ANN characterization of printed reflectarray elements

The design of printed reflectarray antennas (RAs) could be quite complex and computationally expensive, since the need of providing high performances and satisfying requirements that could be also in contrast each other could require the use of a large number of re-radiating advanced element configurations. A possible strategy for the RA design could be therefore of carrying it out adopting an evolutionary optimization tool. In this work, an artificial neural network (ANN) model of the RA single element is presented as convenient interface between antenna design and global optimization algorithms. In order to prove the effectiveness of the model, it will be used in the design of a dual-band dual-layer reflectarray