Theoretical,Experimental And Simulation Study On Traffic Oscillation Evolution And Suppression

Traffic flow theory is a theory that analyzes and studies the movement laws of motor vehicles,non-motor vehicles and pedestrians on the road to reduce travel delays,reduce accidents,and improve the efficiency of road traffic facilities.It can help understand and express the characteristics of traffic flow.It uses analytical methods to explain the mechanism of traffic phenomena and helps us better understand traffic phenomena and their nature.Only by fundamentally explaining the cause of traffic congestion,can we take effective traffic control measures from the source to provide a sustainable road for the development of economy,environment,and people’s livelihood.Traffic oscillation is an important characteristic of traffic flow.It refers to the repeated acceleration and deceleration of vehicles in the traffic flow,rather than maintaining a stable driving state.Traffic oscillation will lead to traffic capacity reduction,driving discomfort,additional fuel consumption,emissions and pollution,and bring more potential safety hazards.In the past,it was believed that traffic oscillation usually formed at the bottleneck or ramp of the road.However,recent experiments have shown that traffic oscillation may occur even without bottleneck and lane change.Using the relevant tools of system science can help us better study the relationship between individuals and systems,explore the evolution and suppression mechanism of traffic oscillation from the perspective of system,reveal the rules of traffic oscillation evolution process.Thus,we can better understand the traffic phenomenon and its essence,solve traffic congestion,improve driving safety,and provide theoretical basis for traffic engineering.In this context,from the perspective of system science,this dissertation mainly uses system theory,control theory,signal processing and other tools to carry out theoretical analysis of traffic oscillation evolution and suppression,carries out large-scale experimental research.Specifically,the main contents of this dissertation can be summarized as follows:(1)Theoretical analysis of traffic oscillation evolution under variable speed leading vehicle.The stochastic linear car-following model proposed by predecessors only considers the case where the leading vehicle moves constantly.In this dissertation,the speed of the leading car is extended to arbitrary function,the expression of speed standard deviation of each vehicle is derived.The influence of different leading car speed patterns on the evolution of traffic flow oscillation can be analyzed by theoretical analysis instead of numerical simulation.Which reveals: If the underlying deterministic car-following model is stable,the speed variance yielded by stochastic linear car-following model will converge to a constant value,which depends on the stochasticity strength,but is independent of the leading speed pattern;for deterministic linear car-following model,the traffic flow stability does not change if the speed time series of leading vehicle is scaled;the amplitude of traffic oscillations might change monotonically or nonmonotonically,which depends on the leading speed pattern.(2)Experimental study and simulations of traffic oscillations evolution under variable speed leading vehicle.In the field car-following experiment in which the leading vehicle drives with variable speeds,the first vehicle in the platoon is instructed to accelerate or decelerate to control the frequency and amplitude of speed oscillations,hence,the traffic oscillation evolution can be observed more clearly.By observing the number changes of the stripes in spatiotemporal diagram,it can be found that in some experiments,the spontaneous formation,propagation,and development of oscillations occurred the speed standard deviation showed a concave growth pattern;in some other experiments,the oscillations are dispersed.To better explain this phenomenon,this dissertation introduces the concept of central frequency.It is found that the decrease of central frequency corresponds to the formation of oscillations,while the increase of central frequency corresponds to the dispersion.The parameters of the deterministic carfollowing model and the stochastic car-following model are calibrated based on the experimental data.Compared with the deterministic model,the simulation results of the stochastic model are closer to the experimental ones,indicating that the stochasticity in the car following plays an important role in the evolution of traffic oscillations.(3)Experimental study and simulation of traffic oscillation suppression with speed limitation.To eliminate the interference factors,we carried out a field car-following experiment under constant speed leading vehicle.This experiment can control the exogenous disturbance and further study the endogenous evolution and propagation characteristics of the traffic oscillations.Combined with simulation,the evolution characteristics of traffic oscillations and the suppression effect under speed limit are revealed.The results show that the simulation results conducted by speed-limiting mechanism proposed under the 2D-IDM model can fit better with the experimental data;the linear stochastic model has certain limitations in describing the formation of traffic oscillations compared with the nonlinear stochastic model.