Dual-Indexed Ensemble Kalman Filtering-Based Anti-Islanding Detection Methods for AC Microgrids

Abstract

This paper presents a novel Ensemble Kalman Filter (EnKF)-based passive anti-islanding method designed to enhance the reliability and stability of AC microgrids amid increasing integration of distributed energy resources (DERs). The dynamic and nonlinear characteristics of AC microgrids pose significant challenges to conventional passive islanding detection methods. To address these limitations, the proposed approach employs the EnKF as a state observer to accurately estimate the point of common coupling (PCC) voltage. In this framework, two robust indices are generated: 1) the Ensemble Kalman Filter residual (EnKFR), derived from the discrepancy between the estimated and measured PCC voltages, and 2) the 3rd harmonic distortion (3rdHD), computed from the 3rd harmonic signal estimated by the EnKF. By applying an OR operation to both the EnKFR and the 3rdHD, the proposed method reliably detects islanding events while effectively differentiating them from non-islanding events. Extensive simulations were conducted on various standards such as the IEEE and UL-1741 microgrid test networks under a range of operating conditions. Results reveal that the EnKF-based method delivers enhanced detection accuracy, swift response times, a minimized non-detection zone (NDZ), and robust immunity to false positives. The findings underscore the superiority of this approach over conventional methods, with successful detection demonstrated under both balanced and unbalanced load and generation conditions. This novel scheme offers a rigorous and innovative solution to islanding detection, providing substantial improvements in microgrid stability and operational reliability.

Keywords

• Anti-islanding
• ensemble Kalman filter
• microgrids
• renewable energy sources
• smart cities
• smart grids