详细信息
基于文化算法的新型Elman网络的过程建模方法 ( EI收录)
Process Modeling of A New Improved Elman Neural Network Based on Cultural Algorithm
文献类型:期刊文献
中文题名:基于文化算法的新型Elman网络的过程建模方法
英文题名:Process Modeling of A New Improved Elman Neural Network Based on Cultural Algorithm
作者:齐仲纪[1];刘漫丹[1]
机构:[1]华东理工大学信息科学与工程学院,上海200237
年份:2008
卷号:34
期号:5
起止页码:745
中文期刊名:华东理工大学学报(自然科学版)
外文期刊名:Journal of East China University of Science and Technology
收录:CSTPCD;;EI(收录号:20084811740435);Scopus;北大核心:【北大核心2004】;CSCD:【CSCD2011_2012】;
基金:国家高技术研究发展计划(2007AA042171;2007AA042164)
语种:中文
中文关键词:文化算法;Elman网络;裂解深度;建模
外文关键词:cultural algorithm; Elman network; cracking severity; modeling
摘要:文化算法从微观(种群空间)和宏观(信念空间)两个层面上模拟文化的双重进化继承过程,为进化搜索机制和知识存储的结合提供一个构架。建立基于输入输出数据生产过程的统计模型时,参数估计是其中的关键,文化算法为此提供了有效途径。本文在Elman神经网络的基础上提出了一种新的改进型Elman网络模型——OAIF-Elman(Output-Add-Input Feedback Elman)网络来建立乙烯装置中裂解深度软测量模型,并结合文化算法来优化其网络权值。实验表明:文化算法比标准遗传算法搜索性能更优,搜索时间更快,同时也得到了满意的裂解深度模型。
Cultural algorithm(CA) depicts cultural evolution as a process of dual inheritance from both the micro evolutionary level (population space) and the macro-evolutionary level (belief space). It provides a framework for the combination of evolution searching and knowledge storage. CA provides an effectual approach for parameter estimation, which is the key step of process statistical model based on input data and output data. A new improved Elman neural network, OAIF-Elman(Output Add-Input Feedback Elman) network, is presented based on the Elman neural network in this paper. It is applied to set up ethylene cracking severity soft sensor model, and the parameters of network are optimized by CA. The simulation result indicate that CA have better search performance and shorter search time than normal genetic algorithm. Meanwhile, the satisfying model of cracking severity is obtained.
参考文献:
正在载入数据...
