Study On The Nonlinear Modeling And Parametric Excitation Of Microscopic Traffic Flow

Transportation has been playing increasingly important role in the development ofsociety and economy, the study of traffic flow theory has been given much more attentionas the traffic congestion situation becoming worse and worse along with the accelerationof the process for urbanization. The macro-character of traffic flow is the collection ofmicro-characters of vehicle operation. The increase of flow reduces the independence ofvehicles, while their interactions become stronger. Under the impact of internal/externalfactors, the operation of individual vehicles will change, and it may lead to the formationand dissemination of traffic congestion. Hence, for the guidance of engineering practice, itis worthwhile to recognize the relationship between the internal/external excitationparameters of the traffic system and the traffic conditions. This thesis will build modelsand analyze the car-following as well as lane-changing behavior based on the optimalvelocity model, and study the impact of excitation parameters’ change about people, car,road, and environment to the operation of traffic flow.This thesis firstly introduces the basic parameters for traffic flow, and analyzes therelationship between velocity and headway, and characteristics of fundamental graphs.Meanwhile, the thesis reviews the nonlinear situations for traffic congestions, builds andclassifies concepts of excitation parameters, and proposes the basic method for the studyof parametric excitation.Secondly, we focus on existing problems of optimal velocity model, analyze thefunction mechanism of velocity difference, and build the cooperative car-following modelby regarding the various impacting factors during car operation process as effects of forces.Parameter standardization and fitting test show that this model is relatively advanced,which is able to describe the formation process of traffic congestion. Otherwise, impactsof individual difference of drivers, proportion of slow cars, and bad weather conditions(such as rain, snow, and fog) to the single lane traffic flow are analyzed, and somecorresponding suggestions for the traffic flow control are proposed.Thirdly, the cooperative car-following model is adapted to multi-lanes in this thesis,and the velocity and space superiority based lane-changing model according to driver’sreacting and recognizing process are proposed. Simulation results show that frequent lane-changing is an important factor that leads to traffic congestion, and it may also leadto the stop and go traffic situation for multi-lane condition. What’s more, cooperatingwith some normal situations, we also study the traffic flow changing characters underconditions such as ping pong lane-changing, lane reduction, and speed limitation.Finally, the relationship between traffic congestion situations for the merge zonearterials, queue length on ramps and traffic flow as well as the impact difference ofacceleration lane’s installation to the whole system and ramps are discussed. Thedeployment conditions for the acceleration lane which provides guidance for the mergezone traffic control are presented.