPS attack prediction method based on path analysis.In order to analyze the attacker's path when the power system encounters an attack.The probabilistic attack graph model is built based on the common attack graph definition and probabilistic knowledge.Cross-domain attack probability and cross-domain average attack time are used to quantify the difficulty of exploiting power in CPS and attack proficiency in offensive and defensive combat.Based on the real-time attack behavior detection,the improved Dijkstra algorithm is used to enumerate the subsequent possible attack paths,and two quantitative indicators are analyzed to obtain the most threatening attack path.(2)Taking the power physics system as the background of the collaborative attack of information physics,this paper uses the load to remove the physical impact of the quantized information physical coordination attack on the power system.Starting from the bounded rationality situation of offensive and defensive confrontation,the offensive and defensive stochastic game model is constructed and the value iterative algorithm is used to solve the Nash equilibrium problem.Based on the analysis of offensive and defensive strategies,a defense resource allocation method was proposed.The validity of the proposed model and method was verified by the IEEE 9-bus system.
Keywords/Search Tags:Cyber Physical System, Probabilistic Attack Graph, Attack Prediction, Attack Defense Random Game, Defense Resource Allocation
PDF Full Text Request
Related items