电子学报 ›› 2018, Vol. 46 ›› Issue (2): 308-314.DOI: 10.3969/j.issn.0372-2112.2018.02.008

• 学术论文 • 上一篇    下一篇

基于全局无偏搜索策略的精英人工蜂群算法

杜振鑫1,2, 刘广钟1, 韩德志1, 余学山1, 贾建鑫1   

  1. 1. 上海海事大学信息工程学院, 上海 201306;
    2. 韩山师范学院计算机与信息工程学院, 广东潮州 521041
  • 收稿日期:2017-05-09 修回日期:2017-07-03 出版日期:2018-02-25
    • 通讯作者:
    • 韩德志
    • 作者简介:
    • 杜振鑫,男,1976年出生,博士研究生,主要研究方向为进化计算,智能信息处理.E-mail:duzhenxinmail@163.com;刘广钟,男,1962年出生,博士,教授、博导,主要研究方向为无线传感器网络、智能信息处理.E-mail:gzhliu@shmtu.edu.cn
    • 基金资助:
    • 国家自然科学基金 (No.61373028,No.61672338)

Artificial Bee Colony Algorithm with Global and Unbiased Search Strategy

DU Zhen-xin1,2, LIU Guang-zhong1, HAN De-zhi1, YU Xue-shan1, JIA Jian-xin1   

  1. 1. College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China;
    2. School of Computer Information Engineering, Hanshan Normal University, Chaozhou, Guangdong 521041, China
  • Received:2017-05-09 Revised:2017-07-03 Online:2018-02-25 Published:2018-02-25
    • Corresponding author:
    • HAN De-zhi
    • Supported by:
    • National Natural Science Foundation of China (No.61373028, No.61672338)

摘要: 针对精英人工蜂群算法(ABC_elite)加速收敛和早熟停滞现象的矛盾,提出一种改进算法(EABC_elite).该算法通过在雇佣蜂阶段引入全局最优解加速收敛,同时通过普通个体平衡全局最优解过大的引导作用,因此全局最优解与普通个体的信息都能得到利用而算法仍能较好平衡.在观察蜂阶段引入普通个体避免算法早熟收敛.改进算法在加速收敛与防止早熟之间取得很好的平衡,总体上没有偏向任何方向,增强了ABC_elite的全局搜索能力.实验表明,改进算法的性能显著好于ABC_elite以及最近提出的几种较高水平的改进人工蜂群算法.

关键词: 人工蜂群, 平衡策略, 搜索公式, 函数优化

Abstract: To settle the contradictory between convergence speed and precocity in artificial bee colony with elite-guided equations (ABC_elite), an enhanced ABC_elite (EABC_elite) is proposed. In the employed bee phase of EABC_elite, the global best (gbest) individual is introduced to the search equation to accelerate convergence, while the ordinary individual is also introduced to the search equation to balance the gbest's great lead ability, thus the information of gbest and ordinary individuals can all be used and the balance between exploration and exploitation can still be maintained. In the onlooker bee phase of EABC_elite, the ordinary individual is also introduced into the search equation to avoid the precocity. EABC_elite can keep good balance between accelerating convergence and averting precocity. It has no bias to any search directions and the global search ability of ABC_elite is enhanced. The experimental results demonstrate that EABC_elite performs significantly better than ABC_elite and other recently-developed state-of-the-art variants of artificial bee colony algorithm.

Key words: artificial bee colony, balance strategy, search equation, function optimization

中图分类号: