An efficient structural optimisation algorithm using a hybrid version of particle swarm optimisation with simultaneous perturbation stochastic approximation

Abstract

An efficient structural optimisation method based on a combination of simultaneous perturbation stochastic approximation (SPSA) and particle swarm optimisation (PSO) is presented. This new method is called the SPSA–PSO method. The operation of SPSA–PSO includes three phases. In the first phase, a preliminary optimisation is accomplished using the SPSA. In the second phase, an efficient initial swarm is produced using the first phase results. In the last phase, the PSO is employed to find the optimum design using the efficient initial swarm. Various benchmark truss examples are considered to demonstrate the effectiveness of the suggested method. The numerical results indicate that the SPSA–PSO is a powerful and computationally efficient method for solving the structural optimisation problems when comparing with other methods found in the literature.

Keywords:

• structural optimisation
• simultaneous perturbation stochastic approximation
• particle swarm optimisation
• structural analysis reduction