摘要
国内高压输电线路机器人巡检存在运行环境复杂、传感器信息量大且多样和越障效率低等问题。本文以极寒条件下500 kV输电线路典型障碍物为研究对象,提出一种利用图像处理技术预处理提取目标特征,机器学习SVM分类决策算法分类,再与结构约束结合从而可以高效准确地识别高压输电线路障碍物,并通过准确性和有效性验证。结果表明,该算法可以准确地识别定位防震锤等障碍物,对我国高压输电线路智能机器人巡检技术的发展提供了参考和借鉴。
There are some problems such as complex operating environment,large and diverse sensor information and low obstacle surmounting efficiency.Based on the 500 kV transmission lines under the condition of cold typical obstacles as the research object,put forward a kind of pretreatment of the technology of image processing to extract target feature,machine learning classification SVM classification decision algorithm,then combined with structural constraints which can efficiently and accurately identify high voltage transmission line obstacles,and through the accuracy and validation.The results show that the algorithm can accurately identify and locate obstacles such as shockproof hammers,which provides a reference for the development of intelligent robot patrol technology of high-voltage transmission lines in China.
作者
高春辉
李方
秘立鹏
王泽禹
GAO Chun-hui;LI Fang;MI Li-peng;WANG Ze-yu(Institute of State Grid East Inner Mongolia Electric Power Co.Ltd.,Hohhot 010020,China;Guangdong Keystar Intelligence Robot Co.Ltd.,Foshan 528300,China;State Grid East Inner Mongolia Maintenance Company,Tongliao 028000,China)
出处
《山东农业大学学报(自然科学版)》
北大核心
2020年第3期525-528,共4页
Journal of Shandong Agricultural University:Natural Science Edition
关键词
输电线路
机器人巡检
High voltage transmission line
robot inspection