The invention discloses a highway traffic flow prediction method. The method comprises the following steps of: S1, acquiring data collected by each monitoring point to construct a graph structure, reconstructing time series data according to time units, acquiring a data set for model training and testing, and then acquiring each piece of time series information and space information through time series convolution, graph convolution and time series convolution operations; S2, embedding the external feature data influencing the flow prediction result into a low-dimensional space to obtain low-dimensional influence features; and S3, combining spatial-temporal features with the low-dimensional influence features to obtain a traffic flow prediction result through an attention mechanism. According to the invention, traffic flow prediction is carried out by combining the spatial-temporal features and the low-dimensional influence features of the data of each monitoring point with an attention mechanism, so that the accuracy of traffic flow prediction is improved.
本发明公开了一种高速公路交通流量预测方法,其中方法包括以下步骤:S1:获取各监测点所采集的数据来构建图结构并按时间单位重构时序数据,获得用于模型训练及测试的数据集,然后通过时序卷积、图卷积、时序卷积操作得到各个时序信息和空间信息;S2:将影响流量预测结果的外部特征数据嵌入至低维空间,得到低维影响特征;S3:时空特征与所述低维影响特征结合并通过注意力机制来获得交通流量预测结果。本发明通过各监测点数据的时空特征与低维影响特征结合注意力机制来进行交通流量预测,提高了交通流量预测的准确性。
Highway traffic flow prediction method
一种高速公路交通流量预测方法
2020-10-23
Patent
Electronic Resource
Chinese
Highway traffic flow prediction method based on weather conditions
European Patent Office | 2023
|Highway traffic flow speed prediction method based on traffic factor state network
European Patent Office | 2020
|Intelligent highway traffic flow prediction method based on continuous monitor
European Patent Office | 2022
|