Screening and response surface method applied to the numerical optimization of electromagnetic devices

Abstract

The solution of a design problem of an electrical motor shows how the experimental design method can be combined with the numerical simulation and how this approach is well adapted to the optimization process. Firstly, the screening of significant geometric parameters is presented. Then the Response Surface Methodology is used to build an analytical model that is finally employed as a constraint in an obvious optimization process. As a result, the performance of the motor is increased.