Research On High-speed Railway Network Delay Propagation Based On Bayesian Networks

In recent years,high-speed railway has become more popular among long-distance travelers because of its advantages of high speed,high safety and large transportation capacity.However,it is vulnerable to a variety of factors in the operation process of high-speed railway,so highspeed trains could not arrived on time and then delay propagation makes more stations in a state of delay in the high-speed railway network,which has a serious impact on passenger travel and normal operation of highspeed railway organization and management.It is a practical problem for organization and management department of high-speed railway operation to discover the mechanism of delay propagation in high-speed railway network from the network level and prevent the delay spreading from local area to whole network.In order to discover the delay propagation mechanism of high-speed railway network,this thesis focuses on a series of researches including structural modeling of stations’ delay propagation Bayesian network and identification of the important delay propagation dependencies between stations in the high-speed railway network.Laying the conceptual basis of high-speed railway delay,in accordance with the research idea of “Model Construction-Important Delay Propagation Relationship Identification-Network Station’s Delay State Prediction”,this thesis firstly analyzes the operation characteristics and delay characteristics of high-speed railway based on the actual operation data and train schedule data.Secondly,the Bayesian network model is established to identify the delay propagation dependencies between stations in high-speed railway network.Then,in order to identify the important delay propagation dependencies in the network,this thesis combines with the procolation theory in complex network to find and analyze the core delay propagation clusters in the percolation threshold to reveal the core delay propagation mechanism in the network.The results show that stations can be divided into 3 categories,namely,delay generator,delay mediator and delay absorber according to their characteristic in the aspect of delay propagation.Delay generator can not only spread the delay to stations close to them,but also spread the delay to the far away stations through delay propagation chain,which makes the network taking part of the delay generator as the center of delay propagation and spreading delay to the delay mediator and the delay absorber in a radiational delay propagation mode.Finally,this thesis establishes a Bayesian network model used for station delay state prediction and analyzes a wide range of delay propagation in the network by visualizing the prediction results of real cases.The experimental results show that the proposed model can accurately capture the dynamic change trends of delay propagation in the network and has a high prediction accuracy,which can provide a certain support for the study of delay propagation in the high-speed railway network.