SURROGATE BASED OPTIMIZATION OF FUNCTIONALLY GRADED PLATES USING PSO AND DE

Resumo

Optimization procedures are often employed in the design of functionally graded structures to achieve
a superior efficiency. Thus, this paper aims at comparing the Particle Swarm Optimization and the Differential
Evolution algorithms when applied to the optimization of FG plates. The optimization is carried out using the
Isogeometric Analysis to evaluate the structural responses and Radial Basis Functions are used as a surrogate
modeling technique to lower the computation cost. A Sequential Approximate Optimization approach is used
to improve the surrogate model during the optimization process. The two optimization algorithms will also be
compared in terms of accuracy and computational efficiency.