摘要
设备的安全稳定运行是石化企业实现安全稳定生产的重要因素。大机组等动设备结构复杂、故障形式多样,传统的以设备或测点为对象的监测技术应用难以满足不断增长的安全保障要求。特别是单一的参数指标不能全面体现设备故障特征,导致设备故障不能提前预警预测,即便出现异常也不能快速定位,亟需研究新的预警预测诊断技术,实现大机组等动设备异常状态的预警预测,异常状态下的快速分析定位。文章结合企业遇到的问题,基于工业大数据分析技术,提出了一种用于大机组等动设备运行状态预警预测的解决方案。通过采用数据驱动的建模方法建立设备运行状态特征模型,并通过历史数据进行学习训练,实现了基于大数据分析技术的运行状态的预警预测,并结合应用案例说明了该方案的实施效果。从应用实践情况看,该技术实施周期短、实用性强,可在一定程度上解决企业大机组等动设备运行管控中存在的难题。
The safe and stable operation of equipment is an important factor for petrochemical enterprises to achieve safe and stable production.Due to the complex structure and various fault forms of large units and other dynamic equipment,the traditional monitoring technology based on equipment or measuring point is difficult to meet the ever-increasing safety requirements.In particular,the single parameter index cannot fully reflect the characteristics of equipment failure,resulting in that the failure of equipment cannot be predicted in advance.Even if an abnormality occurs,it cannot be quickly located.It is urgent to study new warning,prediction and diagnosis techniques to realize early warning and prediction of abnormal conditions of large units and other dynamic equipment,and to quickly analyze and locate the abnormal status.This article proposes a solution for early warning and prediction of the operating status of large units and other dynamic equipment based on industrial big data analysis technology.By using the data-driven modeling method to establish the characteristic model of the running state of the equipment,and through historical data for machine learning and training,the early warning and prediction mode of the operating state based on big data analysis technology is realized,and the implementation effect of the solution is illustrated with application cases.From the perspective of application practice,this technology has a short implementation period and strong practicability,which can solve the problems existing in the operation control of large units and other dynamic equipment in enterprises to a certain extent.
作者
贺宗江
He Zongjiang(China Petrochemical Corporation Refinery Department,Beijing 100728,China)
出处
《当代石油石化》
CAS
2020年第6期35-40,共6页
Petroleum & Petrochemical Today
关键词
工业大数据
设备预警预测
关联分析
industrial big data
equipment early warning and prediction
correlation analysis