计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (31): 108-111.

• 数据库、信号与信息处理 • 上一篇    下一篇

多目标自适应和声搜索算法

陈莹珍,高岳林   

  1. 北方民族大学 信息与系统科学研究所,银川 750021
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-11-01 发布日期:2011-11-01

Multi-objective self-adaptive harmony search algorithm

CHEN Yingzhen,GAO Yuelin   

  1. Institute of Information and System Science,North University for Nationalities,Yinchuan 750021,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-11-01 Published:2011-11-01

摘要: 提出了一种利用Pareto支配来求解多目标优化问题的自适应和声搜索算法(MOSAHS)。该算法利用外部种群来保存非支配解,为了保持非支配解的多样性,提出了一种基于拥挤度的删除策略,这个策略能较好地度量个体的拥挤程度。用5个标准测试函数对其进行测试,并与其他多目标优化算法相比较。实验结果表明,与其他的算法相比,提出的算法在逼近性和均匀性两方面都有很好的表现,是一种有效的多目标和声搜索算法。

关键词: 多目标优化, 和声搜索算法, 拥挤度

Abstract: A self-adaptive harmony search algorithm incorporate Pareto dominance to solve multi-objective optimization problems is presented.The algorithm adopts an external archive to keep non-dominated solutions.In order to maintain the diversity of the non-dominated solutions,a crowding measure is proposed in this article.The crowding strategy can measure the crowding degree accurately.The experiments are performed using five benchmark test functions and compared with other multi-objective optimization algorithms.The experiment results show that,the proposed MOSAHS algorithm is an effective multi-objective harmony search algorithm with fine performance in both convergence and diversity.

Key words: multi-objective optimization, harmony search algorithm, crowing degree