The invention discloses a traffic flow prediction method based on transformer and multi-graph fusion convolution, and the method comprises the steps: inputting traffic flow data into a transformer module in a preset multi-graph convolution network model for processing, and obtaining long-term time correlation feature data; processing the traffic speed data based on a multi-graph fusion module in a preset multi-graph convolutional network model to obtain static and dynamic spatial correlation feature data in the road network; performing feature aggregation processing on the long-term time correlation feature data and the static and dynamic space correlation feature data according to a space-time diagram convolution module in a preset multi-diagram convolution network model to obtain space-time feature data; and analyzing and processing the spatial-temporal characteristic data through an output layer in a preset multi-graph convolutional network model to obtain traffic flow data in the next hour. The problem that dynamic spatial correlation in a road network cannot be captured in the prior art can be solved, and prediction errors are reduced.

    本申请公开了一种基于transformer和多图融合卷积的交通流量预测方法,包括:将交通流量数据输入预设多图卷积网络模型中的transformer模块进行处理,得到长期时间相关性特征数据;基于预设多图卷积网络模型中的多图融合模块对交通速度数据进行处理,得到道路网络中的静态与动态空间相关性特征数据;依据预设多图卷积网络模型中的时空图卷积模块对长期时间相关性特征数据和静态与动态空间相关性特征数据进行特征聚合处理,得到时空特征数据;通过预设多图卷积网络模型中的输出层对时空特征数据进行分析处理,得到未来一小时内的交通流量数据。本申请能解决现有技术无法捕获道路网络中的动态空间相关性的问题,减小预测误差。


    Access

    Download


    Export, share and cite



    Title :

    Transform and multi-graph fusion convolution-based traffic flow prediction method


    Additional title:

    一种基于transformer和多图融合卷积的交通流量预测方法


    Contributors:
    LIU YIJUN (author) / HUANG HAO (author) / YE WUJIAN (author) / TAN DONGJUN (author)

    Publication date :

    2024-07-19


    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 / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    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

    Traffic flow prediction method based on double graph convolution

    HUI BO / GONG JING / ZHANG LIZONG et al. | European Patent Office | 2024

    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 graph attention convolution network

    ZHENG HONG / ZHANG SIKAI / LIU JIAMOU et al. | European Patent Office | 2020

    Free access