The invention discloses a traffic flow data interpolation method and system based on tensor completion and a graph attention network, and the method comprises the steps: obtaining traffic flow data, taking a traffic information detector in a road network as a graph node, and constructing a data matrix; performing preliminary data completion by using a low-rank tensor completion algorithm; according to the acquired traffic flow data, applying a Pearson's correlation coefficient to construct a spatial correlation road network; taking the traffic flow data after preliminary complementation as input data, and combining the trained Graph SAGE and GAT with a neighbor node set of nodes in a spatial correlation road network to perform road network feature extraction; and inputting the extracted road network features into the trained generative adversarial network, and generating traffic flow data after data interpolation by using the generative adversarial network. According to the method, the accuracy and efficiency of traffic flow data interpolation are improved, and important support is provided for development of an intelligent traffic system and research and application in related fields.
本发明公开一种基于张量补全和图注意力网络的交通流量数据插补方法及系统,该方法包括:获取交通流量数据,并将道路网络中的交通信息检测器视为图节点,构建数据矩阵;使用低秩张量补全算法进行初步数据补全;根据获取的交通流量数据,应用皮尔逊相关系数构建空间相关性路网;将初步补全后的交通流量数据作为输入数据,结合训练后的Graph SAGE和GAT,以及空间相关性路网中节点的邻居节点集合,进行路网特征提取;将提取的路网特征输入训练后的生成对抗网络,利用生成对抗网络生成数据插补后的交通流量数据。本发明不仅提高了交通流数据插补的准确性和效率,还为智能交通系统的发展及相关领域的研究和应用提供了重要的支持。
Traffic flow data interpolation method and system based on tensor completion and graph attention network
一种基于张量补全和图注意力网络的交通流量数据插补方法及系统
14.06.2024
Patent
Elektronische Ressource
Chinesisch
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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung |
Missing Traffic Data Imputation based on Tensor Completion and Graph Network Fusion
Transportation Research Record | 2025
|Traffic flow data completion method and system based on tensor chain decomposition
Europäisches Patentamt | 2023
|Neural tensor ring fusion method for traffic flow data completion
Europäisches Patentamt | 2025
|A tensor-based method for missing traffic data completion
Online Contents | 2013
|