基于激光图像处理的输电线路故障定位识别  

Fault location and recognition of transmission lines based on laser image processing

作　　者：冯善强 赵永强 余膺昊 谭炜豪 FENG Shan-qiang;ZHAO Yong-qiang;YU Ying-hao;TAN Wei-hao(Southern Power Grid Electric Power Technology Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]南方电网电力科技股份有限公司,广东广州510000

出　　处：《激光与红外》2025年第4期513-519,共7页Laser & Infrared

基　　金：广东省重点领域研发计划项目(No.2021B0101420002)资助。

摘　　要：研究基于激光图像处理技术的特高压输电线路故障定位识别方法,提升故障定位识别效果。利用单点激光扫描成像仪,采集特高压输电线路激光图像;通过维纳滤波算法,去噪处理特高压输电线路激光图像;利用改进多显著性集成算法,结合K-means聚类算法,在去噪后激光图像内,提取显著性颜色特征与结构特征,并融合颜色特征与结构特征,得到特高压输电线路显著性故障区域;通过多尺度轮廓结构元素形态学,处理显著性故障区域,得到故障区域边缘图像;以统计分析边缘图像颜色分布方式,构建颜色判决模型,凸显故障区域,并提取故障区域的边缘轮廓;在原始特高压输电线路激光图像内绘制故障区域边缘轮廓,便可精准定位识别故障位置。实验证明:该方法可有效采集特高压输电线路激光图像,并完成去噪处理;该方法可有效提取颜色与结构特征,并有效提取故障区域边缘图像,完成故障定位识别;在不同过渡电阻时,该方法均可精准定位识别输电线路故障位置。The fault location and identification method of UHV transmission lines based on laser image processing technology is studied to enhance the fault location and identification effect.Firstly,a laser image of UHV transmission line is captured by single point laser scanning imager,and the laser image of UHV transmission line is denoised by Wiener filter algorithm.Subsequently,by using the improved multi-significance integration algorithm and K-means clustering algorithm,the significant color features and structural features are extracted from the denoised laser image,and the color features and structural features are fused to obtain the significant fault region of UHV transmission lines.Next,through the morphology of multi-scale contour structure elements,the significant fault region is processed and the edge image of the fault region is obtained.Based on the statistical analysis of edge image color distribution,the color decision model is constructed to highlight the fault region and extract the edge contour of the fault region.Finally,the fault location can be accurately identified by drawing the edge outline of the fault area in the origi-nal UHV transmission line laser image.Experimental results show that this method can effectively acquire the laser im-age of the UHV transmission line and complete the denoising process.It also successfully extracts color and structural features and generates the edge image of the fault area,thereby completing fault location identification.This method enables accurate identification of the fault location in transmission lines,regardless of the transition resistors involved.

关 键 词：激光图像处理 特高压 输电线路 故障定位识别 维纳滤波 形态学 

分 类 号：TN249[电子电信—物理电子学] TM726.1[电气工程—电力系统及自动化]