Modeling And Analysis On Cascading Failure Of Cyber Physical Power System Based On Complex Network Theory

The automation of power system is increasing quickly with the rapid development of the smart grid in global area,which lead to the great increase of sensor quantity?decision making unit and information network size.The power system is evolving into cyber physical power system(CPPS),which is composed of power grid and information network.Due to the deep coupling of power grid and information network,the flexibility and efficiency of the power system is improved,and as well as the complexity of grid structure.As a result,the new hidden dangers are brought into the power system.When cascading failure occurs on power system,any node failure of CPPS is likely fueled the fault propagation.Past researches on cascading failure of the power system have shown that traditional analysis method is tend to study the cascading failure from the point of isolated network,which ignored the coupling character of CPPS.Therefore,it is of great theoretical and practical significance to explore the interaction theory,build the coupled network model and study cascading failure model of CPPS.This thesis aimed to research the model and analysis method on cascading failures of CPPS based on the complex network theory.The main contents are presented as follows.(1)As an important statistical feature in complex networks,community structure is widely used in complex network.As a typical complex network,CPPS has obvious community structure.In order to study the effect of community structure on cascading failures of CPPS,three kinds of network coupling modes were constructed,which are the random coupling with communities,assortative coupling in communities and assortative coupling with communities,a cascading failure model of CPPS is proposed based on load-capacity model.This thesis investigated the relationship between cascading failures and the intra-modular rewiring probabilities and inter-modular connections under different attack modes,namely intra-attacks and inter-attacks.Numerical results indicated that increasing intra-modular rewiring probabilities and inter-modular connections can reduce the scale of cascading failures and improve the robustness of the CPPS under intra attacks and inter attacks.The global and local load loss curves showed that the mode of assortative coupling in communities can effectively reduce the scale of cascading failures.(2)Studies have shown that some key components in the power grid play an important role in the spread of cascading failures.Therefore,it is of great significance to identify and protect these nodes to ensure the safe and steady operation of CPPS.A novel method,which considered the global and local characteristics,was proposed to evaluate the node importance of network in this thesis.A cascading failure model of power grid based on DC power flow model is established to verify the effectiveness of the proposed method.Simulation results showed that the high degree nodes in power grid are not always the vital node,both the node efficiency and the degree of adjacent node contribute to the order in power grid.(3)Large-scale abnormal transfer of power flow is one of the main causes of cascading failures in power system.Therefore,it is of great significance to study the cascading failures model that considering the power flow characteristics of power network.Firstly,a cascading model of CPPS based on a DC power flow model was proposed in this thesis.Furtherly,the influence of attack mode on cascading failures of CPPS was investigated,it was found that compared with high degree attack,high betweenness attack and random attack,the attack on the vital nodes can lead to much more load loss.Finally,the influence of coupling mode,failure probability between networks and information network topology on the cascading failures of CPPS was studied,it is found that the failure probability has little effect on the cascading failure of CPPS,the inner self-similarity coupled network and the homogeneous topology of information network can effectively reduce the occurrence probability of cascading failures of CPPS.(4)Power system has put forward higher requirements to the information network with the rapid development of CPPS.Although,much attention has been payed to the isolated power grid or information network in recent years,the studies ignored the coupling characteristics of CPPS.In order to study the influence of the dynamic topology of the information network on the cascading failures of CPPS.Firstly,the factors,which affect the dynamic topology of the information network,namely,the routing strategy of the information network and the random failure of the optical cable,were analyzed in this thesis.Sencodly,an improved congestion aware routing algorithm was proposed to study the effect of different routing strategies on the transmission efficiency of information network.Finally,the CPPS cascading failures model that considering the information network routing strategy and the random fault of optical fiber cable was established respectively.The influence of routing strategy and routing strategy parameters on CPPS cascading failures was investigated.Simulation results showed that the optimal routing strategy and parameters of routing strategy can effectively reduce the occurrence probability of cascading failures of CPPS and improve the robustness of the power system.The influence of random failure ratio of optical cable of information network on CPPS cascading failures was studied under different coupling modes,different failure probability,different information network topology and different location of dispatching center.Simulation results showed that the irnfluence of information network on cascading failure varies from strength to weakness,eventually to stability with the increase of random failure ratio of optical cable,and not only the scale of CPPS cascading failures can be reduced,but also the robustness of power system can be improved by optimizing the location of the dispatching center.