基于图像识别和数值天气预报的输电线覆冰智能监测预警及风险评估

孟令博; 朱红杰; 杨兰均; 董立凡; 陈立言

doi:10.16516/j.ceec.2025-330

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于图像识别和数值天气预报的输电线覆冰智能监测预警及风险评估

极端气候事件下的能源与交通韧性 | 更新时间：2026-10-14

- 基于图像识别和数值天气预报的输电线覆冰智能监测预警及风险评估
- Research on Transmission Line Icing Monitoring, Early Warning and Risk Assessment Technology Based on Numerical Weather and Machine Learning
- 南方能源建设 2026年13卷第3期页码：56-68
- 作者机构：
  
  1. 西安交通大学电气工程学院,陕西,西安,710048
  2. 国网宝鸡供电公司,陕西,宝鸡,721000
  3. 咸阳市气象局,陕西,咸阳,712000
- 作者简介：
  
  孟令博，ORCID: 0009-0002-4489-59882000-，男，工程师，硕士，主要从事架空输电线路覆冰相关研究工作（e-mail）1204733903@qq.com
  朱红杰，1990-，男，工程师，学士，主要从事供电公司应急管理及配网研究工作（e-mail）714384280@qq.com
  杨兰均，ORCID: 0009-0006-1247-91881968-，男，教授，主要从事脉冲功率及放电等离子体科学与技术和电力系统过电压及雷电防护技术等研究工作（e-mail）yanglj@mail.xjtu.edu.cn
  董立凡，ORCID: 0009-0000-9011-61001991-，男，工程师，博士，主要从事电力及新能源气象研究工作（e-mail）donglf0508@163.com
  陈立言，2003-，男，硕士，主要从事架空输电线路在线监测研究工作（e-mail）506964275@qq.com
- 基金信息：
  
  国网陕西省电力有限公司科技项目“基于气候微环境数值建模和‘气候-电力’联合仿真的气象预报和灾害预警技术研究”（5226BJ250005）
- DOI：10.16516/j.ceec.2025-330
  中图分类号： TM7;P49
- 纸质出版：2026
- 稿件说明：
移动端阅览
孟令博, 朱红杰, 杨兰均, 等. 基于图像识别和数值天气预报的输电线覆冰智能监测预警及风险评估[J]. 南方能源建设, 2026, 13(3): 56-68.

.
孟令博, 朱红杰, 杨兰均, 等. 基于图像识别和数值天气预报的输电线覆冰智能监测预警及风险评估[J]. 南方能源建设, 2026, 13(3): 56-68. DOI： 10.16516/j.ceec.2025-330.

. DOI： 10.16516/j.ceec.2025-330.

摘要

目的

为解决输电线路覆冰导致的设备故障与电网安全问题，以宝鸡翻越秦岭区域线路为研究对象，开展覆冰监测、预测与风险评估技术研究。

方法

监测环节采用机器视觉技术，通过图像预处理、导线像素提取和宽度测量算法，结合“观冰精灵”图像传感器，实现覆冰厚度实时监测；预测环节构建周边气象站数据库，基于Haversine公式获取匹配度高的气象数据，基于插值法对气象数据进行精细化处理，将处理后数据输入WRF数值天气模型模拟关键气象要素，用模拟结果驱动Makkonen覆冰模型计算覆冰厚度增长量，结合实时监测值得到预测结果；风险评估环节融合覆冰跳闸样本、杆塔静态数据及气象数据，经特征工程处理后构建XGBoost模型，设计差异化动态预警阈值。最终，基于Python实现以上3个模块功能，并开发可视化监测预警与风险评估系统。

结果

研究表明，覆冰厚度监测结果与人工测量结果对比误差≤3%，覆冰厚度预测结果误差≤5%；风险评估方面，能实现1～3 h和3～6 h覆冰跳闸概率精准预测与分级预警。

结论

开发的可视化监测预警与风险评估系统，能够集成监测、预测、评估3个功能，为宝鸡电网输电线智能运维提供决策支持，有效降低覆冰灾害损失。

Abstract

Objective

To address the equipment failure and grid safety issues caused by ice accumulation on transmission lines

this study takes the lines crossing the Qinling Mountains in Baoji as the research object and conducts research on ice accumulation monitoring

prediction

and risk assessment technologies.

Method

In the monitoring phase

machine vision technolo

gy was adopted. Through image preprocessing

conductor pixel extraction

and width measurement algorithms

combined with the "Ice Observation Spirit Spirit" image sensor

real-time monitoring of ice thickness was achieved. In the prediction phase

a database of surrounding meteorological stations was constructed. Based on the Haversine formula

meteorological data with high matching degrees were obtained. Then

the meteorological data were refined through interpolation methods. The processed data were input into the WRF numerical weather model to simulate key meteorological elements. The simulation results drove the Makkonen ice accumulation model to calculate the increase in ice thickness. Combined with real-time monitoring values

the prediction results were obtained. In the risk assessment phase

ice accumulation tripping samples

tower static data

and meteorological data were integrated. After feature engineering processing

an XGBoost model was constructed

and differentiated dynamic early warning thresholds were designed. Ultimately

the functions of the three modules were implemented based on Python

and a visualization monitoring

early warning

and risk assessment system was developed.

Result

The research shows that the error between the ice thickness monitoring results and the manual measurement results is ≤ 3%

and the error of the ice thickness prediction results is ≤ 5%. In terms of risk assessment

it can accurately predict the probability of ice accumulation tripping within 1~3 hours and 3~6 hours and issue graded early warnings.

Conclusion

The developed visualization monitoring

early warning

and risk assessment system can integrate monitoring

prediction

and assessment functions

providing decision support for the intelligent operation and maintenance of transmission lines in the Baoji power grid and effectively reducing the losses caused by ice accumulation disasters.

关键词

Keywords

references

WANG W S, XIE H B, YANG Z, et al. An introductory analysis of electric meteorology: basic concepts and research directions [J]. Proceedings of the CSEE, 2024, 44(18): 7440-7452. DOI: 10.13334/j.0258-8013.pcsee.241858.

ZHANG J, YAN Z H, ZHANG D Q, et al. Design and application research of an interactive fusion system of multi-source data of electric power and meteorology [J]. Power systems and big data, 2025, 28(2): 83-92. DOI: 10.19317/j.cnki.1008-083x.2025.02.010.

JIA X Q, YANG Y B, DU J, et al. Study on uncertainty operation optimization of virtual power plant based on intelligent prediction model under climate change [J]. Power generation technology, 2023, 44(6): 790-799. DOI: 10.12096/j.2096-4528.pgt.23094.

YANG L, HE S J, CHEN Y F, et al. Effect of droplet deformation on discharge at icicle tip of ice-covered insulators during melting period [J]. Electric power systems research, 2022, 213: 108723. DOI: 10.1016/J.EPSR.2022.108723.

WANG S, ZHU Y C, LIU Y P, et al. Analysis on aerodynamic characteristics and stability of ice-covered lines in wind farm [J]. Inner Mongolia electric power, 2025, 43(2): 45-53. DOI: 10.19929/j.cnki.nmgdljs.2025.0021.

JIN Y, MA S L, FENG J K. On-line monitoring method for icing on high-voltage transmission lines based on machine vision technology [J]. Wireless internet technology, 2023, 20(15): 97-99. DOI: 10.3969/j.issn.1672-6944.2023.15.029.

ZHANG Y, SHU Z Y, CHEN L, et al. Research on line ice cover quality estimation based on image stitching [J]. China measurement test, 2023, 49(5): 108-116. DOI: 10.11857/j.issn.1674-5124-2022120142.

YANG J, GAO Y F, ZHANG K, et al. A conductor icing monitoring method based on image edge detection and normal direction detection [J]. Power system and clean energy, 2023, 39(2): 24-32. DOI: 10.3969/j.issn.1674-3814.2023.02.004.

LI C X. Research on prediction of transmission line icing thickness based on WRF and a hybrid model of machine learning [D]. Lanzhou: Lanzhou University, 2023. DOI: 10.27204/d.cnki.glzhu.2023.003743.

HAN X B, CHEN Z M, XING B, et al. Prediction model of conductor icing based on basic environmental parameters [J]. Journal of Chongqing University, 2023, 46(11): 69-77. DOI: 10.11835/j.issn.1000-582X.2023.12.008.

SUN P J. An icing conductor model based on meteorological parameters and its verification [D]. Nanjing: Nanjing University of Information Science and Technology, 2012.

LI Q R. Micro-meteorology characteristics and icing disaster prediction of power transmission lines in micro-geography environments [D]. Tianjin: Tianjin University, 2020. DOI: 10.27356/d.cnki.gtjdu.2020.003258.

SI X D, YANG H L, LI Y H, et al. Analysis of icing risk assessment algorithm for transmission lines based on beidou data-driven positioning [J]. Electronic technology, 2025, 54(2): 404-406. DOI: 10.3969/j.issn.1000-0755.2025.02.186.

CHEN Y F, WANG R J, ZHANG Y, et al. Multi-factor fusion-based risk assessment and tiered early warning method for icing faults in overhead transmission lines [J]. Power systems and big data, 2025, 28(6): 50-59. DOI: 10.19317/j.cnki.1008-083x.2025.06.006.

XU C Q, WANG D, SU F, et al. Image recognition method of transmission line safety risk assessment based on multidimensionaldata coupling [J]. Computer science, 2023, 50(Suppl. 1): 220500032. DOI: 10.11896/jsjkx.220500032.

CHOI H, KOO G, KIM B J, et al. Weakly supervised power line detection algorithm using a recursive noisy label update with refined broken line segments [J]. Expert systems with applications, 2021, 165: 113895. DOI: 10.1016/j.eswa.2020.113895.

MA F Q, LI J L, JIA R, et al. A fine identification method for transmission line ice coating types based on bilateral feature-fused semantic segmentation [J]. Smart power, 2025, 53(10): 17-25. DOI: 10.20204/j.sp.2025.10003.

HE Y J, SUN G L, ZHANG B , et al. Research on quantitative evaluation method for galloping risk of overhead transmission lines [J]. Shandong electric power, 2025, 52(7): 97-104. DOI: 10.20097/j.cnki.issn1007-9904.2025.07.010.

YAO L, CHI X J, ZHANG L, et al. Melting technology for high-voltage overhead transmission lines based on small mobile robot [J]. Electrical measurement instrumentation, 2024, 61(12): 34-41. DOI: 10.19753/j.issn1001-1390.2024.12.005.

DENG L Y, WANG H H, WANG L Z, et al. Meteorological services in the new power system: status, challenges, and future directions [J]. Southern energy construction, 2026, 13(3): 1-19. DOI: 10.16516/j.ceec.2025-248.

WU J R, HUANG H, JIANG S, et al. Evolutionary strategy-projection pursuit algorithms for risk assessment of transmission line icing disaster considering multi-factors [J]. Science technology and engineering, 2020, 20(21): 8601-8606. DOI: 10.3969/j.issn.1671-1815.2020.21.026.

WENG B J. Icing status detection of transmission line based on machine vision [D]. Fuzhou: Fuzhou University, 2021. DOI: 10.27022/d.cnki.gfzhu.2021.000445.

DONG L F, MENG L B, ZHOU H L, et al. Diagnostic analysis of an icing event on high-voltage transmission lines in southern Shanxi using multivariate data and model coupling [J]. Desert and oasis meteorology, 2025, 19(3): 26-34. DOI: 10.12057/j.issn.1002-0799.2404.15005.

ZENG Y B. Research on state assessment and maintenance optimization strategy of transmission lines [D]. Guangzhou: Guangdong University of Technology, 2021. DOI: 10.27029/d.cnki.ggdgu.2021.001132.

MAKKONEN S. Modeling of ice accretion on wires [J]. Journal of applied meteorology and climatology, 1984, 23(6): 929-939. DOI: 10.1175/1520-0450(1984)0230929:MOIAOW2.0.CO;2.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于主客观赋权的电力市场风险综合评价研究

基于风险链和模糊影响图的风险定性评估方法

IEC与GB防雷标准中建筑物直击雷防护对比分析

基于切圆布气的人工硐室壁温抑制技术

基于双工质气体压缩储能的供气系统