混沌动态种群数粒子群优化算法  

Chaotic dynamic population size particle swarm optimization

作　　者：张寅[1] 曹德欣[1] 

机构地区：[1]中国矿业大学理学院,江苏徐州221008

出　　处：《计算机工程与应用》2011年第35期38-40,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(No.70901073);中央高校基本科研专项基金项目(No.2010LKSX06)~~

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。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 popu- lation, 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.

关 键 词：粒子群优化算法 全局最优值 混沌 种群数 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]