The invention discloses a method for extracting a TMGCN traffic flow prediction model by combining graph Fourier transform. The method comprises the following steps: S1, collecting data to form a data set; s2, a traffic flow Pearson correlation coefficient matrix is calculated, a convex optimization model of a graph Laplacian matrix L is constructed and solved, traffic flow signals changing along with time are mapped to a frequency domain space, and time domain hidden variables which traffic flow space depends on are obtained; s3, inputting a graph convolutional neural network GCN to obtain a frequency domain hidden variable on which the traffic flow space depends, and mapping the frequency domain hidden variable back to a time domain space; s4, extracting a time-dependent hidden variable of the traffic flow time sequence in the data set; s5, transversely splicing and integrating the time-domain hidden variables and the time-dependent hidden variables of traffic flow space dependence; and S6, fusing the space hidden variables and the time hidden variables, and outputting predictive variables. According to the extraction method, the long-time traffic flow prediction precision can be improved, and the evolution trend of the urban road network can be predicted in advance through timely and accurate traffic flow prediction.

    本发明公开了一种结合图傅里叶变换TMGCN交通流预测模型的提取方法,步骤为:S1:采集数据形成数据集;S2:计算交通流量Pearson相关系数矩阵,构建并求解图拉普拉斯矩阵L的凸优化模型,再将随时间变化的交通流信号映射到频域空间,得到交通流空间依赖的时域隐变量;S3:输入图卷积神经网络GCN,得到交通流空间依赖的频域隐变量,并映射回时域空间;S4:提取数据集中的交通流量时间序列的时间依赖隐变量;S5:横向拼接整合交通流空间依赖的时域隐变量和时间依赖隐变量,S6:融合空间与时间隐变量,输出预测变量。该提取方法能提高长时间交通流预测精度,且及时精准的交通流预测能提前预知城市路网的演化趋势。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Method for extracting traffic flow prediction model by combining graph Fourier transform TMGCN


    Weitere Titelangaben:

    结合图傅里叶变换TMGCN交通流预测模型的提取方法


    Beteiligte:
    REN XIN (Autor:in) / LI JUN (Autor:in) / ZHU BILIANG (Autor:in) / HAO BENMING (Autor:in) / ZHOU XIONG (Autor:in)

    Erscheinungsdatum :

    2023-04-28


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Traffic flow prediction method based on graph space-time Transform model considering human mobility

    KONG XIANGJIE / ZHAO ZHENZHEN / SHEN GUOJIANG et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Highway traffic flow prediction method based on Transform and graph attention network

    JIANG CONG / SONG YUN / DENG ZELIN | Europäisches Patentamt | 2023

    Freier Zugriff

    Double-end graph convolution traffic flow prediction method with graph learning

    REN QIANQIAN / LI ZILONG / ZHANG YANG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Adaptive traffic signal control method combining traffic flow prediction and reinforcement learning

    PI JIATIAN / YANG XINMIN / WU CHANGZHI | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic flow prediction method based on Fourier-recursive neural network

    WANG CUN / DONG XIAOFEI | Europäisches Patentamt | 2020

    Freier Zugriff