一种基于蚁群思想的动态路径寻优算法的实现及仿真  

Implementation and Simulation of a Dynamic Path Optimization Algorithm Based on Ant Colony Thoughts

作　　者：周秀娟[1] 花嵘[1] 史娟[1] 高玉励 

机构地区：[1]山东科技大学信息科学与工程学院,山东青岛266590 [2]青岛黄海学院,山东青岛266427

出　　处：《山东科技大学学报（自然科学版）》2016年第4期114-120,共7页Journal of Shandong University of Science and Technology(Natural Science)

基　　金：山东省科技发展计划项目(13GGX10118);青岛经济技术开发区重点科技发展计划项目(2013-1-65)

摘　　要：为了提高路径寻优算法的效率和实时性,本文实现了一种名为DPO-AC的基于蚁群思想的动态路径寻优算法(the ant colony algorithm with dynamic path optimization),在改进蚁群算法的基础上结合神经网络的实时预测方法和限定区域的搜索方式,解决算法在大型网络路径寻优时实时性差、收敛慢的问题。仿真实验表明DLACO算法有比较好的稳定性、收敛性和实时性。In order to improve the efficiency and instantaneity of path optimization algorithms,a dynamic path optimi- zation algorithm named the ant colony algorithm with dynamic path optimization（DPO-AC） was proposed. On the basis of the improved ant colony algorithm and combined with neural network real-time prediction and limited area search, the proposed algorithm solved the problem of poor real-time performance and slow convergence of large net- work path optimization algorithms. Simulation results show that DL-ACO has better stability, convergence and in- stantaneity.

关 键 词：路径寻优 算法 蚁群 限定区域 阻塞矩阵 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]