Distributed Agent-Based Intrusion Detection for the Smart Grid
Date
January 19, 2018
Description
We study the impact of cyber attacks on the power grid control system and adapt well-studied control systems theories to detect and mitigate these cyber attacks. This work specifically focuses on multi-agent systems as a distributed paradigm for intrusion detection and mitigation in smart grids. Agents breakdown the complex system and interactively monitor for cyber intrusions. To coordinate their results, agents run the consensus algorithm to update neighboring agents in a time-bound manner. To demonstrate the agent-based intrusion detection strategy we address false data injection attacks and show that the probability of undetectable false data injection attacks can be as low as 0.01.