Design of frequency selective surfaces using a novel MoM-ANN-GA technique

Abstract

This work describes a new application of genetic algorithm (GA) for frequency selective surfaces (FSS's) design. The work's innovation was the introduction of an artificial neural network (ANN) as a fitness function into GA. The ANN model was trained with accurate data provided by FSS's simulations using the method of moments (MoM). The founded advantages in the design of FSS devices using the technique through of genetic algorithms and neural networks are discussed and the results are presented as way to validate the new proposed technique.