面向复杂城市路网结构的交通流预测理论与方法

With the increasingly serious contradiction between urban road supply and demand, the problem of constant urban traffic congestion, frequent traffic accidents, increased energy consumption, and increased exhaust emissions have become increasingly prominent. Traffic state forecasting and control are valued by governments of various countries because they can effectively alleviate these problems without changing the existing road structure. The traffic state prediction performance for a complex road network structure is the core foundation for ITS to provide accurate feed forward control, and it is also a difficult and hot topic for research. This project aims to propose a traffic state prediction theory and method for complex road network, and conducts research from three aspects: basic theory, key technologies, and practical example. First of all, the theory of vector pattern representation for complex road network traffic flow is.put forward. Secondly, key technologies such as the lack of road network traffic data restoration, the analysis of complex road network structure and traffic conditions, and the prediction methods of complex road network traffic conditions are studied to improve the forecast performance of traffic conditions in complex road networks. Finally, develop a prototype system of traffic condition prediction.for complex road network. Taking the traffic condition prediction of the road network in Suzhou as an example, verify the theory and key technologies proposed. Through the above studies, it is of great significance to explore the theory and the key technologies of traffic state prediction for complex road network structure, which is of great significance for the theoretical research and engineering application of ITS.

随着城市道路供需矛盾日趋严重，导致城市交通拥堵常态化、交通事故频发、能源消耗增多、尾气排放加剧等问题日益突出。交通状态预测和控制因可在不改变道路结构情况下有效地缓解上述问题而受到各国政府的重视。复杂路网结构的交通状态预测性能是智能交通系统能否提供精准前馈控制的核心基础，也是研究的难点和热点。本项目旨在提出一种面向复杂路网结构的交通状态预测理论与方法，从基础理论、关键技术和实例验证三方面开展研究。首先，提出面向复杂路网交通流矢量图模式表示理论；其次，研究路网交通数据缺失修复、复杂路网结构与交通状态的关联分析、复杂路网交通状态预测方法等关键技术，实现复杂路网结构交通状态预测性能的提升；最后，研发复杂路网交通状态预测原型系统，以苏州市路网交通状态预测为例，验证提出的理论和关键技术。通过上述研究，探索面向复杂路网结构的交通状态预测理论与关键技术，对智能交通系统的理论研究和工程应用具有重要意义。

随着城市道路供需矛盾日趋严重，导致城市交通拥堵常态化、交通事故频发、能源消耗增多、尾气排放加剧等问题日益突出。交通状态预测和控制因可在不改变道路结构情况下有效地缓解上述问题而受到各国政府的重视。复杂路网结构的交通状态预测性能是智能交通系统能否提供精准前馈控制的核心基础，也是研究的难点和热点。本项目旨在提出一种面向复杂路网结构的交通状态预测理论与方法，从基础理论、关键技术和实例验证三方面开展研究。首先，提出面向复杂路网交通流矢量图模式.表示理论；其次，研究路网交通数据缺失修复、复杂路网结构与交通状态的关联分析、复杂路网交通状态预测方法等关键技术，实现复杂路网结构交通状态预测性能的提升；最后，研发复杂路网交通状态预测原型系统，以苏州市路网交通状态预测为例，验证提出的理论和关键技术。通过上述研究，探索面向复杂路网结构的交通状态预测理论与关键技术，对智能交通系统的理论研究和工程应用具有重要意义。