Özet:

Abstract In the power system, incorporated distributed energy resources (DERs) bring the benefits of environmental, economic and technical. In addition to these benefits, it elevates some technical concerns when the penetration level of DERs is high. The major issues are islanding detection and it is necessary for equipment protection and personal safety. In this research, the phase angle of positive sequence voltage (PAOPSV) proposed a passive islanding detection scheme. The proposed hybrid method is the combined performance of dwarf mongoose optimizer (DMO) and random forest algorithm (RFA), hence commonly called as DMO-RFA method. From the strict analysis of 13 different factors, the PAOPSV is selected.  In addition to this, for islanding in solar photovoltaic (SPV) based distributed generation (DG) system. The comparative analysis shows that among all other parameters, the PAOPSV has better accuracy and sensitivity for islanding parameters.  In order to verify the efficiency  of proposed method, a comprehensive case study consider all worst-case scenarios, which  is accomplished the differentiates islanding events from the non-islanding events, like load, motor, capacitor, various fault type switching. The proposed system is implemented on MATLAB/Simulink, and the proposed work is performed on Intel Core 2 Quad CPUQ6600 at 2.40 GHz using 2 GB random access memory (RAM), MATLAB R2013a (8.1.0.604) 32-bit and Version 3.7.2.

Anahtar Kelimeler: