基于视觉识别的配网线路无人机智能巡检  

Power Distribution Network Intelligent Inspection of UAV Based on Visual Recognition

作　　者：刘翀 LIU Chong

机构地区：[1]国网湖南省电力有限公司长沙市湘江新区供电分公司,湖南长沙410000

出　　处：《电力系统装备》2024年第7期111-113,共3页Electric Power System Equipment

摘　　要：配网线路作为电网的重要组成部分,其运维质量至关重要。为实现配网线路智能化巡检,提升配网线路数字化运检质效,提出基于视觉识别的配网线路无人机自适应巡检方案。采用视觉识别引导,自动规划航迹,自动寻塔,识别确定航线方向上的下级杆塔。通过采用基于深度学习的目标检测算法,轻量化模型和卷积神经网络,使线路缺陷识别更加高效准确。As an important component of the power grid,the power distribution network lines have wide coverage area,poor environment,long distance,large workload and high quality requirements of operation and maintenance.To achieve intelligent inspection of distribution network lines and improve the quality and efficiency of digital operation and inspection of distribution network lines.An adaptive inspection scheme base on visual recognition for power distribution network lines is proposed in this paper.Using visual recognition guidance,automatically planning the trajectory,automatically searching for towers,and identifying lower level towers in the direction of the determined route.By adopting object detection algorithms base on deep learning,lightweight models,and convolutional neural networks,defect recognition of power network is more efficient and accurate.

关 键 词：无人机 配网线路 智能巡检 视觉识别 

分 类 号：TM76[电气工程—电力系统及自动化]