健康度驱动的炼油设备动态风险评价方法研究  

Research on dynamic risk assessment method for refining equipment driven by health degree

作　　者：许述剑 王庆锋[2] 刘晓金 刘曦泽 宁志康 XU Shujian;WANG Qingfeng;LIU Xiaojin;LIU Xize;NING Zhikang(SINOPEC Research Institute of Safety Engineering Co.,Ltd.,Qingdao Shandong 266100,China;不详)

机构地区：[1]中石化安全工程研究院有限公司,山东青岛266100 [2]北京化工大学,北京100029

出　　处：《工业安全与环保》2024年第9期28-34,共7页Industrial Safety and Environmental Protection

基　　金：中国石油化工集团公司科技部资助项目(319022-1)。

摘　　要：针对传统的设备健康状态评价是基于振动烈度等单一性能指标表征,未全面描述设备性能状态并归一化,以及传统的设备风险评价取决于失效后果和基于历史数据概率统计获得的失效概率,属于静态风险模型,难以评价设备的运行风险,提出一种实时监测数据健康度驱动的设备动态风险评价模型和方法。创新构建了振动信号谱距离指标/互相关/凝聚系数的转动设备、基于失效敏感特征参数的承压设备的健康度模型,首创实时监测数据健康度驱动的设备动态风险评价方法,依据健康度及风险状况指导企业及时开展设备预知性维修。最后,采用工程案例验证了本文提出的模型和方法,并在试点企业93台设备上开展工业应用,表明本模型比传统的固定阈值限报警提前不低于28 h,预警准确率92.98%,避免了多次非计划停机。Traditional equipment health status evaluation is based on a single performance indicator such as vibration intensity,which does not fully describe and normalize equipment performance status,and traditional equipment risk evaluation depends on failure consequences and failure probability obtained based on historical data probability statistics,which belongs to the static risk model and is difficult to evaluate the operation risk of equipment,a real-time monitoring data health degree-driven model and method for dynamic risk assessment of chemical equipment is proposed.Health degree models of the cross-correlation coefficient,aggregation coefficient,spectral distance index coefficient for rotating equipment and of failure sensitive characteristic parameters for pressure equipment are innovatively constructed.The dynamic risk assessment methods driven by health degree of real-time monitoring data are first developed,and then according to equipment health and risk conditions,guide enterprises to carry out the predictive maintenance.Finally,an engineering case is used to verify the model and method proposed in this paper,and the industrial application is carried out on 93 equipment in pilot enterprises.The application shows that the model is no less than 28 hours ahead of the traditional fixed threshold alarm,the early warning accuracy rate is 92.98%,and many unplanned shutdowns are avoided.

关 键 词：炼油设备 健康度 动态风险 评价方法 数据驱动 

分 类 号：TE96[石油与天然气工程—石油机械设备]