Intelligent Classfication Method of Composite Insulator Surface Hydrophobicity Based on Target Detection Algorithm

HE Xiao; SUN Jian; DENG Yeqiang; WANG Yu; FU Zhongxiang; JIANG Jiongting; WEN Xishan

doi:null

您当前的位置：

首页 >

文章列表页 >

Intelligent Classfication Method of Composite Insulator Surface Hydrophobicity Based on Target Detection Algorithm

更新时间：2025-06-18

- Intelligent Classfication Method of Composite Insulator Surface Hydrophobicity Based on Target Detection Algorithm
- High Voltage Apparatus Pages: 1-11(2025)
- 作者机构：
  
  1.武汉大学电气与自动化学院，武汉　 430072
  2.国网老河口供电公司，湖北　襄阳　 420682
  3.国网宁波供电公司，浙江　宁波　 315010
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation Youth Project(52207168);National Natural Science Foundation of China(51977162)
- DOI：
  CLC：
- Published：
- 稿件说明：
移动端阅览
HE Xiao, SUN Jian, DENG Yeqiang, et al. Intelligent Classfication Method of Composite Insulator Surface Hydrophobicity Based on Target Detection Algorithm[J/OL]. High voltage apparatus, 2025, 1-11.
DOI：

HE Xiao, SUN Jian, DENG Yeqiang, et al. Intelligent Classfication Method of Composite Insulator Surface Hydrophobicity Based on Target Detection Algorithm[J/OL]. High voltage apparatus, 2025, 1-11. DOI：

摘要

目前基于卷积神经网络将复合绝缘子喷水图像进行整体憎水性分类的方法对于图像局部的憎水性关注度不足，因此本文提出了一种基于目标检测算法的复合绝缘子表面憎水性判别方法。首先取样不同形态的喷水图片共5 800张，根据水珠形貌和接触角提出了单独针对水珠的分类标准。之后采用以SE-Resnet为骨架网络的Faster R-CNN对表面水珠进行分类，并获得了基于目标检测算法的21个水珠局部特征参数。为了兼顾图片全局特性，同时基于数字图像处理构建了12个与水珠亮斑面积和形态相关的全局参数。最后通过特征筛选，建立了基于BP神经网络的26参数憎水性等级自动判别模型，实现了检测过程的自动化。在±1误差允许范围内，模型判别准确率达到了99.13%。

Abstract

The current method of classifying the overall hydrophobicity of composite insulator spray images based on convolutional neural network does not pay enough attention to the local water repellency of the images

so this paper proposes a method to discriminate the water repellency of composite insulator surface based on target detection algorithm. Firstly

5 800 water spray images with different morphologies are sampled

and the classification criteria for water droplets are proposed according to their morphology. After that

Faster R-CNN with SE-Resnet as the backbone network is used to classify the surface water droplets

and 21 local feature parameters of water droplets based on the target detection algorithm are obtained. In order to take into account the global characteristics of the image

12 global parameters related to the bright spots and morphology of water droplets are also constructed based on digital image processing. Finally

a BP neural network-based 26-parameter hydrophobicity automatic classification model is constructed after feature screening

which realizes the automatic classification process. The accuracy of the model reached 99.13% with ±1 error tolerance.

关键词

Keywords

references

绝缘子表面湿润性测量导则：GB/T 24622—2022 [S ] . 2022 .

Guidance on the measurement of wettability of insulator surfaces: GB/T 24622—2022 [S ] . 2022 .

BERG M , THOTTAPPILLIL R , SCUKA V . Hydrophobicity estimation of HV polymeric insulating materials. Development of a digital image processing method [J ] . IEEE Transactions on Dielectrics and Electrical Insulation , 2001 , 8 ( 6 ): 1098 - 1107 .

TOKORO T , OMOTO Y , KOSAKI M . Image analysis of hydrophobicity of polymer insulators using PVM [C ] // 2001 Annual Report Conference on Electrical Insulation and Dielectric Phenomena) . Kitchener : IEEE , 2001 : 581 - 584 .

TOKORO T , NAGAO M , KOSAKI M . Image analysis of hydrophobicity of silicone rubber insulator [C ] // 1999 Annual Report Conference on Electrical Insulation and Dielectric Phenomena . [S.l.] : IEEE , 1999 : 763 - 766 .

刘彪 , 袁文海 , 董小顺 , 等 . 基于改进边缘连接Canny算法的绝缘子憎水性图像分割研究 [J ] . 高压电器 , 2022 , 58 ( 1 ): 162 - 169 .

LIU Biao , YUAN Wenhai , DONG Xiaoshun , et al . Research on hydrophobic image segmentation of insulator based on improved edge connection canny algorithm [J ] . High Voltage Apparatus , 2022 , 58 ( 1 ): 162 - 169 .

颜伟韬 . 基于图像处理的复合绝缘子憎水性判别方法研究 [D ] . 长沙 : 湖南大学 , 2018 .

YAN Weitao . Research on the discrimination method for the hydrophobicity of composite insulators based on image processing [D ] . Changsha : Hunan University , 2018 .

邱志斌 , 于小彬 , 霍锋 , 等 . 基于一致性测度区间分类的复合绝缘子喷水图像处理与憎水性智能识别 [J ] . 高电压技术 , 2020 , 46 ( 9 ): 3008 - 3017 .

QIU Zhibin , YU Xiaobin , HUO Feng , et al . Spray image processing of composite insulators based on interval classification of uniformity measure and intelligent identification of hydrophobicity [J ] . High Voltage Engineering , 2020 , 46 ( 9 ): 3008 - 3017 .

GIRSHICK R , DONAHUE J , DARRELL T , et al . Rich feature hierarchies for accurate object detection and semantic segmentation [C ] // 2014 IEEE Conference on Computer Vision and Pattern Recognition . [S.l.] : IEEE , 2013 : 580 - 587 .

GIRSHICK R . Fast R-CNN [C ] // Proceedings of the IEEE International Conference on Computer Vision . [S.l.] : IEEE , 2015 : 1440 - 1448 .

REN Shaoqing , HE Kaiming , GIRSHICK R , et al . Faster R-CNN: Towards real-time object detection with region proposal networks [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2017 , 39 ( 6 ): 1137 - 1149 .

杨秋玉 , 王栋 . 卷积神经网络在复合绝缘子憎水性智能识别中的应用 [J ] . 高电压技术 , 2022 , 48 ( 2 ): 603 - 611 .

YANG Qiuyu , WANG Dong . Application of convolutional neural network in intelligent classification of hydrophobicity of composite insulators [J ] . High Voltage Engineering , 2022 , 48 ( 2 ): 603 - 611 .

谢军 , 肖朝轩 , 张思刚 , 等 . 基于迁移学习和特征融合的复合绝缘子憎水性等级判别方法 [J ] . 电网技术 , 2021 , 45 ( 10 ): 3964 - 3971 .

XIE Jun , XIAO Chaoxuan , ZHANG Sigang , et al . A determination method for hydrophobicity class of composite insulator based on transfer learning and feature fusion [J ] . Power System Technology , 2021 , 45 ( 10 ): 3964 - 3971 .

汪然然 , 娄联堂 . 基于图像分析和深度学习的复合绝缘子憎水性分级 [J ] . 武汉工程大学学报 , 2021 , 43 ( 5 ): 580 - 585 .

WANG Ranran , LOU Liantang . Hydrophobicity classification of composite insulators based on image analysis and deep learning [J ] . Journal of Wuhan Institute of Technology , 2021 , 43 ( 5 ): 580 - 585 .

黄杰 . 基于深度学习的绝缘子憎水性识别与故障检测方法研究 [D ] . 长沙 : 湖南大学 , 2021 .

HUANG Jie . Research on hydrophobicity recognition and fault detection of insulators based on deep learning [D ] . Changsha : Hunan University , 2021 .

HU Jie , SHEN Li , ALBANIE S , et al . Squeeze-and-Excitation networks [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2020 , 42 ( 8 ): 2011 - 2023 .

卜文斌 , 游福成 , 李泉 , 等 . 一种基于大津法改进的图像分割方法 [J ] . 北京印刷学院学报 , 2015 , 23 ( 4 ): 76 - 78 .

BU Wenbin , YOU Fucheng , LI Quan , et al . An improved image segmentation method based on Otsu [J ] . Journal of Beijing Institute of Graphic Communication , 2015 , 23 ( 4 ): 76 - 78 .

宋爽 , 任洪娥 , 官俊 . 基于Sobel梯度模板的多阈值实时边缘检测方法 [J ] . 计算机工程与应用 , 2015 , 51 ( 23 ): 199 - 202 .

SONG Shuang , REN Honge , GUAN Jun . Multi-threshold and read-time edge detection method based on Sobel gradient template [J ] . Computer Engineering and Applications , 2015 , 51 ( 23 ): 199 - 202 .

张重远 , 闫康 , 汪佛池 , 等 . 基于图像特征提取与BP神经网络的绝缘子憎水性识别方法 [J ] . 高电压技术 , 2014 , 40 ( 5 ): 1446 - 1452 .

ZHANG Chongyuan , YAN Kang , WANG Fochi , et al . Insulator hydrophobic identification based on image feature extraction and BP neural network [J ] . High Voltage Engineering , 2014 , 40 ( 5 ): 1446 - 1452 .

XIAO He , YU Wang , YEQIANG Deng , et al . Hydrophobicity classification of composite insulators based on faster R-CNN object detection algorithm [C ] // 2022 IEEE International Conference on High Voltage Engineering and Applications . [S.l.] : IEEE , 2022 : 1 - 4 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Identification of Typical Construction Defects in 10 kV Cable Intermediate Joints Based on Canny Algorithm and Image Processing

Investigation into Vacuum Arcs with High Speed Image Sampling Computer System

On-line Monitoring of Icing-thickness on Transmission Line with Image Edge Detecting Method

Development and Application of UV Imaging Detection Management System for High Voltage Electrical Equipment

Ultraviolet Image Facular Area Vary Characteristic and Apparent Discharge Estimation of Suspension Composite Insulator Corona

Related Author

YOU Haixing

YU Qiao

JIANG Quancai

FANG Chunhua

CHEN Zhi

王尔智

王季梅

王毅

Related Institution

China Southern Power Grid, Zhongshan Power Supply Company

School of Electrical and New Energy, China Three Gorges University

沈阳工业大学

大连理工大学

⁰