Random user balance day-by-day dynamic traffic flow prediction method considering influence of overlapping paths

Abstract

The invention provides a novel random user balance day-by-day path traffic flow prediction method considering the influence of overlapping paths. The method can simulate a day-by-day evolution processof traveler path flow under random user balance. According to the method, through an initial network path flow mode obtained through investigation, a path flow mode on an arbitrary N(th) day thereafter can be predicted, so that guiding opinions are provided for formulation and implementation of traffic control measures in a future period of time. Through the method, the influence of path overlapping on a traveler path selection behavior is considered by defining overlapping cost, therefore, a model is more approximate to the actual situation, and the reliability of flow prediction is improved; besides, the value of a flow regulation ratio can be regulated automatically along with development of the evolution process; and therefore, the trouble of investigation is omitted, and the method better conforms to the actual situation.

Description

technical field [0001] The invention belongs to the technical field of traffic flow distribution forecasting, and in particular relates to a random user equilibrium daily dynamic traffic flow forecasting method considering the influence of overlapping paths. Background technique [0002] The traffic distribution model is used to predict the link flow or path flow in the traffic network under the equilibrium state. The traditional traffic allocation model assumes that the traveler's route selection behavior satisfies the user equilibrium allocation model (User Equilibrium referred to as UE) or the stochastic user equilibrium allocation model (Stochastic User Equilibrium referred to as SUE). Among them, the SUE model assumes that people's perception of the travel time of the route is erroneous. Therefore, in the SUE solution point, no traveler can reduce his understanding of the travel time by unilaterally changing the route. [0003] The traditional traffic allocation model ...

AI Technical Summary

Problems solved by technology

[0004] Among them, the daily dynamic evolution model can be divided into path model and road section model according to the different carriers of traffic flow evolution. The path model uses path flow as a variable. As traffic is a variable, this type of model cannot accurately describe the individual behavior of travelers, but can only reflect the macroscopic situation of traveler's route choice

[0005] Since the traditional stochastic user daily dynamic evolution model has the independence of uncorrelated alternatives, that is, it does not consider the overlap (similarity) problem between pat

Images