改进的引力搜索算法求解约束优化问题

发布时间：2018-01-22 05:06

本文关键词： 引力搜索算法约束优化违反约束度全局优化精确罚函数法　出处：《渤海大学》2017年硕士论文　论文类型：学位论文

【摘要】：约束优化问题在许多领域中都有广泛的应用,而且随着社会的发展,约束优化问题越来越复杂,很难求出全局最优解。本文对现今求解约束优化问题的方法进行了大致的总结,并分析了其发展状况与优缺点。在此基础上提出了两种新的求解约束优化问题的方法。论文的主要研究工作如下:1、提出求解约束优化问题的双质量引力搜索算法。该算法中每个粒子具有两个质量,即“可行质量”和“不可行质量”。如果粒子在可行区域中的位置被更新,则使用可行质量。否则,使用不可行质量。这个想法是基于可行的粒子被吸引向更好的可行粒子,不可行的粒子被吸引向可行区域。最后,提出的算法在10个基准测试函数上进行测试,并与其他算法进行比较分析。数值结果表明,所提出的算法在解决约束优化问题中具有更好的性能。2、提出混合引力搜索算法。通过模拟退火精确罚函数法构造一个约束子问题,然后用引力搜索算法求解该约束子问题,最后通过与其他四种算法进行比较,证明了其可行性。
[Abstract]:Constrained optimization problems are widely used in many fields, and with the development of society, constrained optimization problems become more and more complex. It is difficult to find the global optimal solution. In this paper, the methods for solving constrained optimization problems are summarized. On the basis of this, two new methods for solving constrained optimization problems are proposed. The main research work of this paper is as follows: 1. A two-mass gravitational search algorithm for constrained optimization problems is proposed, in which each particle has two masses, that is, "feasible mass" and "unfeasible mass", if the particle's position in the feasible region is updated. Otherwise, use the unfeasible mass. This idea is based on the idea that viable particles are attracted to better viable particles, and infeasible particles to feasible regions. Finally. The proposed algorithm is tested on 10 benchmark functions and compared with other algorithms. The numerical results show that the proposed algorithm has better performance in solving constrained optimization problems. A hybrid gravity search algorithm is proposed. A constraint subproblem is constructed by simulated annealing exact penalty function method. Then the constraint subproblem is solved by gravity search algorithm. Finally, it is compared with the other four algorithms. The feasibility has been proved.
【学位授予单位】：渤海大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18;O224

【相似文献】