The invention relates to a metropolitan area traffic flow prediction method based on a knowledge graph and depth space-time convolution. According to the method, a space-time diagram convolutional neural network is adopted for traffic flow prediction, and compared with a traditional prediction method, the method has the advantages that time and space are strongly associated, so that the accuracy of traffic flow prediction is improved. According to the method, conventional convolution and recursion units are not adopted, problems are established on the graph, and a model with a complete convolution structure is established, so that a higher training speed can be brought by using fewer parameters, and the computing power requirement of traffic flow prediction is greatly reduced; new data generated by a road network are added into a training set, iterative optimization is continuously carried out on the model, and the model performance is improved.
本发明涉及一种基于知识图谱和深度时空卷积的城域交通流量预测方法。本发明采用时空图卷积神经网络进行交通流量预测,相比传动的预测方法,将时间和空间进行强关联,提高了交通流量预测的准确性;本发明没有采用常规的卷积和递归单位,而是在图上建立问题,并建立具有完整卷积结构的模型,使得可以用更少的参数带来更快的训练速度,极大了降低了交通流量预测的算力需求;本发明将路网产生的新数据加入到训练集中,不断对模型进行迭代优化,提升模型性能。
Metropolitan area traffic flow prediction method based on knowledge graph and depth space-time convolution
基于知识图谱和深度时空卷积的城域交通流量预测方法
2021-04-20
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 / 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 double graph convolution
European Patent Office | 2024
|Traffic Flow Prediction Based on Dynamic Time Slot Graph Convolution
Transportation Research Record | 2025
|Traffic flow prediction method based on interactive space enhanced graph convolution model
European Patent Office | 2024
|European Patent Office | 2025
|European Patent Office | 2024
|