摘要
为解决传统遗传算法在一维多峰函数优化中容易陷入局部极值、收敛概率低、稳定性不理想等问题,提出了一种新型的自适应遗传算法。结合自适应差分进化算法流程,提出了一种基于种群适应度变化程度而变化的非线性交叉算子和变异算子,使算法跳出局部极值,寻找到全局最优解,提升最优值迭代效率。函数测试实验表明,在一维多峰函数优化中,该算法在函数收敛概率、最优值迭代效率以及稳定性上比已有算法均有提高。
Traditional genetic algorithm is easy to fall into the local extremum,its convergence probability and stability in one-dimensional multi-peak function optimization are also low.This paper presents an improved adaptive genetic algorithm.This algorithm uses a nonlinear crossover operator and mutation operator,which are based on the variation degree of population fitness values.Furthermore,the algorithm combines the process of the adaptive differential evolution algorithm.Two modifications can make the algorithm escape from the local extremum,find the global optimal solution and increase the iterative efficiency of the optimal value.Compared with other existing algorithms,the results of experiments show that the improved algorithm has good performance in improving the probability of convergence,iterative efficiency of optimal value and algorithm stability.
作者
张大科
钱谦
ZHANG Da-ke;QIAN Qian(Faculty of Information Engnieering and Automation,Kunming University of Science and Technology;Yunnan Key Laboratory of Computer Technology Applications,Kunming 650500,China)
出处
《软件导刊》
2018年第6期85-87,91,共4页
Software Guide
关键词
遗传算法
自适应
函数优化
变异概率
交叉概率
genetic algorithm
adaptive
function optimization
mutation probability
crossover probability
