Research On Modeling For Macroscopic Traffic Flow Based On Self-Stabilizing Control Strategy

The work of this dissertation aims to study the feasibility of control strategies to solve the problem of traffic road congestion: Based on the existing intelligent transportation application scenarios,this dissertation introduces traffic information(vehicle driving information or road traffic information)into vehicle control,using the theoretical research method of traffic flow,analyzing the causes of traffic congestion and other problems,and exploring the ability of vehicle control strategies to improve traffic problems after optimization,to alleviate the current situation of congestion from traffic flow substantially.In order to solve the implementation difficulties of cooperative driving control,this dissertation attempts to introduce the self-stabilizing control using the vehicle’s own information into the macroscopic traffic flow modeling.Considering the validity of existing models and practical application scenarios,improved macroscopic models are proposed for single lane,multi-lane and single lane with ramp groups traffic flow.Through the corresponding theoretical analysis and numerical simulation,it is of great theoretical significance and application value to discuss the impact of the introduction of self-stabilizing control on traffic flow.The details are as follows:(1)Due to the high reliability and real-time performance of data communication in cooperative driving control,this dissertation studies the improvement ability of macroscopic traffic flow stability by using the self-stabilizing control of the vehicle’s own historical data,and conducts in-depth research on macroscopic traffic flow modeling with single historical speed self-stabilizaing control.On the one hand,based on the introduction of the vehicle’s own single historical velocity information at the same time delay,a single historical velocity self-stabilizing homogeneous macroscopic traffic flow model is proposed.The linear stability analysis and nonlinear stability analysis of the model are carried out theoretically,and the neutral stability conditions of the traffic system and the KdV-Buergers equation describing the critical congestion and its corresponding solutions are obtained.Results show that the introduction of speed self-stabilizing at a single historical moment has a significant stabilizing effect on the instability of traffic flow caused by small perturbations.At the same time,the introduction of historical data has a good compensating effect on the variable delay of the traffic system.On the other hand,considering the difference in the time delay of vehicle selection of a single historical speed information,a heterogeneous macroscopic traffic flow model with long and short dual-delay speed self-stabilizing is proposed.The research results show that the higher the proportion of long-term self-stabilizing feedback vehicles in the mixed traffic flow,the more obvious the effect of self-stable control on the stability of the traffic flow.The average feedback duration of vehicles in the traffic system is the same,and the self-stable control has the same effect on the stability of traffic flow.(2)The stability control of a single historical data is easy to fail when data is lost,and the anti-interference ability in practical application scenarios is poor.This dissertation studies the comprehensive consideration of speed information at multiple historical moments as the information source of vehicle stability control.While the selfstable control improves the stability of the traffic flow,it also enhances the executable in the actual application process.A macroscopic traffic flow model for multi-history speed self-stabilizing control is proposed,and three weight distributions are used for the multi-history speed datasets,respectively.Three scenarios are considered: complete record of historical data,random loss of part of historical data,and compensation mechanism for lost data.The stability conditions under each scenario are obtained through linear stability analysis.The results show that the multi-history stabilization control of the reverse Poisson distribution shows better traffic flow stabilization ability in the three historical data set states.(3)In view of the frequent traffic bottlenecks formed by urban expressway ramps,the influence of self-stable control using single historical flow data on the stability of traffic system in and out of traffic is studied.A lattice hydrodynamic traffic flow model with a combination of one on-ramp and one off-ramp is proposed.The neutral stability condition of the model is obtained by using linear analysis theory.The effects of the ramp spacing,the flow of vehicles entering and leaving the main road on the ramp,and the stability control gain and feedback delay of the flow data at a single historical moment on the improvement of traffic flow stability are respectively studied.The results show that the out and on-ramps influence each other,resulting in traffic congestion.The introduction of historical flow self-stabilizing control can effectively improve the stability of traffic flow.(4)In order to further study the stabilization effect of the stabilization control on the traffic flow in the multi-lane traffic environment,this dissertation adopts the speed data and headway data of a single historical moment and proposes a macroscopic multilane traffic flow model based on the self-stabilizing control.Considering the limitations of the traditional multi-lane model lane changing conditions,this dissertation optimizes the multi-lane macro traffic flow model by considering the characteristics of the threelane traffic environment and taking the density of the entire road as one of the lane changing conditions.The research results show that the mechanism of historical speed auto-stabilization and historical headway self-stabilizing on traffic flow stability is different,and the introduction of the historical headway information of vehicles can further improve the stabilizing ability of speed self-stabilizing to traffic flow.The model considers the density of three lanes to effectively reduce the frequency of drivers changing lanes,thereby reducing the risk of collision and improving traffic safety,making the overall density evolution of the road more reasonable.