The invention relates to the technical field of traffic flow prediction, in particular to a highway flow prediction method and system based on multi-source data fusion, and provides a flow prediction deep learning model based on adaptive spatial-temporal correlation modeling to realize more accurate multi-time-step flow prediction. According to the model, the spatial correlation of a flow sequence in a road network is more comprehensively captured based on a graph convolution network of an attention mechanism and a gating mechanism, and then the nonlinear correlation of the flow sequence at different time steps is adaptively modeled by using a multi-head self-attention mechanism; the complex space-time correlation in the road network information is further captured through a gating fusion module, and finally the flow of multiple time steps in the future is predicted through an encoder-decoder model. According to the accurate flow prediction value of the multiple time steps of the expressway, a traffic management department can be assisted to implement measures for ensuring smooth operation of the expressway, road network congestion can be assisted to be relieved, the travel satisfaction degree of tourists is improved, and the travel efficiency is improved.

    本发明涉及交通流量预测技术领域,尤其涉及一种基于多源数据融合的高速公路流量预测方法及系统,提出一种基于自适应时空相关性建模的流量预测深度学习模型实现更准确的多时间步流量预测,该模型基于注意力机制和门控机制的图卷积网络更全面地捕获路网中流量序列的空间相关性,然后利用多头自注意力机制自适应地建模流量序列在不同时间步的非线性相关性,通过门控融合模块进一步捕获路网信息中复杂的时空相关性,最后通过编码器‑解码器模型预测未来多时间步的流量。根据高速公路多时间步准确的流量预测值,可以辅助交通管理部门实施保障高速公路畅通运行的措施,并且可以辅助减轻路网拥堵、提高游客出游满意度、提升出行效率。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Expressway flow prediction method and system based on multi-source data fusion


    Weitere Titelangaben:

    一种基于多源数据融合的高速公路流量预测方法及系统


    Beteiligte:
    TENG ZHIWEI (Autor:in) / XU HUA (Autor:in) / YANG JIANGUO (Autor:in) / JIANG DEHONG (Autor:in) / DUAN HONGLIN (Autor:in) / WANG ZHENHUA (Autor:in) / WANG YANAN (Autor:in) / CAI CAN (Autor:in) / MA HAO (Autor:in) / HUANG YUGUANG (Autor:in)

    Erscheinungsdatum :

    2023-09-12


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

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



    Expressway toll station exit flow prediction method and system based on multi-source data fusion

    ZHANG JIAN / ZHOU KAICHENG / QIAN PINZHENG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Expressway lane-level speed prediction method based on multi-source data fusion

    SUN DIHUA / ZHAO MIN / ZHAO WENZHU et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Urban expressway traffic state estimation method based on multi-source data fusion

    LONG KEJUN / LIU YANG / WU WEI et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Expressway flow prediction method and system

    MA QIANG / DU ZHICHENG / LI JIN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Expressway traffic flow prediction method

    JIANG YUNLIANG / FAN JINBIN / ZHANG XIONGTAO et al. | Europäisches Patentamt | 2020

    Freier Zugriff