基于Tent映射的自适应混沌混合多目标遗传算法

Adaptive chaos hybrid multi-objective genetic algorithm based on the Tent map

提出一种Tent映射在计算机上实现的改进算法,有效解决了受计算机字长影响,Tent映射存在不动点和小周期的情况.将改进的Tent映射应用于混沌优化算法中,对基本NSGA-Ⅱ算法进行改进.使用混沌序列对初始种群赋值,提高算法收敛能力;使用改进的混沌搜索增强种群多样性;分别使用基本算法和改进算法对标准测试函数进行数值仿真.统计结果显示:改进的算法可以在保持高效率求解的同时,得到的非劣解在收敛性和多样性指标上均优于基本NSGA-Ⅱ算法.

A modified algorithm which realized the tent map on the computer was proposed to deal effectively with the problem of fixed point and the small periodic of the tent map, which affected by the finite wordlength of computer. The chaos research algorithm based on the modified Tent map was introduced to the multiobjective genetic algorithm. Firstly, the chaos sequence was applied to assign the initial population value to enhance the convergence ability. Then, the chaos search optimization algorithm was adopted to improve the di- versity of the population. The benchmark problems were tested by the basic and the modified algorithm respectively. The statistical results show that the improved algorithm could seek the solution more efficiently, and a- chieve better convergence and diversity than basic NSGA-II.