Hybrid Particle Swarm Optimization Algorithm Based on the Simplex Method

To overcome the premature convergence and low execution efficiency of the Particle Swarm Optimization (PSO) algorithm, this paper presents a hybrid particle swarm optimization algorithm based on the simplex method. This algorithm firstly initializes particles according to uniform distribution to improve the diversity of the particles. When selecting initial particles, personal and global optimum particles are chosen by a set probability. Local optimization will be performed by the simplex method and then the original individuals will be replaced in every iteration. The strong local search function of the simplex method provides an effective mechanism for PSO algorithms to escape from the local optimum, which avoids the prematurity of the algorithm. Simulation results show that this algorithm features a function of stronger global search and faster speed of convergence than conventional PSO, so that the optimization process can be improved remarkably.