Robust Optimization Utilizing the Second-Order Design Sensitivity Information

Abstract

This paper presents an effective methodology for robust optimization of electromagnetic devices. To achieve the goal, the method improves the robustness of the minimum of the objective function chosen as a design solution by minimizing the second-order sensitivity information, called a gradient index (GI) and defined by a function of gradients of performance functions with respect to uncertain variables. The constraint feasibility is also enhanced by adding a GI corresponding to the constraint value. The distinctive feature of the method is that it requires neither statistical information on design variables nor calculation of the performance reliability during the robust optimization process. The validity of the proposed method is tested with the TEAM Workshop Problem 22.