域约束优化问题的普适免疫进化算法
Universal Immune Evolutionary Algorithm for Interval-Constrained Optim ization Problem s
-
摘要: 在免疫进化算法的基础上,针对域约束优化问题,提出了一种普适算法.通过区间变换,该算法在保证所 产生的个体分量均能满足相应的区间约束的同时,消除了参数设置的随意性,不仅提高了计算效率,而且增强了 算法的统一性,克服了其它进化算法采用罚函数处理域约束问题的不足.多峰函数优化和遗传算法欺骗问题的 测试结果表明:与采用罚函数处理域约束问题的免疫进化算法相比,普适算法不仅易于编程,而且能以更快的速 度稳健地收敛到全局最优解.Abstract: Based on the current immune evolutionary algorithm ( IEA) and aimed at interval- constrained problems, a universal immune evolutionary algorithm was proposed. W ith this universal algorithm, any individualofevery generation canmeet the requirementof interval constraintsbymeans of interval transition, and the subjectivity ofparameters setting can be further eliminated. As a resul,t the computation efficiency is greatly raised, the unitarity of the universal algorithm is improved, and the disadvantages of the IEA adopting penalty for similar problems are also avoided. In addition, the universal algorithm was applied to the optimization ofmulti-model functions and the test of a GA (genetic algorithm) deceptive problem. The results show that compared with the IEA adopting penalty, programming is easy for the universal algorithm, and the global optimal solution can be obtained at a fast and stable speed.
点击查看大图
计量
- 文章访问数: 1606
- HTML全文浏览量: 68
- PDF下载量: 203
- 被引次数: 0