文献类型：期刊文献

中文题名：S-Zorb反应器的机理建模与基于改进鲸群算法的参数估计

英文题名：Mechanism Modeling of S-Zorb Reactors and Parameter Estimation with Improved Whale Algorithm

作者：贾华宇[1];杜文莉[1];范琛[1];杨明磊[1]

机构：[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

年份：2018

卷号：32

期号：6

起止页码：1395

中文期刊名：高校化学工程学报

外文期刊名：Journal of Chemical Engineering of Chinese Universities

收录：CSTPCD;;EI(收录号：20190906549166);Scopus;北大核心:【北大核心2017】；CSCD:【CSCD2017_2018】；

基金：国家自然科学基金重大项目(61590923);国家自然科学基金(61422303;61590922);中央高校基本科研业务费专项基金(222201717006)

语种：中文

中文关键词：S-Zorb;脱硫率;辛烷值;参数估计;鲸群算法

外文关键词：S-Zorb;desulfurization rate;octane number;parameter estimation;whale optimization algorithm

摘要：汽油吸附脱硫(S-Zorb)装置是我国催化裂化汽油的主要脱硫装置之一。针对S-Zorb工业模型缺乏,实验室模型无法预测产品油辛烷值等问题,作者提出了一种基于过程机理的反应器建模方法。该方法将催化裂化汽油划分为5个集总,在集总基础上构建了反应动力学模型、辛烷值关联模型等。另外,针对传统群优化算法参数估计耗时长的缺点,提出了一种改进的鲸群算法,并成功应用于S-Zorb模型。通过与工业数据对比验证结果表明,所提模型能够有效地预测脱硫率、辛烷值等关键指标,从而为后续的优化生产提供基础。
Gasoline adsorption desulfurization units(S-Zorb)are one of the main devices for FCC(fluid catalytic cracking)gasoline desulfurization in China.However,S-Zorb lacks industrial prediction models and laboratory models cannot predict product octane number.A reactor model based on process and mechanism was developed by dividing the catalytic cracking gasoline into five lumps.The reaction kinetics model and octane correlation model were constructed based on these lumps.In addition,an improved whale algorithm was proposed to overcome the disadvantages of traditional group optimization algorithm,and was successfully applied for the S-Zorb model.The proposed model was compared with industrial data and the results show that it can effectively predict key indexes such as desulfurization rate and octane number.Therefore,the model provides a new method for the optimization of S-Zorb production.