Multi-objective genetic algorithm optimization of composite sandwich plates using a higher-order theory

The main purpose of this paper is multi-objective optimization of soft-core composite sandwich plates using a known accurate high-order sandwich plate theory. In this three-layer theory, high-order kinematic assumptions are dedicated to each face sheet and the core layer. This theory considers the transverse flexibility of the soft core and satisfies transverse shear stresses continuity conditions and zero transverse shear stresses conditions at the upper and lower surfaces of the plate. The governing equations for bending and buckling analyses are derived using Hamilton’s principle, and their analytical solutions are presented for cross-ply plates. Two multi-objective optimization problems consist of weight/deflection optimization and weight/buckling load optimization of the unidirectional and cross-ply sandwich plates being studied. The thicknesses of the core and the face sheet layers are set as problems variables. The genetic algorithm is employed to find the optimal solutions for the problems, and the results are presented in form of the Pareto front. The accuracy of the present modeling and optimization method is evaluated for special case of buckling load maximization of laminated composite plates. For each problem, the optimization process is continued for more than 200 generations, but the results don’t change after 100 generations and the optimized results are obtained. The results confirm that there is no significant difference between the optimal solutions of unidirectional and cross-ply sandwich plates in both optimization problems. The process statistics show that cross-ply layup optimization takes about 25% more time than the unidirectional layup.