Membrane computing based particle swarm optimization algorithm and its application

Intelligent heuristic algorithms have been paid more and more attention in solving large-scale, complex optimization problems. Membrane computing is a new branch of natural computing with the features of distribution and great parallelism. PSO is also a simple and effective intelligent computing method. Considering the features of membrane computing and PSO, a hybrid algorithm MCBPSO is proposed in this paper. In MCBPSO, PSO is introduced into the computing model of membrane system. Meanwhile, cooperation and mutation strategy are also established in the hybrid algorithm to improve the performance. The mechanism of cooperation can help the algorithm improving the efficiency. Mutation operations help the algorithm to jump out of local minima and improve the precision. An application of MCBPSO is also presented. MCBPSO and LS-SVM are used together in soft sensor modelling of the components of Texaco gasifier syngas. Simulation results shows that MCBPSO has the best performance in the comparing test.