Vehicle Queue-Length-based Urban Road Traffic Flow Macroscopic Fundamental Diagrams

With the rapid development of urban economy and the continuous advancement of urbanization,the living standard of the urban residents is rising,then the number of vehicles is also increasing.Consequently,the traffic congestion,traffic safety problem and other problems are becoming more and more severe,that have seriously affected the travel quality of urban residents and restricted the city development.In recent researches,the macroscopic fundamental diagram(MFD)has been proved to be a powerful tool for large urban network modelling and control.Considering the fact that queue length is usually regarded as an important index for evaluating the performance of intersections,this paper proposes a novel concept called queue-length-based MFD(QMFD).The existence of QMFD is verified based on floating car data(FCD)in Beijing in this paper.In addition,a discriminant model for traffic condition is established with respect to queue length.The specific research work is as follows:Firstly,the FCD in Beijing is preprocessed and the map matching is carried out based on FCD in this paper.The data preprocessing is to eliminate abnormal or erroneous data;The map matching is to determine the location of the floating cars.Because FCD contains many instantaneous velocity data points whose value are zero,and most of the direction angles are inaccurate in these zero-speed point data.Therefore,the map match for zero-speed point data and non-zero-speed point data adopts to different algorithms,respectively.Secondly,a method based on FCD for estimating the queue length in the intersection entrance lane is developed in this paper.Because the queue length is related to the distance between the floating car and the parking line when the floating car stop at the intersection entrance lane.The method obtains the statistics distribution of distance data for different time periods.Then,this method trains the BP neural network model according to the sample data pairs of queue length and distance data with different floating car proportions and during different statistical time periods based the Vissim which is a traffic simulation tool.Finally,the queue length is estimated based on the trained model.Thirdly,on the basis of accurate map matching with the FCD,this paper calculates the parameters such as speed,density,flow,and queue length of each road segment,studies the relationship between the average queue length and other macroscopic traffic parameters such as average flow,average speed,average density at the entrance of urban road network intersection,that is QMFD.According to the FCD in Beijing in September,2014,the existence of QMFD is verified from different perspectives,and the relationship between MFD and QMFD is analyzed.Finally,a discriminant model of traffic condition with respect to queue length is established in this paper.In order to clearly describe the traffic condition,this paper redefines the traffic flow considering the travel speed of vehicles passing through the road.According to the traffic data provided by traffic simulation software Vissim,it is found that the relationship between traffic flow and queue length is represented with a single peak cureve which has low scatter when the cycle length of traffic lights and the length of green lights remain unchanged.It is also found that there is a certain functional relationship between the critical queue length and the green time ratio of the traffic signal.