危化品企业人员作业安全与风险智能管控平台建设研究  

Development of Intelligent Management and Control Platform of Site Operation Safety Risk for Hazardous Chemical Enterprises

作　　者：穆波 刘超[3] 王廷春 慕璘 王辉 王云龙 李千登 MU Bo;LIU Chao;WANG Tingchun;MU Lin;WANG Hui;WANG Yunlong;LI Qiandeng(SINOPEC Research Institute of Safety Engineering,Qingdao 266012,China;State Key Laboratory of Safety and Control for Chemicals,Qingdao 266012,China;HSE Supervision,China Petroleum Electric Energy Co.,Ltd.,Daqing163455,China;Department of Chemical Engineering,Tsinghua University,Beijing100084,China;Shandong Qiwangda Petrochemical Co.,Ltd.,Zibo 255400,China)

机构地区：[1]中国石油化工股份有限公司青岛安全工程研究院,山东青岛266012 [2]危险化学品安全控制国家重点实验室,山东青岛266012 [3]中国石油集团电能有限公司HSE监督站,黑龙江大庆163455 [4]清华大学化学工程系,北京100084 [5]山东齐旺达石油化工有限公司,山东淄博255400

出　　处：《安全与环境工程》2020年第5期134-140,共7页Safety and Environmental Engineering

基　　金：国家重点研发计划项目(2018YFC0809300);国家危险化学品风险预警与信息共享服务关键技术研究与示范;山东省重点研发计划项目(2018YFJH0802);化工企业人员作业行为智能识别与管控技术研究。

摘　　要：近年危化品企业安全事故带来了巨大的经济损失和人员伤亡,其中大多数都发生在现场作业环节。依托物联网、深度学习等智能技术,开发了适用于危化品企业的人员作业安全与风险智能管控平台,包括工作台、人员位置管控、违章行为监测等9个模块,实现了对现场作业过程中人员位置、人员状态、人员行为、监督管理和作业环境五个方面风险信息的采集、感知、处理、预警、风险分级和闭环处置。通过将平台在某化工企业的混合芳烃加氢装置进行应用表明:该平台能够准确地识别现场作业人员典型违章行为等风险,其综合识别准确率大于80%,可以有效规范人员行为和提高作业安全管理水平。Recently safety accidents of hazardous chemical enterprises cause huge economic losses and casualties.Most of the accidents occur in the site operations.With intelligent technology such as Internet of Things and deep learning,this paper develops an intelligent management and control platform for site operation safety.This platform includes nine modules including workbench,personnel position control,and violation monitoring.It realizes the collection,perception,processing,early warning,risk grading and closed-loop management of risk information in the aspects of personnel position,personnel status,personnel behavior,supervision and management,and operating environment during the operation.Its application to the mixed aromatic hydrocarbon hydrogenation device in a chemical enterprise shows that the platform can accurately identify the dangers such as typical violations of on-site operators,and the comprehensive recognition accuracy rate is greater than 80%,which can effectively regulate personnel behaviors and improve operational safety management.

关 键 词：危化品企业 作业行为 作业安全 风险管控 人工智能 

分 类 号：X924.3[环境科学与工程—安全科学] TQ086[化学工程]