Research On Railway Traffic Distribution Model And Its Improved Grey Wolf Solving Algorithm

In our country,traffic distribution and route optimization is a very important basic work.On the one hand,it can realize the matching between the line capacity in the road network and the actual transportation demand,and on the other hand,it can make the running route of the traffic flow get optimization,thereby greatly reducing the detour distance and transit time of the traffic flow.At the same time,with the development of the economy,the competition in the transportation market has become increasingly fierce,and how to optimize the railway traffic distribution plan has become the key to improving the competitiveness of railway transportation.In this context,this paper studies the traffic distribution in railways from the perspectives of mathematical modeling and algorithm solution.Through Mathematical modeling,improving models,designing algorithms,and example verification to find an optimized method for traffic distribution in railway transportation,so as to provide a certain theoretical basis and basic support for the further planning and adjustment of the railway network.The actual work and group research contents of the thesis are as follows:(1)The basic theory of railway traffic distribution and route optimization is expounded and summarized.We have concluded the definition of traffic distribution and traffic path,and analyzed the principles and influencing factors of traffic distribution in actual production.Finally,we have generalized and summarized commonly used methods in traffic distribution.(2)The basic model of railway traffic distribution and route optimization was studied,and the model is improved by combining the solution principle of swarm intelligence algorithm.Firstly,the structure and basic model of railway vehicle flow path are described,then the difficult constraints that limit the solution of the model by the swarm intelligence algorithm are found,and the road network and the model are improved by introducing virtual arcs and penalty functions,so as to achieve the effective solution to the difficult constraints and infeasible flows in the model.deal with.Finally,Finally,we verify the effectiveness of the model improvement strategy through an example.After the model was improved,the swarm intelligence algorithm can effectively solve it.(3)In order to facilitate the solution of the model,we apply the improved gray wolf algorithm to the solution of the improved model in(2).First,the basic principle of the gray wolf algorithm is expounded,and then an improved strategy is proposed from the three aspects of the gray wolf algorithm’s convergence factor,step size and weight.At the same time,the classic test function is used to conduct simulation experiments.The traditional gray wolf algorithm,particle swarm algorithm,differential evolution algorithm and other classic swarm intelligence algorithms are compared,and the effectiveness of the proposed improved strategy is verified from three aspects of algorithm solution accuracy,convergence speed and stability.Finally,combined with the characteristics of the model in(2),the improved gray wolf algorithm is applied to the solution of the traffic distribution model.(4)In order to verify the practicability of the research contents of(2)and(3)in road networks of different scales,two different examples of small-scale road networks and large-scale road networks are designed to verify the improvement strategies of the model and algorithm.analyze.Firstly,the structure of railway freight network is introduced,and based on this structure,two different scales of road network and OD traffic flow are designed.Then the improved model in(2)and the algorithm in(3)are applied to the solutions of the two cases.Finally,through the comparative analysis of the results,it is confirmed that the improved strategy of the model proposed in(2)and the improved strategy of the algorithm proposed in(3)are equally applicable under different scales of road networks.It provides a theoretical basis for the future railway network traffic distribution and route optimization.