Fast Hybrid Design Approach for Optimization of Inductors

The optimal design of magnetic components is fundamental for the overall optimization of power converters. This work presents a fast hybrid design approach for the optimization of inductors to be employed in such converters. The presented design process applies Geometric Programming (GP) to formulate the optimization problem, which is a type of non-linear convex optimization whose objective function and constraints have a particular form. The GP optimization is looped with an update of the inductor winding resistance obtained by direct interpolation of a multidimensional surface pre-computed using the finite element method (FEM).