Dynamic Attacks Against Inverter-Based Microgrids

In this paper, dynamic stealthy data integrity attacks are introduced as a class of attacks against frequency droop controllers in the inverter-based microgrid networks. The impact of adversarial actions on transient performance of frequency control schemes in the inverter-based microgrid networks is investigated. Each inverter is modeled as a single integrator and vulnerability of the standard frequency droop controller to dynamic integrity attacks on active power measurements is evaluated through a control theoretic analysis. The measurement residuals are utilized to detect data integrity attacks on sensor measurements in a power grid. It is shown that the attacker can destabilize the network frequency and bypass bad data detector (BDD) by a carefully designed coordinated attack on sensor measurements. The effects of the stealthy attack on the stability of the network frequency control are modeled as a pole placement optimization problem and the solution is obtained by leveraging the coordinate descent algorithm.