The invention provides an expressway traffic flow prediction method. The expressway traffic flow prediction method comprises: constructing data collected by all monitoring points into a graph structure, reconstructing time series data according to time units, and obtaining related space-time characteristics through time series convolution and graph convolution operation; embedding other data influencing the flow prediction result into a low-dimensional space by means of an embedding technology to obtain low-dimensional influence characteristics; combining low-dimensional influence characteristics, using an attention mechanism for endowing time series data weights under different time units, and finally obtaining a prediction result through a full connection layer by integrating other factor characteristics. Multi-modal data are adopted, neural network training is used, and the obtained traffic flow prediction result can accurately predict the high-speed traffic flow in the future timeslice according to the historical time slice.
本发明提出了一种高速公路交通流量预测方法包括:将各监测点所采集数据构建成图结构并按时间单位重构时序数据,并分别通过时序卷积和图卷积操作得到相关时空特征;借助嵌入技术将其他影响流量预测结果的数据嵌入至低维空间,得到低维影响特征;结合低维影响特征并运用注意力机制以赋予不同时间单位下时序数据权重,最后综合其他因素特征通过全连接层得到预测结果。采用多模态数据并使用神经网络训练,所得到的交通流量预测结果能够较为准确的根据历史时间片段预测未来时间片段高速交通流量。
Expressway traffic flow prediction method
一种高速公路交通流量预测方法
2020-09-25
Patent
Electronic Resource
Chinese
Expressway interleaving area traffic accident prediction method
European Patent Office | 2021
|Fluctuation in Expressway Traffic Flow
British Library Conference Proceedings | 2005
|