A new parallel approach for the exploitation of the search space based on PSO algorithm
Maria Zemzami, Norelislam Elhami, Mhamed Itmi, Nabil Hmina · 2016
The objective of this paper is to propose a new parallelization approach of Particle Swarm Optimization (PSO). PSO is an intelligent optimization algorithm based on swarm intelligence, it is not only simple, easy to implement but also can be used to solve various complex problems that are difficult by using traditional calculation methods. In this work, a novel approach is presented for avoiding premature convergence to local minima based on maintaining a balance between exploration and exploitation of the search space, besides, the reduction of computational costs by using parallel computation. This algorithm is applied to various benchmark problems including multimodal test functions. The effectiveness of our algorithm is discussed.