The invention relates to a road traffic flow prediction method and system based on deep learning, and relates to the technical field of intelligent traffic flow prediction. The method comprises the following steps: constructing a network topology structure diagram of urban roads, establishing a training model for predicting traffic flow, taking node features and edge features obtained by feature extraction as input, and training the training model by using feature representation fused with a space-time self-attention mechanism to obtain a prediction model; inputting node feature data at a to-be-predicted moment into the prediction model, and outputting the predicted traffic flow at the moment; and evaluating the prediction model, optimizing the prediction model according to an evaluation result, and adjusting the structure and parameters of the prediction model. According to the method, flow prediction can be carried out on a single traffic node, prediction can also be carried out in the whole road network range, high universality and applicability are achieved, and the precision and robustness of traffic flow prediction can be improved.
本发明涉及一种基于深度学习的道路交通流量预测方法及系统,涉及智慧交通流量预测技术领域。方法包括:构建城市道路的网络拓扑结构图,建立预测交通流量的训练模型,将特征提取得到的节点特征和边特征作为输入,利用融合了时空自注意机制的所述特征表示对所述训练模型进行训练,得到预测模型;将待预测时刻的节点特征数据输入所述预测模型,输出预测的该时刻的交通流量;对所述预测模型进行评估,根据评估结果对所述预测模型进行优化,调整所述预测模型的结构和参数。该方法不仅可以在单个交通节点上进行流量预测,还可以在整个路网范围内进行预测,具有较强的普适性和适用性,能够提高交通流量预测的精度和鲁棒性。
Road traffic flow prediction method and system based on deep learning
一种基于深度学习的道路交通流量预测方法及系统
2024-08-30
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Urban road traffic state prediction system based on deep learning
European Patent Office | 2020
|Short‐term traffic flow prediction of road network based on deep learning
Wiley | 2020
|Urban area road network traffic flow prediction method and system based on mixed deep learning model
European Patent Office | 2021
|