Genetic Algoritim For Multiple Membrane Computing And Its Applications

As the youngest branch of natural computing,membrane computing(P system)mainly study how to build computing model from structure and function of cells,tissue of the organism,structure and function of organs.A lot of research showed that membrane computing model has computing power of the Turing machine.Membrane computing model can solve the difficult problem of computing in the allowed time range.Application of the membrane computing generally adopt optimization methods from the design of P system,which is designed to solve some problems in the practical application.Mainly in combination with genetic algorithms?particle swarm optimization?ant colony algorithm?simulated annealing algorithms,solving the traveling salesman problem?knapsack problem?robot path planning problem.But most studies do not take into account the exchange of communications between the systems.Du to the different concentration of cells and the environment,cells passive transport of chemical substances each other.This article will use this means of communication,and three characteristics of membrane computing apply to traditional genetic algorithm,devised a new optimization algorithm-genetic algorithm for multiple membrane.As described in this article,multiple membrane genetic algorithm,for three separate membrane structure,based on the exchange of communications between cells.The variable population genetic algorithms(VPGA)is applied to three membrane system.And changed the three operators of SGA from serially to parallel.This communication system based on concentration,which population can be distributed equally.VPGA can not make the population converge too fast,and can not waste computing resources.Not only improve the quality of the search but also take into account the effect of quick optimization.By using real-coded manner,this new algorithm program compiled in matlab environment.Test and verify the effectiveness of new algorithm through several classic test functions,and compare to the GA.Finally,after verify the effectiveness of multiple membrane genetic algorithm.Optimize the parameters of non-minimum phase system with this new algorithm.In the matlab environment,through combining multiple membrane GA and simulink simulation,achieve the parameter optimization of the control system and have get ideal control result.