基于多代竞争遗传算法的数值函数优化

Function Optimization Problems Based on Multi-Generation Competitive Genetic Algorithm

目的对标准遗传算法的过早收敛问题进行改进。方法保持种群的多样性,将上几代个体中的一部分与本代共同参与竞争,提出多代竞争遗传算法,通过理论分析和对数值函数优化证明该算法的有效性。结果推导出多代竞争遗传算法的模式定理,经验证明显优于标准遗传算法。结论多代竞争遗传算法有利于保持种群的多样性,避免了过早收敛。

Objective To improve the prematurity of standard Genetic Algorithm. Methods In order to avoid prematurity, we kept the diversity of population, that was, some schemas of previous generation compete with schemas of present generation. Results In this paper, we present multi-generation competitive genetic algorithm and the schema theorem which were tested by theory analysis and applied practice. We took the De Jong testing functions to evaluate the performance of the multi-generation competitive genetic algorithm. Conclusion Multi-generation competitive genetic algorithm can avoid prematurity by keeping the diversity of population.