Freeway Traffic Flow Modeling And Ramp Control

Accurate model of traffic flow is important not only for better understanding of the collective behaviour of vehicles, but also for analyzing flow conditions, planning traffic networks, and designing efficient control strategies. On-ramp control can make traffic flow move in optimum conditions by regulating the number of vehicles entering a freeway entrance point, thus improve the passing capability of the mainline, and avoid traffic jams and congestion. Freeway traffic flow modeling and on-ramp metering have been presented and discussed in detail. The main contributions can be stated as follows.1. The wavelet transform is used to eliminate traffic noise and disturbance, and the traffic flow model is built based on a recurrent neural network in this paper. First, the freeway macroscopic traffic flow model is analyzed. Then the noise elimination method of wavelet transform is formulated, and Elman recurrent network is used for traffic flow modeling. The weights of the Elman network are obtained with an improved algorithm. Finally, a freeway with five segments, an on-ramp and an off-ramp is simulated. BP and RBF neural networks are chosen in contrast to the Elman network. The results show that the Elman network has the fewest training epochs, the smallest error and the best generalization ability.2. Nonlinear feedback methods are proposed for freeway on-ramp metering. The freeway traffic flow dynamic model is built. Based on the model and in conjunction with advanced PID control, three nonlinear feedback ramp controllers are designed: artificial immune controller, single neuron network self-adaptation controller, and fuzzy-immune controller. The above controllers are simulated seperately in MATLAB software. Simulation results show that nonlinear feedback is effective to the on-ramp control.3. The hierarchical structure and CMAC (Cerebellar Model Articulation Controller) are used for freeway ramp metering. The macroscopic model to describe the evolution of freeway traffic flow is first established. The algorithm of the composition controller of PID and CMAC is then studied. There are two layers in this control architecture: the coordinated control layer is responsible for computing the desired state of each segment, and the direct control layer is in charge of the ramp metering rate via the composition controller of PID and CMAC. Finally, the control system is simulated in MATLAB software and fuzzy logic control is also chosen in contrast to the composition control. The result shows that the composition controller improves evidently on the aspects of response speed and dynamic performance. The method can effectively eliminate traffic jams, andmake vehicles travel more efficiently and safely.