基于三维激光点云数据的电力电缆绝缘缺陷识别  

Insulation defect identification based on 3D laser point cloud data of power cables

作　　者：黄绪勇[1,2] 林中爱 唐标 赵李强 HUANG Xuyong;LIN Zhongai;TANG Biao;ZHAO Liqiang(Department of Electrical Engineering,Huazhong University of Science and Technology,Wuhan 430030,Hubei,China;Electric Power Research Institute,Yunnan Power Grid Co.,Ltd.,Kunming 650012,Yunnan,China;Power Grid Visualization Platform Business Department,Kunming Nexus Technology Co.,Ltd.,Kunming 650012,Yunnan,China)

机构地区：[1]华中科技大学电力工程系,湖北武汉430030 [2]云南电网有限责任公司电力科学研究院,云南昆明650012 [3]昆明能讯科技有限责任公司电网可视化平台事业部,云南昆明650012

出　　处：《沈阳工业大学学报》2024年第4期409-415,共7页Journal of Shenyang University of Technology

基　　金：国家重点研发计划项目(2021YEE0204200);云南电网有限责任公司电力科学研究院项目(2021-048)。

摘　　要：针对电力电缆绝缘缺陷识别精准度较低的问题,提出一种基于三维激光点云数据的电力电缆绝缘缺陷识别方法。利用剪裁法增强原始三维激光点云数据,通过区域生长法与最小二乘法完整获取电缆绝缘材料的三维结构面信息。借助Canny边缘检测方法求解电缆绝缘表面缺陷与内部缺陷边缘信息,自动识别出电力电缆绝缘的缺陷位置及缺陷类别。结果表明,所提方法可以精准识别电缆绝缘表面的划痕缺陷、电缆外屏蔽表面起泡和孔洞缺陷,识别耗时短,鲁棒性较优,具有较高实际应用价值。A method for identifying insulation defects based on 3D laser point cloud data of power cables was proposed to address the issue of low accuracy in identifying insulation defects of power cables.The original 3D laser point cloud data were enhanced by using clipping method.The 3D structural surface information of cable insulation material was obtained completely through region growing method and least square method.The edge information of surface defects and internal defects of cable insulation were solved by Canny edge detection method,and the defect location and type of power cable insulation were automatically identified.The results indicate that the as-proposed method can accurately identify scratches,blistering and holes on the insulation surface of cables,with short recognition time and relatively high robustness.It has higher practical application value.

关 键 词：电力电缆 绝缘缺陷 三维激光点云数据 平均法向量 高斯滤波 梯度幅值 CANNY边缘检测 孔洞缺陷 

分 类 号：TP998[自动化与计算机技术]