The Research On The Inflow Control On Toll Collection Freeway Networks

The dissertation attempts to control inflow in expressway network based on bi-level programming model. The aim is to avoid traffic congestion in expressway network and improve the utilities of expressway network and realize maximum of social welfare, and take users' route choice in response to strategy the upper-level problem carries out who want their travel cost as little as possible.At first, the dissertation simply reviews the traffic assignment problem and its solution strategies , solution algorithms of bi-level programming mode at present,> traffic control methods , network toll collection and the systemic configuration of network toll collection. Then the dissertation considers the relation and intention of manager and users in the problem of inflow control on toll collection freeway network, and builds the bi-level programming model to solve the problem of freeway inflow control. The upper-level problem represents the behavior of manager, and the lower-level represents the users' travel choices. So we propose multi- vehicle and multi-criteria network equilibrium models with entry-exit based on toll road of the network to describe route choice of users, namely user equilibrium model with fixed demand. Bi-level programming problems are generally difficult to solve, so we put forward a new algorithm named GAA algorithm combining the genetic algorithm and Alopex algorithm.At last, the dissertation gives a simplified expressway network, we firstly chose the best objective function to research the inflow control of expressway network, and compare and contrast the outcomes of various vehicle with the case of simple vehicle type, and investigate the corresponding system utilities under the optimal inflow in the condition of the fixed traffic demand. The results are reported by genetic algorithm and GAA algorithm respectively. The conclusion drawns from the above comparison is that the objective value in the case of various vehicle types is less than that in the case of simple vehicle type and improving the expressway network's utilities. In addition, the GAA algorithm excels to the genetic algorithm in solving bilevel programming problem judging from the results and running time of the two algorithm.