Research on State Recognition Technology of High-voltage Disconnectors Based on Image Fusion

ZHANG Ping; YANG Tian; LI Zhijun

doi:10.13296/j.1001-1609.hva.2025.10.015

您当前的位置：

首页 >

文章列表页 >

Research on State Recognition Technology of High-voltage Disconnectors Based on Image Fusion

Digital Twin Technology and Applications for Multi-scenario in New Power System | 更新时间：2025-10-21

- Research on State Recognition Technology of High-voltage Disconnectors Based on Image Fusion
- High Voltage Apparatus Vol. 61, Issue 10, Pages: 139-148(2025)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2025.10.015
  CLC：
- Received：12 April 2025，
  
  Revised：2025-07-11，
  
  Published：16 October 2025
- 稿件说明：
移动端阅览
ZHANG Ping, YANG Tian, LI Zhijun. Research on State Recognition Technology of High-voltage Disconnectors Based on Image Fusion[J]. High voltage apparatus, 2025, 61(10): 139-148.
DOI：

ZHANG Ping, YANG Tian, LI Zhijun. Research on State Recognition Technology of High-voltage Disconnectors Based on Image Fusion[J]. High voltage apparatus, 2025, 61(10): 139-148. DOI： 10.13296/j.1001-1609.hva.2025.10.015.

摘要

针对高压隔离开关状态识别中单一模态图像受环境干扰大、误检率高

的技术难题，文中提出一种基于多模态图像融合与轻量化检测的高压隔离开关状态（分闸/合闸、触头温度异常、瓷瓶裂纹）识别方法。借助CNN-Transformer分层交互嵌入模块（CTHIE）融合CNN局部特征提取能力与Transformer全局依赖建模优势，运用动态重聚合特征表示（DRFR）模块实现跨模态特征自适应对齐，通过全感知前向融合（FPFF）机制重建高保真融合图像。在此基础上，提出轻量化检测模型YOLO-MECD，以高效多尺度注意力（EMA）机制替代原C2PSA模块增强小目标特征聚焦能力，采用基于部分卷积的CSPPC模块替换C3K2结构，并引入最小点距离交并比（

MPDIoU

）损失函数优化边界框回归过程。实验结果显示，改进的CNN-Transformer融合模型在视觉和图像质量评价指标均优于比较流行的融合方法；YOLO-MECD检测模型在高压隔离开关数据集上精度（

）达85.6%、召回率（

）达83.32%、平均精度均值（

mAP

）达88.6%，模型体积压缩至4.66 MB（减少74.4%）。

Abstract

In view of the technical problem of large environmental interference and high false detection rate in the status recognition of high-voltage disconnectors using single-modal images

in this paper a recognition method for the status of high-voltage disconnectors (opening/closing

abnormal contact temperature

porcelain cracks) based on multimodal image fusion and lightweight detection is proposed. By leveraging the CNN-Transformer hierarchical interactive embedding (CTHIE) module to integrate the local feature extraction capability with transformer’s global dependency modeling

the dynamic re-aggreation feature representation (DRFR) module is employed to achieve adaptive cross-modal feature alignment

while the full-perception forward fusion (FPFF) mechanism is adopted to reconstruct high-fidelity fused images. On this basis

a lightweight detection model YOLO-MECD is proposed

which replaces the original C2PSA module with an efficient multi-scale attention (EMA) mechanism to enhance the feature focusing ability for small targets

uses a CSPPC module based on partial convolution to replace the C3K2 structure

and introduces a minimum point distance intersection over union (

MPDIoU

) loss function to optimize the bounding box regression process. The experimental results show that the improved CNN-Transformer fusion model outperforms popular fusion methods in both visual and image quality evalu

ation indicators；theYOLO-MECDdetection model achieves a precision (

) of 85.6%

a recall (

) of 83.32%

and a mean average precision (mAP) of 88.6%on the high-voltage disconnector dataset

with the model volume compressed to 4.66 MB (reduced by 74.4%) .

关键词

Keywords

references

DOMOROSHCHYN S , SAKHNO A . Estimating the failure probability of an instrument transformer cell in the high voltage distributing device using an expert-statistical method [J ] . Eastern-European Journal of Enterprise Technologies , 2020 , 4 ( 2 ): 70 - 81 .

刘子英 , 张靖 , 邓芳明 . 基于BP神经网络的高压隔离开关分合闸监测识别 [J ] . 电力系统保护与控制 , 2020 , 48 ( 5 ): 134 - 140 .

LIU Ziying , ZHANG Jing , DENG Fangming . Monitoring and identification of state of opening or closing isolation switch based on BP neural network [J ] . Power System Protection and Control , 2020 , 48 ( 5 ): 134 - 140 .

腾云 , 雷丞 , 李洪涛 , 等 . 基于HOG和SVM的高压隔离开关分合闸状态自动识别技术研究 [J ] . 高压电器 , 2020 , 56 ( 9 ): 246 - 252 .

TENG Yun , LEI Cheng , LI Hongtao , et al . Research of automatic identification for the on/off state of high-voltage disconnector based on HOG and SVM [J ] . High Voltage Apparatus , 2020 , 56 ( 9 ): 246 - 252 .

张伟 , 张咪 , 金佳齐 , 等 . 基于ResNet深度学习网络的隔离开关状态识别研究 [J ] . 高压电器 , 2024 , 60 ( 6 ): 99 - 106 .

ZHANG Wei , ZHANG Mi , JIN Jiaqi , et al . Research on status identification of disconnector based on ResNet deep learning network [J ] . High Voltage Apparatus , 2024 , 60 ( 6 ): 99 - 106 .

ZHANG Mi , BAO Zhe , WU Zefeng , et al . Status recognition method of high voltage disconnector based on image enhancement and improved neural network [C ] // The Proceedings of the 19th Annual Conference of China Electrotechnical Society . Singapore : Springer Nature Singapore , 2025 : 750 - 763 .

JIAO Runnong , LIU Jiefeng , ZHOU Zikai , et al . Real-time segmentation of overheating faults in disconnectors using YOLOv7 [J ] . IEEE Transactions on Instrumentation and Measurement , 2025 ( 74 ): 1 - 9 .

王永平 . 高压开关设备红外图像识别与故障诊断方法研究 [D ] . 重庆 : 重庆理工大学 , 2020 .

WANG Yongping . Research on the infrared image recognition and fault diagnosis method of high-voltage switchgear [D ] . Chongqing : Chongqing University of Technology , 2020 .

张靖 , 单长吉 , 周丽 , 等 . 基于图像融合的高压隔离开关分合闸状态识别 [J ] . 红外技术 , 2024 , 46 ( 5 ): 539 - 547 .

ZHANG Jing , SHAN Changji , ZHOU Li , et al . Recognition of highvoltage isolation switch opening and closing state based on image fusion [J ] . Infrared Technology , 2024 , 46 ( 5 ): 539 - 547 .

王泽天 . 基于双光图像融合的高压隔离开关状态监测系统研究 [D ] . 西安 : 西安石油大学 , 2023 .

WANG Zetian . Research on the state monitoring system of highvoltage disconnectors based on dual-light image fusion [D ] . Xi'an : Xi'an Shiyou University , 2023 .

LI Huafeng , LIU Junyu , ZHANG Yafei , et al . A deep learning framework for infrared and visible image fusion without strict registration [J ] . International Journal of Computer Vision , 2024 , 132 ( 5 ): 1625 - 1644 .

WANG Chishe , SONG Xingqing , WANG Jie , et al . An improved YOLOv11 algorithm for small object detection in UAV images [J ] . Signal Image and Video Processing , 2025 , 19 ( 6 ): 507 .

张焕龙 , 齐企业 , 张杰 , 等 . 基于改进YOLOv5的输电线路鸟巢检测方法研究 [J ] . 电力系统保护与控制 , 2023 , 51 ( 2 ): 151 - 159 .

ZHANG Huanlong , QI Qiye , ZHANG Jie , et al . Bird nest detection method for transmission lines based on improved YOLOv5 [J ] . Power System Protection and Control , 2023 , 51 ( 2 ): 151 - 159 .

杨乐 , 郭一鸣 , 霍勇博 , 等 . 改进YOLOv5在电力生产违规穿戴检测中的应用 [J ] . 电力系统保护与控制 , 2023 , 51 ( 14 ): 160 - 168 .

YANG Le , GUO Yiming , HUO Yongbo , et al . Application of improved YOLOv5 for illegal wearing detection in electric power construction [J ] . Power System Protection and Control , 2023 , 51 ( 14 ): 160 - 168 .

刘富宽 , 罗素云 , 何佳 , 等 . FVIT-YOLOv8:基于多尺度融合注意机制的改进YOLOv8小目标检测 [J ] . 红外技术 , 2024 , 46 ( 8 ): 912 - 922 .

LIU Fukuan , LUO Suyun , HE Jia , et al . FVIT-YOLOv8:Improved YOLOv8 small object detection based on multi - scale fusion attention mechanism [J ] . Infrared Technology , 2024 , 46 ( 8 ): 912 - 922 .

刘雄彪 , 杨贤昭 , 陈洋 , 等 . 基于CIoU改进边界框损失函数的目标检测方法 [J ] . 液晶与显示 , 2023 , 38 ( 5 ): 656 - 665 .

LIU Xiongbiao , YANG Xianzhao , CHEN Yang , et al . Object detection method based on CIoU improved bounding box loss function [J ] . Chinese Journal of Liquid Crystals and Displays , 2023 , 38 ( 5 ): 656 - 665 .

胡凯涛 , 马向华 , 孙向宇 , 等 . 融合Res2Net和部分卷积的带钢表面缺陷检测算法 [J ] . 计算机工程与应用 , 2025 , 61 ( 5 ): 334 - 343 .

HU Kaitao , MA Xianghua , SUN Xiangyu , et al . Strip surface defect detection algorithm integrating Res2Net and PConv [J ] . Computer Engineering and Applications , 2025 , 61 ( 5 ): 334 - 343 .

GHANDOUR C , EL-SHAFAI W , EL-RABAIE S , et al . Comprehensive performance analysis of different medical image fusion techniques for accurate healthcare diagnosis applications [J ] . Multimedia Tools and Applications , 2024 , 83 ( 8 ): 24217 - 24276 .

DILEEP R , JAYANTH J , VEDHA A S , et al . Enhancing image fusion via optimized BAT algorithm in brovey transform for remote sensing images [C ] // 2024 International Conference on Knowledge Engineering and Communication Systems(ICKECS) . India : IEEE , 2024 : 1 - 6 .

HUSSEIN D Y , MAKKEY M , A ABDELRAHMAN S A . Hybrid fusion approach for Alzheimer's disease progression employing IHS and wavelet transform [J ] . Menoufia Journal of Electronic Engineering Research , 2024 , 33 ( 1 ): 17 - 23 .

TRIVEDI G J , SANGHVI R . Optimizing image fusion using modified principal component analysis algorithm and adaptive weighting scheme [J ] . International Journal of Advanced Networking and Applications , 2023 , 15 ( 1 ): 5769 - 5774 .

JI Chuanming , ZHOU Wujie , LEI Jingsheng , et al . Infrared and visible image fusion via multiscale receptive field amplification fusion network [J ] . IEEE Signal Processing Letters , 2023 ( 30 ): 493 - 497 .

TALAAT F M , ZAINELDIN H . An improved fire detection approach based on YOLO-v8 for smart cities [J ] . Neural Computing and Applications , 2023 , 35 ( 28 ): 20939 - 20954 .

ARDAÇ F B K , ERDOGMUS P . Car object detection:Comparative analysis of YOLOv9 and YOLOv10 models [C ] // 2024 Innovations in Intelligent Systems and Applications Conference (ASYU) . Romania : IEEE , 2024 : 1 - 6 .

WANG Ao , CHEN Hui , LIU Lihao , et al . Yolov10:Real-time end to-end object detection [J ] . Advances in Neural Information Processing Systems , 2024 ( 37 ): 107984 - 108011 .

何胜红 , 吴小平 , 王俊波 , 等 . 基于背景数据增强和改进YOLOv4的断路器试验机器人接线定位方法 [J ] . 电力科学与技术学报 , 2023 , 38 ( 2 ): 196 - 204 .

HE Shenghong , WU Xiaoping , WANG Junbo , et al . Wiring locating method for circuit breaker test robot based on background augmentation and improved YOLOv4 [J ] . Journal of Electric Power Science and Technology , 2023 , 38 ( 2 ): 196 - 204 .

彭曙蓉 , 刘登港 , 何洁妮 , 等 . 基于改进YOLOv4的架空线路电力设备故障检测 [J ] . 电力科学与技术学报 , 2023 , 38 ( 5 ): 169 - 176 .

PENG Shurong , LIU Denggang , HE Jieni , et al . Fault detection for overhead line power equipment based on improved YOLOv4 [J ] . Journal of Electric Power Science and Technology , 2023 , 38 ( 5 ): 169 - 176 .

严宇平 , 杨秋勇 , 谢翰阳 , 等 . 基于改进YOLOv7的输电线路异物检测模型 [J ] . 南方电网技术 , 2024 , 18 ( 9 ): 47 - 58 .

YAN Yuping , YANG Qiuyong , XIE Hanyang , et al . Foreign object detection model of transmission line based on improved YOLOv7 [J ] . Southern Power System Technology , 2024 , 18 ( 9 ): 47 - 58 .

叶采萍 , 陈炯 , 马显龙 , 等 . 改进YOLOv5的变电站反无人机目标检测算法 [J ] . 南方电网技术 , 2024 , 18 ( 2 ): 89 - 97 .

YE Caiping , CHEN Jiong , MA Xianlong , et al . Anti UAV target detection algorithm for substation based on improved YOLOv5 [J ] . Southern Power System Technology , 2024 , 18 ( 2 ): 89 - 97 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Development and Practical Application of Auxiliary Device for Mechanical Characteristic Test of High-voltage Apparatus

Mechanical Fault Diagnosis Method of High-voltage Disconnector Based on Empirical Modal Decomposition and Support Vector Machine

Research of Automatic Identification for the on/off State of High-voltage Disconnector Based on HOG and SVM

Design Analysis and Characteristic Research of Composite Insulator for High-voltage Disconnector

Switch State Recognition of Substation Based on NSST Image Fusion

Related Author

LI Weidang

YANG Le

ZHANG Xin

SUN Bo

CHEN Chen

ZHANG Yongtao

HE Ming

ZHANG Yiming

Related Institution

Xi'an XD High Voltage Apparatus Co., Ltd.

State Grid Gansu Electric Power Company Ultra High Voltage Company

Pinggao Group Co., Ltd.

School of Electrical Engineering,Beijing Jiaotong University

State Grid Jiangsu Electric Power Research Institute

⁰