Approach for identification and classification of HIFs in medium voltage distribution networks

Abstract

The modern power system operation is faced with numerous challenges related to the power quality improvements such as identification and classification of power distribution network (PDN) faults. The recent advances in the area of signal processing allow the development of new algorithms and methods which can be used for fault identification and classification in PDN. This study presents a comparison of two approaches for identification and classification of high-impedance faults (HIFs) in medium-voltage PDN. The first approach is based on the voltage phase difference algorithm, whereas the second approach is based on the combination of discrete wavelet transform and artificial neural networks algorithm. The proposed algorithms are tested on models of a real distribution network, which represents a typical PDN currently used in Bosnia and Herzegovina. It was demonstrated that the proposed methods are capable to accurately detect and classify HIF in PDN. This study makes a contribution to the existing body of knowledge by developing, testing and comparing two methods for HIF classification and identification, whose application represents an improvement when compared with the capability of the existing protection devices.