Research On Cascading Failure And Immune Methed Of Grid Cyber Physical Systerm

With the rapid development of smart grid,the power system becomes the most important infrastructure component in today’s society,the power system and information system gradually evolve into a fused Grid Cyber-Physical System(CPS)with a certain connection,however,the deep integration between the two systems also brings potential problems.There is no unified modeling method for the vast network formed by GCPS.The frequent interaction of information flow and energy flow,information systems are vulnerable to network-initiated attacks.Failure of the information layer may be propagated at the physical level and increased probability of cascade failure,caused a power outage causing a blackout.Therefore,we study the network unified modeling method and analyze the cascading failure principle of power-information coupling network,finding ways to improve the network’s resistance to destruction and finding strategies to effectively suppress network risks have become problems to be solved.This article aims at the above problems and analyzes and explains the content of the research based on complex network theory and propagation dynamics,the main research content is as follows:Firstly,considering the actual power and information communication network has a scale-free network under the concept of a complex network and certain community structure characteristics,this paper analyzes the structure of power and information networks and proposes a scale-free network model based on community structure,the model was built to improve the fixed-power law of the traditional BA model with the intermediate effect of the node and roulette method,the defect of the clustering coefficient was small,and the community structure was used to improve the uncertainty of the local boundary,weakening the emergence of super nodes,the ability of the network to defend against deliberate attacks from outside,through theoretical verification,the degree distribution of the proposed model satisfies the power law distribution,can meet the power and information network modeling needs.Border node attacks have greater losses than internal node attacks.Second,build a corresponding power-information coupling connection,four coupling methods are proposed to represent the influence of community structure on the coupling relationship of nodes.Establishing a cascading failure protection model based on community structure,analyze cascaded failure process under the fusion of information network and power network.Discovered through experimental simulation,under the controllable network construction cost,appropriate increase in node load distribution index and boundary node capacity second patience value can increase the invulnerability of the coupled network.Increase the resistance of the network by increasing the connection points of the community boundary points.Having a community structure to some extent inhibits the spread of network failures,compared with other coupling methods,Community Inner Assortative Link coupling shows greater resistance to damage.The proposed cascading failure model based on community structure is more practical.Finally,due to community boundary points causing larger cascade failures,Rapid identification of these key nodes and immunization plays an important role in inhibiting risk transmission.This paper,in the case of unknown global information,this paper proposes a boundary-center-point immune algorithm based on local immune,the algorithm finds and immunizes the network boundary center node and its large neighbors according to the principle of self-avoiding random walk.The original risk propagation model ignores the disadvantage that normal nodes can also gain immunity through artificial means.This article is based on the original risk propagation model that ignores the drawback that normal nodes can also be obtained by artificial means,combined with cellular automata.The experimental verification shows that the proposed algorithm is faster,can identify the target node accurately,and has better immunity,it can provide for the control and prevention of risk propagation under the condition of incomplete structural information for large-scale complex networks.