Research On The Model And Optimization Algorithm Of Inter-city High-speed Railway Timetable

As a short-distance transportation mode for commuting passengers around the city circle,the intercity high-speed railway directly displays the passenger transportation services provided to passengers through the train timetable.In order to improve the service quality,enhance the attraction of passengers,meet the travel needs of passengers and maximize the benefits of enterprises,it is necessary to compile and optimize the train timetable.Mining the potential of railway transportation fully and realize the optimal allocation of transportation resources.This paper aims at the optimization of the intercity high-speed railway timetable.In order to solve the complex problems of the comprehensive optimization process,it adopts the divide-and-conquer,two-stage optimization method and the experimental verification of the Shanghai-Hangzhou Intercity High-speed Railway.The main work of the thesis is as follows:1.Study and analysis of the key elements,conflict resolution techniques and optimization algorithms that affect the train timetable.The study found that the nature of the run-line paving is based on the large-scale combinatorial optimization problem of the train's sequence in the interval.The operation line conflict resolution is the key to affect the quality and efficiency of the train timetable.Study and analyzed for different operational conflict,its resolution strategy is propsed.Futhermore,for the intensive transmission of intercity lines,how to establish the balance between the efficiency and stability of the train operation system is the goal of train timetable optimization.2.Based on the evolutionary algorithm,an improved algorithm for feasible train timetable is proposed.According to the essence of running line paving,the algorithm designs the two-dimensional coding method of “train-interval running sequence” based on the genetic evolution algorithm,and the conflict is solved by using the panning running line and expanding the initial time domain.The decoding algorithm of breadth-first loop routing is propsed.In order to improve the search efficiency of the algorithm,an adaptive mechanism is introduced in the process of individual evolution.Experiments show that the accuracy of the algorithm is 0.75% higher than that of the basic genetic algorithm,and the convergence speed is improved by 55.42%.It can be seen that the adaptive evolutionary algorithm has high efficiency and can quickly converge to the optimal solution,thus making efficient Conflict-free train timetable.3.Based on NSGA-II algorithm,an improved algorithm for robust optimization of train timetable is proposed.The algorithm introduces the virtual train parameters,adopts the improved congestion distance calculation and elite control strategy,and further studies the preparation of the feasible train timetable to optimize the train operation map structure and improve the robustness of the train timetable.The simulation of the sudden disturbance is carried out.The results show that with the increase of the number of virtual trains,the total running time of the train increases by 1%-2%.However,after the occurrence of operational disturbances,the volatility of the train operation plan is reduced by 0.5%-1.5%,the train's punctuality rate is increased,and the average late time of the terminal station and the intermediate station is reduced.