计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (35): 38-40.

• 研究、探讨 • 上一篇    下一篇

混沌动态种群数粒子群优化算法

张 寅,曹德欣   

  1. 中国矿业大学 理学院,江苏 徐州 221008
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-12-11 发布日期:2011-12-11

Chaotic dynamic population size particle swarm optimization

ZHANG Yin,CAO Dexin   

  1. School of Science,China University of Mining and Technology,Xuzhou,Jiangsu 221008,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-12-11 Published:2011-12-11

摘要: 针对粒子群优化算法在整个迭代过程中粒子极易陷于局部极值区域,提出一种混沌动态粒子数的粒子群优化算法,也即在判定全局最优值处于停滞时,以混沌策略对粒子进行位置初始化后加入种群,从而有效地保证了粒子群的多样性。用4个测试函数验证了该算法具有很好的寻优能力和较高的搜索精度。

关键词: 粒子群优化算法, 全局最优值, 混沌, 种群数

Abstract: For the particle swarm optimization,the particles are easily trapped in the local extremum region in the whole iterative process.This paper proposes a particle swarm optimization based on chaotic dynamic population size.When the global optimum is judged in stagnation,the location of particles are initialized using the chaotic strategy,then be added to the population,thus can effectively ensure the diversity of the particle swarm.Four test functions verify that the algorithm has very good optimizing ability and high precision of the search.

Key words: particle swarm optimization, global extremum, chaos, population size