| With the rapid development of high-speed trains in China,traveling by high-speed trains has become one of the popular modes of transportation.When an emergency occurs in the process of train running,the train dispatcher adjusts manually by virtue of experience.However,the manual adjustment method to deal with complex delayed scenes still has its limitations,and it is difficult to achieve the optimal from a global perspective.In order to reduce the work intensity of dispatchers,reduce the impact of human factors on train scheduling and minimize the impact of delays,it has become a technical development trend of high-speed railway scheduling command system to introduce computer optimization algorithm to assist in generating train operation adjustment plan.In this thesis,the ant colony algorithm is used to optimize the train operation adjustment model.Firstly,taking the minimum train delay time as the objective function,the TSPLIB algorithm library is designed to solve the problem on the basis of the construction of the original ant colony algorithm solution space,and its shortcomings are analyzed.Secondly,the solution space structure of the ant colony lgorithm is reconstructed to optimize the minimum objective function of the total delay time of trains.The optimization results of the ant colony algorithm in the solution space are compared with those of other methods to prove the advanced performance of the ant colony algorithm proposed in this paper.Finally,a multi-objective ant colony algorithm was designed based on the construction of the solution space to optimize the two objective functions of minimizing the total train delay time and balancing the train delay time,so as to further improve the comprehensive performance of the optimization results.The main research contents of this paper include the following parts:(1)Firstly,by analyzing the advantages and disadvantages of domestic and foreign literature in solving the problem of train operation adjustment,the reasons for using ant colony algorithm to solve the problem of train operation adjustment are drawn out Secondly,it describes the related concepts of train operation adjustment,understands the relevant definitions and characteristics of high-speed railway train operation adjustment from the overall perspective,and analyzes the principles and measures of train operation adjustment,so as to make more reasonable use of computeraided algorithm for optimization.Analyzing the reason of train delay and the principle of delay propagation is convenient for model construction.In this paper,the train operation adjustment problem is deeply analyzed and the relevant constraints of train operation adjustment are symbolized.In order to improve the comprehensibility of the train operation adjustment scheme,based on the objective function commonly used in the train operation adjustment to minimize the total train delay time,the train delay time equilibrium objective function is innovatively proposed,and multi-objective optimization is carried out in the subsequent chapters.(2)The problem of train operation adjustment is deeply analyzed,and the relevant constraints of train operation adjustment are symbolized.According to the relevant constraints of train operation adjustment,the train operation adjustment model is established with the objective function of minimizing the total train delay time.(3)The selection of arrival and departure time for train operation adjustment is transformed into a two-dimensional traveling salesman problem through coordinate transformation,and the TSPLIB algorithm library is improved to solve the problem.This paper analyzes the disadvantages of using the traditional construction of the solution space of ant colony algorithm to solve the problem of train operation adjustment,and points out the necessity of reconstructing the solution space of ant colony algorithm(4)According to the characteristics of train operation adjustment problem,the solution space structure of ant colony algorithm is reconstructed,and the corresponding heuristic information is designed.The minimum objective function of train total delay time is optimized,and the optimization results are compared with those of FCFS and FSFS,which are commonly used in manual scheduling,to prove the advancements of ant colony algorithm.The results are compared with integer programming and particle swarm optimization.(5)Based on the solution space construction of the algorithm in this thesis,the objective function of train delay time equalization is added,and the multi-objective optimization is carried out.The heuristic information and pheromone formula of ant colony algorithm were redesigned for multi-objective problems,and the final nondominated solution set was optimized.In order to analyze the advantages and disadvantages of the multi-objective,a solution of the non-dominated solution set was selected to generate the adjusted train schedule,and the stability and anti-interference ability of the scheme were comprehensively analyzed.In this thesis,the improved ant colony algorithm is used to recover and adjust the train delay caused by an emergency,so as to better assist the train dispatcher to conduct traffic scheduling,and provide a theoretical basis for the realization of intelligent train operation adjustment,which has certain practical significance... |
