Analysis Of Power System Vulnerability Under Cyber Attack Threat

Power system takes an important responsibility of delivering power energy for society. It directly affects the economy and people’s livelihood whether power system could operate safely and reliably. And then power system needs to ensure stability and reliability of its operation. In recent years, with the development of smart grid technology, information technology and military technology, the possibility of cyber attacks occurring in electric secondary system increases. In contrast to random attacks, optimized cyber attacks cause a greater threat to power system, and therefore the research of power system vulnerability under cyber attack threat demonstrates its theoretical and practical value.This paper introduces cyber attacks, research methods and achievement and features which electric secondary system exhibits in the attack process. And then this paper describes principles and mechanisms of the false data injection attack. After that, This research proposes a bi-level optimization model of which upper and lower levels separately describe behaviors and characteristics of the attacker and system operator. In the upper level, cyber attackers with limited resources attack the measurements of the power system. While in the lower level, the power system operator performs optimal power system dispatching based on the output which state estimator obtains after processing the measurement with false data.In order to handle the complexity of the bi-level optimization model, this paper presents a hybrid optimization combining genetic algorithms, nonlinear least squares method and optimal power flow to solve the problem. This hybrid optimization solves the the upper level problem by algorithm genetic algorithm, and embeds least squares method and nonlinear OPF into genetic algorithm respectively solving the lower level problem. Numerical examples demonstrate the effectiveness of the proposed approach.