Optimal design of stiffened laminate composite cylinder using a hybrid SFL algorithm

Abstract

Shuffled frog-leaping algorithm (SFLA), a memetic meta-heuristic, is proved to be a successful combinatorial optimization algorithm and applied to several engineering problems. However, its potential is not explored in the context of laminate stacking sequence optimization of composite structures. In this article, we propose a hybrid version of SFLA for solving the combinatorial optimization problem associated with lay-up sequence optimization of laminate composite structures. In order to improve the computational performance as well as reliability of the optimal solutions, a customized neighborhood search algorithm and an adaptive search factor are incorporated in to the SFL algorithm to accelerate the convergence characteristics. Apart from this, a crossover operator is suitably incorporated in the proposed hybrid SFLA. Numerical experiments have been carried out by first considering the problem of buckling and failure load optimization of composite plate and later, optimal design of stiffened composite cylindrical shells. Superior convergence characteristics and robustness of the proposed hybrid SFLA is demonstrated by comparing with other popular meta-heuristic algorithms including genetic algorithms.