Study Of Train Traffic Control And Simulation Based On Mixed Logic Dynamic System

With the rapid developing of railway transportation business in China,it’s becoming a difficult problem to be solved for railway transportation departments to ensure trains running safety and punctuality because of the increasing of trains operation speed and train operation density and the complexity of the railway transportation environment.Train Traffic Control is the system which controls trains to use the various railway resources reasonably and ensures trains operation safety and punctuality and the efficiency of railway transport.Based on the study of the experts at home and abroad,through using Mixed Logic Dynamic system theory(MLD),the train traffic control problem is studied in this paper,which focuses on building a MLD model for train traffic control.The model unites the continuous behaviors and discrete behaviors into a unified framework which is benefit to analyze the train traffic control problem.And using predictive control theory the optimization predictive control model is built under the non-extreme situations.The main research work includes:(1)The study of mixed logic dynamic systemThe equivalence transformation between propositional logic and mixed integer linear inequalities is the core of mixed logic dynamic system.Based on the former researching,the transforming methods from the propositional logic to mixed integer linear inequalities were summarized,and the process of modeling systems using mixed logic dynamic system is described in detail by some examples.(2)Summarizing and analyzing the events in the process of train runningThe events occurring in the process of train operation always cause some impacts of trains normal running state,such as the event of trains entering station makes trains brake,outbound event makes trains speed up,etc.Through the analysis of the events in the process of train running,the events are classified into normal events and abnormal events,and it is pointed out that train traffic control problem is a typical hybrid dynamic system,and the essence of train traffic control is to eliminate or reduce the effects produced by abnormal events in the train normal operation process.(3)Building MLD model of Train Traffic ControlSome extreme situations would happen in the process of train running,such as earthquake,lines break,etc.For these extreme situations,the only means whichdispatchers can take is to adopt the predefined plan to ensure trains safety and punctuality.In this paper,under the non-extreme situation,namely,the abnormal events only can lead to the trains delay,the train traffic control model was established based on the type of moving block by using MLD modeling theory,the discrete events and continuous process are unified in this model.Based on this model,it was pointed out that the essence of train traffic control is to eliminate or reduce the impacts produced by abnormal events in the train operation process using the normal events,and to change trains acceleration at some time to restore train operation order.(4)Study on the optimal control method for train traffic controlBased on the MLD model,taking the deviation between actual timetable and plan timetable as the optimization target,an optimal control model of train traffic control was established by using predictive control theory,which was solved using receding horizon optimization algorithm and ensured that every adjustment plan was optimal or suboptimal.(5)Simulation analysisTaking a dispatch section including 6 stations and 5 sections as an example and using C# computer programming language,the train traffic control simulation platform was built,which was used to simulate the train traffic control built in this paper.Finally,the simulation results proved the correctness and the validity of the model.