Research On Train Station Scheduling Problem Based On Ant Colony Optimization Algorithm

The rapid development of railway traffic in our country has brought great challenges to the scheduling process.In the process of railway dispatching,different station types and surrounding environments of stations make the scheduling process,algorithm and the system structure are all different.Therefore,the research of scheduling model and corresponding scheduling algorithm for different station scheduling target diversity and performance requirements has certain theoretical significance and strong practical value.The scheduling of receiving-departure line is a typical multi-objective scheduling problem.In order to meet the diversified optimization goals of different stations,this paper presents a multi-objective scheduling model which can adjust the weights of each optimal target,and uses the hybrid behavior ant colony algorithm to solve the problem.This paper describes the process of train dispatching in the receiving-departure line of railway station,analyzes the time occupation of train dispatching to the receiving-departure line and the turnout group in throat area,and extracts the corresponding constraint rules.On this basis,the three different scheduling optimization targets,which are based on the train occupancy to the receiving-departure line equilibrium,the optimal train operation time and the priority based on train docking to the receiving-departure line,are analyzed respectively.By using the principle of multi-objective programming,a multi-objective platform scheduling model with corresponding constraints is established.Aiming at the above problems,the ant colony algorithm based on hybrid behavior is designed,the method of total pheromone adjustment is improved,and the pheromone volatilization factor adaptive algorithm is proposed to optimize the multi-objective scheduling problem.Finally,taking the train-to-departure dispatching of a station as an example,the influence of the adjustment of each target's weight on the overall scheduling results is analyzed.The analysis of optimization results verifies the feasibility of the proposed optimization model and algorithm.