The invention discloses a traffic flow prediction method based on variational self-encoding learning. The method comprises the steps of data pre-processing, training set construction, model initialization, variational self-encoder learning and model iteration. According to the method, an encoder and decoder structure is utilized, a variational time attention learning mechanism of an encoder and a decoder is expanded based on bidirectional GRU so as to automatically learn the implicit representation and the nonlinear correlation depth features in multivariable traffic flow correlation sequence data, and compared with a traditional method, the prediction result is more accurate; meanwhile, in the output reconstruction process of the encoder and the decoder, the attention mechanism of multivariable traffic flow sequence data is introduced, and the variational self-encoding learning is combined, so that the deep features in the traffic flow related data can be captured more accurately, and the traffic flow prediction under the nonlinear correlation influence condition of the multivariable traffic sequence data is more accurate.

    本发明公开了基于变分自编码学习的交通流预测方法,其步骤有:数据预处理、构造训练集、模型初始化、变分自编码器学习和模型迭代。该方法利用编码器解码器结构,基于双向GRU并扩展编码器解码器的变分时间注意力学习机制,以自动学习多变量交通流相关序列数据中的隐含表示和非线性相关深度特征,相比传统方法预测结果更准确;同时,在编码器解码器输出重构过程中,引入了多变量交通流序列数据的注意力机制,结合变分自编码学习,使其能更准确地捕获交通流相关数据中的深层次特征,使得多变量交通序列数据非线性相关影响条件下的交通流预测更为准确。


    Access

    Download


    Export, share and cite



    Title :

    Traffic flow prediction method based on variational self-coding learning


    Additional title:

    基于变分自编码学习的交通流预测方法


    Contributors:
    FENG XUAN (author) / CAI YOUBAO (author) / JIAO WEICHAO (author) / WANG SILIANG (author)

    Publication date :

    2022-02-18


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    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 / 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 via Variational Bayesian Inference-based Encoder-Decoder Framework

    Kong, Jianlei / Fan, Xiaomeng / Jin, Xue-Bo et al. | ArXiv | 2022

    Free access

    Traffic flow prediction method based on self-attention mechanism and deep learning

    JIE YURUI / HAO CHUNLIN / ZHANG JIAN | European Patent Office | 2024

    Free access

    A Variational Bayesian Inference-Based En-Decoder Framework for Traffic Flow Prediction

    Kong, Jianlei / Fan, Xiaomeng / Jin, Xuebo et al. | IEEE | 2024


    Traffic flow prediction method based on deep learning

    XIE GANG / WANG HAIYING / XIE RUIQI | European Patent Office | 2024

    Free access