Intelligent islanding detection in smart microgrids using variance autocorrelation function-based modal current envelope

Abstract

Islanding detection is a critical issue in grid-connected distributed microgrid systems. Distributed generation in the current power system has caused many challenges. Consequently, detecting quick and effective islanding is the most critical issue to minimise equipment failure, avoid danger, and maintain grid safety. There are various techniques for islanding identification in microgrids. Three classifications have been applied to categorise these strategies, which are: active, passive, and hybrid. This paper proposes and demonstrates an efficient and accurate approach to islanding detection based on the Variance Autocorrelation Function of a Modal Current Envelope (VAMCE) technique. Demodulation techniques including synchronous real demodulation, square law demodulation, asynchronous complex square law demodulation, and the quadrature demodulation technique are employed to detect the envelope of the 3-phase current signal. The VAMCE methodology is better suited for islanding detection because of its response to current sensitivity under islanding scenarios but not under normal conditions. Several simulations under various settings, including normal and islanded scenarios are used to analyse this method. These simulations have demonstrated different situations, such as when the system works normally and when it does not. The VAMCE along with the quadrature demodulation technique outperforms the others. The proposed solution is not only more accurate but also much faster compared to other methods. The proposed approach can identify normal and islanded situations in just 0.4 s.