A Geometric Approach for Cyber-Attack Detection in DC Microgrids

In this paper, a distributed method for detecting cyber-attacks is proposed based on linear geometric control theory. While dynamic consensus strategies are increasingly employed in the secondary control of dc microgrids, the communication links between energy sources introduce potential vulnerabilities to cyber-attacks. To address this, a simplified model of the de network that reduces the a priori information required between sources is first developed. Building on this model, a reduced-order observer is designed to detect and identify cyber-attacks that compromise secondary control. Finally, the effectiveness of the approach is validated through simulations using a detailed dc microgrid model.