Short-time forecasting method for traffic flow based on urban macroscopic road network model

Abstract

The invention relates to a short-time forecasting method for a traffic flow based on a macroscopic road network model, which comprises the following steps: (1) obtaining the input flow of the source node of a road network at a forecasting period, extracting the average speed of each road section at a previous forecasting period and determining the flow ratio of different turning directions at each intersection; (2) calculating the time of vehicles running to the tail of a queued vehicle queue, which are input on a road so as to obtain the number of the vehicles arriving at the tail of the queue at an iterative period; (3) determining the number of the vehicles which are correspondingly turned to leave the intersections by the conditions of the number of the vehicles queued at the intersections, the saturated leaving flow and the like; (4) accumulating to obtain the total number of the vehicles leaving the intersections at one forecasting period and converting to obtain the traffic flow within the forecasting period; and (5) updating the number of the queued vehicles as known data for iterative forecasting at the next time. The short-time forecasting method for the traffic flow based on the macroscopic road network model aims at the defects that the adaptability of the road network is poor, a great deal of training data are needed, the operation quantity in a microscopic model is large and the like, which exist in the prior art. The spatial information of an urban road network is fully utilized. The short-time forecasting method for the traffic flow based on the macroscopic road network model is based on a macroscopic traffic flow model, and the forecasting of the traffic flow of a road with high accuracy and good real-time property can be realized. Moreover, the short-time forecasting method for the traffic flow based on the macroscopic road network model is suitable for most of urban road networks.

Description

technical field [0001] The invention relates to a short-term prediction method for urban road traffic flow, in particular to a short-term prediction method for traffic flow based on an urban macroscopic road network model. Background technique [0002] With the rapid development of urban traffic, traffic congestion and other phenomena are becoming more and more common and serious, and the demand for intelligent urban traffic guidance system is becoming more and more urgent. As a key basic technology of the guidance system, urban road traffic flow in short Prediction has been a hot issue in the field of intelligent transportation systems in recent years. Short time generally refers to 5 to 15 minutes, and the forecasting indicators are generally average traffic, average speed or average delay. The current forecasting methods can generally predict the above three indicators, and there is no essential difference. [0003] In recent years, the academic community has proposed ma...

AI Technical Summary

Problems solved by technology

Traditional mathematical statistical models can no longer meet the requirements of high-precision forecasting. Although neural networks and support vector machines can achieve high precision, they require a large amount of data training and have poor portability.

Most of the above methods use the information in the time domain or frequency domain of traffic flow to process and predict, and do not make good use of road space information. The modeling of road network is divided into macro model and micro model. The micro model is very good. Considering randomness, car-following model, vehicle distribution and other microscopic information, it has high precision. The current mainstream simulation software such as VISSIM and CORSIM all use advanced microscopic models, but their computational complexity is large and the speed is slow, which cannot meet the prediction requirements. Dynamic release of information to meet the needs of traffic flow induction

Images