Hybrid-YOLO for classification of insulators defects in transmission lines based on UAV

被引:79
作者
Souza, Bruno Jose [1 ,2 ]
Stefenon, Stefano Frizzo [3 ,4 ]
Singh, Gurmail [5 ]
Freire, Roberto Zanetti [1 ]
机构
[1] Pontif Catholic Univ Parana PUCPR, Polytech Sch EP, Ind & Syst Engn Grad Program PPGEPS, Rua Imac Conceicao 1155, BR-80215901 Curitiba, Brazil
[2] Pumatronix Elect Equipment Ltd, Bartolomeu L de Gusmao 1970, BR-81650050 Curitiba, Brazil
[3] Fdn Bruno Kessler, Digital Ind Ctr, Via Sommar 18, I-38123 Trento, Italy
[4] Univ Udine, Dept Math Comp Sci & Phys, Via Sci 206, I-33100 Udine, Italy
[5] Univ Wisconsin Madison, Dept Comp Sci, 1210 W Dayton St, Madison, WI 53706 USA
关键词
Deep learning; Power grid inspection; You only look once; FAULT-DIAGNOSIS; RECOGNITION; INSPECTION; SEGMENTATION;
D O I
10.1016/j.ijepes.2023.108982
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Transmission power lines are essential to supply electrical energy to consumption centers. Keeping a reliable transmission system requires the early identification of faults. Image-based inspection of transmission lines makes fault identification faster and more accessible since it can be carried out using unmanned aerial vehicles (UAVs) in hard-to-reach places. In this paper, it is proposed to use a hybrid version of the You Only Look Once (YOLO) using ResNet-18 classifier, for power system inspection based on real images of failed components recorded by UAVs. This work assumed a dataset including 1,593 power grid inspection pictures for a supervised training. Based on YOLOv5x, with an mAP of 0.99262, the proposed method was superior to YOLOv5n, YOLOv5s, YOLOv5m, and YOLOv5l for object detection. For the multiclassification task, with an F1_score result of 0.96216, the proposed Hybrid-YOLO was superior to distinct architectures as the VGG-11, VGG-13, VGG-16, VGG-19, ResNet-34, ResNet-50, ResNet-152, DenseNet-121, DenseNet-161, DenseNet-169, DenseNet-201, YOLOv5, YOLOv6, and YOLOv7 versions.
引用
收藏
页数:12
相关论文
共 99 条
[1]  
Anjum S, 2014, THESIS U WATERLOO
[2]   Safety Distance Analysis of 500kV Transmission Line Tower UAV Patrol Inspection [J].
Chen, De-Quan ;
Guo, Xiang-Hua ;
Huang, Po ;
Li, Feng-Hang .
IEEE LETTERS ON ELECTROMAGNETIC COMPATIBILITY PRACTICE AND APPLICATIONS, 2020, 2 (04) :124-128
[3]   Research on Recognition Method of Electrical Components Based on YOLO V3 [J].
Chen, Haipeng ;
He, Zhentao ;
Shi, Bowen ;
Zhong, Tie .
IEEE ACCESS, 2019, 7 :157818-157829
[4]  
Chen X., 2021, P 2021 5 IEEE ASIAN, P53
[5]   Insulator Recognition Method for Distribution Network Overhead Transmission Lines Based on Modified YOLOv3 [J].
Chen, Zhihao ;
Xiao, Yewei ;
Zhou, Yan ;
Li, Zhiqiang ;
Liu, Yang .
2020 CHINESE AUTOMATION CONGRESS (CAC 2020), 2020, :2815-2820
[6]  
Cheng X, 2022, 2022 7 INT C INT COM, P661, DOI [10.1109/ICSP54964.2022.9778822, DOI 10.1109/ICSP54964.2022.9778822]
[7]   Evaluation of Methods for Electric Field Calculation in Transmission Lines [J].
Corso, M. ;
Stefenon, S. ;
Couto, V. ;
Cabral, S. ;
Nied, A. .
IEEE LATIN AMERICA TRANSACTIONS, 2018, 16 (12) :2970-2976
[8]   Classification of Contaminated Insulators Using k-Nearest Neighbors Based on Computer Vision [J].
Corso, Marcelo Picolotto ;
Perez, Fabio Luis ;
Stefenon, Stefano Frizzo ;
Yow, Kin-Choong ;
Garcia Ovejero, Raul ;
Quietinho Leithardt, Valderi Reis .
COMPUTERS, 2021, 10 (09)
[9]   Unmanned aerial vehicle for transmission line inspection using an extended Kalman filter with colored electromagnetic interference [J].
da Silva, Mathaus Ferreira ;
Honorio, Leonardo M. ;
Marcato, Andre Luis M. ;
Vidal, Vinicius F. ;
Santos, Murillo F. .
ISA TRANSACTIONS, 2020, 100 :322-333
[10]   Yolo V4 for Advanced Traffic Sign Recognition With Synthetic Training Data Generated by Various GAN [J].
Dewi, Christine ;
Chen, Rung-Ching ;
Liu, Yan-Ting ;
Jiang, Xiaoyi ;
Hartomo, Kristoko Dwi .
IEEE ACCESS, 2021, 9 :97228-97242