Traffic operation state prediction method and system for congested area

Abstract

The invention relates to the technical field of traffic prediction, in particular to a traffic operation state prediction method and system for a congested area. The method comprises the following steps: S1, acquiring regional traffic data information; S2, generating a plurality of congestion subareas according to the regional traffic data information; S3, performing road section speed predictionon each road section in each congestion subarea according to the prediction time period to obtain a road section prediction speed of the corresponding prediction time period; S4, carrying out the OD prediction according to the obtained prediction speed of the road section to obtain OD prediction parameters; S5, simulating the regional traffic operation state according to the road section prediction speed and the OD prediction parameters of the congestion subarea so as to predict the regional traffic operation state. According to the method, the dynamic characteristics of the road section association relationship are reserved by constructing the congestion subareas, the problem that a previous traffic simulation area is fixed is solved, the problem of simulation flow parameter calibration under data loss is solved by utilizing the generalization performance of the meta-learning model, and the traffic operation state of the congestion area can be better predicted.

Description

technical field [0001] The invention relates to the technical field of traffic prediction, in particular to a method and system for predicting traffic operation status in a congested area. Background technique [0002] With the rapid development of cities, traffic congestion is becoming more and more normal, and traffic management and control are facing major challenges. Due to the contradiction between limited road resources and people's growing travel needs, the urban road network often appears overloaded, and the resulting traffic congestion, traffic pollution and traffic safety problems have become the basic problems of urban development. [0003] In urban smart transportation, people realize the management and control of urban traffic through three stages: acquisition of basic urban traffic data, extraction of urban traffic operating status, and analysis and prediction of urban traffic operating status. The first two stages are for the analysis and prediction of traffic...

AI Technical Summary

Problems solved by technology

[0004] There are two deficiencies in the simulation prediction of the current traffic state. First, the traffic simulation is often aimed at the designated area and cannot be dynamically changed according to the change of the traffic state.

Due to the time-varying nature of the traffic state, the relationship between the road sections is also changing all the time. The simulation prediction of the designated area may ignore the important related road sections, resulting in misjudgment of the operating state; The data base, especially the cross-section flow, cannot be simulated for traffic detection equipment failures or partially missing road networks, or relies on huge parameter training, which takes a long time and has poor applicability

Unable to meet the needs of smart management and control of urban road networks

Images