End-to-End Partial Discharge Detection in Power Cables via Time-Domain Convolutional Neural Networks

Mohammad Azam Khan, Jaegul Choo, Yong Hwa Kim

Research output: Contribution to journalArticle

Abstract

The analysis of partial discharge (PD) signals has been identified as a standardized diagnostic tool in monitoring the condition of different electrical apparatus nowadays. In this paper, we propose a novel data-driven approach to detect PD pulses in power cables using one-dimensional convolutional neural networks (CNNs), a successful deep neural network approach. Applying this deep learning method, an end-to-end framework has been proposed considering the propagations of PD signals and noises in power cables. The proposed method uses PD pulses as input, automatically extracts meaningful features for waveforms of PD pulses, and finally detects PD. Most of the existing methods, which use traditional classifiers, such as support vector machines (SVMs) and multi-layer perceptron (MLP) have mainly focused on improving feature representation and extraction manually for this task. However, the proposed CNN-based detection algorithm captures important latent features for waveforms of PD pulses with the help of its automatic feature extraction capability from raw inputs. Our experimental results show that the proposed method is better than conventional SVMs and achieves 97.38% and 93.23% detection accuracies in end-to-end settings based on our theoretical model-generated and empirical real-world PD signals, respectively.

Original languageEnglish
JournalJournal of Electrical Engineering and Technology
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Partial discharges
Cables
Neural networks
Support vector machines
Multilayer neural networks
Feature extraction
Classifiers
Monitoring

Keywords

  • Convolutional neural networks
  • Deep neural networks
  • Fault detection
  • Partial discharges
  • Power cable

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

@article{6cf9ad99cad44fb4b1ab36c18d064d80,
title = "End-to-End Partial Discharge Detection in Power Cables via Time-Domain Convolutional Neural Networks",
abstract = "The analysis of partial discharge (PD) signals has been identified as a standardized diagnostic tool in monitoring the condition of different electrical apparatus nowadays. In this paper, we propose a novel data-driven approach to detect PD pulses in power cables using one-dimensional convolutional neural networks (CNNs), a successful deep neural network approach. Applying this deep learning method, an end-to-end framework has been proposed considering the propagations of PD signals and noises in power cables. The proposed method uses PD pulses as input, automatically extracts meaningful features for waveforms of PD pulses, and finally detects PD. Most of the existing methods, which use traditional classifiers, such as support vector machines (SVMs) and multi-layer perceptron (MLP) have mainly focused on improving feature representation and extraction manually for this task. However, the proposed CNN-based detection algorithm captures important latent features for waveforms of PD pulses with the help of its automatic feature extraction capability from raw inputs. Our experimental results show that the proposed method is better than conventional SVMs and achieves 97.38{\%} and 93.23{\%} detection accuracies in end-to-end settings based on our theoretical model-generated and empirical real-world PD signals, respectively.",
keywords = "Convolutional neural networks, Deep neural networks, Fault detection, Partial discharges, Power cable",
author = "Khan, {Mohammad Azam} and Jaegul Choo and Kim, {Yong Hwa}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s42835-019-00115-y",
language = "English",
journal = "Journal of Electrical Engineering and Technology",
issn = "1975-0102",
publisher = "Korean Institute of Electrical Engineers",

}

TY - JOUR

T1 - End-to-End Partial Discharge Detection in Power Cables via Time-Domain Convolutional Neural Networks

AU - Khan, Mohammad Azam

AU - Choo, Jaegul

AU - Kim, Yong Hwa

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The analysis of partial discharge (PD) signals has been identified as a standardized diagnostic tool in monitoring the condition of different electrical apparatus nowadays. In this paper, we propose a novel data-driven approach to detect PD pulses in power cables using one-dimensional convolutional neural networks (CNNs), a successful deep neural network approach. Applying this deep learning method, an end-to-end framework has been proposed considering the propagations of PD signals and noises in power cables. The proposed method uses PD pulses as input, automatically extracts meaningful features for waveforms of PD pulses, and finally detects PD. Most of the existing methods, which use traditional classifiers, such as support vector machines (SVMs) and multi-layer perceptron (MLP) have mainly focused on improving feature representation and extraction manually for this task. However, the proposed CNN-based detection algorithm captures important latent features for waveforms of PD pulses with the help of its automatic feature extraction capability from raw inputs. Our experimental results show that the proposed method is better than conventional SVMs and achieves 97.38% and 93.23% detection accuracies in end-to-end settings based on our theoretical model-generated and empirical real-world PD signals, respectively.

AB - The analysis of partial discharge (PD) signals has been identified as a standardized diagnostic tool in monitoring the condition of different electrical apparatus nowadays. In this paper, we propose a novel data-driven approach to detect PD pulses in power cables using one-dimensional convolutional neural networks (CNNs), a successful deep neural network approach. Applying this deep learning method, an end-to-end framework has been proposed considering the propagations of PD signals and noises in power cables. The proposed method uses PD pulses as input, automatically extracts meaningful features for waveforms of PD pulses, and finally detects PD. Most of the existing methods, which use traditional classifiers, such as support vector machines (SVMs) and multi-layer perceptron (MLP) have mainly focused on improving feature representation and extraction manually for this task. However, the proposed CNN-based detection algorithm captures important latent features for waveforms of PD pulses with the help of its automatic feature extraction capability from raw inputs. Our experimental results show that the proposed method is better than conventional SVMs and achieves 97.38% and 93.23% detection accuracies in end-to-end settings based on our theoretical model-generated and empirical real-world PD signals, respectively.

KW - Convolutional neural networks

KW - Deep neural networks

KW - Fault detection

KW - Partial discharges

KW - Power cable

UR - http://www.scopus.com/inward/record.url?scp=85064669814&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064669814&partnerID=8YFLogxK

U2 - 10.1007/s42835-019-00115-y

DO - 10.1007/s42835-019-00115-y

M3 - Article

JO - Journal of Electrical Engineering and Technology

JF - Journal of Electrical Engineering and Technology

SN - 1975-0102

ER -