Research And Application Of Particle Swarm Optimization Algorithm Based On P System

Membrane Computing(MC)is a new branch of biomimetic computing.It was proposed by Professor G.P?un,a member of the Romanian Academy of Sciences and the European Academy of Sciences.The theoretical model of membrane computing is also called Membrane System or P system.The model idea is inspired by the biological world,and is devoted to analyzing and exploring how to abstract theoretical calculation models from the internal structure and function of biological cells and to conduct in-depth research.Membrane systems have theoretically proven to be a distributed and highly parallel computing system.With today's computer hardware conditions not meeting the real-world requirements for high-performance parallel computing capabilities,the membrane system with a distributed structure and extremely parallel computing capabilities has evolved into a hot area of biomimetic computing.Particle Swarm Optimization(PSO)was proposed by Kennedy and Eberhart in 1995.The algorithm abstracts the theoretical calculation model of the optimization algorithm by simulating the flight foraging behavior of birds in nature.In the particle swarm optimization algorithm,there are a speed update formula and a position update formula to adjust the flight direction of the particle itself to ensure that the particle is oriented toward the food.All the particles in the group have the ability to remember,according to their personal best position and the global best position of all the particles in the population,constantly adjusts its position dynamically,and iteratively optimizes to find the optimal value.Through the mutual cooperation between individuals in the population,each particle learns by continuously learning the personal best position and the global best position in the population,and finally obtains the optimal solution.This is a typical Swarm Intelligence(SI)iterative optimization algorithm.In view of the advantages of few parameters,simple implementation flow and fast convergence speed,the particle swarm optimization algorithm has been widely used in function optimization,cluster analysis,neural network and many other fields.Therefore,based on the above theory and the membrane system model,improved particle swarm optimization algorithms based on P system are proposed.The improved particle swarm optimization algorithm can combine the membrane rules of membrane system to realize parallel computing and inter-membrane communication.It can improve the convergence speed and search accuracy of the algorithm itself.In summary,the main research contents and innovations of this paper are as follows:(1)Use Logistic chaotic mapping theory to initialize the population,and the average of all particle positions?personal best and global best are jointly introduced into the speed update formula and set the adaptive adjustment strategy to make the particles adaptively update,called Chaotic Self Adaptive Particle Swarm Optimization(CSAPSO),and combined with the cell like P system to achieve the optimal particle interaction between multiple populations while the algorithm performs parallel computing;(2)In the middle and late iteration,when the position of the particle is close to the individual optimal value and the global optimal value,the velocity of the particle tends to zero and it is easy to fall into the local optimum and it is not easy to jump out of the local optimal.Gaussian Particle Swarm Optimization(GPSO)is proposed.Gaussian function is introduced in the algorithm to randomly select the particle position.When the algorithm falls into local optimum,it helps the algorithm to jump out of local optimum.Combined with the tissue like P system not only can guarantee the population diversity of the algorithm but also can accelerate the convergence rate;(3)Applying the proposed TP-CSAPSO algorithm into partition clustering to carry out research and comparison experiments,and testing the performance and analysis of the algorithm through an artificial data set and three UCI dataset;The experiment results with different clustering datasets show that the proposed CP-CSAPSO algorithm can achieve better clustering results.