改进的蚁群算法求解连续性空间优化问题  

A randomized ant colony algorithm for continuous function optimization

作　　者：王育平[1] 亓呈明[1] 

机构地区：[1]北京联合大学自动化学院,北京100101

出　　处：《辽宁工程技术大学学报（自然科学版）》2010年第5期903-906,共4页Journal of Liaoning Technical University (Natural Science)

摘　　要：为了克服基本蚁群算法搜索时间过长,易陷于局部最优等缺点。引入了随机算法,并提出了一种采用随机模式调整信息素的改进蚁群算法RACA(Randomized Ant Colony Algorithm)。采用随机地计算部分点的函数值,并对当前最优、次优解的部分解启用新的信息素调整规则。该算法保持了解的多样性,增强了全局搜索的能力。运算结果表明新的蚁群算法对求解连续函数优化问题有较好的改进效果。In order to overcome the problem of costly search and premature convergence with continuous function optimization,an improved randomized ant colony algorithm （RACA） is presented in this paper. The new algorithm can compute partial function values and adjust pheromone according to generated random number. The algorithm can enhance the ability of global search by increasing the diversity of solutions. A simulation demonstrates that the proposed algorithm can efficiently find out the minimum value,and has a strong global optimization capability.

关 键 词：连续函数优化 随机算法 蚁群算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]