An urban traffic prediction method and device based on space-time mixed graph convolution are used for predicting traffic flow of urban road sections, and the method comprises the following steps: firstly, collecting traffic road network data information, such as a speed time sequence of the road sections and an adjacency matrix for describing an adjacency relation between the road sections, and analyzing and modeling the collected data to obtain a model; dividing into a training set and a test set; secondly, constructing a mixed graph convolution dense connection block by using a graph attention network and graph convolution, sending an extracted spatial feature vector to an Informer long sequence prediction model after function fusion, and extracting spatial and temporal features of a traffic road network; and finally, the training set is sent into the space-time mixed graph convolution model for training, model parameters are adjusted, so that the model achieves the optimal prediction performance, and the model performance is evaluated by using the test set. The method can enhance the traffic management capability and reduce the traffic congestion cost by extracting the spatial-temporal information of the urban traffic road network and deeply mining the spatial features of the road network, and can also be applied to other spatial-temporal prediction tasks.

    基于时空混合图卷积的城市交通预测方法和装置,用来对城市路段的交通流进行预测,其方法包括:首先,收集交通路网数据信息,如路段的速度时序,描述路段间邻接关系的邻接矩阵,并对所收集的数据进行分析和建模,划分为训练集和测试集;其次,利用图注意力网络和图卷积构建混合图卷积密集连接块,经过融合函数后将提取的空间特征向量送到Informer长序列预测模型中,提取交通路网的时空特征;最后,将训练集送入时空混合图卷积模型中训练,调整模型参数,使得模型达到最佳预测性能,并用测试集对模型性能进行评估。本发明通过对城市交通路网时空信息的提取,深度挖掘路网空间特征,可以增强交通管理能力,减少交通拥堵成本,还能应用于其他时空预测任务。


    Access

    Download


    Export, share and cite



    Title :

    Urban traffic prediction method and device based on space-time mixed graph convolution


    Additional title:

    基于时空混合图卷积的城市交通预测方法和装置


    Contributors:
    GUO HAIFENG (author) / XU HONGWEI (author) / ZHOU ZISHENG (author)

    Publication date :

    2023-12-12


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Traffic prediction method based on dynamic graph convolution

    FAN JIN / WENG WENCHAO / TIAN HAO et al. | European Patent Office | 2023

    Free access

    Traffic flow prediction method based on interactive space enhanced graph convolution model

    LI QIN / XU PAI / ZHENG ZUOCAI et al. | European Patent Office | 2024

    Free access

    Dynamic graph convolution traffic speed prediction method

    LIU QILIANG / YUAN HAOTAO / YANG LIU et al. | European Patent Office | 2020

    Free access

    Traffic flow prediction method based on fusion of space-time adaptive graph learning and dynamic graph convolution

    ZHANG HONG / CHEN LINBIAO / CHEN LINLONG et al. | European Patent Office | 2024

    Free access

    Traffic flow prediction method based on multi-graph space-time convolution and dynamic measurement fusion

    SHI ZHENQUAN / FENG JI / GUO CHANG et al. | European Patent Office | 2025

    Free access