A Novel Passive Islanding Detection Methods Using Wavelet Transform for Grid Connected PV System

This paper presents new passive islanding detection methods for a grid-connected PV (GCPV) system. In the first proposed approach, the 3-phase voltage (p.u) at the point of common coupling (PCC) is extracted and calculates the mean voltage samples. This is passed through the wavelet transform, decomposed up to the level-6(d6) coefficient and then it obviously localizes the event and as a result detects the islanding condition. Finally, to calculate the energy and standard deviation (STD) at level-6(d6) for 2-cycles mean voltage samples it will discriminate islanding condition from the non-islanding condition. The second proposed approach, the negative sequence voltage is considered at PCC. This is passed through the wavelet transform and successfully detects the islanding state. This method also calculates the energy and standard deviation (STD) at level-4(d4) for one cycle of negative sequence voltage it will discriminate islanding situation from the non-islanding situation. The proposed passive methods are tested on a GCPV system with islanding and non-islanding (normal operation and abrupt load change) situation. These techniques are turn out the extremely successfully in islanding detection at PCC for GCPV system. The test GCPV system and simulations are developed in MATLAB/Simulink Software.