Synthesis of crossed dipole frequency selective surfaces using genetic algorithms and artificial neural networks

Abstract

This work presents the synthesis of crossed dipole frequency selective surfaces (FSSs) using a genetic algorithm (GA) whose fitness function is composed by an artificial neural network (ANN). The ANN model was trained by the resilient backpropagation (RPROP) algorithm, through the use of accurate data provided by a parametric study developed to investigate some of the geometric parameters of the FSSs. The founded advantages in the design of FSS devices using this optimization technique are discussed and the results are compared to those obtained with simulations using the Ansoft Designertrade commercial software, which is based on the method of moments (MoM).