| As a significant national strategic project,high-speed railways provide essential support in terms of promoting economic construction,coordinating sustainable urban development,as well as serving passenger operations.Along with the continuous growth of passenger traffic and its increasing demand for service quality for high-speed railways,how to effectively guarantee the efficiency of traffic operation and line passing capacity is facing an unprecedented huge challenge.At present,in the actual operation of high-speed railways,dispatchers are faced with time-varying nonlinear and complex strongly coupled train operation status information.The dispatchers require considering the different needs of railway operation,passenger service,train rescheduling costs,etc.The dispatchers adjust the timetable based on the manual experience,and the degree of automation needs improvement.Although many scholars at home and abroad have researched the train rescheduling problem,most of them focus on a single objective and lack synergistic consideration of different optimization objectives.In this dissertation,from the actual demand of recovering train delay,the target speed profile,train platform scheme,and train timetable are used to study the situation of blockage and temporary speed limit along the line.The dissertation focuses on developing a multi-objective optimization method of train rescheduling under the coordination of train operation situations and platform utilization with evolutionary algorithms.The main contents of the dissertation are as follows.Firstly,a train operation situation analysis method based on dynamics characteristics is proposed under regional temporary speed restriction for the train delay scenario.The coupling mechanism of interval buffer time and arrival delay is studied.The constraints of train arrival and departure time at stations and tracking interval time in the block section are constructed.The train dynamics equations are established considering train traction and braking characteristics.The on-time performance,impact rate,and train delay are used as the optimization objectives to calculate the train running acceleration,speed,passing moment,and the actual interval running time other operating situation information.The train arrival delay at the station is predicted for train rescheduling.Secondly,considering the train platform scheme rescheduling problem at the high-speed railway station with train arrival delay and failure occupation in some arrival and departure tracks,a multi-objective optimization model is formulated to minimize the train delay and the number of train platform scheme adjustments.A mixed-encoding scheme with integer and permutation encoding is proposed to deal with the complex constraints,including safe interval time for inbound and outbound operations constraint,arrival and departure route conflict constraint,etc.According to different conflict scenarios,a conflict resolution method,a train platform assignment,and a train departure/arrival time decoding scheme are proposed.Population initialization,individual crossover,and mutation strategies are designed for the mixed-encoding scheme.Experiments show that train platform rescheduling schemes can be obtained in a short time,which effectively predicts the train departure delay for train timetable rescheduling.Finally,considering the train timetable rescheduling problem with train arrival and departure delay,a multi-objective optimization model is formulated to minimize the total train delay and the number of train arrival/departure time adjustments.A non-dominated sorting genetic algorithm with a multi-permutation-based encoding scheme is proposed.A multi-permutation encoding scheme is designed to deal with the complex constraints,including train arrival and departure interval time constraint,etc.According to the train order in different sections,a train departure/arrival time decoding scheme is proposed.Experiments show the effectiveness of the proposed algorithm,which helps provide quick and rational train timetable rescheduling schemes. |
