计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (15): 181-186.DOI: 10.3778/j.issn.1002-8331.1602-0083

• 模式识别与人工智能 • 上一篇    下一篇

自适应种群更新策略的多目标粒子群算法

翁理国,王  骥,夏  旻,纪壮壮   

  1. 南京信息工程大学 江苏省大气环境与装备技术协同创新中心,南京 210044
  • 出版日期:2017-08-01 发布日期:2017-08-14

Multi-objective particle swarm optimization algorithm using adaptive population updating strategy

WENG Liguo, WANG Ji, XIA Min, JI Zhuangzhuang   

  1. Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing University of Information Science and Technology, Nanjing 210044, China
  • Online:2017-08-01 Published:2017-08-14

摘要: 针对粒子种群较差的局部搜索能力,提出了一种自适应种群更新策略的多目标粒子群算法。该算法在每次种群进行迭代时,根据种群的多样性测度以及每个粒子的适应度值,自适应地改变速度权重,以此来提高种群粒子在局部搜索时的活性,使算法具有较强的局部搜索能力同时又保留了足够的全局搜索能力。最后利用多组经典测试样例进行仿真,并与传统的粒子群算法以及速度线性衰减算法做比较,在单目标优化中,自适应粒子群算法能够更快地寻找最优位置;在多目标优化中,自适应粒子群算法能够更快速地收敛于帕累托最优边界。

关键词: 粒子群优化算法, 搜索能力, 局部最优, 自适应策略

Abstract: In order to solve the poor local search ability of the particle population, a multi-objective particle swarm optimization algorithm based on adaptive population updating strategy is proposed. At each population iteration, according to the population diversity measure and each particle’s fitness value, the algorithm changes the speed weight adaptively to improve the local search activity of particle population, and the algorithm retains enough global search ability. Finally, multiple groups of classical test samples are used for simulation, compared with the traditional particle swarm algorithm and speed linear attenuation algorithm. In single objective optimization, adaptive particle swarm optimization algorithm can find the optimal location faster. In multi-objective optimization, adaptive particle swarm optimization algorithm can convergence to Pareto optimal boundary more quickly.

Key words: particle swarm optimization, search capabilities, local optimal, adaptive strategy