Research On Phase Diagrams Of Traffic States In The Intersection Traffic Bottleneck And Cellular Automata Traffic Models

Traffic congestion has become a national urban public nuisance. It has resulted in the increase of energy consumption, environmental pollution and has become a ma-jor bottleneck restricting the sustainable development of large and medium-sized cities in China. Consequently, it has a great impact on economic development. The fun-damental way to solve the traffic congestion problem is to develop the transportation science and technology and carry out the study of traffic flow theory. The objective of the study of traffic flow theory is to establish traffic flow model which can describe general characteristics of actual traffic, find the basic law of traffic flow and then reveal mechanism of traffic congestion. Only in this way, we can achieve the transformation from "symptoms" to "disease" and provide basic protection for sustainable and rapid development of modern city. Therefore, it is necessary to study traffic flow theory.Cellular automata is a mathematical model which is discrete in space, time and variables. It has many great characteristics, such as simple algorithm, clear physi-cal pictures and high degree of parallelism. Cellular automata methods and relevant modeling techniques are powerful way to describe, perceive and simulate behaviors of complex system. Nowadays, cellular automata have been widely used in research of the transport system and a variety of cellular automata traffic flow models have also been proposed. Due to the fact that it is easy to consider various actual traffic factors in the model, it is widely used to study the various complex phenomena of traffic flow.Based on the complexity and richness of the traffic dynamics research, we first survey domestic and overseas relevant research developments of traffic flow theory, and then study a number of popular dynamics issues. The research contents include modeling and simulations of one-dimensional, two-dimensional traffic flow; computer simulations and theoretical analysis of traffic flow in the intersection bottlenecks.First of all, this dissertation has studied a one-dimensional cellular automata traffic flow model: 1. In the evolutionary rules of the classical Nagel-Schreckenberg model, the ran-domization rule captures natural speed fluctuations due to human behavior or varying external conditions. This rule introduces overreactions of drivers when braking, pro-viding the key to the formation of spontaneously emerging jams. In this dissertation, the randomization rule is improved as follows:compared with the case that velocities are equal to or greater than2, it is notable that vehicles moving at their lower velocity1are relatively safe and the randomization rule may be neglected (not be considered). Here, none of vehicles is permitted to overtake in one lane, so the vehicle velocities must be less or equal to their corresponding gaps. That is to say, vehicle needs to slow down when their velocities equal their corresponding gaps and are no less than2. Ac-cording to simulation results, it has been found that the structure of the fundamental diagram of the new model is sensitively dependent on the values of the delay probabil-ity.In comparison with the Nagel-Schreckenberg model, one notes that the maximum flow value of the fundamental diagram in our model is more consistent with the results measured in the real traffic,and the velocity distributions of our model are relatively reasonable. Thus, our model is more fit for simulating actual traffic.2. Using the cellular automaton traffic flow model, we have studied the vehicle gap distribution of the mixed traffic flow. By computer simulations, one notes that the vehicle gap distribution varies with the delay probability in the case of low and intermediate vehicle density for the system with given mixing rate. Furthermore, one notes that the mixing rate has great effect on the vehicle gap distribution in the case of low vehicle density for system with given delay probability.Secondly, this dissertation has mapped out phase diagrams of traffic states in the intersection bottlenecks:1. This dissertation has studied phase diagrams of traffic flow at an unsignalized intersection consisting of two perpendicular one-lane roads where periodic boundary conditions are adopted and parallel update rules are employed. One simulates the mo-tion of vehicles by using deterministic Fukui-Ishibashi cellular automata traffic model. Vehicles are not allowed to turn at the intersection. At the intersection, if conflict hap-pens, the yielding dynamics will be implemented. Based on the principles for making phase diagrams, we have presented the phase diagrams for the cases of various maxi-mum vehicle velocities, and it is noted that the phase diagrams have several different topology structures which are shaped by maximum velocities. The flow formulas in all regions in the phase diagram have been derived. The results of theoretical analysis are in good agreement with simulation ones.2. Similar to the studies above, the following study allows vehicles to adopt the evolutionary rules of the deterministic Nagel-Schreckenberg cellular automata traffic model. We have also presented the phase diagrams for the cases of various maximum vehicle velocities. The flow formula in all regions in phase diagrams have also been deduced by the same way of the theoretical method. The theoretical analysis approach used in this dissertation is not only simple but also may be widely used in the study of other bottlenecks.Last but not least, this dissertation has put forward a two-dimensional cellular automata traffic flow model where vehicles can change lanes:Based on the classical Biham-Middleton-Levine model, this dissertation has es-tablished a two-dimensional urban traffic flow model where vehicles can change lanes. Comparatively speaking, this model can simulate actual traffic better. By using numer-ical simulations on the square and rectangular grid traffic systems, some new config-uration graphs have been found in free flow regime. In addition, some intermediate stable phases, where jams and freely flowing traffic coexist, have been discovered and stay within larger density range.