A New Approach for Machine Learning-Based Fault Detection and Classification in Power Systems

Authors

H. A. Tokel, R. A. Halaseh, G. Alirezaei, R. Mathar,

Abstract

        The detection and classification of faulty conditions in power systems is a task of crucial importance for a reliable
operation. Recently, the use of high-resolution synchronized phasor measurements has been proposed by several researchers for fault detection and classification. Unlike the proposed approaches available in the literature, the central idea in this work is to leverage the delay information of phasor measurement streams to enable a faster recognition of faulty operation. In this work, therefore, we focus on the effect of the communication network delays on the fault detection time, and propose a novel training technique for fault detection and classification which takes delayed measurements into consideration. The performance of the proposed approach is verified using simulated power system data, where artificial neural networks are used for fault detection and classification.

Keywords

power system;fault detection; fault classification; machine learning; artificial neural network}

BibTEX Reference Entry 

@inproceedings{ToHaAlMa0,
	author = {Halil Alper Tokel and Rana Al Halaseh and Gholamreza Alirezaei and Rudolf Mathar},
	title = "A New Approach for Machine Learning-Based Fault Detection and Classification in Power Systems",
	booktitle = "2018 {IEEE} Power Energy Society Innovative Smart Grid Technologies Conference (ISGT)",
	address = {Washington DC, USA},
	note = {preliminary metadata entry, documents might follow soon — stay tuned!}
	}

Downloads

 Download bibtex-file

Sorry, this paper is currently not available for download.