摘要
先叙述了遗传算法的基本原理和操作流程,并结合实际问题具体分析其特点,针对遗传算法易早熟、局部搜索能力弱以及易受参数影响等缺陷,从适应值函数、变异概率和自适应调节三个方面对算法进行了改进研究,并通过实例进行仿真验证,结果表明改进的算法有较好的寻优性能。参数间的不同组合将影响算法的寻优性能,尤其是收敛性,研究交叉概率和变异概率之间的交互作用,得出交叉概率和变异概率的建议取值区间。
First describes the basic idea of genetic algorithm, mathematics principle and operation process, and analysis their char- acteristic combine with the actual problem. In, view of the weak of local searching ability of the genetic algorithm is easy to pre- mature and wait for blemish. From the fitness function , mutation probability and adaptive control algorithm is improved in the aspect of research, and through the example to simulation, the results show that the improved algorithm has better optimization performance, parameters between different combination would have a great influence on the performance of genetic algorithm, particularly affect the convergence of the algorithm, by studying the interaction between the crossover probability and mutation probability, and concluded that the advice of the crossover probability and mutation probability values range.
作者
刘帅
彭业飞
冯智鑫
张维继
LIU Shuai , PENG Ye-fei, FENG Zhi-xin, ZHANG Wei-ji （Naval University of Engineering, Wuhan 430033, China）
出处
《电脑知识与技术》
2016年第2期182-185,共4页
Computer Knowledge and Technology
关键词
遗传算法
早熟
局部搜索
交叉概率
变异概率
Genetic Algorithm
precocious
local searching
crossover probability
mutation probability
