乙烯裂解炉机理模型在线预报系统的设计与实现  

System design and realization of on-line prediction of ethylene pyrolyzer based on mechanism model

作　　者：康少栋[1] 熊智华[1] 吕宁[1] 王京春[1] 邵杰峰[2] 钟向宏[2] 

机构地区：[1]清华大学自动化系,北京100084 [2]中国石化集团茂名石油化工公司,广东茂名525011

出　　处：《计算机与应用化学》2010年第10期1313-1316,共4页Computers and Applied Chemistry

基　　金：国家高技术研究发展计划(863)(2007AA041402).

摘　　要：乙烯裂解炉裂解产物收率的在线预报对于指导乙烯生产具有十分重要的意义。本文基于Kumar分子反应动力学模型,针对某大型乙烯厂SL-Ⅱ型裂解炉建立了该炉辐射段的裂解数学模型;采用C#和MATLAB混合编程技术开发了1套乙烯裂解炉产物收率在线预报软件系统;针对长周期运行的情况加入了模型在线调整功能,以随机搜索方式调整1次反应选择性系数,使模型能够及时与当前原料性质匹配。现场运行效果表明,系统的实时性和准确性满足生产要求,平均预报时间在10秒左右,乙烯等主要产物收率的系统预报值与工厂化验值的误差约2%;同时,系统具有良好的实用性,操作简便且运行稳定。该系统基于规范化的软件系统框架实现了乙烯裂解机理模型的实时运行和在线预报,以间接方式获得主要裂解产物收率的实时数据,为进一步指导生产过程优化、提高经济效益奠定了基础。On-line prediction of ethylene pyrolysis product yields makes great significance for guiding ethylene production.Based on Kumar＇s molecular dynamics model,a mechanism model is built for an SL-Ⅱethylene pyrolyzer in a petrochemical company;A software system of product yields on-line prediction is developed using C# and MATLAB.Considering a long-term operation of the system,an on-line model adjusting function, which uses a stochastic method to adjust the selectivity coefficient ofthe primary reaction,is also added to make the model matched with current feed oil property in time.The field application demonstrates that the real-time performance and accuracy of the system meets production requirements,the average predictive time is around 10 seconds and the error of the predictive value of main product yields sr ch as ethylene is around 2%,compared with the experimental value.Besides,the practical system is easy to use and runs steadily.Based on standardized frame of software engineering,the system realize real-time operation and on-line prediction of the ethylene pyrolysis mechanism model.The indirect obtaining of main pyrolysis product yields data in quasi real-time lay a basis for further guiding optimization of production process and improving economic benefit.

关 键 词：乙烯裂解炉 机理模型 混合编程 实时预报 

分 类 号：TQ015.9[化学工程] TP391.9[自动化与计算机技术—计算机应用技术]