摘 要:为了解决粘性二进制粒子群算法在优化过程中易陷入局部最优、全局搜索能力弱、后期收敛性能差的弊端,提出了一种非线性因子的粘性二进制粒子群算法(NFSBPSO)。NFSBPSO 算法采用非线性递减策略优化粘性权重,平衡全局与局部探索能力;同时,利用混沌策略对种群进行初始化,在每个迭代过程中对最差和最优粒子进行优化,以提高种群质量,扩大种群的全局探索能力。将新算法与 3 个对比算法在基准函数上进行测试,实验表明新的算法能较好地跳出局部最优,提高了算法的稳定性和收敛能力。
关键词:非线性因子;混沌策略;粒子优化;粘性二进制粒子群算法
DOI:10.19850/j.cnki.2096-4706.2023.03.014
中图分类号:TP301 文献标识码:A 文章编号:2096-4706(2023)03-0061-05
A Nonlinear Factor Sticky Binary Particle Swarm Optimization
CHENG Qianqian
(Taiyuan Normal University, Jinzhong 030619, China)
Abstract: In order to solve the disadvantages of sticky binary particle swarm optimization, which is easy to fall into local optimum, weak global search ability and poor late convergence performance in the process of optimization, a nonlinear factor sticky binary particle swarm optimization (NFSBPSO) is proposed. NFSBPSO algorithm uses nonlinear decreasing strategy to optimize the sticky weight and balance the global and local exploration ability. At the same time, the chaos strategy is used to initialize the population, and the worst and best particles are optimized in each iteration process to improve the quality of the population and expand the global exploration ability of the population. The new algorithm is tested on the benchmark function with three comparison algorithms. The experimental results show that the new algorithm can jump out of the local optimum better and improve the stability and convergence ability of the algorithm.
Keywords: nonlinear factor; chaos strategy; particle optimization; sticky binary particle swarm optimization
参考文献:
[1] KENNEDY J,EBERHART R. Particle swarm optimization [C]// IEEE International Conference on Neural Networks,Perth:IEEE, 1995:1942-1948.
[2] KENNEDY J,EBERHART R C. A discrete binary version of the particle swarm algorithm [C]//International conference on Systems. Orlando:IEEE,1997:4104-4108.
[3] 张鹏威 . 采用正弦映射与扩张算子的二进制粒子群优化算法 [J]. 小型微型计算机系统,2019,40(6):1160-1164.
[4] 赵远东,方正华.带有权重函数学习因子的粒子群算法 [J].计算机应用,2013,33(8):2265-2268.
[5] NGUYEN B H,XUE B,ANDREAE P,et al. A new binary particle swarm optimization approach:momentum and dynamic balance between exploration and exploitation [J].IEEE Transations on Cyberntics,2021,51(2):589-603.
[6] 姜磊,刘建华,张冬阳,等 . 一种自适应变异二进制粒子群算法 [J]. 福建工程学院学报,2020,18(3):273-279.
[7] PAN A Q ,WANG L,GUO W A. A diversity enhanced multiobjective particle swarm optimization [J].Information Sciences: An International Journal,2018,436-437:441+465.
[8] 徐浩天,季伟东,孙小晴,等 . 基于正态分布衰减惯性权重的粒子群优化算法 [J]. 深圳大学学报:理工版,2020,37(2):208-213.
[9] 李真,王帆,王冉珺 . 一种结合灰狼算法的粒子群优化算法 [J]. 计算机测量与控制,2021,29(10):217-222.
[10] 李浩君,张广,王万良 . 一种惯性权重与种群多样性协同调整的二进制粒子群优化算法 [J]. 小型微型计算机系统,2018,39(3):529-533.
[11] 孙一凡,张纪会 . 基于模拟退火机制的自适应粘性粒子群算法 [J]. 控制与决策,1-10[2022-09-03].https://kns.cnki.net/ kcms/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST& filename=KZYC20220512001&uniplatform=NZKPT&v=SFfl7Z nyzm7G6kYM3mYnqTigMmGf3EFhaf0-oqT7JD8EXhm3pVm_ PJiZON1c2qsz.
[12] 王皓,欧阳海滨,高立群 . 一种改进的全局粒子群优化算法 [J]. 控制与决策,2016,31(7):1161-1168.
[13] 邱飞岳,王京京 . 自适应学习因子的混沌二进制粒子群优化算法 [J]. 浙江工业大学学报,2020,48(4):411-417.
[14] 王越,邱飞岳,郭海东 . 一种带变异算子的自适应惯性权重二进制粒子群优化算法 [J].小型微型计算机系统,2019,40(4):733-737.
[15] 李浩君,吴嘉铭,戴海容 . 基于多维关联本体的学习资源推荐方法 [J]. 浙江工业大学学报,2021,49(4):374-383.
作者简介:程倩倩(1997—),女,汉族,山西运城人,硕士研究生在读,研究方向:教育软件开发技术。