信息素S型更新的耦合ACO算法及其应用  

Coupled ACO algorithm with the S-type pheromone update and its application

作　　者：高雷阜[1] 王飞[1] GAO Leifu WANG Fei(College of science, Liaoning Technical University, Fuxin 123000, China)

机构地区：[1]辽宁工程技术大学理学院,辽宁阜新123000

出　　处：《辽宁工程技术大学学报（自然科学版）》2017年第3期316-319,共4页Journal of Liaoning Technical University (Natural Science)

基　　金：辽宁省教育厅基金项目(L2015208)

摘　　要：针对蚁群算法存在易过早收敛、出现停滞现象、陷入局部极值的问题,提出S型信息素更新策略与Alopex算法相耦合的改进蚁群优化算法(IACO).该算法定义全新的S型动态自适应信息素全局更新函数,使信息素增量随迭代次数和目标函数值变化而动态变化,同时耦合Alopex算法以提高算法的局部搜索能力.将IACO算法应用于支持向量机参数的优化中,构成IACO-SVM模型.利用UCI标准数据集进行数值实验.研究结果表明:IACO算法具有较强的寻优性能,IACO-SVM模型具有较高的平均分类准确率和较好的稳定性.Aiming at the problems existing in the ant colony algorithm which are premature convergence, occurring stagnation phenomenon and dropping into local optimum, this paper presented an improved ant colony algorithm （IACO） by coupling the S type pheromone update strategy and Alopex algorithm. The IACO defines a new S type dynamic adaptive pheromone global update function to make the pheromone changes dynamically, coupled with the Alopex algorithm to improve the local search capability. The public datasets from UCI are employed for evaluating the proffered algorithm. The experimental results show that the IACO has stronger optimal performance, and the IACO-SVM model has high average classification accuracy and good stability.

关 键 词：蚁群算法 信息素 ALOPEX算法 支持向量机 参数优化 分类准确率 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]