Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations

Halil Ibrahim Uckol; Suat Ilhan; Aydogan Ozdemir

doi:10.1109/ICHVE49031.2020.9279399

Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations

Halil Ibrahim Uckol, Suat Ilhan, Aydogan Ozdemir

Department of Electrical Engineering

Istanbul Technical University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

12 Citations (Scopus)

Abstract

This paper presents a method based-on Convolutional Neural Network (CNN) to identify the partial discharge (PD) sources located in medium voltage (MV) cable terminations. Five different defect types due to improper workmanship are artificially fabricated in 36 kV cross-linked polyethylene (XLPE) cable terminations. Phase-resolved Partial Discharge (PRPD) patterns of the aforementioned defects are recorded for 30 seconds (1500 cycles of the input voltage) to generate one pattern. Different cable sets are utilized in the training and testing steps to test the robustness of the model, and 600 patterns are extracted for each cable group. Every PRPD data is converted into a Red-Green-Blue (RGB) image to be inputted to the proposed CNN algorithm. The algorithm is used for both feature extraction and classification steps. Effects of hyperparameters on the performance of the method are evaluated and optimized.

Original language	English
Title of host publication	7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728155111
DOIs	https://doi.org/10.1109/ICHVE49031.2020.9279399
Publication status	Published - 6 Sept 2020
Event	7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Beijing, China Duration: 6 Sept 2020 → 10 Sept 2020

Publication series

Name	7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings

Conference

Conference	7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020
Country/Territory	China
City	Beijing
Period	6/09/20 → 10/09/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Keywords

convolutional neural network
medium voltage cable terminations
partial discharges
PRPD pattern image

Access to Document

10.1109/ICHVE49031.2020.9279399

Cite this

Uckol, H. I., Ilhan, S., & Ozdemir, A. (2020). Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations. In 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings Article 9279399 (7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICHVE49031.2020.9279399

Uckol, Halil Ibrahim ; Ilhan, Suat ; Ozdemir, Aydogan. / Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations. 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. (7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings).

@inproceedings{75f4c96dcd9a4fce9fc862de96edd06d,

title = "Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations",

abstract = "This paper presents a method based-on Convolutional Neural Network (CNN) to identify the partial discharge (PD) sources located in medium voltage (MV) cable terminations. Five different defect types due to improper workmanship are artificially fabricated in 36 kV cross-linked polyethylene (XLPE) cable terminations. Phase-resolved Partial Discharge (PRPD) patterns of the aforementioned defects are recorded for 30 seconds (1500 cycles of the input voltage) to generate one pattern. Different cable sets are utilized in the training and testing steps to test the robustness of the model, and 600 patterns are extracted for each cable group. Every PRPD data is converted into a Red-Green-Blue (RGB) image to be inputted to the proposed CNN algorithm. The algorithm is used for both feature extraction and classification steps. Effects of hyperparameters on the performance of the method are evaluated and optimized.",

keywords = "convolutional neural network, medium voltage cable terminations, partial discharges, PRPD pattern image",

author = "Uckol, {Halil Ibrahim} and Suat Ilhan and Aydogan Ozdemir",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 ; Conference date: 06-09-2020 Through 10-09-2020",

year = "2020",

month = sep,

day = "6",

doi = "10.1109/ICHVE49031.2020.9279399",

language = "English",

series = "7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings",

address = "United States",

}

Uckol, HI, Ilhan, S & Ozdemir, A 2020, Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations. in 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings., 9279399, 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020, Beijing, China, 6/09/20. https://doi.org/10.1109/ICHVE49031.2020.9279399

Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations. / Uckol, Halil Ibrahim; Ilhan, Suat; Ozdemir, Aydogan.
7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. 9279399 (7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations

AU - Uckol, Halil Ibrahim

AU - Ilhan, Suat

AU - Ozdemir, Aydogan

PY - 2020/9/6

Y1 - 2020/9/6

N2 - This paper presents a method based-on Convolutional Neural Network (CNN) to identify the partial discharge (PD) sources located in medium voltage (MV) cable terminations. Five different defect types due to improper workmanship are artificially fabricated in 36 kV cross-linked polyethylene (XLPE) cable terminations. Phase-resolved Partial Discharge (PRPD) patterns of the aforementioned defects are recorded for 30 seconds (1500 cycles of the input voltage) to generate one pattern. Different cable sets are utilized in the training and testing steps to test the robustness of the model, and 600 patterns are extracted for each cable group. Every PRPD data is converted into a Red-Green-Blue (RGB) image to be inputted to the proposed CNN algorithm. The algorithm is used for both feature extraction and classification steps. Effects of hyperparameters on the performance of the method are evaluated and optimized.

AB - This paper presents a method based-on Convolutional Neural Network (CNN) to identify the partial discharge (PD) sources located in medium voltage (MV) cable terminations. Five different defect types due to improper workmanship are artificially fabricated in 36 kV cross-linked polyethylene (XLPE) cable terminations. Phase-resolved Partial Discharge (PRPD) patterns of the aforementioned defects are recorded for 30 seconds (1500 cycles of the input voltage) to generate one pattern. Different cable sets are utilized in the training and testing steps to test the robustness of the model, and 600 patterns are extracted for each cable group. Every PRPD data is converted into a Red-Green-Blue (RGB) image to be inputted to the proposed CNN algorithm. The algorithm is used for both feature extraction and classification steps. Effects of hyperparameters on the performance of the method are evaluated and optimized.

KW - convolutional neural network

KW - medium voltage cable terminations

KW - partial discharges

KW - PRPD pattern image

UR - http://www.scopus.com/inward/record.url?scp=85099365198&partnerID=8YFLogxK

U2 - 10.1109/ICHVE49031.2020.9279399

DO - 10.1109/ICHVE49031.2020.9279399

M3 - Conference contribution

AN - SCOPUS:85099365198

T3 - 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings

BT - 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020

Y2 - 6 September 2020 through 10 September 2020

ER -

Uckol HI, Ilhan S, Ozdemir A. Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations. In 7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2020. 9279399. (7th IEEE International Conference on High Voltage Engineering and Application, ICHVE 2020 - Proceedings). doi: 10.1109/ICHVE49031.2020.9279399

Partial Discharge Pattern Classification based on Deep Learning for Defect Identification in MV Cable Terminations

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this