Home :: Academic Members :: News

view:38048   Last Update: 2020-1-8

Ali Amiri

Seyyed Ali Chavoshi, Reza Noroozian, Ali Amiri
Islanding detection of synchronous distributed generation resources using AdaBoost algorithm
تشخیص جزیره ای منابع تولید پراکنده همگام با استفاده از الگوریتم AdaBoost
Abstract


This paper presents a novel approach for intelligent islanding detection in a multi-machine distribution network using the Adaptive Boosting (AdaBoost) algorithm. The AdaBoost algorithm is an iterative procedure that improves the accuracy of classification models by combining multiple base classifiers. In this paper, two different case studies are specified so as to assess the performance of the AdaBoost in islanding detection. In order to train the AdaBoost algorithm, various scenarios are implemented in the case studies. A number of electrical parameters including frequency and voltage are utilized as a feature vector to describe the designed scenarios. In order to determine the accuracy of AdaBoost in detecting the islanding phenomenon, the proposed method is subsequently tested in further scenarios. Response time and non-detection zone of the suggested method are also evaluated in the case studies as the two other main criteria. The results indicate that this method can successfully detect islanding operation in a variety of operating states in the networks under study. All the simulations presented here are produced in the Simulink toolbox of MATLAB.

 

 

Copyright © 2022, University of Zanjan, Zanjan, Iran
master[at]znu.ac.ir