Analysis And Protection Of Cascading Failures In Controlled Cyber-Physical Power Systems

With the proposal of the Energy Internet and the development in the past decade,the generation,transmission and distribution system of the contemporary smart grid is deeply integrated with information technology.Compared with the traditional power system,the smart grid,which integrates various communication technologies and control systems,has a high degree of informatization,automation and intelligence.It can improve the comprehensive utilization efficiency of energy,the access rate of new energy,promote the optimal allocation of resources,and make the power system operate more reliably,safely,economically and efficiently.However,a large number of power system regulation and stability maintenance operations and wide-area power coordination and interaction are handed within the intelligent control center of the power system,making the system introduce new forms of failure production and propagation,which posing new requirements and challenges for the stable operation of the power system.In addition,modern power systems span hundreds or even thousands of kilometers,resulting in longer time delays in communication network transmission,data processing and physical control.As the response of the power grid control system becomes faster and more precise,the control time delay also brings new challenges to its transient stability.Base on these practical problems faced by the current power grid development,this paper focuses on the reliability evaluation and optimization of the interdependent network of the controlled Cyber-Physical Power System(CPPS),and has achieved the following main results:(1)Firstly,this paper establishes a mathematical model for the interdependent structure and intra-network characteristics in the CPPS by combining the control mechanism of the information layer and the power flow operation mechanism of the physical layer.This paper studies the relevant control mechanisms of load frequency control in the CPPS,unifies and abstracts a control architecture based on the Network Control System(NCS),as an important dependency mechanism from the information layer to the physical layer.Similarly,the energy dependence is modeled as an dependency mechanism from the physical layer to the information layer.Then,in the information layer,this paper develops a service flow based communication network model.In the physical layer,this paper uses a dynamic transient power flow model to describe the power flow redistribution after network failure.Compared with other static power flow models,the transient power flow can describe the power flow changes in real time,which is more accurate when the physical layer fails.Then,based on the modeling of the information layer and the physical layer,this paper proposes the interdependent cascading failure mechanism,and establishes a real-time cascading failure model of the controlled CPPS.Finally,the proposed model is implemented and simulated in various power networks,and the characteristics and differences of failure propagation are analyzed and discussed.Simulation results show that the transient power flow fluctuation caused by cascading failures has a great impact on the stability of the power grid.When the control system is introduced into the power grid,the transient stability of the power grid will be significantly improved.(2)Modern power systems can usually span hundreds or even thousands of kilometers.Under the circumstance of long distance and high coupling,time delay can not be ignored in the modeling of power grid control system.This paper then studies and analyzes the influence of time delay on the stability of the CPPS.Based on the model of the CPPS established in the first part of the research,the control time delay related parameters are added as a part of differential equations,so that the model has the ability to simultaneously model the control time delay and power transient power flow in a relatively short time scale.The power flow fluctuation with control time delay is obtained by simulation,and the stability of the system in case of cascading failure is analyzed.In addition,we studied the influence of some network control parameters,including time-invariant delay,time-varying delay,and line capacity,on the cascading propagation in IEEE power system topologies and a real world power system topology in China.The simulation shows that in the process of cascading failure,the time delay plays a vital role in the dynamic control of power flow and has a significant impact on the transient stability of the CPPS.In general,this part of the work emphasizes the influence of the communication system on the power transient process in the cascading failure,and provides the model and method for simulating cascading failures.(3)According to the research in the previous part,if the control time delay is large,the control system may produce delayed response and misresponse problems.This paper studies how to reduce the risk of system stability caused by the control time delay through some protection measures of the control system when the delay is inevitable.In this paper,the intelligent generation control system(SGC)in the information layer of the CPPS is taken as the research object.At first,the control objectives and performance of the traditional PID controller are studied,as well as some control defects in face of cascading failures.Then the protection measures for the stability of SGC control system are proposed from two dimensions.One is to solve the problem that the delay response and error response generated by the control system are amplified when the time delay is too large.This paper proposes a delay compensation method to limit the inaccurate response of the control system by reducing the adjustment sensitivity of the controller when the time delay is too large.Secondly,aiming at the problem that the control target of traditional PID controller based on the reference frequency of the power grid may aggravate the failure spread,a control strategy based on the relative frequency is proposed to reduce the risk of failure spread caused by the improper control target of the controller during the cascading failure propagation.