Optimization For Dynamic Train Timetabling On High-speed Railway

With the large-scale construction of China’s high-speed railway network,the demand for passenger flow is increasing and differs significantly from season to season,which leads to increasingly prominent requirements for refined operations.In order to promote the development of railway passenger transportation from an internal production type to a market-oriented type,"daily train timetabling" came into being.At present,the National Railway Group adopts the "quarterly train timetabling" mode,although to a certain extent,the ability of railways to adapt to market changes has been improved.However,there is still a problem that this compilation model does not match the dynamic changes of passenger flow,and it is still in the initial exploration stage of the train timetabling based on "daily train timetabling ".Based on the idea of "daily train timetabling ",this paper studies the dynamic train timetabling mode of high-speed railway,that is,on the basis of the basic train timetable,the train timetable for the implementation day is compiled in combination with factors such as passenger flow demand and market changes.The mode of train timetabling can realize the effective matching between transportation capacity and passenger flow demand,improve social service level and take into account the benefits of railway operation,thus forming a virtuous circle of win-win.The main research work is as follows:1.Research on seasons and time periods division of high-speed railway passenger flow.Analyze the dynamic characteristics of passenger flow.Considering the fluctuation of annual passenger flow,the firefly-affinity propagation algorithm is used to divide the passenger flow of high-speed railway into seasons as the basis of the basic train timetabling.Considering the fluctuation of daily passenger flow,the sequential clustering method is used to divide the high-speed railway passenger flow into time periods,which is the basis for implementing daily train timetable adjustment.2.Research on the matching and evaluation of high-speed railway passenger flow and train timetable.Based on the interaction between high-speed railway passenger flow and train timetable,this paper studies the matching evaluation methods from two aspects: the whole-day matching evaluation and the time-based matching evaluation to judge the matching degree.Considering the adaptability related indicators between train schedule and passenger flow demand,a train timetable evaluation index system is constructed,and the principal component and clustering method is used to evaluate the train timetable,and the adjustment mode of the train timetable is judged according to the evaluation results.3.Research on local adjustment and optimization problem of dynamic train timetable based on implementation day.Analyze the principles and contents of local adjustment of train timetable.On the basis of the original basic train timetable,considering constraints such as out-of-service train restrictions,additional train restrictions,and safety interval time restrictions,a comprehensive optimization model is constructed to minimize passenger flow demand versus train timetable capacity deviation,minimize train travel time,and minimize original basic train timetable adjustments.Finally,the model is solved using an adaptive large-scale search.4.An example analysis of dynamic train timetabling based on implementation day.Taking the Beijing-Shanghai high-speed railway as an example,the train timetable for the implementation day is compiled and the results are analyzed.The results show that the local adjustment model of the dynamic train timetable based on the implementation day constructed in this paper can better meet the dynamic passenger flow changes.