The invention discloses a flow prediction method for an intelligent traffic control system. The method comprises the following steps: constructing an urban road network structure by using a graph structure; road space features are extracted by using a graph convolutional network, preliminary time feature extraction is performed through a bidirectional long-short-term memory network, and strong correlation time features are filtered and extracted by using a backbone network; generating a feature matrix based on the road space features and the time features; on the basis of the feature matrix, a traffic data time sequence predicted in the future is generated through a mapping function; and setting a network training target, and performing network training by using an absolute value loss function. According to the method, the bidirectional long-short-term memory network is utilized, so that the time feature extraction capability is enhanced, and the capability of capturing the dynamic change of traffic data on a road by the flow prediction method is improved; a residual network jumper structure is adopted, so that the problem of gradient disappearance caused by neural network model stacking is effectively relieved, and the prediction method has better prediction performance; and by setting an absolute value loss function, the overall prediction precision is improved.

    本发明公开了一种面向智能交通控制系统的流量预测方法包括:利用图结构构建城市道路网络结构;利用图卷积网络提取道路空间特征,通过双向长短期记忆网络进行初步时间特征提取,并利用主干网络过滤提取强相关时间特征;基于道路空间特征和时间特征生成特征矩阵;基于特征矩阵,利用映射函数生成未来预测的交通数据时间序列;设置网络训练目标,利用绝对值损失函数进行网络训练。本发明利用双向长短期记忆网络,增强了时间特征提取能力,提升了流量预测方法捕捉道路上交通数据的动态变化的能力;采用残差网络跳线结构,有效缓解神经网络模型堆叠导致的梯度消失问题,使得预测方法具有更好的预测性能;通过设置绝对值损失函数,提高了整体预测精度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Flow prediction method for intelligent traffic control system


    Weitere Titelangaben:

    一种面向智能交通控制系统的流量预测方法


    Beteiligte:
    XU XIAOLONG (Autor:in) / SHI QIUHAO (Autor:in)

    Erscheinungsdatum :

    2022-12-13


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    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 method and system in intelligent traffic system

    DING FENG / XIAO TIAOZONG / GUO ZHIWEI et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction system and method for intelligent traffic

    LI YUEYING / ZHANG SHUWEN / QIN SHIJIE et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic light intelligent control method and system based on traffic flow prediction

    YAN JUN / HUO JIANJIE | Europäisches Patentamt | 2024

    Freier Zugriff

    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

    2-dimensional flow prediction based intelligent urban traffic-flow prediction method

    ZHAO LEI / SU QINGGANG / WANG ZHENYI et al. | Europäisches Patentamt | 2016

    Freier Zugriff