Study On Multi-Objective Optimization Method Of Medium-Speed Maglev Timetable

As a new type of urban railway transportation,the medium-speed maglev(MSM for short)with speed of 200 km/h serves commuter passenger flow in urban and suburban areas.With national policy support,it has great prospects for development.Timetable is an important bridge to communicate transportation supply and demand.In order to achieve the goal of high-efficiency and energy-saving,research on optimization problem of MSM timetabling is developed.The main work includes following aspects.(1)Analyze problems related to multi-objective optimization of MSM timetable.Firstly,discuss relationships among subsystems,then analyze characteristics of passenger demand and train operation.Consider the minimum passenger travel time and the minimum energy consumption as optimization objectives.Meanwhile,the restrictive relationship between objectives is studied.Finally,sort out the correlation between key variables and objectives by analyzing influence factors,which lays the foundation for modeling.(2)Establish the multi-objective optimization model of MSM timetable.According to Rolling Horizon Procedure method,total operating period is divided into several segments.Each of segment is a basic unit for timetabling,named "partial time windows".In each partial time windows,mathematical expressions of relationship between timevarying passenger flow and energy consumption of train operation are developed.The model takes the sum of general cost of passenger travel time and train operation energy consumption as objective function.Besides,train departure time and running time in each section are considered as decision variables.With consideration of constraints from different aspects,a multi-objective optimization model of MSM timetable is established.(3)Construct a multi-objective-oriented optimization algorithm of MSM timetable.With complexity analysis of the model,a non-dominated sorting genetic algorithm with elite strategy is employed.It takes the sum of passenger travel time and general cost of train operation energy consumption as fitness function.Moreover,specific genetic operators are designed.The model is solved through limited iterations.(4)Verify the correctness of model and effectiveness of algorithm.Firstly,input data,such as lines,trains,and passenger flow statistics,are brought into the model according to variable structure designed by algorithm.The operator parameters of crossover and mutation are set for iterations.Then the set of initial timetables and optimized result after iterating are obtained.One of the better individuals in the initial population is selected as control group,compared with an optimized timetable,which is randomly chosed as well.The total energy consumption and passenger travel time indicators are improved by 1.2%and 1.9% respectively,which proofs the effectiveness of model and algorithm.