Research On Parameter Calibration Of Car-following Model Based On Simulation Detector Data

Traffic flow theory is an important part of traffic engineering.The purpose is to analyze traffic characteristics,reveal the basic laws of traffic flow,and better improve road planning and design and system construction.Micro-traffic flow model takes a single vehicle as the research object,focusing on the micro-driving behavior of vehicles,which plays an important role in exploring the interaction between vehicles.Car-following model is one of the representatives of microscopic traffic flow model.By calibrating parameters of the car-following model,the actual traffic situation can be described reasonably,and the application of the car-following model from theory to practice can be realized.However,due to the influence of drivers,vehicle conditions and road environment,and the lack of appropriate detection equipment,the micro-behavior of vehicles can not be recorded,which brings some difficulties to the parameter calibration of car-following model.The research object of macro-traffic flow model is traffic flow,speed,density and other macro-parameters.It has the advantages of simple expression,clear physical meaning and simple calculation,and the macro-traffic flow parameters can be obtained directly by vehicle detectors on the road.Therefore,the parameter calibration of macro traffic flow model is more convenient and efficient than that of direct calibration of car-following model.Under the above background,a new method for calibrating parameters of car-following model using detector data is proposed.Based on the existing micro-car-following model GHR,a density-flow traffic flow model is designed by theoretical analysis and mathematical derivation to describe the macro-traffic flow in the car-following state.Then we use VISSIM,which is a simulation software,and use the vehicle detector function in the simulation software to obtain traffic flow parameters such as traffic volume,speed and so on,as the original data of parameter calibration.Finally,the optimization theory is used to solve the values of the parameters to be calibrated in the model,and the parameters of the model are calibrated,which verifies the scientificity and operability of the method proposed in this paper.The specific contents of this paper are as follows:(1)A density-flow traffic flow model based on GHR model is designed.The relationship between headway spacing,speed difference and acceleration of front and rear vehicles in the car-following model is transformed into the relationship between traffic flow density and flow in the section.The car-following model is described by density and flow,and the transition from micro model to macro model is realized.Conversion and docking.(2)On the basis of demonstrating the rationality of VISSIM simulation and conducting traffic investigation,the road traffic flow is simulated by using VISSIM software,and traffic flow parameters such as traffic volume,occupancy and speed are obtained through data acquisition point module.(3)Using optimization theory and method to calibrate the parameters of density-flow model based on GHR model proposed in this paper,using mathematical method to solve the parameters in the model,and verifying the accuracy of the parameters and the rationality of the model through error analysis.