具有重组学习和混合变异的动态多种群粒子群优化算法
作者:
作者单位:

1.武汉科技大学汽车与交通工程学院;2.武汉理工大学 物流工程学院

作者简介:

通讯作者:

中图分类号:

TP18

基金项目:

面向多移动智能物流资源调度的动态渐进群集智能优化方法研究,国家自然科学基金,No.61603280;三峡枢纽航运突发事件船舶交通应急调控方法研究,国家自然科学基金,No.71874132


Dynamic Multi-population Particle Swarm Optimization Algorithm with Recombined Learning and Hybrid Mutation
Author:
Affiliation:

1.School of Automobile and Traffic Engineering, Wuhan University of Science and Technology;2.School of Logistics Engineering,Wuhan University of Technology

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为解决粒子群优化算法中种群多样性和收敛性间的矛盾,提出了一种具有重组学习和混合变异的动态多种群粒子群优化算法。该算法动态划分多种群并融入重构粒子作为引导因子,在增加种群多样性的同时保留优秀粒子的空间信息;在算法执行阶段对最优个体施加混合变异,基于时变概率实施反向学习策略或者邻域扰动操作,帮助粒子快速跳出局部困境,加强对附近区域内的精细搜索。基于14个多类型标准测试函数,并与其他的改进粒子群算法进行对比,验证了几种改进措施的有效性和叠加影响;其次,为进一步探究概率性混合变异策略的敏感性,对变异方式及参数设置进行仿真实验,结果表明所采用的极值扰动策略具有显著的优势,合理地控制学习强度可以充分发挥反向学习的作用,并给出了影响参数的建议取值范围。上述实验结果表明本文所提出的算法能够更好地平衡种群的开发与勘探的能力,提高求解精度和收敛性能。

    Abstract:

    To solve the contradiction between population diversity and convergence in Particle Swarm Optimization, an improved particle swarm optimization which called dynamic multi-population particle swarm optimization algorithm with recombined learning and hybrid mutation was proposed. In the proposed algorithm, a population was divided dynamically and a new particle was reconstructed as a guiding factor. It retained the spatial information of the excellent particles while increasing population diversity. During the execution of the algorithm, a hybrid mutation strategy was applied to adjust the optimal solution. A opposition-based learning and a neighborhood-disturbance operations were implemented based on a time-varying probability. It helped the particles jump out of the local dilemma quickly, and strengthened good searching in the nearby areas. The effectiveness and superposition effects of several proposed improvement operations compared with several improved particle swarm algorithms based on a set of 14 multi-type benchmark functions were verified. In order to further explore the sensitivity of probability-based hybrid mutation strategy, a large number of simulation experiments were carried out to analyze the mutation mode and parameter settings. The results showed that the disturbed extreme strategy had significant advantages. Controlling the learning intensity reasonably can make the opposition-based learning show better performances, furthermore, a suggersted value range was given. Finally, experimental results indicated that the proposed algorithm can get a better balance betweetn the exploitation and exploration for the swarm searching and improve the solution accuracy and convergence performance.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-07-05
  • 最后修改日期:2020-09-23
  • 录用日期:2020-09-25
  • 在线发布日期:
  • 出版日期: