Assessment of Robustness of Power Systems From a Network Perspective

In this paper, we study the robustness assessment of power systems from a network perspective. Based on Kirchhoff's laws and the properties of network elements, and combining with a complex network structure, we propose a model that generates power flow information given the electricity consumption and generation information. It has been widely known that large scale blackouts are the result of a series of cascading failures triggered by the malfunctioning of specific critical components. Power systems could be more robust if there were fewer such critical components or the network configuration was suitably designed. The percentage of unserved nodes (PUN) caused by a failed component and the percentage of noncritical links (PNL) that will not cause severe damage are used to provide quantitative indication of a power system's robustness. We assess robustness of the IEEE 118 Bus, Northern European Grid and some synthesized networks. The influence of network structure and location of generators are explored. Simulation results show that the connection with short average shortest path length can significantly reduce a power system's robustness, and that the system with lower generator resistance has better robustness with a given network structure. We also propose a new metric based on node-generator distance (DG) for measuring the accessibility of generators in a power network which is shown to affect robustness significantly.