摘要
提出了一种新的正交遗传算法(OBGA),算法的特点是利用正交数组产生初始种群,它比随机产生的初始种群更均匀分布在解空间中,而且在正交设计的基础上提出了一种新的杂交算子,与高斯变异算子相结合,提高了种群的多样性和算法的局部搜索能力,最后对6个多峰函数进行了测试。数值实验结果表明,新算法正确高效,稳定性好。
A new orthogonal-based genetic algorithm （OBGA） is proposed. Orthogonal array is applied to generate initial population, which is scattered more uniformly over the feasible solution space than randomly generating initial population. In addition, a new crossover operator is proposed and the Gaussian mutation operator is used, which enhance population＇s diversity and algorithm＇s local search ability. The new algorithm solves 6 multimodal functions. The numerical experimental results show that the new algorithm is high-efficiency and high-stabilization.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第13期3413-3415,3418,共4页
Computer Engineering and Design
基金
"十一五"民用航天基金项目(C5220061318)
关键词
正交设计
正交数组
遗传算法
高斯变异
函数优化
orthogonal design
orthogonal array
genetic algorithm
Gaussian mutation
fimction optimization
