Innovative Sandia Frequency Shift Islanding Detection for Microgrids Using Interval Type-II Fuzzy Logic and VP-PSO Optimization

This paper presents an innovative islanding detection method (IDM) that combines interval type-2 fuzzy logic with the Sandia frequency shift (SFS) technique for enhanced microgrid protection. The study examines a hybrid microgrid incorporating photovoltaic and wind farm distributed generation sources, explicitly accounting for uncertainties in solar irradiation, wind speed, and load demand through probabilistic modeling. Using probability density functions (PDFs), multiple uncertainty scenarios were generated, with representative scenarios selected via Monte Carlo simulation (MCS) for IDM development. The proposed approach further integrates a variable-parameter PSO (VP-PSO) algorithm to optimize both fuzzy membership functions and controller parameters. Comprehensive simulations demonstrate 100% detection accuracy for both PV and wind farm islanding scenarios, significantly outperforming conventional methods: standard SFS (84-86%), type-1 fuzzy SFS (89-91%), and neural network approaches (94-95%). This consistent 12-16% performance improvement establishes the method’s superior reliability for practical microgrid applications.