Blind False Data Injection Attack Using PCA Approximation Method in Smart Grid

Abstract

Accurate state estimation is of paramount importance to maintain normal operations of smart power grids. However, recent research shows that carefully produced attacks with the knowledge of the grid topology, i.e., Jacobian matrix, can bypass the bad data detection (BDD) system. The BDD is used to ensure the integrity of state estimation to filter faulty measurements introduced by device malfunctions or malicious attacks. However, to construct the false data injection attack vectors, a common assumption in most prior works on false data injection attacks is that the attacker has complete knowledge about the power grid topology and transmission-line admittances. By contrast, this paper studies the general problem of blind false data injection attacks using the principal component analysis approximation method without the knowledge of Jacobian matrix and the assumption regarding the distribution of state variables. The proposed attack is proven to be approximately stealthy. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> The performance of the proposed attack is analyzed. Simulations confirm the performance of the proposed method.