Amir Hossein Kimiai Asadi - Engineering Department, Hamedan Branch, Islamic Azad University, Hamedan, Iran
Hasan Meyar Naeimy - Engineering Department, Hamedan Branch, Islamic Azad University, Hamedan, Iran

In this paper, a Hybrid Islanding Detection Method (HIDM) is proposed for Multiple Distributed Generations (MDGs) in distribution networks. This method is able to Islanding Detection (ID) independently type of DGs, network mode of operation and topology. The method is based upon two criteria, including: 3rd decomposition level of unbalance voltage magnitude by using Discrete Wavelet Transform (DWT) with Daubechies mother wavelet and also phase angle difference between negative sequence voltage and current. While, in conventional methods: voltage magnitude, phase displacement and frequency change, islanding conditions can be detected easily by monitoring these parameters. However, in case of small changes in loading the conventional methods have some difficulty in detecting such a particular islanding condition. The performance of the method, i.e. accuracy, speed and reliability are evaluated by using a sample IEEE 13 nodes standard distribution network and considering three scenarios, including ground fault, load rejection and capacitor bank switching, then performance of this method compared with Rate Of Change Of Frequency (ROCOF) and Rate Of Change of Phase Angle Difference (ROCPAD) methods. The PSCAD/EMTDC is sued to simulate these scenarios. Simulation results show that the proposed method can detect the islanding operation correctly and the Non Detection Zone (NDZ) reduced, remarkably in comparison to conventional methods and also ROCOF and ROCPAD relays.

Islanding detection, Discrete wavelet transform, Unbalance voltage, Rate Of Change Of Frequency, Rate Of Change of Phase Angle Difference