A traffic flow prediction method based on a Transform dynamic graph convolutional network comprises the following steps that original data are divided into long-term data and adjacent data, the long-term data comprise day periods and week periods of multiple time periods, and the day period and the week period of one time period are input into a Transform model to be preprocessed; performing period trend feature extraction on the preprocessed day period and week period through a period feature extractor; performing long-term feature extraction on the long-term data through a long-term extractor; inputting the two extracted features and the adjacent data into a feature fusion module for fusion; and inputting the fused feature trend into the GCN for prediction processing so as to generate a final prediction result. According to the method, multiple features are fused, so that the stable mode and abnormal fluctuation in the data can be recognized, the model can quickly respond to short-term change, and the recognition capability on a complex traffic mode and the resistance to noise are improved.

    一种基于Transformer动态图卷积网络的交通流预测方法,步骤如下:将原始数据分为长期数据和临近数据,其中长期数据包括多个时段的日周期和周周期,将其中一个时段的日周期和周周期输入Transformer模型预处理;将预处理后的日周期与周周期通过周期特征提取器进行周期趋势特征提取;将长期数据通过长期提取器进行长期性特征提取;将上述所提取出两项特征和临近数据输入至特征融合模块进行融合;将融合后的特征趋势输入GCN进行预测处理,以生成最终的预测结果。本发明通过融合多种特征,不仅能够识别出数据中的稳定模式和异常波动,使得模型能够迅速响应短期变化,而且提高了对复杂交通模式的识别能力以及对噪声的抵抗力。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Transform dynamic graph convolutional network-based traffic flow prediction method


    Weitere Titelangaben:

    一种基于Transformer动态图卷积网络的交通流预测方法


    Beteiligte:
    ZHU YANING (Autor:in) / ZHOU YI (Autor:in) / TONG LIHONG (Autor:in) / LI PEIYU (Autor:in)

    Erscheinungsdatum :

    24.06.2025


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung



    Traffic flow prediction method of graph convolutional network based on double-wavelet transform

    NIU BINGXIN / YU KEXIN / MIAO RUIHENG et al. | Europäisches Patentamt | 2025

    Freier Zugriff

    Traffic flow prediction method based on multi-view dynamic graph convolutional network

    HUANG XIAOHUI / YE YUMING / LING JIAHAO et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Traffic flow prediction method based on multimode dynamic memory graph convolutional network

    HUANG XIAOGE / YANG WENZHUO / ZHOU ENZHOU et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on adaptive dynamic fusion graph convolutional network

    ZHANG SHUAI / YU WANGZHI / LEE HAE KWANG et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on graph convolutional network

    XU HUI / MENG FANYU / REN QIANQIAN et al. | Europäisches Patentamt | 2025

    Freier Zugriff