Signal Timing Models Based On The Optimization Of Urban Transportation Network Efficiency

The amount of the possessed automobiles and the various kinds of traffic demands are increasing rapidly along with the continuous high speed development of national economy and the acceleration of urbanization. Consequently, in our country many big or middle cities, in which the number of people beyond one million, are facing up with the problem of very serious traffic congestion. Traffic congestion may result in the invalidation of the whole or local function of the city, so the problem of urban traffic has changed in to the bottleneck which restricting the urban economy development and sustained development of the society. It not only leads to great loss of people's life and property, but also lowers the life quality of the people. However, along with the perfection of urban transportation network day by day, the less the land can be used for road construction, the more the construction cost and the fewer the possibility of improve traffic condition through the construction of transportation network will be. Therefore, it isn't insufficient to improve traffic establishment in the congested area in order to relieve traffic congestion.The dissertation relies on Key Program of National Natural Science Foundation"Research on key theories of dynamic traffic management and control and simulation technique for urban road networks". The paper mainly studies signal setting models based on the optimization of urban traffic network efficiency. The full paper is composed of five chapters. The studied emphases are Chapter Three and Four. The primary contents of all chapters are as follows:Chapter One Firstly, the research background is introduced; secondly, the phenomenon of disequilibrium traffic flow in urban transportation network is analyzed and then the research intention and meaning of the paper are pointed out, Thirdly, domestic and overseas research achievements about the topic of which recurring to equilibrium assignment-signal control problem are summarized in details; Finally, the research content of the paper is put forward.Chapter Two The fundamental theories, which are involved in the course of construct model in this paper, are introduced from four aspects, such as signal setting theory, equilibrium assignment models, disequilibrium assignment models and cost function. First of all, basic parameters of signal setting, signal control planand the basic signal setting methods are elaborated in the part of signal setting theory. Secondly, user equilibrium model, system equilibrium model and some expand user equilibrium models are presented in the part of equilibrium assignment theory. Thirdly, four kinds of disequilibrium assignment models, including all-or-nothing assignment, capacity restrained incremental loading assignment model, multi-path assignment model and capacity restrained multi-path assignment model, are given in the third part. Especially the multi-path assignment is elaborated exhaustively in the light of the requirement of this paper. Lastly, the common used lane cost function and the intersection turn delay function are presented in the fourth part of this chapter.Chapter Three In the first section, the ideology of how to built the model is explained. Base on it, the second section, first of all how to express the objective of traffic manager is depicted and maximize the summarization of the multiplication of speed and flow on all lanes in transportation network are utilized to express the optimal situation of network traffic flow. It is taken in to account that the turn delay of different intersection approach is different. So based on it, the user equilibrium model and stochastic user equilibrium model are developed to express the principle of traveler route choice in the second section. In the third section, a bi-level program model based on the optimization of transportation network efficiency is proposed. Due to the difficulty of calculating the bi-level program models, in the light of the similarity between the game theory and the bi-level program, the simple changes are made for the purpose of simplifying the computation of the problem.Chapter Four In the first section, the algorithms of bi-level program model are introduced including the extreme point search method, the Kuhn-tucker method, the descent method, the simulated annealing approach, the genetic method and so on. Firstly, the capacity restrained assignment method; the incremental loading assignment method and the Frank-wolf algorithm to solve the user equilibrium model are elaborated thoroughly in the first section. Secondly, the MSA algorithm for stochastic user equilibrium assignment is presented and the conclusion that different method is used to solve the lower program with different congestion degree of network is summarized. Thirdly, a practical network is designed to validate the bi-level program model based on the optimization of network efficiency.Chapter Five The obtained improvements and the meaning of this paper are summed up. Some problems are put forward to need to be discussed further.Different from the problem of the isolated intersection control, the arterial intersection control and the area-wide control in traffic engineering, a bi-level program model based on the optimization of network efficiency in which the disequilibrium traffic flow in urban transportation network cause by the disequilibrium of traffic supply and demand and the mutual influence between the traveler route choice and signal setting are taken into account is developed. It aims to equilibrate the traffic flow, lower of traffic flow because of local lane capacity and make the best use of the lane capacity. This paper provides a new method for urban traffic control, but how to use it in practical transportation network need further study.