一种基于等位基因原理的NSGA-Ⅱ算法

NSGA-Ⅱ Algorithm Based on Allele Principle

NSGA-Ⅱ算法是通过模拟达尔文进化原理,采用选择、交叉和变异等操作算子,对个体种群进行繁殖和进化,以实现多目标优化。在NSGA-Ⅱ算法进化过程中,变异或交叉操作应用于个体的所有基因。但在真实的自然变异过程中,生物体根据等位基因频率的原理,不会整个基因发生变异,而只有少数基因会发生变异。同时,在交叉操作中该生物体仍有变异的机会。为了完全模拟自然变异过程,笔者提出了一种基于等位基因原理的NSGA-Ⅱ算法,该算法允许个体在变异过程中保持某些基因不变,在交叉操作中却仍有变异的机会。实验结果表明,与其他多目标进化算法相比,该算法显著提高了搜索性能,且具有收敛性强和不易陷入局部极小的特点。

NSGA-Ⅱ algorithm simulates Darwinian evolution principle and uses selection,crossover and mutation operators to evolve individual population in order to solve multi-objective optimization problems.In the evolution process of NSGA-Ⅱ algorithm,all genes of individuals are applied to mutation and crossover operation.But in the real natural mutation process,according to the principle of allele frequency,only a few genes will mutate,not the whole gene.At the same time,in the cross operation,the organism still has the chance of mutation.In order to fully simulate the natural mutation process,a new NSGA-Ⅱ algorithm based on allele principle is proposed.The algorithm allows individuals to keep some of their genes unchanged in the process of mutation,and there is still a chance of mutation in the crossover operation.The experimental results show that compared with other multi-objective evolutionary algorithms,the proposed algorithm can significantly improve the search performance,and has strong convergence and is not easy to fall into local minima.