Discriminative Discordant Element-based Detection of False Data Injection Attacks in DC Microgrids with Distributed Consensus Control

Integrating distributed consensus controls into the secondary control layer for DC microgrids (DCMG) can lead to performance improvement in proportionate current sharing and average voltage regulation. However, cybersecurity risks are apparent when control data transmission exists in the secondary control layer. Specifically, false data injection attacks (FDIAs) may severely degrade regulation performance and compromise microgrids’ stability. Existing discordant element (DE)-based detection methods are found to have limitations in differentiating between sensor tampering and communication link intrusions, which in some way reduce the practical value. This paper focuses on designing and validating a discriminative detection algorithm to complement the DE-based detection method. The enhancement is centered around the concept of developing a logic-based decision using the newly defined global auxiliary consensus variable that is integrated into the DE-based detection flow. The MATLAB/Simulink studies confirm its efficacy in providing definitive discrimination of communication link attacks from sensor attacks for the worst case (multi-point FDIAs) and identification for compromised distributed energy resources (DER).