Research On Dynamic Optimization And Control Techniques For Urban Traffic Signal

Congestion in urban traffic is a problem commonly concerned around the world, and the harmful results that produce are increasingly affecting our daily life and the development of our society. As the distributing center of urban traffic, intersections are the main taches that caused congestion, and rationally controlling the traffic flow in the intersections is not only an important method to relieve the traffic congestion but also a key respect to study urban traffic. In this paper, in allusion to the status quo of the urban traffic in China and based on the design and development of an intelligent traffic signal control system, the dynamic optimization and control techniques for urban traffic signal are mainly studied and its realization is also discussed. The most distinctive parts can be described in the following aspects:1. To the signal optimization problem of urban traffic, a hybrid optimization algorithm is presented. In the algorithm, a real-coded genetic algorithm (RGA) is firstly adopted to globally search for the optimal solution to the optimization problem for some steps, and then a specially designed local optimization algorithm named modifiable search space random search algorithm is adopted to intensify the search for the optimal solution in the local region determined by the RGA so as to improve the search efficiency and the solution quality. In addition, for reinforcing the search efficiency of the RGA, a compute method of self-adaptive crossover probability and mutation probability is also developed. The simulation results indicate the validities of the proposed hybrid optimization algorithm and the compute method of self-adaptive crossover probability and mutation probability.2. For applying the hybrid optimization algorithm to solve the optimization problem with constraints, a self-adaptive penalty strategy for handling the constraints is designed, which is used to convert the optimization problem with inequality constraints and equality constraints to the one only with the upper and lower bound constraints of the decision variables. By introducing the concepts of constraints measure and feasible degree, the extent to which the decision variables subject to the inequality constraints and equality constraints is described and then the self-adaptive penalty function is constructed, whose values are adaptively changed with the constraints feasible degree. The application of this strategy to varied optimization algorithms show that it can effectively handle the constraints contained in theoptimization problem.3. Aiming at the signal control problem of the isolated intersection, a dynamic optimization and control method is put forward, whose objective is to minimize the total delay queue vehicles over the whole control period. According to the typical flow distribution, the dynamic optimization model which adopt cycle length and split as the decision variables is set up, and the simulations are conducted in the condition of light, moderate and heavy traffic demands respectively. The result indicates that the proposed method can greatly improve the traffic condition compared with the traditional fixed-time control method.4. Based on the signal control of the isolated intersection, a dynamic optimization and control method for urban artery is presented, whose objective is to maximize the throughput of artery system during the whole control period. In allusion to the two-way traffic artery, the dynamic optimization model which adopt cycle length, split and offsets as the decision variables is established, and the simulation is conducted in a dynamic traffic demand mode. The results show the validity of the proposed method.5. For the practical demand of urban traffic signal control, an intelligent traffic signal control system is designed. The monitoring and controlling software as well as the remote communication program based on PSTN network are developed using Visual Basic 6.0, and the software and hardware for the human-machine interface of the signal controll...