Research On On-ramp Metering Of Urban Expressway

With the rapid development of urban economy and urban construction, insufficient supply of road resources and poor traffic management resulted in congestion and underutilized of urban expressway, which cannot rely on by expanding the scale of road infrastructure to increase capacity because of lack of urban land, what is more important is making full use of capacity and related services of urban expressway. Ramp metering has been proved to be one of the most effective way in dynamic traffic management to promote the efficiency urban expressway. Reasonable ramp metering can significantly prevent traffic congestion, improve traffic conditions, make full use of the capacity of expressway, and make the coordination of the expressway and associated road so as to improve service level of road.Accurate short-term traffic flow forecasting can predict the traffic environment in advance to support active control traffic, improve the traffic environment. Since the nonlinear and uncertain characteristics of short-term traffic flow, a wavelet neural network based on genetic algorithm optimization is established, which has strong learning ability and to avoid falling into local optimum compared with the conventional neural network, and an example is performed to verify the accuracy of its prediction.By analyzing the characteristics of urban expressway and the combination of expressway ramp to provide the basis for simulation and traffic flow analysis. Then the on ramp control principle is analyzed, and the evaluation of the ramp control will provide the basis for the control effect. Analysis of the different control strategies and methods are classified to provide theoretical support for the establishment of an adaptive ramp control.Based on the thoroughly analysis of ramp metering strategies and methods and in view of the existing ramp metering insufficiency which are ramp queue spillover, and adverse effects on the freeway mainline because of queuing fast cleanup and ramp control lagging, a ramp metering method considering the expressway efficiency, gap acceptance with the combination of short-term traffic flow forecasting is established. Finally, an example of ramp metering is performed with VISSIM and COM interfaces to simulation evaluation, and the results show effective and practical.