The invention discloses a data-driven urban expressway traffic flow modeling method. According to the method, modeling is carried out on speed evolution of expressway macroscopic traffic flow by using time sequence data containing speed information of each road section on a road. Firstly, the speed of each sub-road section serves as the state quantity of a road, and data information of the state quantity on the road is collected; then, fitting the state quantity data within a period of time by using a least square method to obtain an initial linear state space model of state evolution on the road; and finally, updating the system matrix to obtain an update rate of the system matrix along with the obtained new data, and further obtaining a state space model of speed evolution dynamic update on the road. The expressway traffic data is predicted through the dynamically updated model, the traffic flow evolution trend can be accurately predicted, and the effectiveness and accuracy of the modeling method are verified.
本发明公开了一种数据驱动的城市快速路交通流建模方法,该方法使用包含道路上各个路段速度信息的时间序列数据对快速路宏观交通流的速度演化进行建模。首先,将各个子路段的速度作为道路的状态量,收集道路上状态量的数据信息;然后,用最小二乘法对一段时间内的状态量数据进行拟合,得到道路上状态演化的初始线性状态空间模型;最后,随着获取到新的数据,对系统矩阵进行更新得到系统矩阵的更新率,进而得到道路上速度演化动态更新的状态空间模型。通过该动态更新的模型对快速路交通数据进行预测,能准确的预测交通流演化趋势,验证了该建模方法的有效性和准确性。
Data-driven urban expressway macroscopic traffic flow modeling method
一种数据驱动的城市快速路宏观交通流建模方法
03.02.2023
Patent
Elektronische Ressource
Chinesisch
Modeling and Control of Urban Expressway Traffic Flow
ASCE | 2007
|Urban Expressway Traffic Flow Modeling and Control Using Neural Networks
British Library Online Contents | 2004
|Traffic flow control method and system for urban expressway
Europäisches Patentamt | 2024
|Urban Expressway Traffic Flow Modeling and Control Using Artificial Neural Networks
British Library Conference Proceedings | 2003
|