Computer Engineering and Applications ›› 2019, Vol. 55 ›› Issue (22): 53-59.DOI: 10.3778/j.issn.1002-8331.1808-0460

Previous Articles     Next Articles

Hybrid Artificial Bee Colony Algorithm with Adaptive Search Strategy

SONG Xiaoyu, GAO Minghai, ZHAO Ming   

  1. School of Information and Control Engineering, Shenyang Jianzhu University, Shenyang 110168, China
  • Online:2019-11-15 Published:2019-11-13

具有自适应搜索策略的混合人工蜂群算法

宋晓宇,高明海,赵明   

  1. 沈阳建筑大学 信息与控制工程学院,沈阳 110168

Abstract: The basic artificial bee colony algorithm and its search strategy focus on exploration. In order to enhance the exploitation of the algorithm, this paper proposes a hybrid artificial bee colony algorithm with adaptive search strategy and better exploitation. Firstly, the objective function value information and optimal solution guiding information are introduced into the search strategy, so that this paper obtains a new search strategy with strong exploitation and self-adaptive mechanism; At the same time, in order to avoid premature convergence, this paper uses three random food sources and Gaussian distribution to form a new search strategy with good exploration. And then, this paper balances exploration and exploitation by hybridizing the two search strategies in employed bee phase, and enhances exploitation by using the search strategy with strong exploitation in onlooker bee phase. Compared with basic and some representative improved artificial bee colony algorithms in 20 standard test functions, the experimental results show that the proposed algorithm has better search ability and faster convergence speed.

Key words: artificial bee colony algorithm, information of objective function value, self-adaption, improved search , strategy, Gaussian distribution

摘要: 基本人工蜂群算法及其搜索策略侧重探索,为增强算法的开发能力,提出一种具有自适应搜索策略的混合人工蜂群算法。将目标函数值信息和最优解引导信息引入搜索策略,提出具有自适应机制、开发能力强的搜索策略;为防止“早熟”现象,利用三个不同随机食物源和高斯分布,设计出全局探索能力较强的搜索策略。将两个搜索策略在雇佣蜂阶段混合以平衡算法的探索与开发能力,在观察蜂阶段使用具有自适应机制、开发能力强的搜索策略以加快收敛。与基本及具有代表性的改进人工蜂群算法在20个标准测试函数中进行对比实验,结果表明所提算法具有更好的搜索能力和更快的收敛速度。

关键词: 人工蜂群算法, 目标函数值信息, 自适应, 改进搜索策略, 高斯分布