An Model Predictive Control Approach To Coordinated Ramp Metering Strategy

With the development of cities and the rapid increase in traffic demand, traffic congestion has become a common problem in the metropolises of the world. If it is unable to relieve traffic congestion on the city expressway efficiently, the larger traffic jam will ultimately emerge in city road network as a result. On-ramp metering is one of the efficient control methods to alleviate and prevent traffic congestion in urban expressway system.In this dissertation, the control strategy with the measure of ramp metering are sought to deal with traffic congestion. The prevalent model predictive control (MPC) structure is used to formulate the problem. A macroscopic traffic flow model is used as the prediction model of the traffic system. The main contributions of the dissertation are summarized as follows:First, a macroscopic dynamical traffic flow model called METANET is used for the description of traffic flow on expressway. With total time spent (TTS) as the objective, the dynamic nonlinear time discrete MPC control problem is established and the numerical solution is achieved by use of revised gradient based on the minimum principle. By a simulated case study, the efficiency and robustness of the MPC strategy was investigated. The results show that the MPC strategy can obviously alleviate congestion and improve the overall network performance.Second, the original MPC is extended with a controller positioning procedure to conquer the side effect arising from the big number of control variables and reduce the computational effort. Not all but only the necessary ramps are selected as controllers by investigating the gradients of control variables and applying the K-means cluster algorithm. The simulation experiments show that the MPC extended with controller positioning approach is an efficient method to reduce the computation time, as well as to ensure the control effectiveness.Third, the problem of designing ramp metering strategy considering environment benefits is presented. Based on METANET, the macroscopic traffic variables are transformed into microscopic variables, and then the emission amount is calculated by a well developed microscopic emission model. The problem is formulated in MPC structure, in which the objective is a combination of traffic efficiency and environment benefit. Simulation results show that, using the derived optimal ramp metering strategy, traffic efficiency as well as the environment benefit is improved, especially the emission of CO, HC and fuel consumption are greatly reduced.Finally, the ramp control general software is designed in order to compare simulation results of a variety of road network structure or different traffic demands. The software system consists of two parts:parameters setting and simulation analysis. The parameters setting part includes network parameters, model parameters and simulation parameters. The simulation analysis part includes traffic demand, the optimized performance, ramp metering rates and traffic flow state.