Traffic speed prediction method and system and storage medium

Abstract

The invention discloses a traffic speed prediction method and system and a storage medium, and the method comprises the steps: collecting an original traffic speed data set, dividing the data set intoa training set and a test set, mining different spatial relations of the original traffic speed data set, and constructing two road network graphs; fusing the adjacent matrixes of the two road network graphs into a new graph adjacent matrix; taking the training set and the new graph adjacency matrix as input of a traffic prediction model, and training to obtain a prediction model; and predictingthe traffic speed by using the prediction model. The spatial relationship can be fully and comprehensively mined, the complexity of model processing is reduced, and the traffic speed is accurately predicted.

Description

technical field [0001] The invention relates to the field of traffic data processing, in particular to a traffic speed prediction method, system and storage medium. Background technique [0002] In recent years, with the rapid development of intelligent transportation systems and various positioning technologies such as Global Positioning System (GPS), mobile devices, etc., the availability of traffic data is increasing. Mining valuable knowledge from traffic data is crucial to many real-world applications, including smart transportation, urban planning, public safety, etc. Traffic prediction is of great significance to the realization of traffic guidance, travel planning and congestion control. So how to make real-time and accurate traffic forecast has become more and more concerned. [0003] The traffic forecasting problem is defined as predicting the traffic information of a certain time period in the future based on the historical traffic information in the road networ...

AI Technical Summary

Problems solved by technology

However, the existing graph convolutional neural network model does not fully mine the spatial characteristics of traffic speed data when predicting traffic speed. Most of them only consider the single spatial relationship of adjacent road sections or adjacent areas. There are many kinds of relationships. For example, there are road sections that are far apart on the map but have very similar traffic demand patterns or congestion patterns.

Or, although the existing ST-MGCN model has mined a lot of spatial information, there are problems such as high complexity of the prediction model, many parameters, and slow convergence.

Images