一种基于熵的电力通信网络业务资源均匀分配算法  

An Algorithm for Business Resource Uniform Distribution in Power Communication Network Based on Entropy

作　　者：崔力民[1] 孙静月[2] 李珊君[2] 宋广磊[1] 

机构地区：[1]国网新疆电力公司信息通信公司 [2]四川大学电气信息学院,四川省成都市610065

出　　处：《电网技术》2017年第9期3066-3073,共8页Power System Technology

摘　　要：为了均匀电力通信网络业务流量的分布并提高网络吞吐量和传输可靠性,提出了一种基于熵的业务均匀分配算法。根据电力通信网具有的业务特点进行业务流量分析,业务流量的分布情况体现了网络业务运行状态,将业务信息熵作为衡量网络业务分布均匀化的指标,进而引用信息熵作为目标函数得出优化全局业务路由的算法。首先将业务的时延作为约束条件,求得满足业务需求的可用路径集,再将业务信息熵作为优化函数,最后利用量子遗传算法解决多约束路由问题,在适应度评价的过程中添加带宽约束,控制每条业务流量路径走向,求取使得当前网络业务信息熵值最大的业务路径集。仿真结果显示,在基于熵的业务均匀算法下的网络业务分布相对均匀,并有效地控制了流量,实现了优化网络资源和均衡网络负载的目的。In order to uniformly distribute business traffic of power communication network and improve network throughput and transmission reliability, an algorithm for uniform business optimization based on entropy was proposed. Business traffic was analyzed according to business characteristics of electric power communication network. Traffic distribution reflected operation of network business, so business information entropy as an index of network business distribution uniformity was proposed. Then global business routing algorithm referring to the information entropy as objective function optimization was obtained. Firstly, taking business time delay as a constraint, available path set meeting business requirements was acquired. Business information entropy was adopted as optimization function. Finally, quantum genetic algorithm was used to solve the multi-constrained routing problem. With bandwidth constraints added to process of fitness evaluation to control direction of each business flow path, the business path set of maximum information entropy value of current network was worked out. Simulation result showed that, network business under the algorithm based on entropy distribution was relatively well-distributed, business traffic was effectively controlled and network resource optimization and network load balance were realized.

关 键 词：电力通信网 业务均匀分布 信息熵 业务流量 量子遗传算法 

分 类 号：TM721[电气工程—电力系统及自动化]