A two-level surrogate-assisted optimization algorithm is proposed for electric machine design using 3-D finite-element analysis (FEA). The algorithm achieves the optima with much fewer FEA evaluations than conventional methods. It is composed of interior and exterior levels. The exploration is performed mainly in the interior level, which evaluates hundreds of designs employing affordable kriging models. Then, the most promising designs are evaluated in the exterior loop with expensive 3-D FEA models. The sample pool is constructed in a self-adjustable and dynamic way. A hybrid stopping criterion is used to avoid unnecessary expensive function evaluations.
Digital Object Identifier (DOI)
The support of University of Kentucky, the L. Stanley Pigman endowment and the SPARK program, and of ANSYS Inc. is gratefully acknowledged.
Taran, Narges; Ionel, Dan M.; and Dorrell, David G., "Two-Level Surrogate-Assisted Differential Evolution Multi-Objective Optimization of Electric Machines Using 3-D FEA" (2018). Power and Energy Institute of Kentucky Faculty Publications. 37.