试验设计在遗传算法参数配置中的应用

An Application of Experimental Design in Genetic Algorithm

遗传算法是受生物进化论的启发而产生的,其是一种高效的随机搜索与优化算法,具有很好的收敛性,简单通用。试验设计是使用频率最高的统计方法之一,其中正交设计是利用标准化正交设计表进行试验而迅速找到优化方案。均匀设计是考虑让试验点在试验范围尽量均匀地分布,因此试验次数少。本文将这两种试验设计方法应用在遗传算法的参数优化配置中,模拟结果表明采用设计方案后可以提高遗传算法的收敛速度及寻优性能。

Genetic algorithms are widely used, efficient random search methods for the evolution of Darwin's theory. Genetic algorithms have good convergence and used universally. Experimental design is one of the highest frequency of statistical methods, including a lot of experimental design. Orthogonal design uses the standardized orthogonal experiment to quickly find optimization. Uniform designs consider experimental points distributed as uniformly as possible in the test range, having less number of runs. In this paper, the two methods are applied to optimize the parameters of genetic algorithm. Simulation results show the convergence of genetic algorithms is improved..