Laminated composites are highly in demand for the applications where high strength
and stiffness are required at less weight. They generally fail due to buckling, as they
are modeled as thin plates and are loaded compressively. Therefore, the design
parameters of the laminated composite plates are to be optimized for the
multiple-conflicting objectives buckling strength and weight. However, the composite
plates, which are used in real world applications, are to be made with cut-outs and
finite element analysis is required to analyze them. As it makes the optimization
process more complex, a methodology is proposed in this paper to carry out a
multiobjective optimization for the rectangular composite plate made with a
central elliptical cut-out. The nondominated solutions are obtained using
nondominated sorting genetic algorithm (NSGA-II) in which the multilayer
feed-forward neural network is used to replace the time consuming finite
element analysis. The numerical results show that the proposed method finds
the nondominated solutions efficiently and reduces the computational cost
prominently.
Keywords
stacking sequence optimization, artificial neural network,
NSGA-II, finite element analysis