False Data Injection Attacks and Corresponding Countermeasure in DC Microgrid

In this paper, we consider a hierarchical control based DC microgrid (DCmG) equipped with unknown input observer (UIO) based detectors, and investigate false data injection (FDI) attacks against the secondary control layer. First, we point out that the fundamental limitation of the UIO-based detector comes from the unknown inputs, whose dynamics can be utilized to construct zero trace undetectable (ZTU) attacks causing zero impact on detection residuals. Furthermore, we extend ZTU attacks to nonzero trace undetectable (NTU) attacks by utilizing the system noise, under which the detection residuals are still bounded by certain detection thresholds. Then, by approximating primary control loops as unit gains, we theoretically analyze attack impacts of NTU attacks on voltage balancing and current sharing in the DCmG. Moreover, we propose a countermeasure against ZTU and NTU attacks by observing the average PCC voltage deviation. Finally, extensive simulations are conducted in Simulink/PLECS to validate theoretical results and effectiveness of the countermeasure.