Optimization Of Train Scheme Based On Hybrid Particle Swarm Optimization

With the rapid development of China's economy,and the passenger dedicated line of the urban economic exchange center has been completed.To meet the needs of the railway departments and passengers,more optimized train operation plan has to be developed.Train program is a key part of passenger transport organization,Therefore,this research focuses on the train stop scheme optimization.First of all,for traditional stop solution without fully considering the passenger travel time and segment accessibility issues,established a multi-objective train stop plan optimization model.Secondly,in order to solve the complex multi-objective programming problem of train stop plan,combined with particle swarm optimization and quantum genetic algorithm,an improved multi-objective quantum genetic particle swarm optimization algorithm is proposed.Finally,combined with a high-speed passenger flow data,solve the train stop optimized model with the above research improvement algorithm,and implement the auxiliary decision system of the train plan.The main work is as follows:(1)In view of the traditional train stop plan model without fully considering the different passenger train interval time loss and different levels between the station direct sexual problems,set up with least passenger travel time and segment accessibility of the multi-objective optimization model of train stop scheme.First,use LDA(Latent Dirichlet Allocation,LDA)theme model to analysis characteristics of passenger flow,and different passengers on the train and train stop time interval of time loses for passenger travel total time;Secondly,according to the coefficient of station grade to calculate the zone accessibility;Finally,according to the theory of multi-objective optimization put forward the thought of multi-objective optimization algorithm model.(2)In order to solve the multi-objective optimization problem of train stop plan,a multi-objective quantum genetic particle swarm optimization algorithm is proposed.The algorithm aimed at particle swarm optimization(pso)algorithm is easy to fall into local optimal solution,quantum genetic algorithm is introduced,uses the adaptive value ordering method to update the particle inertia weight,and adopt the improved speed update formula to update the quantum rotation gate angle.In choosing a balance optimal solution of the interests of the passengers and railway departments,proposed a construct Pareto optimal solutions method based on position ranking,which makes the algorithmconverge to Pareto frontier.Experiments show that the quality of the improved algorithm is 38.1% higher than quantum genetic algorithm,and the time complexity is decreased by 8.9%.The quality of the improved algorithm is 11.3% higher than that quantum particle swarm algorithm,and the time complexity is reduced by 39.6%.(3)According to the above established multi-objective stop scheme optimization model and the method of multi-objective genetic quantum particle swarm optimization algorithm,designs Gao Tie train stop solution.The experiment shows that the optimized train stop plan can improve the reachability of the section by 9.1%,reduced the travel time by 5.9% and reduced the total number of stoppages by 13.6%,while ensuring the full passenger flow.Finally,formulate strategies of the operation plan is used in passenger auxiliary decision-making system,automatically generated and adjust the train stop scheme,provide auxiliary decision-making for railway department.