文献类型：期刊文献

中文题名：融合copula EDA的AEA算法及其在参数估计中的应用

英文题名：AEA Combined with copula EDA and Its Application on Parameter Estimation

作者：何鹏飞[1];李绍军[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：2015

卷号：29

期号：1

起止页码：177

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：20151300692445);Scopus;北大核心:【北大核心2014】；CSCD:【CSCD2015_2016】；

基金：国家自然科学基金资助(21176072)

语种：中文

中文关键词：Alopex;AEA;copula;EDA;优化;参数估计

外文关键词：Alopex; AEA; copula EDA; optimization; parameter estimation

摘要：着眼于AEA(Alopex-based evolutionary algorithm)算法本身的不足,提高算法寻优性能,今构造出了一种融合了copula分布估计算法(copula EDA)和AEA算法的改进型算法-CAEA。将copula分布估计算法嵌入到AEA中,改进AEA算法中种群的生成方式,保持了种群的多样性。改进后的算法不仅拥有AEA算法启发搜索和确定性搜索的优点,同时还具有copula分布估计算法收敛速度快、包含全局搜索信息的特点。利用CAEA算法对9个标准函数进行测试实验,并与AEA算法、改进的粒子群算法(MSCQPSO)和差分进化算法(ISDEMS)的测试结果进行比较,结果表明CAEA算法无论在精确度还是稳定性方面都具有较大的提高。最后将算法用于发酵动力学模型参数的估计,通过优化得到了较好的结果,充分验证了所提出的算法的可行性和有效性。
An algorithm （CAEA） was proposed which combines copula estimation of distribution algorithm （copula EDA） and AEA together. In order to improve the generation method of population, copula EDA was imbedded into AEA and the population diversity was maintained. The modified algorithm takes advantages of heuristic search and deterministic search of AEA, and inherits the characteristics of rapid convergence and global search information of copula EDA. Moreover, the CAEA algorithm was tested by 9 benchmark functions and the performance of the algorithm was obtained. Compared with AEA, MSCQPSO and ISDEMS, the results of the simulation indicate that the performance of the modified algorithm is significantly improved in both accuracy and stability. Furthermore, the algorithm was applied to parameter estimation of models of fermentation dynamics and satisfactory results were achieved, which indicates that the proposed algorithm is effective and efficient.