摘要
为了提高混合交通微观仿真的运行速度,基于多核并行计算技术构建了一个混合交通微观仿真平台.文中首先从平台功能结构、并行混合交通微观仿真流程设计与实现两个方面对平台进行介绍;其次,设计并实现了该仿真平台的关键并行算法,包括初始路网分割算法、车辆穿越边界算法及动态负载平衡算法;最后,通过仿真实验验证了多核并行计算技术能够大大提高混合交通微观仿真的运行效率.
In order to speed up the microscopic simulation of mixed traffic, a simulation platform based on the multi-core parallel computing technology is proposed. The proposed platform is described in two aspects, namely the functional structure and the design as well as the implementation of the simulation process. Then, some key al- gorithms for implementing the microscopic simulation, namely the initial transportation network segmentation algo- rithm, the crossing border algorithm of vehicles and the dynamic load-balancing algorithm, are designed and imple- mented. Simulated results show that the adoption of multi-core parallel computing technology greatly improves the microscopic simulation efficiency of mixed traffic.
出处
《华南理工大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2012年第2期57-62,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(70972041)
高等学校博士学科点专项科研基金资助课题(20100009110010)
北京交通大学基本科研业务费资助课题(2011YJS246)
关键词
交通运输系统工程
多核并行计算
混合交通
微观仿真
transportation system engineering
multi-core parallel computing
mixed traffic
microscopic simula-tion