多个体参与交叉的遗传算法

Genetic Algorithm with Multiple Individuals Crossover

提出了多个体参与交叉的遗传算法，即采取新的交叉算子使子代个体同时含有多个父代个体的模式．突破了以前遗传算法只有两个个体参与交叉的局限，通过调整参与交叉的父代个体数目和交叉后产生的后代个体数目，实际上提出了遗传算法调试中的两个新参数．通过调整新参数，使得遗传算法可能有更高的计算效率．证明了多个体参与交叉的遗传算法的模式定理．将方差与熵作为描述遗传算法解群多样性的工具．分析了多个体参与交叉的遗传算法对解群方差及熵的影响．

A new genetic algorithm was presented, which proceeds crossover with multiple individuals and permits that an individual in the next generation possesses schemata from multiple different individuals of this generation. In fact, the crossover was presented as two new parameters of genetic algorithm. The schema theorem was proved for the genetic algorithm with multiple individuals crossover. Variance and entropy were proposed as the measures of diversity of population in genetic algorithm. The influence which the genetic algorithm with multiple individuals crossover act upon the variance and entropy was analyzed. The example proves that this genetic algorithm is feasible and efficient.