Low complexity wireless sensor system for partial discharge localisation

Ephraim Iorkyase, Christos Tachtatzis, Pavlos Lazaridis, Ian Glover, Robert Atkinson

Research output: Contribution to journalArticle

Abstract

This study describes a key element of any modern wireless sensor system: data processing. The authors describe a system consisting of a wireless sensor network and an algorithmic software for condition-based monitoring of electrical plant in a live substation. Specifically, the aim is to monitor for the presence of partial discharge (PD) using a matrix of inexpensive radio sensors with limited processing capability. A low-complexity fingerprinting technique is proposed, given that the sensor nodes to be deployed will be highly constrained in terms of processing power, memory and battery life. Two variants of artificial neural network (ANN) learning models (multilayer perceptron and generalised regression neural network) that use regression as a form of function approximation are developed and their performance compared to K-nearest neighbour and weighted K-nearest neighbour models. The results indicate that the ANN models yield superior performance in terms of robustness against noise and may be particularly suited for PD localisation.

LanguageEnglish
Pages158-165
Number of pages8
JournalIET Wireless Sensor Systems
Volume9
Issue number3
Early online date30 Jan 2019
DOIs
Publication statusPublished - 27 May 2019

Fingerprint

Partial discharges
Neural networks
Sensors
Multilayer neural networks
Processing
Sensor nodes
Wireless sensor networks
Data storage equipment
Monitoring

Keywords

  • wireless sensor system
  • wireless sensor network
  • condition-based monitoring
  • partial discharge

Cite this

@article{06d8a541fba14cc1a7cb59274751bb6a,
title = "Low complexity wireless sensor system for partial discharge localisation",
abstract = "This study describes a key element of any modern wireless sensor system: data processing. The authors describe a system consisting of a wireless sensor network and an algorithmic software for condition-based monitoring of electrical plant in a live substation. Specifically, the aim is to monitor for the presence of partial discharge (PD) using a matrix of inexpensive radio sensors with limited processing capability. A low-complexity fingerprinting technique is proposed, given that the sensor nodes to be deployed will be highly constrained in terms of processing power, memory and battery life. Two variants of artificial neural network (ANN) learning models (multilayer perceptron and generalised regression neural network) that use regression as a form of function approximation are developed and their performance compared to K-nearest neighbour and weighted K-nearest neighbour models. The results indicate that the ANN models yield superior performance in terms of robustness against noise and may be particularly suited for PD localisation.",
keywords = "wireless sensor system, wireless sensor network, condition-based monitoring, partial discharge",
author = "Ephraim Iorkyase and Christos Tachtatzis and Pavlos Lazaridis and Ian Glover and Robert Atkinson",
note = "This paper is a postprint of a paper submitted to and accepted for publication in IET Wireless Sensor Systems and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at the IET Digital Library",
year = "2019",
month = "5",
day = "27",
doi = "10.1049/iet-wss.2018.5075",
language = "English",
volume = "9",
pages = "158--165",
journal = "IET Wireless Sensor Systems",
issn = "2043-6386",
number = "3",

}

Low complexity wireless sensor system for partial discharge localisation. / Iorkyase, Ephraim; Tachtatzis, Christos; Lazaridis, Pavlos; Glover, Ian; Atkinson, Robert.

In: IET Wireless Sensor Systems, Vol. 9, No. 3, 27.05.2019, p. 158-165.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Low complexity wireless sensor system for partial discharge localisation

AU - Iorkyase, Ephraim

AU - Tachtatzis, Christos

AU - Lazaridis, Pavlos

AU - Glover, Ian

AU - Atkinson, Robert

N1 - This paper is a postprint of a paper submitted to and accepted for publication in IET Wireless Sensor Systems and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at the IET Digital Library

PY - 2019/5/27

Y1 - 2019/5/27

N2 - This study describes a key element of any modern wireless sensor system: data processing. The authors describe a system consisting of a wireless sensor network and an algorithmic software for condition-based monitoring of electrical plant in a live substation. Specifically, the aim is to monitor for the presence of partial discharge (PD) using a matrix of inexpensive radio sensors with limited processing capability. A low-complexity fingerprinting technique is proposed, given that the sensor nodes to be deployed will be highly constrained in terms of processing power, memory and battery life. Two variants of artificial neural network (ANN) learning models (multilayer perceptron and generalised regression neural network) that use regression as a form of function approximation are developed and their performance compared to K-nearest neighbour and weighted K-nearest neighbour models. The results indicate that the ANN models yield superior performance in terms of robustness against noise and may be particularly suited for PD localisation.

AB - This study describes a key element of any modern wireless sensor system: data processing. The authors describe a system consisting of a wireless sensor network and an algorithmic software for condition-based monitoring of electrical plant in a live substation. Specifically, the aim is to monitor for the presence of partial discharge (PD) using a matrix of inexpensive radio sensors with limited processing capability. A low-complexity fingerprinting technique is proposed, given that the sensor nodes to be deployed will be highly constrained in terms of processing power, memory and battery life. Two variants of artificial neural network (ANN) learning models (multilayer perceptron and generalised regression neural network) that use regression as a form of function approximation are developed and their performance compared to K-nearest neighbour and weighted K-nearest neighbour models. The results indicate that the ANN models yield superior performance in terms of robustness against noise and may be particularly suited for PD localisation.

KW - wireless sensor system

KW - wireless sensor network

KW - condition-based monitoring

KW - partial discharge

UR - https://digital-library.theiet.org/content/journals/iet-wss

U2 - 10.1049/iet-wss.2018.5075

DO - 10.1049/iet-wss.2018.5075

M3 - Article

VL - 9

SP - 158

EP - 165

JO - IET Wireless Sensor Systems

T2 - IET Wireless Sensor Systems

JF - IET Wireless Sensor Systems

SN - 2043-6386

IS - 3

ER -