Research On Day To Day Traffic Allocation Of Congested Traffic Network Under The Environment Of Internet Of Vehicles

In recent years,the continuous increase in the number of cars has led to an increasing trend in the traffic demand of the urban network.However,the existing traffic infrastructure cannot meet the increasing traffic demand,so the urban traffic network becomes congested.With the rapid development of science and technology,traffic information technology has been used to alleviate the problem of traffic network congestion.At the same time,the traffic network planning under the background of the Internet of Vehicles(Io V)technology is also affected by traffic information.Information induces travelers to make route decisions,which affects the daily evolution of the traffic network.However,the impact of information in the actual traffic network on travelers needs to be described by a more realistic modeling system.Therefore,traveler categories are divided according to the types of vehicles used by users and the degree of information compliance in this paper,and further a daily traffic allocation model is established based on traffic network theory,which describes the dynamic evolution process of the traffic network under the premise that nonconnected vehicles and connected vehicles are mixed.(1)According to the types of vehicles used by users,users are divided into non connected vehicle travelers and connected vehicle travelers.Then,considering the compliance degree of connected vehicle users to the information,connected vehicle users are further divided into connected vehicle travelers who fully comply with the information and connected vehicles who do not fully comply with the information.Then,a mixed day-to-day traffic allocation model considering compliance degree is constructed based on the traffic network equilibrium theory.The models are mainly divided into route selection models,traffic flow adjustment models and empirical learning models.In the route selection model,non-connected vehicle users and connected vehicle users follow the stochastic user equilibrium(SUE)principle and the stochastic system optimal(SSO)principle to select routes respectively.The selection probability model of the three types of users is logit model.In the flow adjustment model,after all travelers complete the route selection,the traffic network flow is adjusted according to the proportion of each traveler.In the empirical learning model,In the experiential learning model,the three types of users update their routes according to the forecast information and historical travel information,Io V information,as well as the combination of the previous two historical information and Io V information.(2)In order to describe the characteristic that the road travel time under the congested traffic network increases with the decrease of flow,a more appropriate impedance function of the congested road is adopted,which introduces the vehicle evacuation flow for feature analysis.The mixed day-to-day traffic allocation model considering compliance degree is further improved and the congestion traffic network mixed day-to-day traffic allocation model is constructed,which is also composed of route selection model,flow adjustment model and experience learning model.Combined with a real traffic network example,on the one hand,the random error parameters,the proportion of connected bus users,the dependence degree of historical travel experience information and the compliance degree of Internet of vehicles travel information were adjusted to analyze the characteristics of the equilibrium traffic network and the traveler evolution characteristics.On the other hand,the flow characteristics of the two models under different equilibrium states on the same road network were compared.(3)The model calculation results show that the traffic network finally reaches a stable state under the mixed travel conditions of travelers,which is related to the random error parameters.Compared with the single-principle day to day traffic allocation model,the mixed-flow daily traffic allocation model compared with the SSO principle daily traffic allocation model,the number of days required to reach equilibrium is shortened by 20 days.Compared with the SUE principle daily traffic allocation model,the stability of the equilibrium traffic network is improved by26.22%.In the day to day traffic allocation model of the congested traffic network,the traffic flow and the number of vehicles in the equilibrium congestion state are more evenly distributed than those in the equilibrium state of the smooth traffic network,and the eventually equilibrium traffic network is more stable.(4)Studying the impact of the addition of connected vehicles to urban traffic on traffic network characteristics and travelers’ behavior is conducive to providing a more realistic modeling scheme for future connected vehicle applications,and planning traffic decisions more suitable for mixed travel,thereby further mitigating traffic congestion.