Models And Simulations Of Three-Phase Traffic Flow Theory Based On The Theory Of Cellular Automaton

The purpose of theoretical study on traffic flow is to establish traffic flow models that can well describe the general properties of the real traffic and to disclose the basic laws of traffic flow. With those theoretical discoveries, traffic engineering practice can be well guided and traffic congestions can be thus alleviated and eliminated. As a result, basic traffic support can be provided to the harmonious development of the society. As the hotspot in modern traffic theory research field, Three-Phase Traffic Flow Theory, which is proposed by Kerner, can be used to well explain many measured traffic phenomena. However, people are not so clear about the inner mechanism of Three-Phase Traffic Flow Theory, especially not so clear about the origin of synchronized flow in Three-Phase Traffic Flow Theory. Consequently, more research should be launched to the further study of Three-Phase Traffic Flow Theory. It is thus a good attempt to study the theory with the help of cellular automaton.As a kind of emerging models, cellular automaton based models with flexible rules not only can retain the non-linear characteristics of complex traffic systems, but also are suitable for rapid computer simulations. Consequently, they have been widely adopted by the traffic scientists. After a series of analysis of many existing cellular automaton based models, this paper launches modeling and simulation research toward Three-Phase Traffic Flow Theory with the help of cellular automaton based models. Through computer simulations, it is found that the new models proposed in this paper are able to reproduce many interesting phenomena in the framework of Three-Phase Traffic Theory. The creative work of this paper is as follows.1. A possible mechanism that can result in the appearance of synchronized flow is discovered. It is thought that driver's excessive deceleration can result in the appearance of synchronized flow in this mechanism. A time gap based cellular automaton model is thus presented. Considering that the drivers are sensitive to the factor of time gap, the randomization probability of a single vehicle is calculated according to its time gap. Simulation results show that the model is able to reproduce the characteristics of synchronized flow. After the introduction of the slow-to-start rule into the new model, it is found that both the spontaneous and the man-made transitions from the synchronized flow to the wide moving jam can be observed in the simulations. Moreover, the simulation results are relatively consistent with Three-Phase Traffic Flow Theory.2. According to the time gap based cellular automaton model presented above, simulation and investigation work is launched to study traffic patterns at the locations of on-ramp traffic bottleneck. After a series of analysis of relevant phase diagram and spatiotemporal diagrams, it is pointed out that traffic patterns at the locations of on-ramp traffic bottleneck are well reproduced, which is also consistent with Three-Phase Traffic Flow Theory.3. A fuzzy illation based cellular automaton model is presented. Considering that the reactions of drivers to their velocity and spatial distance are fuzzy, fuzzy illation mechanism is introduced into the randomization process of a single vehicle. Thus both the probabilistic and the fuzzy factor are taken into account during the randomization process. Simulation results show that the model is able to reproduce traffic phenomena such as the stop-and-go wave and the synchronized flow in the framework of Three-Phase Traffic Flow Theory.