An improved particle swarm optimizer for mechanical design optimization problems

TLDR

The study proposes an improved particle swarm optimizer for mechanical design optimization problems involving mixed integer, discrete, and continuous variables and problem‑specific constraints. The algorithm introduces a fly‑back constraint‑handling mechanism to keep the population feasible and extends standard PSO to accommodate mixed variables via a simple scheme. The method solves five benchmark engineering optimization problems, demonstrates better convergence and performance than existing techniques, and is easy to implement. Keywords include evolutionary algorithms, particle swarm optimization, constrained optimization, and mechanical design; the authors thank Dr.

Abstract

Abstract This paper presents an improved particle swarm optimizer (PSO) for solving mechanical design optimization problems involving problem-specific constraints and mixed variables such as integer, discrete and continuous variables. A constraint handling method called the ‘fly-back mechanism’ is introduced to maintain a feasible population. The standard PSO algorithm is also extended to handle mixed variables using a simple scheme. Five benchmark problems commonly used in the literature of engineering optimization and nonlinear programming are successfully solved by the proposed algorithm. The proposed algorithm is easy to implement, and the results and the convergence performance of the proposed algorithm are better than other techniques. Keywords: Evolutionary algorithmsParticle swarm optimizationConstrained optimizationMechanical design Acknowledgement The authors would like to acknowledge Dr. Carlos Coello for his helpful discussions.