A Passive Islanding Detection Method for Power System Connected with Multiple Distributed Generations Using an Unscented Kalman Filter

The integration of distributed generations (DGs) powered by renewable energy sources into the power grid has attracted a lot of interest. Despite the many benefits of distributed energy resources (DER), unintentional islanding remains a severe problem that endangers working employees, causes power quality deviation, poses a risk to power utility equipment, and out‐of‐phase connection may arise because of rapid reclosing. This paper provides an innovative passive islanding detection approach (PIDA) formulated on an unscented Kalman filter (UKF) that is suited for numerous distributed generations. The unscented transformation (UT) overcomes the linear Kalman filter's limitation for higher‐order terms in the Taylor series. The suggested approach computes selected harmonic distortion from three voltage recorded at the point of common coupling (PCC). The performance of the proposed technique is thoroughly investigated using IEEE‐13 bus system simulation in MATLAB/Simulink. In addition, various non‐islanding events with significant inductive load and fault conditions have been observed to test for nuisance tripping. Evaluation of the results reveals that the proposed passive method with less computation burden provides rapid detection with a negligible non‐detection zone and distinguishes islanding and normal state effectively. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.