Abstract: Genetic Algorithm（GA） is an intelligent algorithm by simulation of biological evolution.It has been widely used to solve various problems.Because the Simple Genetic Algorithm（SGA） produces a new value that is only dependent on the mutation operator，it often obtains a solution without high accuracy.To overcome it，an improved selection operator for GA is proposed in this paper.In the new selection operator，chromosomes with similar fitness are divided into the same group，and the new individuals are generated by Gaussian search which is affected by all old chromosomes of the group.According to the experimental results with 11 benchmark functions，it is proved that with the new selection operator，GA will maintain the diversity and get higher accuracy than SGA does.

Key words: evolutionary computation, genetic algorithm, selection operator, roulette method, Gaussian search

摘要： 遗传算法（Genetic Algorithm，GA）是一种模拟生物进化的智能算法，被广泛应用于求解各类问题。简单遗传算法（Simple GA）仅靠变异产生新的数值，常常存在搜索精确度不高的问题。针对这个问题，对SGA的选择算子进行改进，即把相似个体分在同一组中，以组为单位进行选择，并通过该组个体的特点进行高斯搜索生成新的群体。这样使得GA在搜索过程中不仅可以很好地保持个体的多样性，并且可以提高解的精确度。通过对11个函数（单峰和多峰）的仿真实验，证明了采用新的选择算子后，GA在求解问题的精确度上有了很大地改善。

关键词: 演化计算, 遗传算法, 选择算子, 轮盘赌法, 高斯搜索