【摘要】 通过分析发现,二进制编码中不同基因位改变时,所引起优化变量的变化程度不相等。基于此,提出一种改进的交叉和变异策略,对编码串中各个基因位赋予不同的交叉点选择概率和变异率,并随进化调整各位的交叉点选择概率和变异率。仿真结果表明,同标准遗传算法相比,采用改进策略的遗传算法具有寻优精度高、稳定性好、收敛性强等优点;此外,同改进交叉操作相比,改进变异操作能更有效地提高算法的寻优能力。更多还原

【Abstract】 It was found that modification on different genomes in binary encoding would result in large variety on solutions in phenotype.According to this conclusion,modified crossover operation and mutation operation were proposed,in which different crossover point selecting probabilities and mutation probabilities were assigned to each genome in encoded string and were adapted in relation to the number of evolution generation.Experiments on five multimodal functions demonstrated that genetic algorithm with modified crossover operation and genetic algorithm with modified mutation operation found the global optimum with higher precision and got stuck at a local optimum fewer times than standard genetic algorithm did and converged perfectly well,and genetic algorithm with modified mutation operation had improved more efficiently its optimum searching ability.更多还原