Methods And Models For Traffic Congestion Control In An Urban Rail Transit Network

In the condition of network operation, the formation and propagation mechanism of traffic congestion in the urban rail transit network is complex, so that it’s difficult to implement traffic congestion control. In this paper, the urban rail transit network operation simulation system is used for traffic dynamic deduction. On the basic of identifying the key congested nodes and the congestion source backtracking, the method of optimizing network collaborative flow-limiting schemes is studied. When foreknowing the congestion situation in the future hours, it’s designed to prevent the formation and propagation of network-wide traffic congestion by previously implementing the collaborative flow-limiting measures, and achieve the network traffic congestion control in a phased scroll way. The main contents in this paper include the following aspects:(1) This paper analyze the network operation characteristics of urban rail transit, the characteristics of normal traffic and congested traffic, the formation and propagation of traffic congestion, and the influencing factors of propagation, and sum up the general measures of traffic congestion control. Considering the multi-point correlated and effect lagged characteristics of congestion control, the prospective associated congestion control method aimed at the key congested nodes is proposed, the control process and the basic idea of this method is provided as well.(2) After building the urban rail transit network model based on three-level sub-network, a method by combining the structural characteristics with the traffic characteristics to determine the significance of the congested nodes is proposed. The structural index of congested nodes based on the complex network theory and the traffic index of congested nodes based on the power law distribution are established, and then the integrated index is formed with these two kinds of index above, which became a basis for identifying the key congested nodes.(3) For "tracing" the traffic of the key congested nodes, the congestion source backtracking method based on simulation data is proposed and the backtracking algorithm is designed, to obtain the traffic source information on the congested nodes, including source station, source amount and source time, all these provided limiting stations candidate set and related traffic parameters for developing the network collaborative flow-limiting schemes.(4) This paper takes the network collaborative flow-limiting as the main means of congestion control, and propose a two-stage optimization method for network collaborative flow-limiting schemes. Then the initial scheme model and adjustment scheme model of network collaborative flow-limiting are built respectively, and make models solved. Finally, the research methods in this paper are verified by taking the Beijing urban rail transit system as an example to conduct simulation experiments and data analysis.