基于阈值搜索的多目标人工蜂群算法

为了平衡人工蜂群算法局部开发能力和全局搜索能力,提高算法收敛速度,提出一种基于阈值搜索的人工蜂群算法.首先,提出一种混沌镜像初始化方法,保证初始种群的多样性和优异性;然后,利用个体阈值动态调整搜索半径,提高搜索精度和收敛速度,考虑外部档案解的开发次数,合理选择精英解来引导进化.在11种测试函数上与其他几种算法对比的仿真结果表明,所提出算法具有较好的分布性和收敛性.

A multi-objective artificial bee colony based on limit search strategy

In order to balance the local exploitation ability and the global exploration ability of artificial bee colony,and accelerate the rate of convergence, a multi-objective artificial bee colony based on the limit search(LSABC) strategy is proposed. Firstly, a chaotic mirroring strategy is proposed in the initialization phase to ensure diversity and superiority. Then, the radius of search is dynamically adjusted while the employed bees and the onlooker bees searching according to the individual limit, which improves the search accuracy and convergence speed. The development times of the external archive are combined to choose elite solution so as to guide evolution reasonably. Finally, experiments are conducted on a set of 11 benchmark functions, and the results show that the proposed algorithm performs well in convergence and distribution.