| Connected and Automated Vehicles(CAV)has certain benefits in improving traffic operation efficiency and saving energy,but CAV is still a certain distance away from large-scale promotion and needs to go through the stage of mixed driving of CAV and human driven vehicles(HDV),but the mixed traffic flow of CAV and HDV The characteristics of mixed traffic flow of CAV and HDV are not yet clear,and the impact of human-driven vehicles on the road network needs to be explored.In this thesis,we study the multi-user mixed traffic equilibrium under the participation of CAVs,define the market penetration rate from the endogenous perspective,and establish a day-by-day dynamic evolution model of mixed traffic flow under the mixed driving environment,analyze the relevant characteristics of mixed traffic flow from the level of equilibrium evolution,in order to give relevant traffic management measures and suggestions for the new environment.First,considering the multi-dimensional benefits of CAVs in terms of capacity,time value,energy consumption,etc.,the travel cost calculation function under mixed driving is established,and then the SO-UE-SUE mixed traffic equilibrium model under the mixed driving environment of human and machine is constructed.Secondly,this thesis assumes that travelers will consider their own travel benefits in the real situation,reanalyzes the equilibrium market penetration of CAV from the perspective of endogenous variables,builds a model based on the difference of benefits brought by CAV to travelers,and gives an internal and external two-layer iterative algorithm for solving the two models.Numerical analysis shows that both the mixed equilibrium model and the market penetration model have faster convergence and better convergence,and the change of relevant parameters does not affect the convergence of the models.Finally,in order to explore the process of realizing the mixed equilibrium state of multi-user class travelers in the new environment,a day-by-day dynamic evolution model of mixed equilibrium in the human-machine mixed driving environment is constructed,and the learning process of perceived travel cost of three types of travelers is given,and then the path flow adjustment process is analyzed and the solution algorithm is designed based on the nonlinear object scale adjustment model.Numerical analysis proves that the day-by-day dynamic evolution model has good convergence,and sensitivity analysis shows that the higher the sensitivity of travelers’ response,the faster the model converges.This thesis considers the penetration rate of CAVs from the endogenous perspective and proposes a mixed equilibrium day-by-day dynamic evolution model for mixed human-machine driving environment,which provides a theoretical basis for future mixed traffic flow management. |
