计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (15): 41-43.

• 学术探讨 • 上一篇    下一篇

小生境粒子群优化算法

向长城1,2,黄席樾1,杨祖元1,杨 欣1   

  1. 1.重庆大学 自动化学院 导航与制导试验室,重庆 400030
    2.湖北民族学院 数学系,湖北 恩施 445000
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-05-21 发布日期:2007-05-21
  • 通讯作者: 向长城

Niche Particle Swarm Optimization algorithm

XIANG Chang-cheng1,2,HUANG Xi-yue1,YANG Zu-yuan1,YANG Xin1   

  1. 1.College of Automation,Chongqing University,Chongqing 400030,China
    2.Department of Mathematics,Hubei Institute of Nationalities,Enshi,Hubei 445000,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-05-21 Published:2007-05-21
  • Contact: XIANG Chang-cheng

摘要: 针对粒子群算法容易早熟收敛和后期收敛速度慢的缺点,结合进化论中小生境技术,提出了小生境粒子群优化算法。通过粒子之间的距离找到具有相似距离的粒子个体组成小生境种群,然后在该种群里面利用粒子群优化算法进化粒子,所有个体经过其小生境群体的进化之后,找到最优的个体存入到下一代的粒子群中,直到找到满意的适应值为止。最后利用Shaffer函数验证了该算法的性能,并且与其他算法进行比较,结果表明该文算法能获得比较好的解,收敛成功率高,并且代价也比较小。

关键词: 粒子群优化, 小生境, 共享机制, 优化

Abstract: The Niche Particle Swarm Optimization(NPSO) is proposed by the Particle Swarm Optimisation(PSO) and Niche technology,and it solves the PSO’s problem of premature convergence and slow convergence in latter half.Niche population is constituted by the particle which has the similar distance,then every particle is evolved by the PSO in each niche population,and the best individual is preserved in next generation.The algorithm is terminated until the satisfactory fitness value is found.The performance of NPSO is validated by Shaffer function.

Key words: Particle Swarm Optimization(PSO), Niche, share Mechanism, ptimization