Dynamic origin-destination (O-D) flows estimation (DODE), dynamic traffic assignment (DTA), real-time traffic management and traveler information service are fundamental key models of intelligent transportation systems (ITS). Existing step-by-step methods cause uncertain errors and further increase time lags; however, combined models seldom appear in the literature. Based on detected multi-source traffic flow data, this research employs DODE as the foundation, and focuses on accuracy, efficiency and application of dynamic transportation network models using combined modelling approaches. There are three levels in this research: 1) based on fusion of multi-source real-time data from advanced traffic surveillance system, we will construct some combined DODE models under traffic congestion using Bayesian approach and multi-source data; 2) integrating traffic signal control and traffic divergence systems, we will analyze the correlation mechanism between dynamic O-D distribution pattern and real-time traffic management and traveler information systems, and propose combined models of DODE with signal control and traffic divergence systems using revised parameter optimization and kalman filtering methods; 3) furthermore, we will dissect key points correlating DODE and DTA, build the bridge using dynamic network loading technique, and develop the combined dynamic transportation network model of DODE and DTA through variational inequality and cell transmission approaches. Moreover, we will develop some application methods of the combined models integrating with regional traffic signal control and network route guidance systems. Finally, this research will realize on-line applications of combined dynamic transportation network models to intelligent transportation systems.
动态O-D反推、动态交通分配、实时交通管理和信息服务模型是智能交通系统的核心模型,目前独立计算的方式导致误差不确定并增加计算时间,而已有研究很少涉及组合模型。本研究以多源数据为支撑,以动态O-D反推为基础,采用组合模型方法,解决动态交通网络模型的精度、效率和应用等问题,包括三个层次的组合模型:1)融合实时检测的多源交通数据,采用贝叶斯方法建立适用于拥堵的、基于多源数据的动态O-D反推组合模型;2)结合信号控制和交通分流,研究实时交通管理和信息服务措施与O-D分布模式的相互影响机理,利用改进参数优化和卡尔曼滤波方法建立动态O-D反推与信号控制、交通分流的组合模型;3)剖析动态O-D反推与动态交通分配联系的关键点,以动态网络加载为桥梁,利用变分不等式和元胞传输方法建立动态O-D反推与动态交通分配的组合模型。结合区域信号控制和路网交通诱导,研究动态交通网络组合模型的应用方法,最终实现在线应用。
动态O-D反推、动态交通分配、实时交通管理和信息服务模型是智能交通系统的核心模型,目前独立计算的方式导致误差不确定并增加计算时间,而已有研究很少涉及组合模型。本研究以多源数据为支撑,以动态O-D反推为基础,采用组合模型方法,解决动态交通网络模型的精度、效率和应用等问题,包括三个层次的组合模型:1)融合实时检测的线圈、浮动车、雷达等多源交通数据,分路口、高速道路和路网三个层次构建了动态O-D反推模型,其中路口和高速道路层面均包括数学优化与卡尔曼滤波模型,路网层面重点针对交通拥堵引起的车流回溢现象,采用变分不等式方法构造了适用于交通拥挤状态的路网反推模型;2)结合信号控制和交通分流,研究实时交通管理和信息服务措施与O-D分布模式的相互影响机理,并建立了适用于单点和干线的、基于动态O-D反推与多目标优化的信号控制组合模型,以及适用于快速路的、基于动态O-D反推的交通分流组合模型;3)剖析了动态O-D反推与动态交通分配联系的关键点,以动态网络加载为桥梁,综合利用变分不等式和元胞传输方法建立了动态O-D反推与动态交通分配的组合模型,并探讨了其在路网诱导系统中的应用方法。对于上述各类模型均设计了求解算法,并采用实际数据或仿真案例进行了验证,结果表明相关成果可以提高动态交通网络分析的精度和效率,并可有效地应用于交通管理和信息服务系统。
{{i.achievement_title}}
数据更新时间:2023-05-31
Efficient photocatalytic degradation of organic dyes and reaction mechanism with Ag2CO3/Bi2O2CO3 photocatalyst under visible light irradiation
涡度相关技术及其在陆地生态系统通量研究中的应用
Intensive photocatalytic activity enhancement of Bi5O7I via coupling with band structure and content adjustable BiOBrxI1-x
跨社交网络用户对齐技术综述
Influencing factors of carbon emissions in transportation industry based on CD function and LMDI decomposition model: China as an example
面向实时交通管理和信息服务的在线动态O-D反推模型与算法
城市交通网络中运量的动态平衡分配模型与算法
城市交通网络组合出行预测模型及其求解算法设计研究
社会网络环境下可信服务组合动态协同模型与算法研究