计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (29): 55-59.DOI: 10.3778/j.issn.1002-8331.2009.29.016

• 研究、探讨 • 上一篇    下一篇

函数全局优化的改进实数遗传算法

金 芬1,陈小平2   

  1. 1.苏州市职业大学 机电工程系,江苏 苏州 215104
    2.苏州大学 电子信息学院,江苏 苏州 215021
  • 收稿日期:2008-12-30 修回日期:2009-03-16 出版日期:2009-10-11 发布日期:2009-10-11
  • 通讯作者: 金 芬

Modified real-coding genetic algorithm for function global optimization

JIN Fen1,CHEN Xiao-ping2   

  1. 1.Department of Mechano-electronics Engineering,Suzhou Vocational University,Suzhou,Jiangsu 215104,China
    2.School of Electronics and Information Engineering,Suzhou University,Suzhou,Jiangsu 215021,China
  • Received:2008-12-30 Revised:2009-03-16 Online:2009-10-11 Published:2009-10-11
  • Contact: JIN Fen

摘要: 针对有界区域复杂函数的全局优化问题,分析了一般实数遗传算法的不足,提出了一种新的改进实数遗传算法。在改进算法中,个体的适应度值直接按其目标值排序的方法获得,这可避免进化后期陷入局部极值;基于适应度的线性逼近交叉策略,随机遍历抽样选择、最优保存和子代淘汰父代选择结合的混合选择策略及变异概率动态变化的实值变异策略,可使算法以较快的速度收敛于最优值。对12个典型的复杂函数进行优化仿真,结果表明改进算法不仅收敛速度快,鲁棒性好,而且能得到较高的优化精度。

关键词: 遗传算法, 全局优化, 适应度, 混合选择, 交叉策略

Abstract: A modified real-coding genetic algorithm is brought out aiming at the deficiency of general real-coding genetic algorithm in solving complex function global optimization problem on a bounded area.In the modified algorithm,the fitness values of the individuals are gotten directly from the objective values’ orders to refrain from falling into local optimum at the late stage of evolution.The algorithm uses a crossover strategy with a linear approximation concept according to function fitness,a hybrid selection strategy which combines selection of stochastic universal sampling with the elitist method and the selection of offspring replacing the worst individuals of parents,and a real-valued mutation strategy of the changing mutation probability.The modification can converge to the global optimum more quickly.The simulation results of twelve typical complex functions show that the modified algorithm can not only converge fast and have a strong robustness,but also get more precise results.

Key words: genetic algorithm, global optimization, fitness, hybrid selection, crossover strategy

中图分类号: