Real-time Dynamic Forecasting System for Large-Scale Road Network Traffic Flow

Abstract

The invention provides a large-scale road network traffic flow real-time dynamic prediction system, comprising traffic flow detectors, a data fusion module, a traffic flow parameter prediction module, a traffic flow prediction model study module, a traffic geographic information platform and a traffic flow state classified evaluation module, wherein the traffic flow detectors are used for collecting the traffic flow parameters of a plurality of road conditions; the data fusion module is used for fusing the traffic flow parameters; the traffic flow parameter prediction module is used for predicting the road conditions of the next time period according to historical data and fusion data; the traffic flow prediction model study module is used for comparing real-time detected traffic flow parameters and predicted information and adjusting a prediction model according to the comparison result; the traffic geographic information platform is used for displaying the current state and the predicted state of the road traffic flow; and the traffic flow state classified evaluation module is used for evaluating the current and predicted road conditions according to the traffic flow parameters and controlling the roads according to the evaluation result and the predicted information. According to the system provided by the embodiment of the invention, the road condition information is predicted through the real-time collected traffic flow parameters and the time-space relation of the road traffic flow, and the corresponding road segments or areas can be controlled according to the predicted information.

Description

technical field [0001] The invention relates to the technical field of traffic data processing and supervision, in particular to a real-time dynamic prediction system for large-scale road network traffic flow. Background technique [0002] With the development of intelligent transportation systems, the number of traffic flow detectors in cities is increasing, which can provide more and more rich historical and real-time traffic flow information for road traffic managers and travelers. At the same time, using historical and real-time traffic information to predict future traffic conditions, on the one hand, traffic managers can take effective management measures in advance for the upcoming traffic congestion, so as to avoid the occurrence of large-scale congestion or reduce the degree of congestion; on the other hand, it can Make traffic travelers choose travel routes and travel time in a targeted manner, effectively avoid congested areas and congested time, improve travel ef...

AI Technical Summary

Problems solved by technology

[0004] 1) In the input of the prediction model, only a certain type of data is often considered, for example, the real-time data of the current location, but the laws implied in the spatio-temporal relationship of traffic flow are not fully exploited

[0005] 2) The effectiveness of the prediction model is not high, and it has good efficiency when predicting one or two locations, but it is difficult to meet the requirements of traffic management and traffic information services in terms of real-time prediction of hundreds of locations in the transportation network of large cities. real-time requirements

[0006] 3) There is a lack of effective real-time dynamic application systems and technologies, and the solution to these problems will be of great significance for improving the management and service levels of intelligent transportation systems

Images