摘要
在遗传算法中约束条件贯穿于遗传运算的始终,这样必定影响运算效率。因为随着进化过程的进行,适应度较低的一些个体逐渐被淘汰,而适应度较高的个体越来越多,且都集中在最优点附近。基于遗传算法这种优胜劣汰的进化思想,该文提出一种改进的遗传算法——自适应遗传算法。其主要思想是在群体进化若干代后,将弱解空间删除,在以后的进化进程中以同样的群体大小只在强解空间进行群体的繁殖,则可加大强解空间的个体密度,提高解的精度,这样有助于性能优良的个体的产生,并且有可能缩短群体进化过程。将这种自适应遗传算法用于复杂函数的优化,算例结果表明该方法是有效和可靠的。
Restrictive conditions run through the whale course of genetic algorithm, and this way will surely confine the effectiveness of the progress. Individuals with lower fitness value will be eliminated gradually while individuals with higher fitness value will increase with the evolution process going, and all the individuals will be near the merit in the genetic algorithm. This paper proposes an improved genetic algorithm——adaptive genetic algorithm based on the thought of selecting the superior and eliminating the inferior. The main idea is to delete the weak solution space, have only the powerful solution space in which individuals of the same number only breed, thus increasing the density of individuals, improving the precision of the problem, helping breed good individuals, shortening possibly the whale progress. The results got by applying this method to some optimization of complicated functions in the paper show its validity and reality.
出处
《计算机仿真》
CSCD
2006年第1期172-175,225,共5页
Computer Simulation
基金
水利部2002年"948"科技创新计划资助课题
关键词
自适应方法
遗传算法
强解空间
弱解空间
空间收缩
Adaptive
Genetic algorithms
Powerful solution space
Weak solution space
Contracting space