Short-term traffic flow prediction method based on dynamic space-time correlation feature optimization

Abstract

The invention relates to a short-term traffic flow prediction method based on dynamic space-time correlation feature optimization. The method comprises the following steps of: acquiring ETC portal data and a prediction position; constructing a road network topological structure according to the ETC portal frame data; screening out a local road network position set according to the road network topological structure; quantifying traffic flow dynamic space-time correlation of the predicted position, an upstream position and a downstream position; extracting corresponding dynamic space-time correlation characteristics according to the traffic flow dynamic space-time correlation; constructing an original training tensor according to the dynamic space-time correlation features, optimizing the original training tensor by using a principal component analysis method, and determining an input feature vector; and inputting the input feature vector into a traffic flow prediction model, and outputting a corresponding traffic flow prediction value. According to the short-term traffic flow prediction method, the feature that the closure of the road network is relatively high and the traffic flow of a single position and the traffic flow of a local road network have high correlation are utilized, and the method can be combined with congestion early warning, so that reference is provided for management decision, and the dispatching, coordination and management efficiency of the expressway is improved.

Description

Technical field [0001] The present invention relates to the field of traffic management, and more particularly to a short-time traffic flow prediction method based on dynamic space-related feature optimization. Background technique [0002] Short-run traffic flow refers to the number of vehicles passing through a certain road in a short period of time (generally 5 minutes to 20 minutes). A short-time traffic flow prediction refers to a prediction of traffic flows within a particular location or a plurality of specific locations (typically take 5 minutes to 20 minutes). Due to the large fluctuation of the short-time traffic flow, it is more difficult to predict. However, short-term traffic flow is characterized by its timeliness. It is not only for real-time management of transportation systems, but also beneficial to urban residents, it has become increasingly valued, gradually become an important foundation for intelligent transportation systems and urban calculations. [0003] ...

AI Technical Summary

Problems solved by technology

[0004] In view of this, it is necessary to provide a short-term traffic flow prediction method based on dynamic spatio-temporal correlation feature optimization to solve the problem of how to efficiently and accurately predict short-term traffic flow

Images