Train Rescheduling Under The Failures Of The Train Control System For Urban Rail Transit Systems

Urban rail transit system has plenty of advantages,such as large traffic volume,good safety,high reliability,high efficiency and energy conservation,compared with other traffic modes.As a result,urban rail transit system has become an effective method to solve "big city diseases" such as congestion and pollution.As train delays often have adverse effects on passengers,the on-time rate of train operation is recognized as one of the key indexes to evaluate the performance of urban rail transit systems.Because of the simple layout of urban rail transit lines and the small headway,it is difficult to reschedule trains after the failures of the train control system.As a result,even small failures of the train control system may lead to big impact on the normal operation of the system.Therefore,failures of the train control system have become the primary cause of train delays in urban rail transit systems.At present,dispatchers will reschedule trains according to their working experience after failures of the train control system,however,the result is usually not as good as expectation.Therefore,generating the rescheduled timetable automatically according to the effects and handling process of failures of the train control system,which aims to help dispatchers reschedule trains after failures of the train control system effectively,has important practical significance.The research can reduce the reparation pressure of the train control system,improve the automatic level and intelligent level of the failure management of the train control system,and increase the on-time rate and safety level of urban rail transit systems.This dissertation summarizes the normal failure handling process of the train control system in urban rail transit systems to summary the typical failure scenarios,builds a train rescheduling optimization model to reschedule trains under typical failure scenarios of the train control system.In the model,the multi-objective method is used to make multi-objective decision.In addition,a real-time train rescheduling method based on rolling optimization,which can upgrade the rescheduled timetable following the failure handling process is proposed.This dissertation mainly carries out the following work:(1)A train rescheduling optimization model,which conforms to the operational requirements,layout and train operation characteristics of urban rail transit system is proposed.The objective of the model is to minimize the train delays and the number of short-turned rolling stock as well as to ensure a regular service for passengers,while adhering to the departure and arrival constraints,turnaround constraints,station constraints,service connection constraints,inventory constraints,and other relevant railway constraints.In addition,train rescheduling during the whole process of the failure,from the beginning of the failure to the end of the failure,is optimized.(2)The weight of each objective in the train rescheduling optimization model is determined.The weights are calculated by applying the Delphi method and the Analytic Hierarchy Process.In detail,the calculation process mainly includes establishing the hierarchical structure model,discussing with experts and questionnaire survey.(3)The typical failure scenarios of the train control system in urban rail transit systems are modeled and the rescheduled timetable for each failure scenario is calculated.Typical failure scenarios of the train control system are defined according to the operational impact of the failures,and then,the effects of the failures on the train operation and available dispatching measures are modeled.Additionally,the rescheduled timetable under typical failure scenarios of the train control system is calculated by the train rescheduling optimization model to verify the correctness and efficiency of the model.(4)A real-time train rescheduling method based on rolling optimization is proposed.A rolling horizon approach is presented to deal with the problem that the duration of failures of the train control system is likely to have great uncertainty.The rolling horizon approach calculates the rescheduled timetable in stages,which can shorten the solution time of each stage and improve the real-time performance of the algorithm effectively.What is more,the rolling horizon approach can update the rescheduled timetable according to the handling process of the train control system,so as to improve the feasibility and efficiency of the rescheduled timetable.