The Research And Application Of Pso With Membrane Structure Based On Traffic Network Layout Problem

With the rapid development of science and technology in the world today and the increasing demand of application,there are a lot of complicated optimization problems in the field of engineering practice.Especially,the distribution problem of transportation network based on distribution center has become the topic hot of the research.Exact algorithms have some limitations in solving the problems of high complexity.Using bionic algorithms to solve some complexity problems has become the research direction of optimization fields.Particle swarm optimization(PSO)is an algorithm for simulating the foraging behavior of birds.It has the characteristics of simple structure,few parameters,easy realization and strong searching ability.Inspired by these advantages,this paper studies the research and application of particle swarm optimization in traffic network layout problem.PSO has the disadvantages of low efficiency and poor solution when solving large-scale problems.According to the characteristics of the problem of radial features and the randomness of PSO,this paper introduces different strategies to traditional algorithm from from different perspectives: For this problem that traditional algorithm has lower efficiency to adjust the feasibility of feasible solutions by checking factor,this paper uses the improved prim algorithm to producing initial population;Improved prim algorithm can only improve the efficiency of generating the initial population and cannot modify the efficiency of updating particle position in the regeneration stage,therefore an exchange-tree factor is introduced in the stage of updating particle position.The main purpose is to ensure that the algorithm effectively searches individuals satisfying the radial and the length constraints in the solution space through polynomial time,so as to further improve the computational efficiency of the algorithm.This paper studies the research and application of particle swarm optimization algorithm in traffic network layout problem,which is not ideal for solving the efficiency and precision of large-scale examples.Particle swarm algorithm based on membrane system is a bionic optimization algorithm designed by using a membrane system to guide the particles toward a better search space.Membrane algorithm studied in this paper is mainly used to solve the efficiency and precision problems encountered by traditional algorithm.For efficiency problems,this paper modifies membrane rules from the following angles: By introducing membrane techniques,all the velocity components of the particle are updated in a parallel way,which changes series calculation structures of all particle velocities so as to enhance the computing efficiency of membrane algorithm;Changing the series computing pattern of particle positions,the rule of membrane adopts the strategy that the spanning trees are updated according to cotree set,which can further improve the computing efficiency of the algorithm.For the problems of precision,this paper designs a roulette way to remove the deleted edges of circle set generated by exchange-tree operator,so as to ensure that these solutions of improved algorithm can be better than the previous one thus improve the accuracy of the algorithm.In this paper,we design a series of simulation experiments to verify the effectiveness of membrane structure and rules of membrane algorithm in solving the traffic network layout problem.The experimental results shows that the algorithm is competitive by comparing with PSO,GA and other heuristics.