摘要
大规模网络条件下,如何经济高效地构建仿真系统来实现城市轨道交通路网客流分布的动态推演是一个亟待解决的问题。将"运输流状态网络模型"与有限状态机模型相结合,提出了适用于大规模路网条件下多分辨率客流分布推演的仿真建模方法,给出了利用仿真推演引擎遍历状态网络来实现系统状态动态推演的算法。采用混合式并行仿真任务分解策略,给出了基于网络客流量的域分解方案优化方法,以及基于关键同步事件的子域内时钟协调方法,以简单高效的方式实现并行仿真的时钟推进。自主开发了仿真系统,并以北京地铁为背景构建仿真案例,并对提出的方法进行了验证。
For a large scale urban rail transit network, the economical and efficient construction of a simulation system to realize the dynamic deduction of passenger flow distribution in urban rail transit network is urgently needed. Combining the ＂transport flow state network model＂ with the finite state machine model, this paper presented a simulation modeling method applicable to multi resolution passenger flow distribution inference un der the condition of large scale rail transit network. The algorithm to realize dynamic inference of system state using simulation inference engine ergodic state network was given. Under the hybrid parallel simulation task decomposition strategy, the optimization method of domain decomposition scheme based on network passenger flow and the sub domain clock coordination method based on key synchronization events were given to realize the clock advance of parallel simulation in a simple and efficient way. The simulation system was developed in dependently, and a simulation case was constructed against Beijing Metro to verify the proposed method.
作者
蒋熙
冯佳平
贾飞凡
孙捷萍
李春晓
JIANG Xi;FENG Jiaping;JIA Feifan;SUN Jieping;LI Chunxiao(State Key Lab of Rail Traffic Control ＆Safety,Beijing Jiaotong University,Beijing 100044,China)
出处
《铁道学报》
EI
CAS
CSCD
北大核心
2018年第11期9-18,共10页
Journal of the China Railway Society
基金
国家重点研发计划(2016YFB1200402002)
轨道交通控制与安全国家重点实验室自主研究课题(RCS2018ZT005)
