摘要
为快速求解潮流计算问题,文章给出一种基于整体同步并行计算模型(Bulk Synchronous Parallel Computing Model,BSP)的全节点并行迭代算法。在利用佩奇排序实现并行机制的基础上,用图数据模型实现了电力网络全节点的并行的潮流迭代计算。基于实际电网算例和国际标准算例进行验证,其结果说明该方法可以在常规服务器上实现万级节点系统百毫秒级、十万节点秒级、百万节点10 s级的计算速度,极大提升现有EMS系统的计算分析效率。
In order to solve the power flow calculation problem quickly, this paper presents a fullnode parallel iterative algorithm based on BSP model. Parallel calculation is applied using PageRank algorithm and the graph data model is used to realize the parallel power flow iterative calculation in the full-node power network. After that, the method is tested by the actual power grid example and international standard example, the results show that on the normal server, the computing speed of the ten thousand node system is sub-100 milliseconds, one hundred thousand nodes system is sub-seconds, and one million nodes system is sub-10 seconds. The method can greatly improve the computational analysis efficiency of existing EMS systems.
作者
刘克文
张国芳
袁琛
刘广一
LIU Kewen;ZHANG Guofang;YUAN Chen;LIU Guangyi(Global Energy Interconnection Research Institute Co.,Ltd.,Beijing102209,China;State Grid Sichuan Electric Power Company,Chengdu 610041,China;Global Energy Interconnection Research Institute North America,Santa Clara,CA 95134,USA)
出处
《电力信息与通信技术》
2018年第10期19-24,共6页
Electric Power Information and Communication Technology
基金
国家电网公司海外研究院科技项目资助(SGRIJSKJ(2016)800)
关键词
图论
图数据库
图计算
电力系统分析
图数据模型
佩奇排序法
BSP
Graph theory
graph database
graph computing
power system analysis
graph data modeling
PageRank algorithm
bulk synchronous parallel computing model（BSP）
