中国安全科学学报 ›› 2021, Vol. 31 ›› Issue (S1): 19-23.doi: 10.16265/j.cnki.issn 1003-3033.2021.S1.004

• 安全社会科学与安全管理 • 上一篇    下一篇

基于人工智能视频分析的选煤厂安全管理研究

赵亮 高级工程师, 孙魁元** 工程师, 韩宝虎 高级工程师, 安文利 高级工程师, 李国强   

  1. 国能宝日希勒能源有限公司, 内蒙古 呼伦贝尔 021025
  • 收稿日期:2021-09-10 修回日期:2021-11-18 出版日期:2021-12-30 发布日期:2022-06-30
  • 通讯作者: ** 孙魁元(1988—),男,山东鄄城人,硕士,工程师,主要从事选煤厂安全管理方面的工作。E-mail: 715521422@qq.com。
  • 作者简介:赵 亮 (1972—),男,吉林辽源人,本科,高级工程师,主要从事机电工程、信息化、智能化方面的工作。 E-mail:741943@qq.com。

Research on safety management of coal preparation plants based on artificial intelligence video analysis

ZHAO Liang, SUN Kuiyuan, HAN Baohu, AN Wenli, LI Guoqiang   

  1. Baorixile Energy Co., Ltd., CHN Energy, Hulunbuir Inner Mongolia 021025, China
  • Received:2021-09-10 Revised:2021-11-18 Online:2021-12-30 Published:2022-06-30

摘要: 为提高选煤厂安全监控的智能化水平,降低人力值守消耗,利用深度神经网络构建面向选煤厂安全管理的人工智能(AI)视频分析系统,首先,设计深度神经网络算法,实时检测分析典型场景视频目标,构建监控预警管理系统;然后,建立人、机、环的视频安全预警机制,实现对选煤厂现场作业人员不安全行为的视频分析、物的不安全状态的视频识别和环境危险因素的视频判断。结果表明:该系统通过对人、机、环信息的获取、处理和反馈,能够实现及时预警消除安全隐患,提高作业过程的安全系数。

关键词: 人工智能(AI), 视频分析, 选煤厂, 安全管理, 深度学习

Abstract: In order to improve intelligent safety monitoring in coal preparation plants, and to reduce manpower consumption, an AI video analysis system for safety management was developed by using deep neural network. Firstly, deep neural network algorithm was designed to detect and analyze video targets in typical scenes in real time, and a monitoring and early warning management system was established. Then, video security early warning mechanism for human, machines, and environment was developed for video analysis of workers' unsafe behaviors at the plants, video identification of unsafe conditions of objects, and video judgments of environmental hazards. The results show that the system can realize timely warning to eliminate potential safety hazards through acquisition, processing and feedback of human, machine, and environmental information, and improve safety factor of operation process.

Key words: artificial intelligence (AI), video analysis, coal preparation plants, safety management, deep learning

中图分类号: