The invention relates to a traffic flow prediction method based on dynamic space-time correlation. The method comprises the following steps: S1, carrying out data design, preprocessing traffic flow information, inputting a fixed sequence formed after the coding of the traffic flow information into a model, and carrying out definition; S2, carrying out model design, processing spatial and temporal correlation by using a convolutional neural network and a gating loop network, capturing dynamic spatial dependency in a hierarchical structure by using a flow gating mechanism, and then capturing time transfer by using a periodic attention transfer mechanism; S3, performing model training, and evaluating the model in a real traffic data set; and S4, evaluating the performance of the model. According to the method, dynamic similarity between positions is learned by introducing a flow gating mechanism, and time transfer is processed by utilizing a periodic attention transfer mechanism, so that high-accuracy model construction is realized, and the accuracy of traffic prediction data is improved.
本发明涉及一种基于动态时空相关性的交通流预测方法,包括以下步骤:S1、进行数据设计,对交通流信息进行预处理,将交通流信息编码后形成的固定序列输入到模型中,再进行定义;S2、进行模型设计,利用卷积神经网络和门控循环网络处理空间和时间相关性,流门控机制捕捉层次结构中的动态空间依赖性,再通过周期注意力转移机制捕捉时间转移;S3、进行模型训练,并在一个真实的交通数据集中评估模型;S4、对模型性能进行评价。本发明通过引入流门控机制学习位置之间的动态相似性,并利用周期性注意力转移机制来处理时间转移,实现高准确度的模型建设,提高交通预测数据的准确性。
Traffic flow prediction method based on dynamic space-time correlation
一种基于动态时空相关性的交通流预测方法
2021-07-13
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES |
Traffic flow prediction method based on time window division modeling dynamic space-time correlation
European Patent Office | 2024
|Time-space correlation traffic flow prediction method based on deep learning
European Patent Office | 2025
|Traffic flow prediction method based on dynamic space-time similar pyramid network
European Patent Office | 2024
|Traffic flow prediction method and system based on space-time correlation and error compensation
European Patent Office | 2023
|Space-time adaptive dynamic graph convolutional network traffic flow prediction method
European Patent Office | 2024
|