Abstract:

The spatial and temporal characteristics of passenger flow distribution of urban rail transit network are the core and foundation for cooperative transport organizations. On the basis of systematic analysis of passenger travelling behaviors, this paper puts forward a dynamic simulation model based on multiagent modeling techniques. It uses this model to simulate passengers travelling on the network and explores the passenger flow distribution characteristics of network. This model well describes passengers travelling behavior, while maintains the scale of huge passenger flow of urban rail transit system. Finally, it develops an urban rail transit passenger flow simulation system using the actual passenger flow data of Beijing and verifies the accuracy and simulation efficiency.

Key words: urban traffic, dynamic passenger flow distribution, multiagent modeling, urban rail transit, network simulation

摘要：

城市轨道交通网络客流时空分布特征是网络化协同运输组织的基础与核心.本文在系统分析城市轨道交通乘客出行行为的基础上，基于MultiAgent建模技术，构建了网络客流分布动态仿真模型，应用该模型模拟乘客的网络出行过程，统计分析得到路网的客流分布特征.该模型既能反映乘客网络出行行为全过程，又能满足大规模路网仿真需求.最后，开发了城市轨道交通网络客流仿真系统,通过北京市轨道交通网络的实际数据对模型和算法进行验证.

关键词: 城市交通, 动态客流分布, 多主体建模, 城市轨道交通, 网络仿真