Time-frequency transform-based islanding detection in distributed generation

TLDR: In this paper, a new technique for islanding detection in distributed generation (DG) using time-frequency transform such as S -transform is presented, where the spectral energy index (ratio of spectral energy content of the voltage to current signals) is computed to track the islanding situation from non-islanding conditions from sudden load change, tripping of other DG, etc.

Abstract: This study presents a new technique for islanding detection in distributed generation (DG) using time-frequency transform such as S -transform. S -transform is an invertible time-frequency spectral localisation technique that provides the time-frequency contours of the voltage and current signals retrieved at the target DG location. The energy index (ratio of the spectral energy content of the voltage to current signals) is computed to track the islanding situation from non-islanding conditions such as sudden load change, tripping of other DG, etc. Further, to aid the islanding detection scheme, a cumulative sum detector is also computed based on the spectral energy content of the negative sequence components of the current and voltage signals. The results, based on extensive study, indicate that the proposed technique can reliably detect islanding in DGs connected to power distribution network.