摘要
为克服标准遗传算法(SGA)搜索效率低、收敛速度慢等缺陷,文章提出了一种免疫遗传算法(IGA),即在父代优秀个体群的基础上叠加一个服从正态分布的随机变量来产生子代个体,以此综合体现父代优秀个体的遗传性和免疫性。研究表明,IGA对SGA的改进是有效且可行的,显示出稳健的全局优化、计算量少和求解精度高等特点,具有较高的应用价值。
In order to improve the poor efficiency and stability of the simple genetic algorithm(SGA),an immune genetic algorithm(IGA) is studied in this paper. A random variable with normal distribution is added into the excellent forerunner individuals to produce the new generation so as to incarnate both transmissibility and immunity of the forerunner individuals. The results of two examples show that the improvement on the SGA is marked by using the IGA,and the presented IGA has the attributes of steady overall optimization,less calculation and high precision,and it can be applied to solve most of complex optimization problems.
出处
《合肥工业大学学报（自然科学版）》
CAS
CSCD
2004年第7期734-737,共4页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金重大资助项目(50099620)
安徽省自然科学基金资助项目(01045102)
教育部优秀青年教师资助计划([2002]350)
安徽省优秀青年科技基金资助项目(2001)
关键词
遗传算法
免疫算法
正态分布
genetic algorithm
immune algorithm
normal distribution
