计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (26): 41-44.

• 学术探讨 • 上一篇    下一篇

基于密度熵的多目标粒子群算法

宋 武,郑金华   

  1. 湘潭大学 信息工程学院,湖南 湘潭 411105
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-09-11 发布日期:2007-09-11
  • 通讯作者: 宋 武

MOPSO algorithm based on density entropy

SONG Wu,ZHENG Jin-hua   

  1. Institute of Information Engineering,Xiangtan University,Xiangtan,Hunan 411105,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-09-11 Published:2007-09-11
  • Contact: SONG Wu

摘要: 提出了一种基于密度熵的多目标粒子群算法(EMOPSO)。采用一个外部集保存所发现的Pareto最优解(精英),并将外部集作为粒子的全局极值。为保证种群的多样性,当精英大于外部集的大小时采用一种基于密度熵的策略进行分布度保持,从而使所得到的解集保持良好的分布性。最后与经典的多目标进化算法(MOEAs)进行了对比实验,实验结果表明了该算法的有效性。

关键词: 多目标优化, 密度熵, 多目标粒子群优化, 粒子群优化

Abstract: This paper presents a new multi-objective particle swarm based on density entropy.It uses an external archive to preserve the Pareto solutions that find so far,and denote the external archive as the Pbest.In order to make the partical spreads the whole pareto frontier,it uses a density entropy schem to run the external archive when the number of the archive is larger than the fixed size,this scheme can make final obtained solutions better distribution.At the end we compare our partical swarn to other classical MOEAs.The experimental result has indicated this algorithm efficiency.

Key words: multi-objective optimization, density entropy, multi-objective particle swarm optimization, particle swarm optimization