基于改进群搜索算法的裂解炉切料优化方法  

Optimization method of changing feedstock based on improved group search optimization for cracking furnace

作　　者：年笑宇[1] 王振雷[1] 田亮[2] 

机构地区：[1]华东理工大学,化工过程先进控制和优化技术教育部重点实验室,上海200237 [2]中国石化上海石油化工股份有限公司,上海200237

出　　处：《计算机与应用化学》2012年第9期1103-1106,共4页Computers and Applied Chemistry

基　　金：自然科学基金项目重点基金(U1162202);国家优秀青年基金项目(61222303);国家自然科学基金项目(21276078,21206037);点学科建设项目(B504)

摘　　要：本文针对裂解炉进料切换过程中存在的时机难以选择的问题,应用COILSIM1D软件建立了裂解炉切料过程的带约束优化函数模型,约束条件包括运行周期约束、炉管外壁温度约束、每种进料的处理时间约束等。利用本文提出的改进群搜索算法求解上述带约束的优化问题,该优化算法利用双适应值法处理约束并针对其缺陷加入了2点改进:(1)采用可行解占总体解的比例自适应地改变罚函数因子;(2)采用保留"优秀"不可行解的方法在边界附近指导寻优。最后,以某乙烯厂实际的进料切换过程为研究实例进行切料时机的优化,优化后,裂解炉乙烯和丙烯的收率总量提高,为操作人员选择最佳切炉时机提供了理论依据。To solve the problem of the choice for the right time of changing feedstock in the process of cracking furnace, a constraint optimal fimction model which describes the process of switching feedstock is constructed by COILS1M1D. The constraints take into account the whole cycle of operation, the maximum tube metal temperature and the process time of each feedstock. The model is calculated with group search optimization algorithm, while the constraint problem is dealt with the method of double fitness values and two improvements are added to the defect： 1） the factor of penalty function should be changed adaptively by the ratio of feasible solution and general solution; 2） the ＂excellent＂ infeasible solution should be guided to search the optimization by retaining them. Finally, the actual process of switching feedstock in an ethylene plant is illustrated as a case study and optimized. After optimization, the yield of ethylene and propylene is increased observably, which provides the support of the proper choice for time of switching feedstock for the operation staff.

关 键 词：乙烯裂解炉调度 群搜索优化算法 约束优化问题 自适应罚函数 COILSIM1D计算 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]