An Optimization-Based Approach to Recover the Detected Attacked Grid Variables After False Data Injection Attack

Abstract

The increasing integration of communication technologies into the power grids has raised concerns about maintaining their cyber security in recent years. Since the State Estimator (SE) plays a critical role in power grid operation, special attention has been paid to its cyber vulnerabilities. Especially, a class of cyber attacks to this module, the so called False Data Injection Attacks (FDIAs) has received attention. So far, various methods to perform an FDIA on the SE have been proposed in the literature. There are also other researches suggesting different FDIA detection methods. However, such detection methods do not typically provide a mechanism to recover the true values of the attacked grid variables. In this paper, an iterative optimization-based approach is proposed to recover the pre-attack values of the attacked grid variables while trying to make as few changes to the non-attacked ones as possible. Furthermore, a framework to calculate an index called the Recovery Quality Index (RQI) is proposed to assess the performance of the recovery algorithm. The simulation results show the satisfactory performance of the proposed method in terms of its calculated RQI for a large number of simulated attack samples on different IEEE test bus systems.