Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions

Overhead transmission line insulators are non-conductive materials that separate conductors from grounded transmission towers. Once in operation, they frequently experience environmental pollution and electrical or mechanical stress. Since adverse operational conditions can lead to insulation failur...

Full description

Saved in:
Bibliographic Details
Main Authors: Arailym Serikbay, Mehdi Bagheri, Amin Zollanvari, B. T. Phung
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/17/22/5595
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850216556306366464
author Arailym Serikbay
Mehdi Bagheri
Amin Zollanvari
B. T. Phung
author_facet Arailym Serikbay
Mehdi Bagheri
Amin Zollanvari
B. T. Phung
author_sort Arailym Serikbay
collection DOAJ
description Overhead transmission line insulators are non-conductive materials that separate conductors from grounded transmission towers. Once in operation, they frequently experience environmental pollution and electrical or mechanical stress. Since adverse operational conditions can lead to insulation failure, regular inspections are essential to prevent power outages. To this end, this paper proposes a novel technique based on deep convolutional neural networks (CNNs) to classify high-voltage insulator surface conditions based on their image. Successful applications of CNNs in computer vision have led to several pretrained architectures for image classification. To use these pretrained models, a practitioner typically fine-tunes and selects one final model via a model selection stage and discards all other models. In contrast with many existing studies that use such a “winner-takes-all” approach, here, we identify the best subset of seven popular pretrained CNN architectures that are combined by soft voting to form an ensemble classifier. From a machine learning (ML) perspective, this focus is warranted because the convolutional base of each pretrained architecture operates as a feature extractor and an ensemble of them works as a combination of various feature extraction rules. Our numerical experiments demonstrate the advantage of the identified ensemble model to individual pretrained architectures.
format Article
id doaj-art-a06ddfa5f24847509611125a258200b3
institution OA Journals
issn 1996-1073
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-a06ddfa5f24847509611125a258200b32025-08-20T02:08:15ZengMDPI AGEnergies1996-10732024-11-011722559510.3390/en17225595Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface ConditionsArailym Serikbay0Mehdi Bagheri1Amin Zollanvari2B. T. Phung3Department of Electrical and Computer Engineering, Nazarbayev University, Kabanbay Batyr Ave. 53, Astana 010000, KazakhstanDepartment of Electrical and Computer Engineering, Nazarbayev University, Kabanbay Batyr Ave. 53, Astana 010000, KazakhstanDepartment of Electrical and Computer Engineering, Nazarbayev University, Kabanbay Batyr Ave. 53, Astana 010000, KazakhstanSchool of Electrical Engineering and Telecommunications, University of New South Wales, Sydney, NSW 2052, AustraliaOverhead transmission line insulators are non-conductive materials that separate conductors from grounded transmission towers. Once in operation, they frequently experience environmental pollution and electrical or mechanical stress. Since adverse operational conditions can lead to insulation failure, regular inspections are essential to prevent power outages. To this end, this paper proposes a novel technique based on deep convolutional neural networks (CNNs) to classify high-voltage insulator surface conditions based on their image. Successful applications of CNNs in computer vision have led to several pretrained architectures for image classification. To use these pretrained models, a practitioner typically fine-tunes and selects one final model via a model selection stage and discards all other models. In contrast with many existing studies that use such a “winner-takes-all” approach, here, we identify the best subset of seven popular pretrained CNN architectures that are combined by soft voting to form an ensemble classifier. From a machine learning (ML) perspective, this focus is warranted because the convolutional base of each pretrained architecture operates as a feature extractor and an ensemble of them works as a combination of various feature extraction rules. Our numerical experiments demonstrate the advantage of the identified ensemble model to individual pretrained architectures.https://www.mdpi.com/1996-1073/17/22/5595deep learningensemble learningcondition assessmenthigh-voltage insulatorscontamination classification
spellingShingle Arailym Serikbay
Mehdi Bagheri
Amin Zollanvari
B. T. Phung
Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions
Energies
deep learning
ensemble learning
condition assessment
high-voltage insulators
contamination classification
title Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions
title_full Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions
title_fullStr Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions
title_full_unstemmed Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions
title_short Ensemble Pretrained Convolutional Neural Networks for the Classification of Insulator Surface Conditions
title_sort ensemble pretrained convolutional neural networks for the classification of insulator surface conditions
topic deep learning
ensemble learning
condition assessment
high-voltage insulators
contamination classification
url https://www.mdpi.com/1996-1073/17/22/5595
work_keys_str_mv AT arailymserikbay ensemblepretrainedconvolutionalneuralnetworksfortheclassificationofinsulatorsurfaceconditions
AT mehdibagheri ensemblepretrainedconvolutionalneuralnetworksfortheclassificationofinsulatorsurfaceconditions
AT aminzollanvari ensemblepretrainedconvolutionalneuralnetworksfortheclassificationofinsulatorsurfaceconditions
AT btphung ensemblepretrainedconvolutionalneuralnetworksfortheclassificationofinsulatorsurfaceconditions