自适应调整控制参数的差异演化算法.pdfVIP

2012 年 3 月 第 33 卷第 3 期 计算机工程与设计 COMPUTER ENGINEERING AND DF~IGN 自适应调整控制参数的差异演化算法 武志峰1 ，黄厚宽2 (1.天津职业技术师范大学信息技术工程学院，天津 300222; 2. 北京交通大学计算机与信息技术学院，北京 100044) Mar.2012 Vol. 33 No.3 摘 要 z 控制参数选取走包括差异演化在内的演化算法设计时所面临的一个重要问题，对算法的性能有着重大影响。针对 差异演化算法参数选取问题，提出一种利用个体适应度作为参数调整决策依据，并结合一定的调整概率对 F 和 CR进行自 适应调整的方法，解决了手工设置控制参数的不便。同时利用交叉操作生成双子代个体与父代个体竞争形成新一代种群， 加快了算法的收敛。对标准测试函数的仿真实验结果表明，该算法无论在最优解质量和收敛速度上都优于相关算法，尤其 对于高维函数而言。 关键词 g 自适应参数控制;差异溃化算法p 双子代竞争; ~寅化计算 z 函数优化 中图法分类号: TP18 文献标识号:A 文章编号: 1000-7024 (2012) 03-1175-06 Self-adapting control parameters in differential evolution WU Zhi-feng 1 , HUANG Hou-kuan 2 (1. Sch∞I of lnformation Technology and Engineering , Tianjin University of Technology and Education , Tianjin 300222 , China; 2. Sch∞I of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044 , China) Abstract: Parameters setting is an important problem of evolution algorithms, including differential evolution algorithm. It has an effect on the performance of evolution algorithms. Ac∞rding to the problem of parameters of differential evolution, a method is presented , which uses self-adaptive as a scientific evidence to adjust parameters and set F and CR combined with modulated prohability. An algorithm is presented, which depends on the fitness of individual and modulated probability set the parameters F 姐d CR automatically. This method can get the optimal ∞ntrol parameters for different optimization problem without user interaction. Mor，ωver ， two trial vecωrs are created by re∞mbination for increased ∞lony diversity and avoided premat盯e ∞n- vergence. These vectors ∞mpete with the p町ent individual to be the next generation. Experimental results indicate that the pro- posed algorithm is efficient and f，回到hle. It is superior to other related methods such as DE, jDE , FADE, MPDE , DDE both on the quality of solution and on the ∞nvergence rate , especially for high dimension functions. Keyw，创由: seH-adaptive parameter control; differential