Research On Cyber-Physical Power System Reliability Assessment And Cascading Failures Modelling

The CPPS(cyber-physical power system)is becoming the development trend and structure form of the smart grid and beyond,also,it is the backbone technology supporting energy internet.The interdependency and interaction between the physical layer and the cyber layer not only improves the performance of the CPPS,but brings new challenges as well.The interdependency can be the channel of failure risks transmission.The failure in the cyber layer can weaken the reliability of the physical layer,moreover,the failures in both layers can happen simultaneously because of nature disasters or human attacks,which would lead to cascading failures.Thus,research on the influence of cyber failures on physical grid long-term reliability and the cascading failures modelling with its mitigation strategies are meaningful for probing the cyber-physical interaction mechanism and failure risk management.This thesis focuses on the CPPS reliability assessment,and the cascading failures modelling based on improved percolation theory with the multi-perspectives mitigation strategies.The main contributions include following four parts.(1)The static model for the cyber and physical layers are established through complex network theory.The dynamic model for the cyber and physical layers are established through information flow and power flow analysis.The states for the system includes secure state,alert state,emergency state and extreme state,the division of system states also demonstrates the imperative of research on the reliability assessment and cascading failures.(2)The reliability and availability of the cyber layer and its protection and monitoring functions with various topologies are derived based on a RBD method.The availability of the physical layer components are modified via a multi-state Markov chain model,in which the cyber layer topology and functions are simultaneously considered.The modification of the availability decouples the physical parts with their corresponding physical functions.Reliability indices of composite power systems are calculated through non-sequential Monte-Carlo simulation.(3)A model of cascading failures in CPPS based on an improved percolation theory is presented.The dynamic development of cascading failures is divided into several iteration stages,and the interaction mechanism between two layers is interpreted in both the dynamic and static perspectives.The information transmission delay and congestion is simulated for the cyber layer stage,also the power flow and security control are incorporated for the physical layer stage.The resilience indices of the failures reflect the influence of cascading failures on both the topological integrity and operational security.Various cases for the CPPS cascading failures are simulated via the presented model,which realizes the accurate illustration of the failures development.(4)Mitigation strategies in multi-perspectives are proposed to further the research of CPPS cascading failures.In the cyber layer information control perspective,the SDN-based cyber layer OpenFlow dynamic route optimization is proposed.In the energy flow control perspective,the security control with UPFC is proposed.The UPFC expends the amounts and categories of control variables.To solve this optimization problem,the McCormick's envelopes are used.In the system management structure perspective,the autonomous security control algorithm is realized through ADMM.At last,the feasibility and efficiency of the mitigation strategies are verified in the simulation.