[1]张金萍,等.一种动态种群不对称交叉的新型遗传算法[J].南京理工大学学报(自然科学版),2007,(04):444-448.
 ZHANG Jin-ping,LIU Jie,LI Yun-gong.Novel Dynamic Population and Anisomerous Crossover Genetic Algorithm[J].Journal of Nanjing University of Science and Technology,2007,(04):444-448.
点击复制

一种动态种群不对称交叉的新型遗传算法
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
期数:
2007年04期
页码:
444-448
栏目:
出版日期:
2007-08-30

文章信息/Info

Title:
Novel Dynamic Population and Anisomerous Crossover Genetic Algorithm
作者:
张金萍1 2 刘 杰1 李允公1
1. 东北大学机械工程与自动化学院, 辽宁沈阳110004; 2. 沈阳化工学院机械工程学院, 辽宁沈阳110142
Author(s):
ZHANG Jin-ping12LIU Jie1LI Yun-gong1
1.School of Mechanical Engineering and Automation,Northeastern University,Shengyang 110004,China;2.Department of Mechanical Engineering,Shenyang Institute of Chemical Technology,Shenyang 110142,China
关键词:
遗传算法 实数编码 交叉操作 寻优 收敛
Keywords:
genet ic algo rithm real coding crossover opt im ization convergence
分类号:
TP18
摘要:
在分析实数编码遗传算法各操作步骤的实质和不足的基础上,提出了以提高算法柔性为目的、以动态种群和不对称交叉为主要特点的新型遗传算法。在遗传寻优的每一代中,父辈个体的繁殖次数在限定的范围内随机波动,种群规模随之动态变化,依据生态平衡的原理,通过选择和复制将新一代种群规模限定于某一波动均值处。为提高新生个体的多样性及其在参数空间中的遍布性,提出并设计了不对称交叉的具体方法。针对新型算法,提出了双重选择的选择方法。经典型算例验证,所提算法具有收敛快、成功率高、抗早熟能力强的显著特点。
Abstract:
The shortcom ings and the essent ia ls o f genetic algorithm are discussed. A novel genetic a lgorithm is proposed whose popu lat ion is dynam ic and whose crossover is anisomerous. In every propagate process, the propagate tmi es of parents random fluctuates in lmi itat ive range, and the size of population is dynam ic. A ccord ing to the principle of balance nature, the size of population is kept at the average of w ave by the crossover and se lection. In order to mi prove the variety and universa lity of the new populat ion in the parameter space, themethod of anisomerous crossover is designed. Based on the new a lgorithm, the method of double select ion is presented. Calculating some typica l instants and comparingw ith other genetic algorithms, the results show that the convergence rate, success rate and ant-i premature of the nove l genetic algorthm are superior.

参考文献/References:

[ 1] 李纯莲, 王希诚, 赵金城. 基于浮点数编码的信息 熵控制多种群遗传算法[ J]. 南京理工大学学报, 2004, 28( 5): 453- 456.
[ 2] SahabM G, Ashour A F, To ropov V V. A hybrid genetic algor ithm fo r re inforced concrete flat slab bu ildings [ J]. Computers and S tructures, 2005, 83 ( 8 - 9): 551- 559.
[ 3] Tam SM, Cheung K C. Genetic a lgo rithm based defect identifica tion sy stem [ J]. Expert Sy stem s w ith Applications, 2000, 18 ( 1): 17- 25.
[ 4] Re id D J. Genetic a lgor ithms in constrained optim ization [ J ]. M a them atical and Computer M odelling, 1996, 23 ( 5): 87- 111.
[ 5] 任子武, 伞治. 实数遗传算法的改进及性能研究 [ J] . 电子学报, 2007, 35( 2): 269- 274.
[ 6] Fe rentinos K P, A lbr ight L D. Optim a l design of plant lighting system by gene tic algor ithm s [ J]. Eng ineer ing Applications o f Artificial Inte lligence, 2005, 18 ( 4): 473- 484.
[ 7] W ang H siaofan, W u Kuangyao. H ybr id genetic a lgor ithm for optim ization prob lem s w ith pe rmuta tion property [ J]. Compu ters and Operations Research, 2004, 31 ( 14): 2453- 2471.
[ 8] R iechm ann T. Gene tic a lgor ithm learn ing and evo lutionary gam es [ J]. Journa l o f Econom icDynam ics and Contro ,l 2001, 25 ( 6- 7): 1 019- 1 037.
[ 9] H ansen J V. Genetic search m ethods in air tra ffic contro l [ J] . Computers and Operations Research, 2004, 31 ( 3): 445- 459.
[ 10] 董颖, 刘欢杰, 许宝栋, 等. 一种基于实数编码的改 进遗传算法[ J]. 东北大学学报, 2005, 26 ( 4): 219 - 221.

备注/Memo

备注/Memo:
基金项目: 辽宁省教育厅高等学校科学研究项目( 2004D089)
作者简介: 张金萍( 1977- ) , 女, 河南尉氏人, 博士生, 主要研究方向: 机电一体化、机器视觉方向, E-m a il: jinping7707_cn@ sina. com;
通讯作者: 刘杰( 1944 - ), 男, 教授, 博士生导师, 主要研究方向: 振动利用与 控制、机电一体化, E-m a il: JIELIU@ me. neu. edu. cn。
更新日期/Last Update: 2007-08-30