Learning to Attack Powergrids with DERs

TL;DR

This paper demonstrates how autonomous agents can learn to execute reactive power attacks on power grids by exploiting their dynamic behavior, even with limited sensory input, highlighting new vulnerabilities.

Contribution

It introduces a realistic simulation framework incorporating transient dynamics and limited attacker sensors, showing that agents can learn effective power grid attacks.

Findings

Agents successfully learned reactive power attack strategies

Attacks remain effective despite independent grid node actions

Simulation reflects realistic transient power grid behavior

Abstract

In the past years, power grids have become a valuable target for cyber-attacks. Especially the attacks on the Ukrainian power grid has sparked numerous research into possible attack vectors, their extent, and possible mitigations. However, many fail to consider realistic scenarios in which time series are incorporated into simulations to reflect the transient behaviour of independent generators and consumers. Moreover, very few consider the limited sensory input of a potential attacker. In this paper, we describe a reactive power attack based on a well-understood scenario. We show that independent agents can learn to use the dynamics of the power grid against it and that the attack works even in the face of other generator and consumer nodes acting independently.