计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (5): 111-116.DOI: 10.3778/j.issn.1002-8331.1609-0248

• 模式识别与人工智能 • 上一篇    下一篇

基于多策略融合的改进人工蜂群算法

魏锋涛,岳明娟,郑建明   

  1. 西安理工大学 机械与精密仪器工程学院,西安 710048
  • 出版日期:2018-03-01 发布日期:2018-03-13

Improved artificial bee colony algorithm based on multi-strategy fusion

WEI Fengtao, YUE Mingjuan, ZHENG Jianming   

  1. School of Mechanical & Precision Instrument Engineering, Xi’an University of Technology, Xi’an 710048, China
  • Online:2018-03-01 Published:2018-03-13

摘要: 针对标准人工蜂群算法存在易陷入局部最优、收敛速度慢等缺陷,提出一种基于多策略融合的改进人工蜂群算法。为了避免陷入局部最优,引入可调压排序选择策略,以保证种群的多样性;同时,通过跟随蜂阶段将线性调整全局引导策略、自适应动态调整因子策略与标准人工蜂群算法的更新策略组成一个动态调整策略集,通过比较食物源的当前质量值与上次迭代质量值对动态策略进行调整,以加快算法的收敛速度。利用标准测试函数进行实验仿真,结果表明该算法不仅提高了求解精度,而且加快了收敛速度,迭代次数明显减少。

关键词: 人工蜂群算法, 可调压排序选择策略, 动态调整策略集, 函数优化

Abstract: To overcome the defects of convergence speed and the local optimum of artificial bee colony algorithm, this  paper  proposes an improved artificial bee colony algorithm based on multi-strategy fusion. In order to maintain the population diversity and avoid the local optimum, this paper imports the strategy of adjustable voltage ranking selection. To accelerate the convergence rate of artificial bee colony algorithm, a dynamic adjustment strategy set is composed of linear adjustment global guidance strategy, adaptive dynamic adjustment factor strategy and updating strategy of standard artificial swarm algorithm in following bee stage. The policy is dynamically adjusted by comparing the current update value of the food source with the last iterative update value. Through the simulation experiment on a suite of standard functions, the results show that the algorithm has a faster convergence rate and higher solution accuracy, and less number of iterations.

Key words: artificial bee colony algorithm, adjustable voltage selection strategy, dynamic adjustment strategy set, function optimization