一种基于基因置换技术的优化算法及其收敛性  

A function optimization method based on the technique of gene replacement and its convergence performance

作　　者：郑小平[1] 丁信伟[2] 冯恩民[3] 唐碧玉[1] 

机构地区：[1]广西大学化工学院,广西南宁530004 [2]大连理工大学化工学院,辽宁大连116012 [3]大连理工大学应用数学系,辽宁大连116012

出　　处：《广西大学学报（自然科学版）》2011年第6期959-966,共8页Journal of Guangxi University（Natural Science Edition）

基　　金：国家自然科学基金资助项目(AD040108;AE040022)

摘　　要：针对实际工程应用中的定精度优化问题,基于基因置换育种技术提出一种局部优化算法。该算法以二进制编码个体作为育种目标,采用基因置换技术完成目标个体进化而实现精确局部搜索。通过对算法的收敛特性进行分析证明,获得优化计算代价的理论上限和经验估计,建立实现定精度优化的算法参数与计算代价的关系。实验结果表明,该算法能够在预定的计算代价内比较可靠地实现局部定精度优化,对函数性状和变量形式无特殊要求,是一种简单有效的通用函数优化方法。Based on the technique of gene replacement,an optimization algorithm identified as gene replacement breeding algorithm（GRBA） is proposed in this paper to achieve optimization of given accuracy to meet the demand of practical engineering applications.In GRBA,an arbitrary binary encoded individual is taken as the breeding objective,and the optimum is obtained through evolutionary operation by gene replacement.The convergence performance and the computation cost of the algorithm is analyzed and estimated theoretically and experimentally.The results show that the current algorithm can approach accurate local optimization in foreseeable number of function calling without any provisos on the form of function and variables,and can be used as a simple and effective universal function optimization method.

关 键 词：优化 育种进化 基因置换 收敛精度 计算代价 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]