Bad data injection as a threat for power system security

Abstract

Power system operation is undeniably affected by system security indications providing system operators with insight into power system status. In this regard, any tampered data would compromise power system security. This paper aims at presenting an approach to scrutinize the impact of cyber attacks on the system security indices. With this in mind, two well-known system security indices representing voltage and loading severities are taken into account. The approach uses genetic algorithm to find the most severe conditions for each security index from financial and technical viewpoints. In the former, bad data injection targets to overestimate security indices resulting in unnecessary preventive actions, thus higher operation costs. In the latter, however, underestimation of security indices may threat system security. The proposed algorithm is applied to the standard IEEE 30-bus system. The obtained results indicate how cyber attacks could manipulate the system security indices, leading to fallacious insight over power system status.