Study On The Characteristics Of Phase Transition In Cellular Automata Models For Traffic Flow And Analysis Of Traffic Empirical Data

With the development of social economy, the contradiction between the growth of traffic flux and relative delay of traffic construction is becoming prominent. Traffic jam, cars following rears bring on crash accidents, which has greatly restricted the development of social economy. Many experts in different countries have concentrated on traffic problem seriously and proposed many models of traffic flow in order to dull traffic jam and reduce traffic accidents. Many scientists widely focus on investigation of Cellular Automaton (CA) traffic flow model. Cellular Automaton model is a mathematical model in which space, time and state value are discrete. It is parallel and simple for computation, so CA is a very good tool for simulating various complex nonlinear phenomena and physical problems.In this dissertation, based on the nonlinear phenomena of jam formation, stop-and-go and synchronized flow traffic, the characteristics of phase transition compared to those of NaSch model are studied via spatiotemporal headway distributions, site correlation function, order parameter of one dimensional CA traffic flow model with "slow-to-start" rule. Having chosen two representative bus stops and measured bus stopping times and analyzed empirical data.This dissertation consists of the following three main parts.1) Because of traffic jamming, cars following leader's tail lead to collision. This is reflected by cars' spatio-temporal headway varying. According to spatio-temporal headway distributions, we can understand traffic congestion to some extent. Compared to those of NaSch model, simulation results show that spatio-temporal headway distribution of three models at free flow traffic is identical and one at...