The invention provides a traffic flow prediction method combining spatial multi-scale and time sequence multi-granularity structures, and belongs to the field of spatio-temporal data mining. The method comprises the following steps: firstly, preprocessing original data, and selecting a proper space-time aggregation method to obtain multi-spatial scale data and multi-granularity time sequence data; secondly, inputting multi-spatial scale data under original data time sequence granularity into a multi-spatial feature mining module, and learning and fusing multi-scale spatial association among traffic units; then, inputting multi-time-sequence granularity data under multiple spatial scales into a multi-time feature mining module, and learning and fusing multi-granularity traffic tense information; and finally, through a multi-spatial-temporal feature fusion module, effectively aggregating multi-spatial-scale and multi-temporal traffic mode features, and outputting a final prediction result. Compared with a traditional method in the embodiment, the traffic flow prediction method has an obvious advantage in prediction precision, the traffic flow prediction efficiency can be effectively improved, and efficient prediction can be achieved.

    本发明提供了一种结合空间多尺度与时序多粒度结构的交通流量预测方法,属于时空数据挖掘领域。首先,对原始数据进行预处理,选取合适的时空聚合方法,得到多空间尺度数据以及多粒度时序数据;其次,将原始数据时序粒度下的多空间尺度数据输入多空间特征挖掘模块,学习并融合交通单元间的多尺度空间关联;然后,分别将多空间尺度下的多时序粒度数据输入多时间特征挖掘模块,学习并融合多粒度交通时态信息;最后,通过一个多时空特征融合模块,有效聚合多空间尺度、多时态的交通模式特征,输出最后的预测结果。通过在实施例中与传统方法比较,本发明在预测精度上有明显优势,而且能够有效提升交通流量预测效率,可以实现高效预测。


    Access

    Download


    Export, share and cite



    Title :

    Traffic flow prediction method combining spatial multi-scale and time sequence multi-granularity structure


    Additional title:

    结合空间多尺度与时序多粒度结构的交通流量预测方法


    Contributors:
    YU WENHAO (author) / ZHAO XINRU (author)

    Publication date :

    2023-12-05


    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Traffic flow prediction method based on multi-scale spatial feature mining

    SONG YUN / BAI XINKE / DENG ZELIN et al. | European Patent Office | 2023

    Free access

    Traffic speed prediction method based on multi-spatial scale space-time Transform

    ZHANG YUE / LIU BAILONG / AN JIYONG et al. | European Patent Office | 2023

    Free access

    Identified Traffic Partition Grooming in Multi-Granularity Optical Network

    Hou, Wei Gang ;Guo, Lei ;Wang, Xing Wei | Trans Tech Publications | 2012


    Road traffic operation situation multi-time scale prediction method

    LYU LIANGXUAN / WU YANHONG | European Patent Office | 2020

    Free access