Optimization Of Timetable Synchronization For Urban Rail Transit Early Train Sets

The key to the problem of network trains timetable compilation is to deal with the connection of trains between different lines at the transfer station.In this paper,by combing the existing literature,the first train schedule optimization is further expanded into simple early train sets schedule optimization and general early train sets schedule optimization.The research content is as follows:The meaning and goal of transfer coordination in network trains schedule optimization is clarified,and the concept of "early train sets" is proposed by analyzing the characteristics of transfer passenger flow demand.And the balanced departure interval of the first train schedule optimization process is adjusted to the unbalanced departure interval of the early train sets schedule collaborative optimization process,and the structure of two early train sets is given.Compare the similarities and differences between the transfer process of the early train sets of different sizes and the first trains,and use graphics to describe the arrival and departure rules of the feeder trains and connecting trains at the transfer station.The relationship between passenger transfer waiting time and redundant transfer waiting time is quantified,and an algorithm for solving passenger waiting time is designedCollaborative optimization model of simple early train sets timetable takes the departure times and balanced departure intervals of the first train of each line as decision variables,and considers three sets of constraints including technical constraints,time parameter upper and lower limit constraints,and passenger transfer waiting time constraints.The objective function minimizes the passengers’ total transfer waiting time.The initial temperature setting method and neighborhood transformation rule of the simulated annealing algorithm are improved,and the neighborhood of the solution is combined with decision variables to design.Design and solve small studies.Comparing the initial train schedule and the schedule of the simple early train sets,the total passenger transfer waiting time obtained by chaos simulated annealing algorithm and simulated annealing algorithm is reduced by 28.2% and 27.9%,respectively.The improved hybrid intelligent algorithm has better solution efficiency,and it takes 64 s and 98 s to obtain a satisfactory solution.In the cooperative optimization model of the schedule of the general early train sets,the trains in the early train sets are issued unevenly at the departure station,the stop time of the train at transfer station is added as a decision variable,and the time-varying characteristics of the passenger flow,to establish a general model to minimize the average passenger waiting time for transfers.Genetic operators are added to the simulated annealing algorithm framework,and the chromosome structure and cross-mutation operation rules are designed in combination with decision variables.The genetic simulated annealing algorithm and the simulated annealing algorithm are used to solve the general model,respectively.Compared with the current train schedule,the average passenger waiting time is reduced by 35.2% and 33.8%,respectively.The efficiency of the two algorithms and the quality of satisfactory solutions are compared.Finally,a large-scale road network example including 14 lines and 25 transfer stations is designed,and the simple model and the general model are solved to verify the applicability of the model under the conditions of a large-scale road network.Comparing the current train schedule and the timetables obtained from the two collaborative optimization models,it is found that in the simple model,the total passenger transfer waiting time obtained by the two algorithms is reduced by 22.0% and 23.2% respectively;in the general model,the two algorithms calculate the average passenger transfer waiting time is reduced by 29.9% and 33.7%,respectively.