电信科学 ›› 2022, Vol. 38 ›› Issue (5): 95-103.doi: 10.11959/j.issn.1000-0801.2022065

• 研究与开发 • 上一篇    下一篇

针对输电线路周边工程机械的防撞在线预警方法

陈超杰1, 谢伟1, 何林巍1, 田强1, 贺润平2, 王哲斐2   

  1. 1 国网上海市电力公司青浦供电公司,上海 201799
    2 上海四量电子科技有限公司,上海 201415
  • 修回日期:2022-04-08 出版日期:2022-05-20 发布日期:2022-05-01
  • 作者简介:陈超杰(1971- ),男,国网上海市电力公司青浦供电公司高级工程师,主要研究方向为电网运维、安全生产管理等
    谢伟(1986- ),男,国网上海市电力公司青浦供电公司工程师,主要研究方向为电力系统及其自动化
    何林巍(1993- ),男,国网上海市电力公司青浦供电公司工程师,主要研究方向为输配电线路运维
    田强(1993- ),男,国网上海市电力公司青浦供电公司工程师,主要研究方向为输电线路运维检修
    贺润平(1990- ),男,上海四量电子科技有限公司助理工程师,主要研究方向为电网运行
    王哲斐(1986- ),男,上海四量电子科技有限公司工程师,主要研究方向为电网运行
  • 基金资助:
    国家电网有限公司科技项目(SGSHQP00HBJS2102101)

Anti-collision online early warning method for construction machinery around transmission line

Chaojie CHEN1, Wei XIE1, Linwei HE1, Qiang TIAN1, Runping HE2, Zhefei WANG2   

  1. 1 State Grid Shanghai Electric Power Company Qingpu Power Supply Company, Shanghai 201799, China
    2 Shanghai Siliang Electronic Technology Co., Ltd., Shanghai 201415, China
  • Revised:2022-04-08 Online:2022-05-20 Published:2022-05-01
  • Supported by:
    Science and Technology Project of State Grid Corporation(SGSHQP00HBJS2102101)

摘要:

输电线路传输距离长、通道环境复杂,大型施工机械造成的撞击事故时有发生。现有的输电线路防撞监测方案主要依靠人工实现,既耗费大量人力资源,也难以保证监测覆盖面。因此,针对性地提出了针对输电线路周边工程机械的防撞在线预警方法,该方法创新地利用智能传感器及压缩感知算法,可实现输电线路通道内的工程机械类威胁异物自动在线预警。利用摄像模块、微波测距模块采集目标的像素分布特征、形状特征以及距离特征,并传输到后台。后台服务器利用建立的离线特征目标数据库,通过压缩感知算法实现针对该目标的研判结果,对具有威胁的目标及时发出预警。结果表明,所提算法的识别精度可达96.34%,能够为输电运检部门提供可靠的预警结果。

关键词: 输电线路防撞, 在线监测, 特征识别, 压缩感知

Abstract:

The transmission line has long transmission distance and complex channel environment.Impacts caused by large construction machinery occur from time to time.The existing anti-collision monitoring scheme of transmission line mainly depends on manual implementation, which not only consumes a lot of human resources, but also is difficult to ensure the monitoring coverage.Therefore, an anti-collision on-line early warning method for construction machinery around the transmission line was proposed.This method innovatively used intelligent sensors and compressed sensing algorithm to realize the automatic on-line early warning of construction machinery threats in the transmission line channel.The pixel distribution characteristics, shape characteristics and distance characteristics of the target were collected by the camera module and microwave ranging module, and transmitted to the background.The background server used the established offline feature target database to realize the research and judgment results of the target through compressed sensing algorithm, and sent out early warning to the threatening target in time.The experimental results show that the recognition accuracy of the proposed algorithm can reach 96.34%, and can provide reliable early warning results for the transmission inspection department.

Key words: anti-collision of transmission line, online monitoring, feature recognition, compressed sensing

中图分类号: 

No Suggested Reading articles found!