Dr. Emre Özkaya, SciComp
Primal-Dual Aggregation Method for Surrogate Modeling of Expensive Functions
Global design optimization using Partial Differential Equations requires very reliable surrogate modeling techniques since the underlying simulations may be computationally very expensive. In general, for such problems, the number of samples that can be generated to train the surrogate models is very limited due to restricted computational resources. On the other hand, recent developments in adjoint methods enable nowadays evaluation of gradient information at a reasonable computational cost for a wide variety of engineering problems. Therefore, a much richer data set can be acquired using an adjoint solver in a Design of Experiment. In the present work, we present a novel aggregation method, which enables the state of the art surrogate models to incorporate gradient information without causing robustness problems. Therefore, accurate surrogate models using few samples and large number of design parameters can be constructed. We also present results from several design optimization studies showing the efficiency and the robustness of the new method.