Designing a robust cyber-attack detection and identification algorithm for DC microgrids based on Kalman filter with unknown input observer

Abstract

In recent years, microgrids (MGs) have attracted much attention from researchers due to the aggregation of distributed generation (DG) sources for visibility purposes, proper connection of these DGs and the power system, as well as more efficient control of the sources. However, high reliance on communication networks makes MGs vulnerable to deliberate cyber-attacks. Also, since the basis of attack detection is to process monitoring and state estimation of the system, this paper presents a Kalman filter (KF) algorithm to monitor the network while modelling MGs in the state space. Moreover, as uncertainty and disturbance are two critical challenges in modelling and estimating states of a system, robust detection of cyber-attacks along with separation of the effect of disturbances and uncertainties are of particular importance. Therefore, in this study, a novel attack reconstruction algorithm is designed using a combination of an unknown input observer method and KF for robust attack detection. Simulation results show excellent performance of the proposed approach in detecting, identifying and neutralizing the effect of cyber-attacks in the presence of disturbance and noise.