两阶段遗传算法的结构及性能分析-河北科技大学学报.pdfVIP

28 1 Vol. 28, No . 1 2007 3 Journal of Hebei University of Science and T ec nology Mar . 2007 : 1008-1542( 2007) 01-0044- 05 1 1 1 2 刘立民, 靳晨霞 , 杨丽芸, 李法朝 ( 1. 河北科技大学理学院, 河北石家庄 050018; 2. 河北科技大学经济管理学院, 河北石家庄 050018 ) : 针对基本遗传算法( 简称BGA) 常常存在局部收敛 及收敛解精度不高等方面的不足, 提 出了一种改进的算法) ) ) 两阶段遗传算法, 给出了算法的结构及具体的实施策略, 进而利用M ark- ov 链理论和仿真技术分析了该算法的收敛性能, 结果表明该算法具有操作简单鲁棒性强等特点, 不仅可 有效地避免寻优过程中的/ 早熟0 现象, 而且在很大程度上能提高最优解精度, 适合于大规 模高精度的优化问题 : 基本遗传算法; 两阶段遗传算法; 收敛性; Markov 链 : T P301. 6 : A Structure of tw o-stage genetic algorit m and its performance analysis 1 1 1 2 LIU L-i min , JIN C en- xia , YANG L-i yun , LI Fa-c ao ( 1. College of Sciences, H ebei University of Science and Tec nology , S ijiaz uang H ebei 050018, C ina; 2. College of Econo- mics and Management , H ebei Universit y of Science and Tec nology, S ijiaz uang Hebei 050018, C ina) Abstract: In view of poor convergence and partially converge of basic genetic algorit m ( BGA) , an new improved algorit m, two-st age genetic algorit m is proposed. T e algorit m structure and its implementat ion strat egies are demonstrated. T en, its convergence is analysed by using Markov c ain t eory and simulation tec nology. All results indicate t at t is new improved al- gorit m w ill elp avoid premature p enomenon, and im prove t e precision of t e optimal solution. It is applicable to large scale optimization problems t at demand ig accuracy . Key words: basic genetic algorit m; tw o-stage genetic algorit m; convergence; Markov c ain [ 1~ 3] , Holland , , , , : 1 , ,