A CLONAL SELECTION ALGORITHM WITH VARYING ORDER FINITE ELEMENTS FOR THE OPTIMIZATION OF MICROWAVE DEVICES

摘要

A real-coded clonal selection algorithm (RCSA), which is a global search method based on the artificial immune systems paradigm, is applied for optimizing microwave devices. We couple RCSA with varying high-order finite elements (FEs) to control the balance between the global and local search efforts in the algorithm. In this scheme, low-order FEs are used in the global search, whereas varying high-order FEs are used in the local search. The proposed technique is investigated in the designs of an E-plane miter bend and a rectangular waveguide impedance transformer. The results illustrate the accuracy and computational savings of this method.