基于自适应变异算子的实数编码遗传算法  

An improved real-coded genetic algorithm based on DNM

作　　者：王剑楠 崔英花[1] WANG Jiannan;CUI Yinghua(School of Information and Communication Engineering,Beijing Information Science&Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学信息与通信工程学院,北京100192

出　　处：《北京信息科技大学学报（自然科学版）》2021年第2期46-51,共6页Journal of Beijing Information Science and Technology University

基　　金：国家自然科学基金资助项目(61340005);北京市自然科学基金面上项目(4202024)。

摘　　要：实数编码遗传算法易在局部极值处收敛,搜索精度难以提高。针对变异操作,提出了一种自适应的基于种群多样性的非均匀变异算子(diversity based non-uniform mutation operator,DNM)。引入种群方差因子和种群熵因子为决策变量,设计了一种基于种群多样性的变异强度控制函数,以保持种群的多样性。测试结果表明,应用改进变异算子的算法能够维持一定的种群多样性,提高遗传算法的求解精度及收敛速度。To solve the problem that the real coded genetic algorithm is easy to converge at the local extremum and difficult to improve the search accuracy,an adaptive mutation operator named diversity based non-uniform mutation operator(DNM)is proposed for mutation operation.This operator introduces population variance factor and population entropy factor as decision variables and designs a variation intensity control function to maintain population diversity.The experimental results show that the algorithm with the improved mutation operator can maintain certain population diversity and improve the solution accuracy and convergence speed of the genetic algorithm.

关 键 词：实数编码 遗传算法 变异算子 变异强度 种群多样性 

分 类 号：TN301.6[电子电信—物理电子学]