Real-Time Estimation of Parameters in Inductive-Resistive Power Networks Using Second-Order Extended Kalman Filter

Abstract

In many control schemes for grid-connected systems, the acquisition of weak grid parameters is critical for robust stability and output quality improvement of grid-connected converter systems. In this paper, the second-order extended Kalman filter (SOEKF) algorithm, which enables recursive computation, is used for real-time estimation of weak grid parameters (including, PCC voltage, grid current, grid voltage, and grid impedance). Considering that the nonlinear dynamics of the inductive-resistive network can lead to rounding errors and subsequently deteriorate the accuracy of the real-time estimation, we reconstruct the inductive-resistive network model to a second-order nonlinear form that can be fully expanded with Taylor series. The theoretical advantages of the SOEKF over the extended Kalman filter (EKF) are dissected in the context of actual inductive-resistive networks. The simulation tests verify that the proposed method can accurately estimate each parameter of the weak grid, and the estimation accuracy is better than that of EKF.