A4 Conference proceedings

Fault detection and classification based on deep learning in LVDC off-grid system


Publication Details
Authors: Demidov Iurii, Pinomaa Antti, Lana Andrey, Pyrhönen Olli
Publication year: 2020
Language: English
Related Journal or Series Information: European Conference On Power Electronics And Applications
Journal name in source: 2020 22nd European Conference on Power Electronics and Applications, EPE 2020 ECCE Europe
Journal acronym: EPE-ECCE Europe
ISBN: 978-1-7281-9807-1
eISBN: 978-9-0758-1536-8
ISSN: 2325-0313
JUFO-Level of this publication: 1
Open Access: Not an Open Access publication

Abstract

The integration of information
and communication technologies (ICT) into energy power systems provides
new applications and possibilities for grid control, operation, and
protection. In this paper, a smart self-sustained off-grid concept based
on photovoltaics- and battery energy storage system and low-voltage
direct current (LVDC) power distribution network is studied. Due to
small electrification and poor internet coverage, Sub-Saharan Africa is
the target area of the concept realization. Focusing on the application
of deep learning, this research presents new approaches to the power
system's protection, as fault detection and classification are one of
the most essential LVDC electricity distribution issues.


KeywordsDeep learning, Distributed generation (DG), Fault-detection, Long short-term memory (LSTM), Low-voltage direct current (LVDC), Off-grid

Last updated on 2020-03-12 at 11:42