基于Yolo v3的特定电力作业场景下的违规操作识别算法
DOI:
作者:
作者单位:

1.广西电网有限责任公司电力科学研究院;2.广西电网有限责任公司

作者简介:

通讯作者:

中图分类号:

TM863

基金项目:

广西电网有限责任公司科技项目:基于机器学习的广西电网电力作业全过程风险管控系统研发及应用


Illegal Operation Recognition Algorithm Based on YOLO v3 in The Specific Power Operation Scenario
Author:
Affiliation:

1.Electric Power Research Institute of Guangxi Power Grid Co,Ltd;2.Guangxi Power Grid Co,Ltd

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    电网作业常处于高空、高压等危险环境,威胁电力作业人员的安全,仅靠人力监管常会出现监管不力的情况,现有的目标检测算法也只能进行简单的安全识别,无法根据特定的电力作业场景识别违规操作行为。针对这一问题,本文提出了一种基于YOLO v3的特定电力作业场景下的违规操作识别算法,选用YOLO v3算法进行目标检测,同时融入场景识别算法,并引用交并比设定逻辑判断函数,检测特定场景下电力作业的违规操作行为。以电焊作业场景为例进行了实验验证,实验结果表明,本模型的检测精确率为82.15%,证明了该方法的有效性,同时本文也对后续优化该模型提出了几点建议。

    Abstract:

    Power grid operations are often in dangerous environments such as high altitudes and high voltages, which threaten the safety of electric power operators. Only human supervision often leads to ineffective supervision, and the existing target detection algorithms can only perform simple safety identification and cannot identify illegal operations based on specific power operation scenarios. To solve this problem, this paper proposes an illegal operation recognition algorithm based on YOLO v3 in the specific power operation scenario. The YOLO v3 algorithm is selected for target detection, and the scene recognition algorithm is incorporated at the same time. The logic judgment function is set by reference to the Intersection over Union to detect the violation of power operations in specific scenarios. After taking the welding scene as an example for experimental verification, the experimental results show that the detection accuracy of this model is calculated to be 82.15%, which proves the effectiveness of the method. At the same time, this article also puts forward several suggestions for subsequent optimization of the model.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-03-19
  • 最后修改日期:2021-04-30
  • 录用日期:2021-05-06
  • 在线发布日期:
  • 出版日期: