The invention provides a spatio-temporal data prediction method and a data acquisition monitoring system, and the method comprises the steps: building a topological space diagram according to collected spatio-temporal data, and recording a diagram signal sequence of each node of the topological space diagram at different moments; constructing a spatio-temporal data prediction model based on the spatio-temporal generative adversarial network, and training the spatio-temporal data prediction model by taking the historical graph signal sequence as a training set; the space-time generative adversarial network comprises a generator and a discriminator, the generator is used for modeling a space-time dependency relationship in the space-time data, and the discriminator is used for regularizing the space-time generative adversarial network; and performing spatio-temporal data prediction by using the trained model to obtain a graph signal sequence of each node in a future preset time step. According to the method, the adversarial loss is introduced into the objective function of the prediction model, so that the uncertainty in modeling data is realized, real spatio-temporal data distribution is learned through the adversarial process, and the ability of the prediction model to learn data representation is enhanced, and therefore, the problem of too fast increase of multi-step prediction errors is relieved.

    本发明提供一种时空数据预测方法及数据采集监控系统,所述方法包括:根据采集的时空数据,建立拓扑空间图,记录不同时刻拓扑空间图各节点的图信号序列;构建基于时空生成对抗网络的时空数据预测模型,以历史图信号序列作为训练集,对时空数据预测模型进行训练;时空生成对抗网络包括生成器和判别器,生成器用于建模时空数据中的时空依赖关系,判别器用于对时空生成对抗网络进行正则化;利用训练后的模型进行时空数据预测,得到未来预设时间步各节点的图信号序列。本发明在预测模型的目标函数中引入对抗损失,以建模数据中的不确定性,并通过对抗过程学习真实时空数据分布,强化预测模型学习数据表征的能力,从而缓解多步预测误差增长过快的问题。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Spatio-temporal data prediction method and data acquisition monitoring system


    Weitere Titelangaben:

    一种时空数据预测方法及数据采集监控系统


    Beteiligte:
    WANG LING (Autor:in) / JIA GAOFENG (Autor:in)

    Erscheinungsdatum :

    05.12.2023


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / 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 spatio-temporal data prediction method, system, device and medium

    XIA YUANQING / YAN TIJIN / GONG HENGHENG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic flow prediction method for traffic flow spatio-temporal data information

    LIU PENG / CHU YUQUAN | Europäisches Patentamt | 2024

    Freier Zugriff

    Spatio-Temporal Prediction of METAR Data for Continental United States

    Thirumalainambi, Rajkumar | SAE Technical Papers | 2011


    Traffic flow prediction method, system and device based on spatio-temporal data and medium

    LING YEHUA / LI YONG / ZHANG YIBO et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Urban traffic prediction method and system based on spatio-temporal data flow fusion analysis

    REN MINGLUN / HUANG XIAODI / CHU WEI et al. | Europäisches Patentamt | 2020

    Freier Zugriff