基于机器视觉的海上平台安全风险自动识别系统
Safety Risk Automatic Identification System for Offshore Drilling and Completion Operations Based on Machine Vision
刘宇沛 1王天昊 1邢紫筱 2李轩 2朱丹丹2
作者信息
- 1. 中海油研究总院有限责任公司,北京 100028
- 2. 中国石油大学(北京)信息科学与工程学院,北京 102249
- 折叠
摘要
针对海上平台的视频监控系统存在全时监测不足、事前预警不足和人工成本高昂等问题,提出了一种基于机器视觉的海上平台安全风险自动识别系统,该系统采用了旋转检测技术,能够在设备倾斜状态下保证检测结果的准确性;通过引入目标跟踪技术,实现对人员和设备的跟踪和测距,从而在复杂环境中有效监控海上平台的运行状态.该系统在某海上平台进行了部署与测试,各项算法均取得了 90％以上的准确率,有效提升了海上钻完井作业的安全性和效率,减少了潜在隐患和安全事故的发生.
Abstract
Aiming at the problem of the video monitoring system on offshore platforms,such as insufficient full-time monitoring,lack of proactive warnings and high labor costs,a safety risk automatic identification system in offshore platform based on machine vision is proposed.The system employs rotation detection technology to ensure the accuracy of detection results even when equipment is tilted.By introducing object tracking technology,tracking and measuring distances of personnel and equipment are realized,thereby the operational status of offshore platforms in complex environments can be effectively monitored.The system is deployed and tested on an offshore platform,and the accuracy rate is over 90％for each individual algorithm.The safety and efficiency of offshore drilling and completion operations are significantly enhanced.The occurrence of potential hazards and safety incidents are reduced.It plays great contributions to improving the safety,efficiency,management of offshore drilling and completion operations.
关键词
海上钻探作业/机器视觉/目标检测/安全生产/智能视频监控Key words
offshore drilling operations/machine vision/object detection/safety production/intelligent video monitoring引用本文复制引用
出版年
2024
NETL
NSTL
万方数据