Estimating Faults in Nonlinear DC Microgrids with Constant Power Loads: A Dual-Extended Kalman Filter Approach

Abstract

This article investigates the problem of detecting and estimating actuator fault in direct current microgrids (DC MGs) with linear and nonlinear constant power loads (CPLs). The actuator fault is modeled by using an additive term in the state-space and highly influences the system response if it is not compensated. An advanced dual-extended Kalman filter (dual-EKF) is proposed to estimate the system states and the accruing actuator fault. Though the presented approach offers a systematic procedure to divide the augmented state vector into two parts and these parts can be estimated in parallel. Thereby, the online computational burden is reduced as it can be implemented by two processes in parallel. The proposed approach does not require restrictive assumptions on the system matrices and is robust against stochastic Gaussian noises. The proposed approach is applied on a practical faulty DC MG benchmark connected to a CPL; and, the results are compared with other state-of-the-art methods from the computational burden and estimation accuracy points of view.