文献类型：期刊文献

中文题名：改进鲸鱼优化算法及其在渣油加氢参数优化的应用

英文题名：Improved whale optimization algorithm and its application in optimization of residue hydrogenation parameters

作者：许瑜飞[1];钱锋[1];杨明磊[1];杜文莉[1];钟伟民[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：69

期号：3

起止页码：891

中文期刊名：化工学报

外文期刊名：CIESC Journal

收录：CSTPCD;;EI(收录号：20184105915756);Scopus;北大核心:【北大核心2017】；CSCD:【CSCD2017_2018】；

基金：国家科技支撑计划项目(2015BAF22B02);国家自然科学基金项目(61422303;61590922);中央高校基本科研业务费专项资金~~

语种：中文

中文关键词：算法;鲸鱼优化算法;渣油加氢;动力学模型;参数估值;优化

外文关键词：algorithm; whale optimization algorithm; residue hydrogenation; kinetic modeling; parameter estimation; optimization

摘要：针对智能优化算法在处理非线性优化问题中存在的容易陷入局部最优和收敛精度差等问题,提出了一种基于结合差分进化和精英反向学习的改进鲸鱼算法(DEOBWOA)。该算法引入对立搜索初始化、精英反向学习,并结合差分进化进行变异修正,显著有效地提高WOA算法的收敛精度和收敛速度,提高其跳出局部最优的能力。之后采用8个标准测试函数进行仿真实验,结果表明:DEOBWOA算法与标准WOA、HCLPSO、DE算法相比,全局搜索能力和收敛速度都有较大提升。最后建立了渣油加氢动力学模型,考虑到渣油加氢过程中存在诸多典型的非线性约束问题,以某炼化厂渣油加氢装置为例,应用DEOBWOA对渣油加氢反应动力学模型参数进行优化,结果表明该算法能较好地处理实际工程优化问题。
An improved whale algorithm(DEOBWOA)based on differential evolution and elite opposition-basedlearning is proposed to solve the problem that the intelligent optimization algorithm is easy to fall into the localoptimum and the convergence precision in dealing with the nonlinear optimization problem is poor.The algorithmuses the opposing search initialization,elite opposition-based learning and combines with differential evolution,which can improve the convergence precision and convergence speed of the whale optimization(WOA)algorithmeffectively and improve the ability to jump out of local optimum.8standard test functions are used to dosimulation experiment.The results show that DEOBWOA algorithm has a better performance than WOA,heterogeneous comprehensive learning particle swarm optimization(HCLPSO)and differential evolution(DE).Finally,the kinetic model of residue hydrogenation was established,but there are many typical nonlinearconstraints in the process of residue hydrogenation.So DEOBWOA was used to optimize the kinetic modelparameters of residue hydrogenation in a refinery residue,which indicates the algorithm can deal with thepractical engineering optimization problem.