基于短时交通流预测的广域动态交通路径诱导方法

为提升车辆通行效率，以预测型诱导策略为基础，以排队长度作为交通诱导的约束条件，利用小波神经网络短时交通量预测预知路段堵死事件发生路段，通过广域诱导时空边界条件对事件路段进行节点分级和诱导周期长度界定，进而建立广域诱导模型；对事件区域路网进行分区，进一步确定该模型诱导起点位置，引入基于路径尺度的Logit 路径选择模型作为诱导路径选择方法，通过流量迭代分配方法实现路网负载均衡. 通过实例验证，该诱导方法能有效地缓解道路交通拥堵，提高路网通行效率.

Wide-area Dynamic Traffic Route Guidance Method Based on Short-term Traffic Flow Prediction

In order to improve the efficiency of vehicle traffic, based on the predictive guidance strategy, this paper puts forward the queue length as the constraint condition of traffic guidance, uses the short- term traffic volume of wavelet neural network to predict the road section where the blocking event occurs, and then uses the time-space boundary conditions of wide area guidance to classify the nodes and define the length of the guidance period of the event road section, and then establishes the wide area guidance. Then, the event area road network is divided into districts to further determine the location of the guidance starting point of the model, and the Logit path selection model based on the path scale is introduced as the guidance path selection method. Finally, the load balance of the road network is realized by the iterative flow distribution method. Through the example, it is proved that the guidance method can effectively alleviate the road traffic congestion and improve the traffic efficiency of the road network.