{"title":"利用语义分割对架空输电线路过热故障进行智能诊断","authors":"Xiangyu Yang, Youping Tu, Zhikang Yuan, Zhong Zheng, Geng Chen, Cong Wang, Yan Xu","doi":"10.1049/hve2.12403","DOIUrl":null,"url":null,"abstract":"<p>The strain clamps and leading wires are important components that connect conductors on overhead transmission lines and conduct current. During operation, poor contact between these components can cause abnormal overheating, leading to electric failures and threatening power system reliability. Recently, the use of unmanned aerial vehicles equipped with infrared thermal imagers for strain clamp and leading wire maintenance has become increasingly popular. Deep learning-based image recognition shows promising prospects for intelligent fault diagnosis of overheating faults. A pre-treatment method is proposed based on dynamic histogram equalisation to enhance the contrast of infrared images. The DeepLab v3+ network, loss function, and existing networks with different backbones are compared. The DeepLab v3+ network with ResNet101 and convolutional block attention module added, and the Focal Loss function achieved the highest performance with an average pixel accuracy of 0.614, an average intersection over union (<i>AIoU</i>) of 0.567, an F1 score of 0.644, and a frequency weighted intersection over union of 0.594 on the test set. The optimised Atrous rates has increased the <i>AIoU</i> by 12.91%. Moreover, an intelligent diagnosis scheme for evaluating the defect state of the strain clamps and leading wires is proposed and which achieves a diagnostic accuracy of 91.0%.</p>","PeriodicalId":48649,"journal":{"name":"High Voltage","volume":"9 2","pages":"309-318"},"PeriodicalIF":4.4000,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/hve2.12403","citationCount":"0","resultStr":"{\"title\":\"Intelligent overheating fault diagnosis for overhead transmission line using semantic segmentation\",\"authors\":\"Xiangyu Yang, Youping Tu, Zhikang Yuan, Zhong Zheng, Geng Chen, Cong Wang, Yan Xu\",\"doi\":\"10.1049/hve2.12403\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>The strain clamps and leading wires are important components that connect conductors on overhead transmission lines and conduct current. During operation, poor contact between these components can cause abnormal overheating, leading to electric failures and threatening power system reliability. Recently, the use of unmanned aerial vehicles equipped with infrared thermal imagers for strain clamp and leading wire maintenance has become increasingly popular. Deep learning-based image recognition shows promising prospects for intelligent fault diagnosis of overheating faults. A pre-treatment method is proposed based on dynamic histogram equalisation to enhance the contrast of infrared images. The DeepLab v3+ network, loss function, and existing networks with different backbones are compared. The DeepLab v3+ network with ResNet101 and convolutional block attention module added, and the Focal Loss function achieved the highest performance with an average pixel accuracy of 0.614, an average intersection over union (<i>AIoU</i>) of 0.567, an F1 score of 0.644, and a frequency weighted intersection over union of 0.594 on the test set. The optimised Atrous rates has increased the <i>AIoU</i> by 12.91%. Moreover, an intelligent diagnosis scheme for evaluating the defect state of the strain clamps and leading wires is proposed and which achieves a diagnostic accuracy of 91.0%.</p>\",\"PeriodicalId\":48649,\"journal\":{\"name\":\"High Voltage\",\"volume\":\"9 2\",\"pages\":\"309-318\"},\"PeriodicalIF\":4.4000,\"publicationDate\":\"2024-01-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://onlinelibrary.wiley.com/doi/epdf/10.1049/hve2.12403\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"High Voltage\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1049/hve2.12403\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"High Voltage","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1049/hve2.12403","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
Intelligent overheating fault diagnosis for overhead transmission line using semantic segmentation
The strain clamps and leading wires are important components that connect conductors on overhead transmission lines and conduct current. During operation, poor contact between these components can cause abnormal overheating, leading to electric failures and threatening power system reliability. Recently, the use of unmanned aerial vehicles equipped with infrared thermal imagers for strain clamp and leading wire maintenance has become increasingly popular. Deep learning-based image recognition shows promising prospects for intelligent fault diagnosis of overheating faults. A pre-treatment method is proposed based on dynamic histogram equalisation to enhance the contrast of infrared images. The DeepLab v3+ network, loss function, and existing networks with different backbones are compared. The DeepLab v3+ network with ResNet101 and convolutional block attention module added, and the Focal Loss function achieved the highest performance with an average pixel accuracy of 0.614, an average intersection over union (AIoU) of 0.567, an F1 score of 0.644, and a frequency weighted intersection over union of 0.594 on the test set. The optimised Atrous rates has increased the AIoU by 12.91%. Moreover, an intelligent diagnosis scheme for evaluating the defect state of the strain clamps and leading wires is proposed and which achieves a diagnostic accuracy of 91.0%.
High VoltageEnergy-Energy Engineering and Power Technology
CiteScore
9.60
自引率
27.30%
发文量
97
审稿时长
21 weeks
期刊介绍:
High Voltage aims to attract original research papers and review articles. The scope covers high-voltage power engineering and high voltage applications, including experimental, computational (including simulation and modelling) and theoretical studies, which include:
Electrical Insulation
● Outdoor, indoor, solid, liquid and gas insulation
● Transient voltages and overvoltage protection
● Nano-dielectrics and new insulation materials
● Condition monitoring and maintenance
Discharge and plasmas, pulsed power
● Electrical discharge, plasma generation and applications
● Interactions of plasma with surfaces
● Pulsed power science and technology
High-field effects
● Computation, measurements of Intensive Electromagnetic Field
● Electromagnetic compatibility
● Biomedical effects
● Environmental effects and protection
High Voltage Engineering
● Design problems, testing and measuring techniques
● Equipment development and asset management
● Smart Grid, live line working
● AC/DC power electronics
● UHV power transmission
Special Issues. Call for papers:
Interface Charging Phenomena for Dielectric Materials - https://digital-library.theiet.org/files/HVE_CFP_ICP.pdf
Emerging Materials For High Voltage Applications - https://digital-library.theiet.org/files/HVE_CFP_EMHVA.pdf