Frame structural sizing and topological optimization via a parallel implementation of a modified particle Swarm algorithm

As a comparatively new developed stochastic method — Particle Swarm Optimization (PSO), it is widely applied to various kinds of optimization problems especially of nonlinear, non-differentiable or non-concave types. In this paper, a Parallel Modified Guaranteed Converged Particle Swarm algorithm (PMGCPSO) is proposed, which is inspired by the Guaranteed Converged Particle Swarm algorithm (GCPSO) proposed by von den Bergh. Details in the algorithm implementation and properties are presented and, an analytical benchmark test and structural sizing and topological test cases are used to evaluate the performance of the proposed PSO variant, PMGCPSO exhibited competitive performance due to improved global searching ability and its corresponding parallel model indicates nice parallel efficiency.