Assessment of An Evolutionary Particle Swarm Optimizer with Inertia Weight

This paper proposes a newly evolutionary particle swarm optimizer with inertia weight (EPSOIW) for obtaining the PSOIW with high performance. Due to the use of meta-optimization, it can systematically estimate appropriate values of parameters in the PSOIW corresponding to a given optimization problem without prior knowledge. Accordingly, the EPSOIW could be expected to not only obtain an optimal PSOIW for efficiently solving a given optimization problem, but also to quantitatively analyze the know-how on designing it. To demonstrate the effectiveness of the proposed method, computer experiments on a suite of multidimensional benchmark problems are carried out. We investigate the intrinsic characteristics of the proposal, and compare the search ability and efficiency with the other methods. The obtained experimental results indicate that the search performance of the PSOIW optimized by the EPSOIW is superior to those of the original PSOIW, OPSO and RGA/E. The EPSOIW is verified to be relatively high in the processing capacity for solving multimodal problems in comparison with the EPSO and ECPSO.