A Deep Learning Game Theoretic Model for Defending Against Large Scale Smart Grid Attacks

Abstract

Power grids that are interdependent with communication networks create more possible modes of failure (e.g., cyberattacks) as well as more complex propagation of failure through the coupled networks. To ensure robust defense of smart grids, it is important to model both attacker and defender as intelligent, a scenario that the framework of game theory provides methods to analyze. However, prior works in applying game theoretic models to smart grid security limit the problem space to a small number targets under threat due to the inability of state-of-the-art methods to scale to large networks. Our method scales to large networks by combining neural networks that use featurized action representations with an approximation of large combinatorial actions to generalize knowledge about the best targets to attack/defend across graphs of various topologies and sizes. Our model's invariance to the size of the input graph allows us to transfer knowledge from games played on small graphs during training to large graphs during evaluation. Our experiments show that our method can learn Nash equilibrium strategies on small networks, and demonstrate low exploitability when generalized to large networks, especially compared to the common heuristics currently used to simulate attacks on large graphs.

Cross-Space Conduction Assessment Method of Network Attack Risk under the Strong Coupling Characteristics of Electric Power Cyber Physics

“…Ubiquitous awareness capabilities of intelligent measurement and control terminals based on power cyber-physical fusion system, edge computing technology and advanced communication network. The power physics system and the information system are deeply integrated and intertwined, and the transmission and conversion relationship between information flow and energy flow can be described by the following three interactive links by Cunningham et al [2], forming a closed-loop control structure of "acquisition-local decision-central coordination-control," as shown in Figure 1.…”

“…Documents by Cunningham et al [2], Zhang et al [3], Xin et al [4], Xu et al [5], Tang et al [6], and Liu et al [7] point out that the deep integration of information system and grid physical system have comprehensively improved the observability, description, and controllability of the grid, but at the same time it has also aggravated the risk of cross-space propagation of faults and expanded the scale of accidents, "12•23 Ukrainian Blackout" and "3.7 Venezuelan Blackout" and other information space network attacks show that the current CPS is facing more and more serious security threats. These blackouts indicate that the information system has failed or been attacked, which will not only affect the information system, but also interfere with the safe and stable operation of the physical system.…”