The invention discloses a traffic flow speed prediction method based on feature attention, and relates to the technical field of time-space sequence data prediction methods, and the method comprises the steps: obtaining traffic flow data, making a data set, and carrying out the preprocessing; obtaining a data time feature matrix and a data space feature matrix, and dynamically fusing the data time feature matrix and the data space feature matrix through a gating fusion mechanism to obtain a spatial-temporal feature matrix; inputting the spatial-temporal characteristic matrix into a characteristic attention network, and updating the spatial-temporal characteristic matrix; inputting the updated spatial-temporal characteristic matrix into a neural network to generate prediction output; training a model based on a feature attention mechanism; and inputting the data into the trained model to obtain a prediction result. According to the method, the problems that part of feature information is lost in convolution pooling operation and errors are large during long-distance prediction are solved, the spatial-temporal features of the data are dynamically fused, the adaptive capacity of the model for obtaining the spatial-temporal information is higher, and the method has the advantages of being more stable in prediction result and higher in prediction precision.

    本发明公开了一种基于特征注意力的交通流速度的预测方法,涉及时空序列数据预测方法技术领域,包括获取交通流数据,制作数据集,并进行预处理;获取数据时间特征矩阵和数据空间特征矩阵,通过门控融合机制动态融合得到时空特征矩阵;将时空特征矩阵输入特征注意力网络,更新时空特征矩阵;将更新后的时空特征矩阵输入神经网络,产生预测输出;训练基于特征注意力机制模型;将数据输入训练好的模型中得到预测结果。本发明克服了卷积池化操作丢失部分特征信息而在长距离预测时误差大的问题,动态地融合数据的时空特征,使模型获取时空信息的适应能力更强,具有预测结果更稳定、预测精度更高的特点。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic flow speed prediction method based on feature attention


    Weitere Titelangaben:

    一种基于特征注意力的交通流速度预测方法


    Beteiligte:
    SUN YONG (Autor:in) / ZHANG ANQIN (Autor:in) / CHEN JINGJING (Autor:in)

    Erscheinungsdatum :

    08.08.2023


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    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




    Traffic flow prediction method based on feature attention decomposition and graph convolution network

    ZHENG LINJIANG / LIU YUMANG / TANG YI et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on LSTM-Attention

    QIN XIAOLIN / LIU JIACHEN / SONG LIXIANG et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Traffic flow speed prediction method based on attention space-time diagram convolutional network

    SUN YONG / ZHANG ANQIN / CHEN JINGJING | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic flow prediction method based on attention mechanism

    SUN LIJUN / LIU MINGZHI / LIU GUANFENG et al. | Europäisches Patentamt | 2022

    Freier Zugriff