Modeling And Simulation Of Cellular Automata Research On Evolution Mechanism Of Traffic Oscillation

Traffic oscillation is a common traffic flow phenomenon on high way and expressway in cities which often causes traffic congestion.Traffic oscillation increases travel delays and fuel consumption during vehicle travel,reduces driver's driving comfort,and increases the probability of traffic accidents.At present,the formation and propagation mechanism of traffic oscillation needs to be further improved.The study of traffic oscillation is a further expansion of the research on traffic flow theory and can also provide decision support for the relief of traffic congestion.This paper extracts the relevant trajectory data of traffic oscillation,analyzes the macroscopic evolution characteristics of traffic oscillation growth and propagation,and analyzes the growth propagation mechanism of traffic oscillation from the perspective of car following behavior.By constructing a cellular automaton car following model considering traffic hysteresis and wave speed differences,the growth and propagation characteristics of traffic oscillation are simulated,and the correctness of the analysis of the evolution mechanism of traffic oscillation is verified in this paper.The paper first uses NGSIM data as the data source,uses wavelet energy analysis to extract the trajectory data related to traffic oscillation,statistically analyzes the standard deviation of vehicle speed in the fleet,and verifies the standard deviation of vehicle speed along the concave growth pattern of the fleet.In addition,it was found that the standard deviation of vehicle speed may decrease during the decay period of traffic oscillation.Further analysis of the speed difference between acceleration and deceleration waves of traffic oscillation verified that in the predecessor stage,the speeds of deceleration and acceleration waves are almost zero;in the development stage,the deceleration waves propagate upstream at a speed slightly less than the acceleration waves.In addition,it was found that when the traffic oscillation showed signs of decline,the deceleration wave propagated upstream at a speed greater than the acceleration wave.According to the growth characteristics and propagation characteristics,the traffic oscillation is divided into the precursor stage,the development stage,the stability stage and the recession stage.Then,based on theoretical and measured data analysis,the positive correlation between the hysteresis amplitude of the traffic hysteresis and the amplitude increase of the traffic oscillation is tested,and the traffic hyster esis is classified into positive hysteresis,general hysteresis,and anti-hysteresis according to the hysteresis amplitude.Drivers are divided into aggressive,Normal and timid types.Statistical analysis of the types of hysteresis of various types of dri vers at various stages of traffic oscillations reveals that aggressive drivers are more inclined to adopt positive hysteresis and this tendency has not changed significantly at all stages;Normal drivers and timid drivers are more prone to anti-hysteresis after the stable stage of traffic oscillation.By calculating the simplified wave speed in each stage,it is found that when the deceleration wave is larger than the acceleration wave,the traffic oscillation increases and vice versa.Comparing the speed of the deceleration wave with the starting wave speed of the deceleration phase,the early reaction phenomenon and the late reaction phenomenon are found,revealing that the driver's early reaction may be the reason why the deceleration wave is larger than the acceleration wave in the recession phase.Finally,by simulating the driver's acceleration and deceleration decision-making process,the positive hysteresis and anti-hysteresis update rules are constructed.The positive hysteresis update rules apply to aggressive drivers,while the anti-hysteresis update rules apply to timid and normal drivers.And according to the acceleration and deceleration wave speed difference in different stages of traffic oscillation,the driver's reaction time is corrected.By simulating the constructed cellular automaton model,the growth and propagation characteristics of traffic oscillation are simulated,and compared with the simulation results of Na Sch model and TSM model,the role of traffic hysteresis and wave speed in simulating the characteristics of traffic oscillation is analyzed.The results show that the model that does not consider traffic lag can't simulate the phenomenon of traffic oscillation.In the model,the driver's reaction time is corrected by substituting the wave speed of the recession phase to trigger the traffic oscillation to enter the recession phase.