| In recent years,the researches on train virtual coupling have been widely carried out at home and abroad.The main characteristics of virtual coupling are: using vehiclevehicle communication technology instead of physical connection to make multiple trains run at the same speed at smaller intervals,thereby treating multiple trains as one train to increase line capacity and transportation flexibility.The realization of train virtual coupling relies on the mature train control system based on vehicle-vehicle communication,so it is planned to design the virtual coupling as a functional module under the train control system based on vehicle-vehicle communication.Since virtual coupling is different from traditional train operation modes in terms of functional requirements,module functions and application scenarios,in order to ensure the correct and reliable realization of the function of train virtual coupling,this thesis is based on the train control system based on vehicle-vehicle communication,with virtual coupling function modules,virtual coupling implementation process and virtual coupling application scenarios as the research objects.A functional model of train virtual coupling is proposed,and on this basis,the virtual coupling function modules and typical scenarios are modelled and verified.The work accomplished in this thesis is as follows:(1)The functional modeling and verification framework for train virtual coupling is constructed.By combining the characteristics of both Sys ML and CPN modeling languages,a top-down hierarchical modeling method is adopted.The formalized modeling and verification process of virtual coupling function is proposed,including scenario description,scenario classification and hierarchical model construction,etc.,and the conversion rules from Sys ML model to colored Petri net model are emphatically designed.Based on CPN Tools used for model simulation and verification,the logic function verification and state space analysis process of the train virtual coupling function model is designed,and then an overall framework for train virtual coupling function and verification based on vehicle-vehicle communication is constructed.(2)The functional module composition of virtual coupling is designed.First,based on the analysis of the similarities and differences between the virtual coupling operation mode and the traditional train operation mode,the virtual coupling functional requirements are proposed;then combined with the structural characteristics of the train control system based on vehicle-vehicle communication,in order to make the virtual coupling function operate normally and to reduce system complexity,the structure of the virtual coupling function module and the interaction between the modules are designed;then,based on the virtual coupling functional requirements,the switchover between the virtual coupling operation mode and the existing train operation mode,as well as the realization process of the virtual coupling,are designed.And for the two scenarios of train emergency braking and passing through the switch,the minimum safe interval of the train virtual coupling is determined.(3)The formalized modelling and verification of typical scenarios of train virtual coupling are completed.Three typical scenarios including virtual coupling application,virtual coupling maintenance,and decoupling are chosen for modeling.Firstly,starting with scenario recognition to construct Sys ML sequence diagram and state diagram models,and to construct hierarchical colored Petri net models according to the conversion rules and hierarchical abstraction methods.Secondly step-by-step simulation and statespace analysis of the model are carried out.Finally the correctness and completeness of the design of the train virtual coupling function are verified.(4)A decision model is designed and the effectiveness of the strategy for the scenarios that may appear in the virtual coupling operation is verified.Starting from the necessity of scene decision-making,the requirements of scenario decision-making are proposed,and the principle of scenario decision-making is designed;on this basis,scenario recognition and decision model construction are carried out based on finite state machine,and the artificial potential field method is used as the transition condition between states.Finally,the results of simulation and modeling verification based on Simulink and Stateflow shows that the decision model can effectively solve the problem.There are 75 graphs,27 tables and 68 references in this thesis. |
