Application of Graph Theory to Intentional Controlled Islanding for Blackout Prevention and System Sectionalising for Parallel Restoration

Abstract

This thesis investigates the application of graph theory to Intentional Controlled Islanding (ICI) for blackout prevention and system sectionalising for Parallel Power System Restoration (PPSR). The outcomes of the thesis are the development of more efficient and more accurate methodologies to partition the power system in a controlled manner for blackout prevention, and the development of methodologies to systematically determine suitable system sectionalising strategies for the purpose of reliable parallel restoration. This research has been conducted as the number and the impact of large-area power system blackouts have considerably increased worldwide over the last two decades, affecting hundreds of millions of people. In this context, ICI has been proposed as an effective corrective control strategy that can be used to minimise and limit the impact of cascading outages leading to large-area blackouts by separating, in a controlled manner, the power system into multiple sustainable and stable islands. Moreover, the duration of these catastrophic and expensive events has lasted for several hours and even weeks in numerous cases. Therefore, PPSR has been suggested as an effective and efficient restoration strategy. This approach sectionalises the entire blackout area into multiple islands and restores the created islands in parallel using their own resources, considerably accelerating the restoration process of large-scale power systems affected by complete blackouts. In the context of ICI, this thesis presents new graph theory-based methodologies to determine islanding solutions. The methodologies developed in this thesis for islanding are computationally more efficient and more accurate compared to existing methods. This thesis also explores the dynamic performance of the electrical network, and introduces the concept of weak areas. These weak areas are defined as the zones where, based on the dynamic coupling between the system buses, network splitting is more likely to occur. In the context of PPSR, this thesis introduces new methodologies that systematically determine suitable system sectionalising strategies for PPSR. These strategies accelerate the restoration process by enabling system operators to simultaneously restore more elements of the electrical network, as the restoration of each island is performed independently. Moreover, they help maintain the steady-state stability of the created islands, as multiple constraints are taken into account during the sectionalising strategies design. All formulations presented in this thesis are validated in different IEEE test systems.

Date of Award	1 Aug 2014
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Vladimir Terzija (Supervisor)