摘要
为了解决蚁群算法在求解连续函数优化问题时,存在局部搜索能力较差的缺陷,提出一种新颖的自适应蜂群—蚁群优化算法。新算法在蚁群优化算法的基础上,设计了一种参数q的自适应机制,进而减少了参数个数,提高了其鲁棒性;根据蜂群算法基本思想,利用雇佣蜂和观察蜂设计了高效的局部搜索算子,从而提升了算法的局部能力。针对五个标准测试函数的仿真实验结果表明:与蚁群优化算法相比,新算法的全局和局部寻优能力均得到了极大的提升。
This paper proposed a novel colony-ant colony optimization algorithm for continuous function optimization problems.The new algorithm was based on ant colony optimization algorithm.There were two improvements in the new algorithm.Firstly,it devised the adaptive mechanism for parameter q to reduce the parameters＇ number and improved the robustness of the new algorithm.Secondly,an efficient local search operator,which used employee bees and observed bees in the artificial bee colony,was devised to enhance the local searching capacity.The simulation results for five benchmark functions show that： compared with those of ant colony optimization,the global and local searching capability of colony-ant colony optimization has been greatly improved.
出处
《计算机应用研究》
CSCD
北大核心
2012年第1期130-134,共5页
Application Research of Computers
关键词
优化问题
蚁群优化
人工蜂群算法
optimization problem
ant colony optimization（ACO）
artificial bee colony algorithm
