In recent years, Information and Communication Technology (ICT) devices with high levels of availability are deployed to facilitate wide-area monitoring and control of Electric Power System (EPS). Historical accidents and malicious cyber-attacks on EPS have not only proven the existence of interdependencies between the auxiliary communication part and the physical energy part under control, but also indicate that cascading failures happening in such interconnected system could lead to severe economic and social impacts. This motivates EPS planning studies to take into account the vulnerability of ICT networks and consider modern EPS as an interconnected electric power-ICT system. Many methods have been used as attempts to explore ways to describe power-ICT interdependencies. There is still a need however, for quantitative investigation that involves more complex characteristics, structures and functions of both sides of the interconnected system to get closer to practical applications. Among them, time variation of different processes is one of the hardest and most important factors to incorporate, which results in either in-depth analysis of delay-sensitive EPS applications or a sophisticated modelling of multiple processes in parallel. This thesis identifies, models, and simulates different forms of dynamic and static interdependencies with different timescales in parallel, within interconnected systems. The aim is to develop a quantification tool for power system planning that assess the reliability of interconnected systems considering vulnerabilities of ICT components. The thesis, divided in four main parts, focuses on modelling and simulating cascading failures induced not only by outages of power system components but also failures of ICT components. The introductory part of the thesis gives an overview of characteristics of interconnected power-ICT systems and specific functions of ICT systems for EPS operation. In the second part of the thesis, mainstream methodologies that have been utilised to conduct interdependencies studies for not just EPS but multi-sector energy systems are introduced and appraised based on five critical aspects for modelling interconnected power-ICT systems. In the main part of the thesis, a three-layer model is developed to describe interdependencies among power system part, ICT system part and system operator part. Probabilistic method for component failures and event-driven method for control dynamics are combined to model multi-timescale processes in parallel. In DC power flow based EPS model, stochastic failures of generators and lines are included as the basic elements of cascading failures while protective and restorative control actions are modelled as tools against cascading effect. ICT system is developed considering both, hierarchy and topology so that unavailability of ICT paths could be used as vulnerability carrier to integrate ICT-induced interdependency into every segment of operation flow of EPS model. Incorporation of operator cooperation model explores the representation of interdependency in multi-region EPS with emphasis on varying time parameters. Linear programming for solving optimisation problems is used as the core method to derive specific operation actions to restore power system from abnormal or emergency system states. A particular iteration flow is designed to consolidate multi-timescale dynamics of the three sub-systems into coherent steps of control operation against component failures. In the last part of thesis, a long-timespan reliability analysis that assesses the sensitivity of interconnected system reliability to different influential factors has been conducted using sequential Monte Carlo simulation method. The impacts of the influential factors are quantified using reliability metrics to evaluate the average performance, and also the frequency distributions of contingencies with cascading failures. The simulation results firstly have proven the possibil
- Critical Infrastructure interdependencies
- Sequential Monte Carlo Simulation
- Cascading failures
- Cyber Physical Power System
- Reliability Analysis
Modelling and Reliability Analysis of Interconnected Electric Power and ICT System
Chen, Y. (Author). 1 Aug 2024
Student thesis: Phd