The invention discloses an urban traffic flow prediction method and system based on graph knowledge distillation, and belongs to the technical field of smart city management. Aiming at the problems that a modern urban road network is complex, and the urban traffic flow is difficult to predict due to the constraint of a traffic road network topological structure and the rule that traffic changes along with time, the invention provides a new graph knowledge distillation model, and a lightweight structure perception MLPs student model is constructed by capturing node structure information and node feature information. And a layer-by-layer mapping distillation strategy is utilized to form a supervision signal of the pre-training teacher model to the lightweight student model, and finally the lightweight and interpretable urban traffic flow prediction method and system are obtained. The spatial structure network of the urban traffic road sections constructed by the invention not only can record the adjacent association information between the road sections, but also effectively expresses the spatial structure relationship between the road sections, and is beneficial to obtaining an urban traffic flow prediction result with higher robustness and interpretability.

    本发明公开了一种基于图知识蒸馏的城市交通流量预测方法及系统,属于智慧城市治理技术领域。针对现代城市路网错综复杂,由于交通路网拓扑结构的约束和交通随时间变化的规律,城市交通流量难以预测的问题,本发明提出一种新的图知识蒸馏模型,通过捕捉节点结构信息和节点特征信息来构建轻量级结构感知MLPs学生模型,并利用逐层映射蒸馏策略形成预训练教师模型对轻量级学生模型的监督信号,最终获得轻量且具可解释性的城市交通流量预测方法及系统。本发明构建的城市交通路段的空间结构网络,不但可以记录路段间的邻接关联信息,还有效表达了它们之间的空间结构关系,有助于获得鲁棒性和可解释性更强的城市交通流量预测结果。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Urban traffic flow prediction method and system based on graph knowledge distillation


    Weitere Titelangaben:

    一种基于图知识蒸馏的城市交通流量预测方法及系统


    Beteiligte:
    DU HANGYUAN (Autor:in) / YU RONG (Autor:in) / WANG WENJIAN (Autor:in) / BAI LIANG (Autor:in) / BAI LU (Autor:in) / JIN DI (Autor:in)

    Erscheinungsdatum :

    2024-03-08


    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 / 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



    Air Traffic Flow Prediction with Spatiotemporal Knowledge Distillation Network

    Zhiqi Shen / Kaiquan Cai / Quan Fang et al. | DOAJ | 2024

    Freier Zugriff

    Metro Traffic Flow Prediction via Knowledge Graph and Spatiotemporal Graph Neural Network

    Shun Wang / Yimei Lv / Yuan Peng et al. | DOAJ | 2022

    Freier Zugriff

    Topological Graph Convolutional Network-Based Urban Traffic Flow and Density Prediction

    Qiu, Han / Zheng, Qinkai / Msahli, Mounira et al. | IEEE | 2021


    Traffic flow prediction method, traffic flow prediction system and urban traffic signal controller control method

    YAN JUN / LIAO FUKUN / ZHANG YUANG | Europäisches Patentamt | 2023

    Freier Zugriff

    Urban traffic flow prediction method

    ZUO HONGNIAN | Europäisches Patentamt | 2021

    Freier Zugriff