Optimal design of frequency selective surface by genetic algorithm

Abstract

In this paper, we describe an autonomous design procedure of frequency selective surface(FSS) with genetic algorithm(GA). With a set of characteristics specified by operators in frequency domain, the proposed system searches the best matching pre-manufactured FSS in the pre-constructed data base(DB), sets initial genes of GA with the searched FSS parameters, and then runs electromagnetic scattering theory on FSS as well as evolution process through GA until the solution settles down to the desired characteristics. DB implemented from manufactured real FSS is utilized as approximated solution set for GA. The system gives tolerable running time, and the results are verified by manufactured FSS.