Improved PSO based on the Uniform Search Strategy

Particle Swarm Optimization (PSO) algorithm is a new optimization approach, which has been widely used to solve various and complex optimization problems. However, there are still some imperfections, such as premature convergence and low accuracy. To address such defects, an improved PSO is proposed in this paper. The improved PSO algorithm introduces a uniform search strategy that makes particles proceed alternately between basic movement and uniform search movement, which can ensure sufficient search over the entire space as well as the convergence of particles. Meanwhile, the learning object of the particle swarm is no longer a single particle, which is helpful to prevent particles from being trapped in local optima. The experimental results on the typical functions demonstrate that the improved algorithm has good performance in terms of precision and convergence when compared with other variants of PSO.