Parallelizing a Modified Particle Swarm Optimizer (PSO)
Recently, a modified Particle Swarm Optimizer (MLPSO) has been succeeded in solving truss topological optimization problems and competitive results are obtained. Since this optimizer belongs to evolutionary algorithm and plagued by high computational cost as measured by execution time, in order to reduce its execution time for solving large complex optimization problem, a parallel version for this optimizer is studied in this paper. This paper first gives an overview of PSO algorithm as well as the modified PSO, and then a design and an implementation of parallel PSO is proposed. The performance of the proposed algorithm is tested by two examples and promising speed-up rate is obtained. Final part is conclusion and outlook.
B. Yang and Q. L. Zhang, "Parallelizing a Modified Particle Swarm Optimizer (PSO)", Advanced Materials Research, Vols. 163-167, pp. 2404-2409, 2011