The invention discloses a traffic flow speed prediction method based on an attention space-time diagram convolutional network. The method is characterized by comprising the following steps: preprocessing a data set; inputting the data set into a dynamic convolution attention module to obtain time features of the data; inputting the data set into a spatial feature capturer to obtain spatial features of the data; dynamically fusing the acquired data time features and spatial features through a gating mechanism to obtain a spatial-temporal feature matrix; inputting the spatial-temporal characteristic matrix into a neural network to generate prediction output; training and storing the model; and inputting the test set into the trained model to obtain a prediction result. According to the method, the capability of capturing the spatial correlation of the traffic flow speed data by the model is enhanced, and the problem of large error in long-term prediction of the road network traffic flow speed data by the existing method is solved.
本发明公开了一种基于注意力时空图卷积网络交通流速度的预测方法,其特征在于:数据集预处理;将数据集输入动态卷积注意力模块,获取数据的时间特征;将数据集输入空间特征扑捉器,获取数据的空间特征;将获取的数据时间特征和空间特征特征通过门控机制动态融合得到时空特征矩阵;将时空特征矩阵输入神经网络,产生预测输出;训练并保存模型;将测试集输入训练好的模型得到预测结果。本发明增强了模型捕获交通流速度数据的空间相关性的能力,解决了现有方法在长期预测路网交通流速度数据存在误差较大的问题。
Traffic flow speed prediction method based on attention space-time diagram convolutional network
一种基于注意力时空图卷积网络交通流速度的预测方法
2023-09-19
Patent
Elektronische Ressource
Chinesisch
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 |
Europäisches Patentamt | 2023
|Traffic flow prediction method based on space-time diagram convolutional network
Europäisches Patentamt | 2023
|Traffic flow prediction method based on space-time diagram convolutional network
Europäisches Patentamt | 2022
|Traffic speed prediction method based on multi-space-time diagram fusion convolutional network
Europäisches Patentamt | 2023
|Traffic flow prediction method of space-time diagram convolutional network
Europäisches Patentamt | 2024
|