Classification of Distribution Network Faults Using Hilbert-Huang Transform and Artificial Neural Network

Abstract

Identification and classification of faults in the electrical power system remain one of the most important tasks for the system operators and managers. In particular, high impedance faults (HIF) identification and classification is an especially challenging task due to the physical properties of the waveforms and low neutral voltage. In recent years, there have been significant technology driven advances in this field of research, owned to the introduction and progress of smart grid technologies. However, this topic still remains an open area of research. This paper presents a method for classification of HIF in a medium voltage distribution network, based on the Hilbert-Huang Transform and Artificial Neural Networks. This method was tested on generated signals based on the model of a realistic distribution system. The results indicated that the proposed algorithm is capable to accurately classify HIF in the distribution system. This paper contributes to the existing research by developing and testing, on a model of realistic distribution system, a HIF classification method which offers very efficient and accurate performance.