计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (15): 14-18.

• 博士论坛 • 上一篇    下一篇

动态分级中心引力约束优化算法及工程应用

吴华伟,陈特放   

  1. 中南大学 信息科学与工程学院,长沙 410083
  • 出版日期:2013-08-01 发布日期:2013-07-31

Central force constrained optimization algorithm with dynamic hierarchical and engineering application

WU Huawei, CHEN Tefang   

  1. School of Information Science and Engineering, Central South University, Changsha 410083, China
  • Online:2013-08-01 Published:2013-07-31

摘要: 结合非固定多段罚函数处理约束条件,提出一种动态分级中心引力优化算法用于求解约束优化问题。该算法利用佳点集初始化个体以保证种群的多样性。在每次迭代过程中将种群分为两个子种群,分别用于全局搜索和局部搜索,根据搜索阶段动态调整子种群个体数目。对几个标准的测试问题和工程优化问题进行数值实验,结果表明该算法能处理不同的约束优化问题。

关键词: 约束优化问题, 中心引力优化算法, 非固定多段罚函数, 工程优化

Abstract: Using non-stationary multi-stage penalty function to deal with the constrained conditions, a modified central force optimization algorithm is proposed for solving constrained optimization problems. Good point set method is used in the initialization of the evolutionary population to ensure its diversity. At each generation, the population is divided into two subpopulations based on the fitness values of particles, which is employed for global and local search respectively. The number of the subpopulation is dynamically adapted according to the search phases. The proposed algorithm has been tested on 4 benchmark problems and engineering optimization problems, and the results show that it can deal with different constrained optimization problems.

Key words: constrained optimization problems, central force optimization algorithm, non-stationary multi-stage penalty function, engineering optimization