Chapter 10 - False data injection attacks and countermeasures in smart microgrid systems

Abstract

During the past decades, many efforts have been made to ensure the cyber security of smart DC microgrid (DCmG) systems. However, there exists almost no work considering false data injection (FDI) attacks in DCmGs, which can cause economic losses or even destabilize the system voltages/currents without being perceived. In this chapter, we consider a hierarchical control-based DCmG equipped with unknown input observer-based detectors, which are widely adopted in the detection and identification of cyber-attacks. We first prove that once the attacker knows the bounds of the initial estimation error and measurement noise, he/she can launch nonzero-dynamics stealthy (NDS) attacks in DCmGs, which can affect the detection residuals but remain stealthy. Thus, NDS attacks are one of the most threatening FDI attacks in DCmGs. Considering the complexity of the multilayer control, we simplify the primary control loops as static unit gains and obtain the systematic dynamics model of DCmG under NDS attacks. Then, we obtain the analytical expressions of the voltage and output current at each point of common coupling (PCC) to show the effects of NDS attacks on voltage balancing and current sharing, respectively. Moreover, an upper bound of the PCC voltage is given, which shows that under NDS attacks, the voltage and current convergences can still be achieved exponentially in DCmG. In addition, corresponding countermeasures are given against NDS attacks. Finally, extensive simulations are conducted in Simulink/PLECS to validate our theoretical analysis.