Research And Application Of The Improved Particle Swarm Optimization Algorithm Based On P Systems

The first membrane computing conference convened on the new century,marking the rise of membrane computing research.Calculation model for the membrane from the nature of living cells and tissues or organs composed of cells out of the structure and function of the model or computing thought.It is a hierarchy model of distributed and parallel computing.Research has shown that the membrane calculation model and Turing machines are equivalent.The PSO algorithm is one of the representative algorithm in the field of swarm intelligence algorithm.Because of that the PSO algorithm proposed only for single objective optimization problem at the begainning,the PSO algorithm can't deal with constraint condition and multi-objectiv.The mechanism that how to introduce the corresponding strategy to enables the algorithm to deal with this kind of problem,will decide whether the algorithm can continue moving forward and be the key to whether it can be transformed into productivity.There are few researchers have pay attention to the improvement and combination of PSO algorithm and the membrane model,but only with the aid of the height parallelism of the membrane system to realization algorithm l under the framework of the membrane system.In this paper,we explore the improvement of particle swarm optimization(PSO)algorithm,the improvemen of cells class P system,and the best structure and corresponding fusion rule of them.On the basis that the algorithm'realization,we research improve the algorithm to make it performant more efficiently.What more,the particles in PSO optimization mechanism will be more perfect,species diversity will be more rich,the structure of membrane system will be more optimized,and high parallelism of the system operation will be more obvious.In this paper,the main research contents are as follows:The one is put forward the NPSO-P system that the P system based on inverse particle swarm optimization algorithm.Also we introduced the design of NPSO-P system from aspects of the data storage structure,experimental framework,the execution of rules.In addition,the new put forward P system's effectiveness is verified by experiment.The second is put forward CODPSO algorithm that based on reverse learning mechanism of particle swarm optimization algorithm.Firstly,we introduced the optimal points,including the parameter design of inertia weight changes,asynchronous learning factor and time factor,as well as the speed limit and rebound strategy to solve the problem of rebound.Moreover,the system based on reverse learning mechanism can increase population diversity,avoid falling into local optimum.Finally,we prove that the optimal points are effective through the experiment of the specific element and entirety,The third is put forward CODPSO-AEPS algorithm that dynamic weighting particle swarm algorithm based on numerical membrane enzyme system.Based on the idea of group collaboration,the CODPSO-AEPS algorithm combines the operation mechanism of CODPSO algorithm and hierarchical structure as well as rules of membrane system.Each particle's position information will be seen as objects in membrane area.The operation of rules can realize information exchange and location optimization between particles,membrane fusion rules can realize the selection of values strategy.So it can accelerate the algorithm convergence speed,makes the optimal solution more accurately.The fourth is applying CODPSO-AEPS algorithm to solve the base station site selection optimization problem.After research the basic idea of the base station planning problem in current TE-LED network,we construct appropriate mathematical model combined with concrete problems.Also we using the proposed algorithm for optimization to get the base station site location information solution set.