Cyber Attack Estimation of Nonlinear DC Microgrids with Noisy Measurements: Spherical Simplex Radial CKF Approach

Abstract

Nowadays, cyber-attacks are immoral ways to disturb power grids and make them devitalized or unstable. To avoid such issues, detection and estimation of cyber-attacks are essential in modern power utilities. This paper investigates a new nonlinear Kalman filter, called Spherical Simplex Radial Cubature Kalman Filter (SSRCKF), as a reliable and accurate estimator to detect cyber-attacks in nonlinear Direct Current (DC) microgrids (MGs) with Constant Power Loads (CPLs). The developed nonlinear Kalman filter offers a high resilient State Estimation (SE) performance in presence of noise as well as desired accuracy, and low computational burden. The SSRCKF is modified for the attack detection issue by augmenting the occurring attacks to the original nonlinear dynamics. The developed SSRCKF approach is deployed to estimate the DC MG states in the presence of Denial-of-Service (DoS) and data integrity attacks. Simulation results show that the presented filter estimates the DC MGs states of the system immediately after the attack.