The invention discloses a traffic flow prediction method based on a bidirectional GRU hypergraph convolution model, and the method employs a dynamic adjacency matrix based on covariance to replace an adjacency matrix between nodes in an actual condition. And the spatial correlation and isomerism of the nodes in the road historical data are extracted from the two aspects of the road network dynamic adjacency matrix and the hypergraph corresponding to the road network dynamic adjacency matrix, so that the prediction accuracy of the model is effectively improved. A traffic prediction network model is adopted to train adaptive network parameters of a target road section through historical traffic data, a road network dynamic adjacency matrix of the historical traffic data and a hypergraph corresponding to the road network dynamic adjacency matrix, traffic data of the target road section in a certain time length before a to-be-predicted moment is obtained, and the trained traffic prediction network model is combined. And obtaining traffic prediction data of the target road section in the target time period. According to the method, the functional spatial correlation is captured from a global angle, the time-varying spatial correlation is captured from a local angle, and the traffic prediction precision is improved.

    本发明公开一种基于双向GRU超图卷积模型的交通流量预测方法,该方法使用基于协方差的动态邻接矩阵替代实际情况下的节点之间的邻接矩阵,并且分别从路网动态邻接矩阵和其对应的超图这两个角度提取道路历史数据中节点的空间相关性和异构性,有效的提升模型的预测准确率。采用交通预测网络模型通过历史交通数据及其路网动态邻接矩阵和其对应的超图对目标路段的适配的网络参数进行训练,获取目标路段待预测时刻之前一定时长的交通数据,结合训练好的交通预测网络模型,得到目标路段目标时段的交通预测数据。本发明方法从全局角度捕捉功能空间相关性和从局部角度捕捉时变空间相关性,提高交通预测精度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic flow prediction method based on bidirectional GRU hypergraph convolution model


    Weitere Titelangaben:

    一种基于双向GRU超图卷积模型的交通流量预测方法


    Beteiligte:
    WANG ZHIZHONG (Autor:in) / ZHANG PING (Autor:in) / ZHENG HAIFEI (Autor:in) / ZHANG XIYANG (Autor:in) / WU JINGUANG (Autor:in) / HUANG TIANBO (Autor:in) / GU JUNHUA (Autor:in)

    Erscheinungsdatum :

    2024-04-26


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    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



    Metro Passenger Flow Prediction via Dynamic Hypergraph Convolution Networks

    Wang, Jingcheng / Zhang, Yong / Wei, Yun et al. | IEEE | 2021


    Traffic prediction method and system based on adaptive hypergraph

    ZHU CHAO / ZHU RUI / LI TONG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic speed prediction method based on geometric algebra and hypergraph

    ZANG DI / LEI JUNTAO / CUI ZHE et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic flow prediction method and device based on adaptive hypergraph convolutional neural network

    LIANG JUN / LOU SHUNJIE / WANG WENHAI | Europäisches Patentamt | 2023

    Freier Zugriff

    Long-time-sequence traffic flow prediction method based on graph convolution-Informer model

    GUO ZIQIANG / CHENG BAOXI / YANG XIAOLEI et al. | Europäisches Patentamt | 2021

    Freier Zugriff