基于并行蚁群记忆查找算法的网络重构  

Network reconstruction based on parallel ant colony memory search algorithm

作　　者：周术鹏 靳松[1] 吕天成 宋丽翠 ZHOU Shu-peng;JIN Song;LYU Tian-cheng;SONG Li-cui(Department of Electronics and Communication Engineering,North China Electric Power University,Baoding 071001,China)

机构地区：[1]华北电力大学电子与通信工程系,河北保定071001

出　　处：《计算机工程与设计》2019年第5期1451-1456,共6页Computer Engineering and Design

基　　金：国家自然科学基金项目(61204027);河北省自然科学基金项目(F2017502043);中央高校基本科研业务费专项资金重点基金项目(2014ZD32)

摘　　要：为节省网络重构的计算资源,提高计算准确度,提出一种并行蚁群记忆查找算法。将前一次迭代得到的最小网损及其对应的网络结构记录到查找表中;若本次迭代计算时蚂蚁找到的结构在查找表中已存在,则通过查找表直接得出对应网损以跳过潮流计算,节约计算资源。在信息素积累阶段,对进程中的信息素浓度进行分步式处理,增加不同进程间的差异值,进而扩大并行蚁群算法的搜索范围,使算法不易陷入局部最优。结合实例,对该算法进行仿真分析,其结果表明了该算法的准确性和高效性。To save the computational resources of network reconstruction and improve the accuracy of computation,the parallel ant colony memory search algorithm was presented.The minimum network loss obtained in the previous ite- ration as well as its corresponding network structure were recorded in the lookup table.If a new structure could be found in the table,the process of power flow calculation skiped and the net loss was calculated directly.In the pheromone accumulation stage,the pheromone concentration in the process was processed step by step,and the differences between different processes were increased,to expand the search range of the parallel ant colony algorithm,which insured that the algorithm was not easy to fall into local optimum.According to the example,the algorithm is simulated and analyzed,and the result shows the accuracy and efficiency of the algorithm.

关 键 词：网络重构 并行计算 蚁群算法 记忆查找算法 查找表 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]