摘要
为了降低连续交通流模型并行计算的实现难度,参考离散交通流模型并行计算的框架,提出了一套连续交通流模型并行计算的框架.该框架包括同步机制、基于网格的区域分解方法和边界缓冲模型,其中同步机制采用与离散交通流模型并行计算相同的主从结构的同步机制,基于网格的区域分解方法和边界缓冲模型分别用来实现任务的分解和提供仿真计算时所需的交互数据.实验中,选用了一个大规模路网进行验证,结果表明应用这一框架,可以方便地实现连续交通流模型的并行计算,提高连续交通流模型运算的速度,从而满足了智能交通系统中大规模交通路网实时、高效和动态仿真的需求.
A new parallel computing framework based on the parallel computing framework for discrete traffic flow model was presented to reduce the complexity of parallel computing for continuous traffic flow model. The framework includes three parts： synchronization strategy, grid domain decomposition and boundary buffer model. The synchronization strategy is the master-slave approach used in the parallel computing for discrete traffic flow model. The grid domain decomposition is used to achieve distributing the workload in parallel computing and the boundary buffer model is to supply the interactive data. A large scale transportation scenario was used in the experiment to prove this framework. The experiment indicated that the parallel computing for continuous traffic flow model is easy to accomplish by using this framework and that the computation speed of continuous traffic model is increased, meeting the demand for realtime, high-efficiency and dynamic simulation of a large-scale traffic network in intelligent transportation systems.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2006年第10期1783-1786,共4页
Journal of Zhejiang University：Engineering Science
关键词
并行计算
交通仿真
交通流模型
parallel computing
traffic simulation
traffic flow model
