Robust Global Optimization of Electromagnetic Devices With Uncertain Design Parameters: Comparison of the Worst Case Optimization Methods and Multiobjective Optimization Approach Using Gradient Index
Ziyan Ren, Minh-Trien Pham, Chang Seop Koh · IEEE Transactions on Magnetics · 2012
The uncertainties in design variables are unavoidable in the optimal design of electromagnetic devices, and there is an imperative demand to find a robust design, which is insensitive to the uncertainties and remains within the feasible region of constraints even perturbed by the uncertainties. In this paper, a gradient-based worst case optimization (G-WCO) algorithm is proposed in a limited uncertainty set to increase the numerical efficiency based on the worst case optimization (WCO) algorithm. Through applications to the robust optimal design of TEAM 22, the performances of the proposed G-WCO, conventional WCO, and multiobjective optimization approach using gradient index (GI) are compared.