文献类型：期刊文献

中文题名：一种基于AEA算法的改进的惩罚函数法

英文题名：Study on an improved penalty function method for constraint problem based on AEA(IAEA) algorithm

作者：李宜萱[1];金治群[1];程军蕊[1];李绍军[1]

机构：[1]r华东理工大学自动化系,上海200237

年份：2014

卷号：31

期号：12

起止页码：1479

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2011】；CSCD:【CSCD2013_2014】；

基金：国家自然科学基金资助项目(21176072)

语种：中文

中文关键词：进化算法;惩罚函数法;松弛参数;交叉搜索

外文关键词：evolutionary algorithm; penalty function method; relaxation parameter; crossover search

摘要：针对约束优化问题的惩罚函数法易陷入局部最优解的缺点,提出了一种新的IAEA算法,通过引入松弛参数μ,保留不可行解中的有用信息,并利用交叉搜索,加大了对可行域边界及其附近区域的搜索几率,使得算法能跳出局部最优,避免早熟现象的发生。在11个标准函数上进行试验,结果表明IAEA的寻优能力要好于其他3个算法。最后,将IAEA算法应用于压力容器设计的工程计算中,得到了更好的结果。
When using penalty function to solve constrained optimization problems,local optimum,not global optimum,is always found.In this paper,a new algorithm named IAEA is proposed,into which a relaxation parameter named μ is introduced to keep the useful information of unfeasible solution,and crossover search is also introduced to increase the chances that IAEA search the optimum around boundary of the feasible solution area.As a result,IAEA algorithm can easily escaped from local optimum and avoid prematurity effectively.It is shown from experimental results on 11 test functions that IAEA can attain better search ability than three other existing algorithms.Finally,IAEA is applied to the engineering calculation of pressure vessel design,and a better result is obtained.