文献类型：期刊文献

中文题名：基于文化算法的生产过程建模法

英文题名：Process modeling based on cultural algorithm

作者：刘漫丹[1];齐仲纪[1]

机构：[1]华东理工大学信息学院,上海200237

年份：2007

卷号：24

期号：9

起止页码：1205

中文期刊名：计算机与应用化学

外文期刊名：Computers and Applied Chemistry

收录：CSTPCD;;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(2007AA04Z171;2007AA04Z164)

语种：中文

中文关键词：文化算法;建模;进化计算;神经网络

外文关键词：cultural algorithm, modeling, evolutionary computation, artificial neural network

摘要：文化算法是一种模拟文化进化过程的优化算法,它由基于个体和群体特性的信念空间和基于个体行为的种群空间组成,为进化搜索机制和知识存储的结合提供一个构架。建立基于生产过程输入输出数据的统计模型时,参数估计是其中的关键,文化算法为此提供了有效途径。本文研究用文化算法实现多变量优化的具体步骤、算法和关键环节的实施。建立裂解炉裂解深度的神经网络模型,并用文化算法优化网络参数,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,同时得到了满意的裂解深度模型。
Cultural Algorithm is a kind of optimizing algorithms according to simulating the cultural evolution model, it is composed of belief space based on traits of individuals or colonies, and population space based on behaviours of individuals. It provides a framework for the combo of evolution searching and knowledge instructing. Cultural Algorithm provides an effectual approach for parameter estimation, which is the pivotal step of process statistical model based on input data and output data. The detail of steps, algorithm, and the actualization of key parts is expatiated for multi-variable optimization problems using cultural algorithm. A Neural Network model for severity of cracking furnace is established, and the parameters of network are optimized by cultural algorithm. The simulation results indicate that cultural algorithm have better search performance and shorter search time than normal genetic algorithm, and the satisfying model of cracking severity is obtained.