摘要： 提出一种改进的遗传算法，为排序选择压力引入自适应调节机制，确保选择压力随种群性状的改变而动态调整，采用新的竞争择优交叉策略提高种群中个体的平均性能。选取典型测试函数进行仿真，结果表明该算法在寻优精度和收敛速度上较原有算法均有较大提高，收敛概率达90%以上。

关键词: 遗传算法, 排序选择, 交叉策略

Abstract: This paper proposes an improved Genetic Algorithm(GA). The ranking selection intensity adopts an adaptive adjusting mechanism, which can adjust the selection intensity dynamically according to the change of the population state. A new crossover strategy which chooses the outstanding individuals according to competition is used to increase the individual average performance of the population. The simulation with the typical test functions indicates that this algorithm can improve the precision of solutions and convergence speed of simple genetic algorithm, and the proportion of convergence can reach more than 90%.

Key words: Genetic Algorithm(GA), ranking selection, crossover strategy

中图分类号: 

• TP18