An Effective Method for Islanding Detection Based on Variational Mode Decomposition

Abstract

DOI : 10.26650/electrica.2019.19007 The present study presents a new islanding detection method based on variational mode decomposition (VMD). In the method, the VMD transform of the voltage at point of common coupling is calculated, and then based on the decomposed modes, an islanding detection index for distinguishing between islanding and normal conditions will be derived. Pursuing this objective, the measured voltage is decomposed into four modes, and a new islanding detection index will be derived from their energy and their variations. Then, thresholds will be adjusted based on values of proposed index in various conditions on normal and islanding operations. The proposed algorithm decomposes the measured signal into a predetermined number of intrinsic modes around central frequencies; the method is insensitive to the noise and sampling frequency. To validate the method, a distribution network has been simulated in PSCAD/EMTDC, and the algorithm is tested on a variety of islanding and normal circumstances. The results illustrate desired performance of the proposed algorithm for conditions such as motor, nonlinear, and large load switching; as normal condition; and islanding condition with a variety of power mismatches between local load and distributed generation. The results demonstrate the desired performance of the method with respect to both speed and accuracy. Cite this article as : Salimi S, Koochaki A. An Effective Method for Islanding Detection Based on Variational Mode Decomposition. Electrica, 2019; 19(2): 135-1145.