Font Size: a A A

Study On Ramp Metering Method Of Freeway And Related Urban Expressway

Posted on:2010-10-17Degree:MasterType:Thesis
Country:ChinaCandidate:H J SongFull Text:PDF
GTID:2132360275973449Subject:Systems Engineering
Abstract/Summary:PDF Full Text Request
The congestion on the freeway and urban expressway sections into and out of cities becomes more and more serious along with the step of urbanization and motorization. Tide type of traffic flow in peak hours and merging and conversion between traffic flows of different characteristics causes the disorder in the periphery-road-network, reduction in the level of service and delay increase.At abroad,the ramp metering is widely used in freeway and urban expressway which are accomplished by eliminating all cross traffic through the use of grade separations and interchanges and have no traffic lights.Thus,it can be used to ease traffic congestion in the joint area of freeway and urban expressway.The main contents and related conclusion are listed as follows:The research achievement and practical application of ramp metering of freeway and urban expressway at home and abroad has been summed up systematically.And related theory and approach have also been presented in the paper.After comparison and analysis of various ramp metering methods,the technical method used in this paper is determined.The paper definites the study object that is the joint area or periphery road network. Taking Jing-jin-tang freeway joint area as an example,the three traffic parameters of typical sections and turning characteristics at grade interchanges are analyzed.Then the method of traffic state classification is discussed.Firstly,fuzzy C-means clustering is used for the initial classification.Then,fuzzy neural network is used to moderate the results absorbing the idea of supervised classification.At last,spatio-temporal correlation characteristics of traffic state of typical sections are analyzed.The paper analyzes the related factors of the ramp metering and establishes ramp metering control model using fuzzy neural network.The model has two input parameters.One is the difference of mainline traffic state and expected state,which can make the mainline state tend to be stable and have a feedback effect on the system.And the other input parameter is ramp traffic state reflecting the changing of travel demand, which is ramp traffic volume divided by ramp capacity.Control rules are set according to the two directions and control method under two situations which are single ramp control and double ramps control are described.Ramp metering is simulated using MATLAB and the data of Jing-jin-tang freeway in joint area.And the results present a satisfying effect.Mainline traffic state after ramp metering has a smoother and better condition than origin.The difference of the mainline traffic state and expected state reduced,and the average queue length is within 5 vehicles and need less time to dissipate.
Keywords/Search Tags:joint area, ramp metering, fuzzy neural network, traffic state
PDF Full Text Request
Related items