Research On Ramp System Structure Decompose And Coordinated Metering For Periphery Road Networks

Traffic congestion in metropolitan periphery road networks are growing seriously, and the corresponding management and control is obviously lacking. Nowadays, trains of strategies and methods on urban traffic congestion prevention and relief which based on diverse transportation sharing and traditional traffic control have been proposed. Mainly of them are suit for downtown area instead of for periphery road networks, for there are giant differences of transportation system, road layout and traffic flow characteristics between downtown area road networks and periphery road networks. Intelligent control is a new development to find the combination of control measures for the best road performance and control effectiveness. Ramp metering has gained general acceptance as an essential part of modern traffic congestion management systems in either preventing or reducing freeway congestion. On considering the realities of periphery road networks and to relieve traffic congestion during peak hours on freeways, this paper designed, discussed and tested suiTable ramp metering control method based on system engineering, various ramp metering models have been attempted for regulation of the inputs to freeway from entry ramps for three kinds of problem that include multi-objects coordinated for turning ramp from freeway to expressway, multi-ramps coordinated metering control for a simple traffic corridor that consists of G2freeway and a set of parallel arterials connected by entrance ramps and interchange congestion metering, all of them formed a integral research system from point, line to plane structure.The main contents, innovations and research result are summarized as follows:(1) In consideration of some weaknesses of current intelligent metering, this paper proposes an innovative concept and associated local ramp metering approach. Inspired by the rationale of equity, we introduce a novel ramp queuing parameter, the accumulative waiting time, to ramp metering. The objective is to avoid long waiting time of multiple cycles for any single vehicle. In calibration of the proposed model, a radical based function support vector machine algorithm is developed for model calibration. From results of a comparative case study with real world traffic flow data in Beijing, the nevel metering approach is proven to significantly output from traditional models, particularly in regards to its effectiveness in minimizing the waiting time of multiple cycles.(2) The idea of "flexible capacity for mainline" and the problem of its reasonable allocation and effective utilization for ramp coordinated metering are considered on considering the traffic characteristics of periphery road. Based on this idea, this paper examines the conditions for which ramp metering can be beneficial to overall systems in terms of ramp queuing length, redundancy waiting time and redundancy waiting cycle number decreasing for a simple traffic corridor that consists of G2freeway and a set of parallel arterials connected by entrance ramps. The focus is on analyzing system state and control relationships to arrive at general analytical results regarding optimal metering policies, discusses the optimal coordinated control metering of multi-ramps, especially the method of enhance the efficiency and equity of "flexible capacity" allocation by modifying the length of control cycle. A modified ramp latency model is posed using the method of queuing theory, and the related redundancy latency cycle amount is designed to evaluate the consequence quantitatively.(3) Continuously, on relieving congestion within interchange networks, the problem of designing control strategies for congested interchange is considered, it is a novel research field with challenges. This paper presented a novel methodology based on graph theory and network flow theory, the former is used to modelling by unfolding a Typical interchange into a directed weighted network with multi origins and destinations, and the latter is used to calculate the dynamic flow spectrum, clarify maximum flow and to certify the controlled ramp arc.(4) In addition, two novel resistance balancing factors aim to increase the fairness of control are designed with ramp queuing lengths and redundancy time, incorporating flow spectrum and based on control theory, a linear model with a quadratic objective function is constructed, and the corresponsive control law algorithm is designed with the adaptive queuing management method.Trains of ramp metering simulation cases are finished. With in-situ traffic flow data, two scenarios of no-control and coordinated with different flexible capacity using strategies are compared, and the result demonstrates the advantage of the new method. For interchange congestion control, a preliminary simulation-based investigation of the virtual coordinated ramp metering control problem using the methodology demonstrates the comparative efficiency and its ability to maintenance network performance while balancing cost distribution is better.