Multi-layer-oriented Integration Of Train Rescheduling And Control For High-speed Railways

As China’s "bright card" in the new era,high-speed railways play a significant role in promoting coordinated regional development,developing a low-carbon economy,and advancing the construction of the "Belt and Road." With the increasing complexity of the high-speed railway network and the continuous demand for operation service quality,one unprecedented major challenge is to extract and excavate the complex multi-dimensional spatio-temporal information of train operation in the high-speed railway signal(HSRS)system.The other challenge is to improve the intelligence level of dispatching decisions and control strategies.Due to the complex and changeable operation environment of China’s high-speed railways,external emergencies quickly lead to a large delay in the network.Dispatchers and drivers need to make dispatching decisions and control strategies by experience timely.Nevertheless,the spatio-temporal information of the HSRS system needs more effective integration,decoupling,and excavating in the dispatching and control process.The tracking operation process of multiple trains in stations and block sections lacks a refined portrayal.The functions of train rescheduling and optimization control lack integrated and collaborative decision-making.As a result,the optimization and automatization of train rescheduling solutions and driving strategies need to be improved.With the objective of the functional integration of train rescheduling and optimization control,this dissertation explores the train spatio-temporal information in three different layers,namely,the dispatching layer,block section layer,and station layer,and proposes a multi-layer-oriented integration approach of train rescheduling and control for high-speed railways.This approach includes the macro-integration of the dispatch layer,the mesoscopic-integration of the block section layer,and the micro-integration of the station layer.The automatically generated train rescheduling solution and multi-train target speed profile assist dispatchers and drivers in adjusting the stage plan and train driving strategy in real time.The main contents of the dissertation are as follows.Firstly,this dissertation addresses the poor HSRS data utilization rate at the dispatching layer and the low automation of the train operation decision scheme.Adjusting the timetable by train dispatchers is regarded as the starting point of the macro-integration for train rescheduling and control in high-speed railways.The information interaction process is analyzed between the high-speed railway dispatching command system and trains.The dissertation constructs the macro-integration minimum system and information interaction model.A macro-integration approach is proposed under the mutual coordination of rescheduling and control functions.With the predicted arrival time of the control layer as input,a real-time train rescheduling approach is proposed to provide an optimal train departure sequence at the blockage station of the dispatching layer based on problem knowledge and the evolutionary algorithm framework.With this optimal train departure sequence of the dispatching layer as input,the calculation approach of the train target speed profile based on the problem knowledge of time-efficient driving strategy is studied to predict the train arrival time of the control layer for the subsequent stations behind the blockage station.The simulation results show that the optimal improved evolutionary algorithm can provide the optimal departure sequence of the affected trains at the blockage station in real time.The distance step of 10 m can effectively ensure the accuracy and efficiency of the train predicted arrival time.Compared with the non-integration approach,the macro-integration decision-maker can provide the train operation decision scheme in real-time,effectively reduce the number of manual operations by train dispatchers and assist in adjusting the stage plan.Secondly,as for the lack of real-time deduction for multi-train operation situations at the block section layer,this dissertation regards optimizing of multi-train tracking driving strategy as the starting point of the mesoscopic-integration of train rescheduling and control for high-speed railways.The mesoscopic-integration minimum system is established with the information transfer process between the high-speed railway dispatching command system,radio block center,and trains.The dissertation establishes a deduction model of multi-train operation situations based on the conversion relationship of train operation phases and kinematic equations.An online deduction approach of train situations based on expert system is proposed to improve the deduction solution efficiency and reduce the traction energy consumption.Two mesoscopic-integration approaches of multi-train micro and macro situation deduction are developed.The former approach can calculate the safety lower bound of the multi-train overspeed protection curve and train rescheduling solution.The latter approach can solve the train rescheduling solution and driving strategy to assist dispatchers and drivers in real-time through the mesoscopic-integration decision-maker.The simulation results show that compared with the non-integration approach,the mesoscopic-integration approach of multi-train macro-situation deduction can effectively reduce the total delay and the total traction energy and improve the efficiency of calculating the train operation decision scheme.Finally,as for the insufficient dynamicity of the route opening time for receiving and sending trains at the station layer,this dissertation considers the station micro-detail and train-route cooperation as the starting point of micro-integration of train rescheduling and control for high-speed railways.Considering the process of opening the route by station autonomous computer,the meaning and connotation of train-route cooperation are proposed,and the micro-integration minimum system is constructed.The model of train-route cooperation is established based on the micro station details and the tracking process of multiple trains arriving at stations.The dissertation proposes a refined calculation method of train tracking interval and a dynamic opening time of the station route.Two micro-integration approaches of multi-train micro and macro situation deduction are developed based on train-route cooperation to calculate the micro-integration train operation decision scheme.The simulation results show that with the increase of trains in the time domain,the micro-integration approach of multi-train macro situation deduction can calculate the refined train operation decision scheme,stopping target curve,and dynamic opening time of the station route under any delay scenario in real-time through the micro-integration decision maker.The micro-integration approach solves the frequent switching of train operation phases and assists train dispatchers in making timely emergency dispatching plans for severe delay scenarios.