Buckling Load Maximization of Composite Laminates Using Particle Swarm Optimization With Various Constraints

The stacking sequence of composite laminates is designed to have maximum buckling load using the particle swarm optimization (PSO) algorithm. The original PSO algorithm is modified to handle the discrete ply angles and the constraints such as stiffness and 4-ply contiguity requirements. For this, the augmented Lagrange multiplier (ALM) method is incorporated into the PSO algorithm. For the verification of the algorithm, the benchmarking problems are solved and the results are compared with the ones from the genetic algorithm or the analytic solutions. And then the laminates under in-plane compressive loadings are optimized for maximum buckling load considering the various constraints. The numerical results show that the algorithm finds the optimum with relatively small number of iterations with satisfying the constraints explicitly. Considering its advantage of derivative-free and simple procedures, the proposed algorithm can be applied to more complex models coupled with finite element analysis and various constraints.

Optimum Stacking Sequence Design of Composite Sandwich Panel Using Genetic Algorithms

In the present study, both critical buckling load maximization and face-sheet laminate thickness minimization problems for the composite sandwich panel, subjected to bi-axial compressive loading under various imposed constraints have been investigated using genetic algorithms. In the previously published work, the optimization of simple composite laminate panels with only even number of laminae has been considered [1, 3]. The present work allows the optimization of a composite sandwich panel with both even and odd number of laminae in the face-sheet laminates. Also, the effects of the bending-twisting coupling terms (D16and D26) in bending stiffness matrix which were neglected in the previous studies [1, 2, 3], are considered in the present work for exact solutions. In addition effect of both balanced and unbalanced face-sheet laminates on the optimum solutions have also been investigated, whereas only balanced laminates were considered in the previous studies [1, 2, 3].