Traffic flow prediction method based on graph discrete attention network, medium and equipment

Abstract

The invention discloses a traffic flow prediction method based on a graph discrete attention network, a medium and equipment, and the method comprises the steps: carrying out the flow statistics of traffic big data, and carrying out the short-time prediction of future traffic flow according to a designed algorithm model. According to the method, the time and space characteristics of traffic flow are comprehensively considered, the space characteristics are represented through a graph discrete attention mechanism, and the time sequence characteristics are represented by using the architecture of the multi-layer encoder sequence to the multi-layer decoder sequence, so that a complete traffic flow model is constructed, and a road traffic flow prediction model can be obtained through a training algorithm model. The result shows that the model constructed by the invention can accurately predict future traffic flow data of the traffic monitoring points and can characterize the dynamic change of the flow between the traffic monitoring points.

Description

technical field [0001] The invention belongs to the field of digital intelligent transportation, and in particular relates to a traffic flow prediction method, medium and equipment based on a graph discrete attention network. Background technique [0002] In the past few decades, the number of cars in my country has increased year by year, and it is expected that in 2022, the number will reach more than 300 million. At the same time, the traffic demand is also increasing day by day, which makes the current road traffic load increase day by day, which brings a series of problems such as congestion and accidents. Although the traffic control department has taken measures to relieve traffic congestion to a certain extent, such as road construction, vehicle number restriction, etc., the traffic congestion status has not been well improved. [0003] Traffic flow prediction and control is the core problem of solving traffic efficiency. Making reasonable decisions in advance accor...

AI Technical Summary

Problems solved by technology

However, these existing advanced traffic flow forecasting models use static and predefined graph structures. The literature Bai, Lei, et al. "Adaptive graph convolutional recurrent network for traffic forecasting." Advances in Neural Information Processing Systems 33(2020 ):17804-17815. Try to change the graph structure into a dynamically updated parameter, which improves the prediction accuracy, but makes the stability of the graph structure information decrease

Images