摘要
针对标准遗传算法和二倍体遗传算法容易陷入早期收敛、末期局部搜索能力差等不足之处,给出了一种忽略等位基因显隐性的二倍体遗传算法的改进方法;模仿二倍体生物繁殖的过程,引入同源染色体交叉、配子重组操作,改进了传统遗传算法的遗传操作过程;在选择过程中采用了结合最优保留的受限选择策略及精英种群方案。仿真结果表明,该改进算法不但能使种群基因保持多样性,有效抑制了算法的早熟收敛,还降低了算法复杂度、提高了搜索精度,使算法能以较快的速度与较高的精度达到全局最优。
A kind of improved method of the diploid genetic algorithm without considering the dominant-recessive of the allele was given direct at the disadvantages of the SGA and diplont genetic algorithm which are easy to fall into premature convergence and have low efficiency in late period local searching.The genetic operation process was improved by imitating the reproductive processes of diplont and introducing the process of gamete reorganization and the chiasma of Homologous chromosomes.The elitist strategies and constrained selection were used in the selection process.Simulation results show that the improved algorithm not only keeps the diversity of population gene,represses the premature convergence effectively,but also reduces the algorithm complexity,improves the precision,which makes the algorithm achieve global convergence with faster speed and higher precision.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2012年第4期816-820,共5页
Journal of System Simulation
基金
黑龙江省博士后科研启动基金项目(LBH-Q08159)
高等学校青年学术骨干支持计划项目(1152G001)
关键词
遗传算法
双倍体/二倍体
受限选择
智能计算
genetic algorithm
diploid
constrained selection
intelligence algorithms
