The invention belongs to the technical field of traffic flow prediction, and particularly relates to a traffic flow prediction method for multivariate traffic flow spatio-temporal data information, which comprises the following steps: acquiring input data, performing weighted summation on the input data based on an attention mechanism to obtain a dynamically weighted input data matrix, respectively masking the dynamic weight matrix by using an out-degree matrix and an in-degree matrix to obtain a corresponding dynamic weight matrix; constructing a self-adaptive matrix and carrying out sparsification on the self-adaptive matrix; constructing a diffusion diagram convolutional network module based on the dynamic weight matrix and the adaptive matrix, constructing a GRU gating unit based on the diffusion diagram convolutional network module, and inputting the input data into the gating unit to obtain a state vector; processing the acquired hidden state information by using a multi-head attention mechanism, and inputting the output of multi-head attention into a neural network layer for prediction; according to the invention, the accuracy of traffic flow prediction is improved, and the influence of traffic flow fluctuation on the prediction result is reduced.

    本发明属于交通流预测技术领域,具体涉及一种面向多元交通流时空数据信息的交通流预测方法,包括获取输入数据,基于注意力机制对输入数据进行加权求和得到动态加权的输入数据矩阵,并利用出度矩阵和入度矩阵分别对其进行掩码得到对应的动态权重矩阵;构建自适应矩阵并对其进行稀疏化;基于动态权重矩阵和自适应矩阵构建扩散图卷积网络模块,在基于扩散图卷积网络模构建GRU门控单元,将输入数据输入门控单元获取状态向量;利用多头注意力机制对获取的隐状态信息进行处理,将多头注意力的输出输入到神经网络层进行预测;本发明提高交通流预测的准确度,同时降低了交通流波动对预测结果的影响。


    Access

    Download


    Export, share and cite



    Title :

    Traffic flow prediction method for multivariate traffic flow spatio-temporal data information


    Additional title:

    一种面向多元交通流时空数据信息的交通流预测方法


    Contributors:
    LI LIN (author) / CHEN HAOJI (author) / CHEN KANG (author) / HE TANG (author) / REN XIN (author) / XIA HELONG (author)

    Publication date :

    2024-02-02


    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 / 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



    Spatio-Temporal AutoEncoder for Traffic Flow Prediction

    Liu, Mingzhe / Zhu, Tongyu / Ye, Junchen et al. | IEEE | 2023


    Traffic flow spatio-temporal data prediction method, system, device and medium

    XIA YUANQING / YAN TIJIN / GONG HENGHENG et al. | European Patent Office | 2023

    Free access

    Global spatio‐temporal dynamic capturing network‐based traffic flow prediction

    Haoran Sun / Yanling Wei / Xueliang Huang et al. | DOAJ | 2023

    Free access

    Global spatio‐temporal dynamic capturing network‐based traffic flow prediction

    Sun, Haoran / Wei, Yanling / Huang, Xueliang et al. | Wiley | 2023

    Free access

    Method for traffic flow prediction based on spatio-temporal correlation mining

    SHI SHIXIONG / YANG SU | European Patent Office | 2017

    Free access