Particle Swarm Optimizer with Shuffled Subpopulations Mutation and Its Application

Particle Swarm Optimizer (PSO) with the idea of SSMPSO (shuffled subpopulations and mutation PSO) was improved. In this algorithm, the swarm is divided into two subpopulations by the fitness of particles and all particles will be shuffled together to be a new swarm again if the terminal conditions don’t satisfied after certain iterations. Some of the particles with poor position will be mutated to other particles with better position of the swarm. The process will be repeated until the terminal conditions to be satisfied. SSMPSO was tested with some benchmark problems and compared with traditional PSO. The results indicate that SSMPSO performs better. Furthermore, SSMPSO is applied to train artificial neural network to construct a soft-sensor of gasoline endpoint of crude distillation unit. The results show that the model constructed by SSMPSO is better than that of PSO.