基于机器视觉的变电生产作业安全风险智能监控方法

Machine Vision-based Method for Intelligent Monitoring of Safety Risks in Substation Operation

为了提高变电生产运行过程中的安全性与高效性,引进机器视觉技术,对变电生产作业安全风险进行智能监测。提取了变电生产作业的风险因素,基于机器视觉技术获取了变电生产作业的图像信息,分析并处理了采集的图像信息;构建了变电生产作业安全风险智能监控模型,实现了变电生产作业风险的实时、自动化监控。从500组图像数据中随机选取100组进行运行时间测试,得出监控时间平均值为3.89 s;随机选取50组安全作业数据和50组风险作业数据进行正确率测试,得出监控结果的正确率为98%。测试结果证明,所提出的监控方法可有效提高变电生产作业安全风险监控的正确率,减少监控时间,具有较好的实时性与精确性,能够为变电生产项目的安全预警提供有效的监控技术支持。

Aiming at improving safety and efficiency during daily operation of substations,the present work focuses on introducing machine vision technology into intelligent safety monitoring of safety risks of substations.In the proposed monitoring scheme,risk factors of substation operation are extracted,and pictorial information of substation operation obtained based on machine vision is collected,analyzed,and treated.Hence the intelligent monitoring model simulating substation safety risks is established,which facilitates real-time and automatic monitoring of safety risks.The experiments achieve an average monitoring time of 3.89 s when performing running time consumption test on 100 groups of image data randomly selected from 500 groups,and a monitoring accuracy of 98%when performing accuracy test on randomly selected 50 groups of safe operation data and 50 groups of risky operation data.It can be proved that,the proposed monitoring method can effectively improve accuracy and reduce time consumption of safety risks monitoring in substation operation,and thereby has better real-time property and accuracy and provides effective monitoring technological support for safety early warning of substations.