Towards a rigorous formulation of the space mapping technique for engineering design

The paper deals with the space mapping (SM) approach to engineering design optimization. We attempt a theoretical justification of methods that have already proven efficient in solving practical problems, especially in the RF and microwave area. A formal definition of optimization algorithms using surrogate models based on SM technology is given. Convergence conditions for the chosen subclass of algorithms are discussed and explained using a synthetic example, the so-called generalized cheese-cutting problem. An illustrative, circuit-theory based example is also considered.