The invention provides a traffic flow data completion method and system based on tensor chain decomposition, and belongs to the technical field of data processing, and the method comprises the steps: collecting traffic flow data, and constructing a preset high-dimensional tensor of the traffic flow data based on time features and spatial features; performing tensor chain decomposition on the preset high-dimensional tensor to obtain a core tensor; constructing an objective function of the core tensor, and optimizing and updating the core tensor by using the objective function to obtain an optimal core tensor; and combining the optimal core tensor to obtain complemented traffic flow data. Aiming at the difficulty that traffic data is difficult to complement, the method effectively extracts the essential characteristics of the data, can accurately complete the data complementation task, and is lower in calculation complexity and more stable in performance in the recovery problem of high-dimensional data compared with other methods adopting tensor decomposition.

    本发明提供一种基于张量链分解的交通流量数据补全方法及系统,属于数据处理技术领域,包括:采集交通流量数据,基于时间特征和空间特征构建所述交通流量数据的预设高维张量;对所述预设高维张量进行张量链分解,得到核心张量;构建所述核心张量的目标函数,利用所述目标函数对所述核心张量进行优化更新,得到最优核心张量;合并所述最优核心张量,得到补全的交通流量数据。本发明针对交通数据难以补全的难点,有效提取出数据本质特征,能准确地完成数据补全任务,相对于其他采用张量分解方法,在高维数据的恢复问题中计算复杂度更低,且性能更加稳定。


    Access

    Download


    Export, share and cite



    Title :

    Traffic flow data completion method and system based on tensor chain decomposition


    Additional title:

    一种基于张量链分解的交通流量数据补全方法及系统


    Contributors:
    ZHU CHENLU (author) / YIN PU (author) / YANG ZECAN (author) / ZHANG LIJIE (author) / XIONG JIE (author) / RUAN MENGXIONG (author) / DENG XIANJUN (author)

    Publication date :

    2023-12-08


    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




    A tensor-based method for missing traffic data completion

    Tan, Huachun / Feng, Guangdong / Feng, Jianshuai et al. | Elsevier | 2012


    Tucker factorization-based tensor completion for robust traffic data imputation

    Lyu, Cheng / Lu, Qing-Long / Wu, Xinhua et al. | Elsevier | 2024


    A Tensor Completion-Based Traffic State Estimation Model

    Tan, Huachun / Song, Li / Cheng, Yang et al. | ASCE | 2014


    Graph Spectral Regularized Tensor Completion for Traffic Data Imputation

    Deng, Lei / Liu, Xiao-Yang / Zheng, Haifeng et al. | IEEE | 2022