摘要
为了解决传统花授粉算法(FPA)收敛速度慢、易陷入局部最优、寻优精度低等缺陷,提出了一种t-分布扰动策略和变异策略的花授粉算法(t MFPA).首先利用混沌映射初始化花朵个体的位置,然后在全局授粉过程中,利用t-分布扰动的随机个体和莱维飞行共同实现个体位置更新,加快收敛速度的同时提高搜索空间的多样性;在局部授粉过程中,加入具有两个差分向量的变异策略和小概率策略,结合两种策略使算法能够跳出局部最优.实验结果表明,t MFPA相比于FPA和其他启发式智能算法具有更好的寻优精度和收敛速度,相对于其他改进算法具有更好的收敛性能.
In order to solve the shortcomings of flower pollination algorithm(FPA)such as slow convergence,easy to fall into local optimum,and low optimization accuracy,a flower pollination algorithm based on t-distribution perturbation strategy and mutation strategy(t MFPA)was proposed.Firstly,chaotic mapping was used to initialize the position of individual flowers.Then in the global pollination process,random individuals with t-distribution perturbations and Levy flight were used to jointly update the individual position,which speeded up the convergence rate and increased the diversity of search space;during the local pollination process,the mutation strategy with two difference vectors and the small probability strategy were added to make the algorithm jump out of local optimization.The experimental results show that tMFPA has better optimization accuracy and convergence speed than FPA and other heuristic intelligent algorithms,and has better convergence performance than other improved FPA algorithms.
作者
宁杰琼
何庆
NING Jie-qiong;HE Qing(College of Big Data&Information Engineering,Guizhou University,Guiyang 550025,China;Guizhou Provincial Key Laboratory of Public Big Data,Guizhou University,Guiyang 550025,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2021年第1期64-70,共7页
Journal of Chinese Computer Systems
基金
贵州省科技计划项目重大专项项目(黔科合重大专项字[2018]3002,黔科合重大专项字[2016]3022)资助
贵州省公共大数据重点实验室开放课题项目(2017BDKFJJ004)资助
贵州省教育厅青年科技人才成长项目(黔科合KY字[2016]124)资助。
关键词
花授粉算法
t-分布扰动策略
变异策略
flower pollination algorithm
t-distribution perturbation
mutation strategy
