Research And Application Of Genetic Algorithm Based On The Principle Of Membrane Computing

Membrane computing is a new branch of natural computing, which is abstracted from the structure and function of cells, tissues and organs out of a parallel computing model, it is a new field of scientific research. The structure and principles of membrane computing are often used to other areas of research for reference or transferred, for example, the membrane computing has been applied to the swarm intelligence algorithm, the combination of membrane computing with particle swarm algorithm(PSO), fish algorithm, ant colony algorithm and genetic algorithms designing a better performance algorithm, but the most combinations are based on the membrane computing structure but the principle of membrane computing. Therefore, this paper simply introduces the basic principle of membrane computing, focusing on the three characteristics of membrane computing principle:synchronization, uncertainty and parallelism, these three characteristics are applied to traditional genetic algorithm, designing a new optimization algorithm-genetic algorithm for a single membrane.Because of the different design concepts and methods, the single membrane GA is divided into two forms:the first form is to use the rules with equal probability in a single membrane genetic algorithm. In the design process of the traditional genetic algorithm, selection, crossover and mutation are changed from serially running to parallel, and the three operators are regarded as the evolutionary rules of the membrane computing, and the three probabilistic operators use the same probability; The second form makes a bit improvement based on the first design method, changing the equal probabilities rules to unequal probabilities. The probability of selection, crossover and mutation rules is uncertain. Through experiments identify the impact of the probability of these three rules on convergence time and convergence results. Then on the matlab software, using real-coded manner completes the compilation process. Finally, through two classic test functions to test and verify the effectiveness and feasibility of the new algorithm.After verifying the effectiveness of the single membrane GA. The single membrane GA is applied to practical problems, verifying the feasibility of solfing problem. The new algorithm is applied in two areas:First, the nonlinear correction parameter optimization on the control system; Second,the price determination on grid bids mechanisms. Through Matlab and Simulink simulation, getting the non-linear correction parameter optimal results, and the price that the power companies and users are satisfied with.