A transgenic genetic algorithm design method that helps to increase the design freedom of metasurfaces

Abstract

While the optimization of design is an effective way to increase the performance of metasurfaces, there is still much to be done to improve optimization algorithms. In this work, a transgenic genetic algorithm (TGA) proposed for metasurface design offers higher performance and more design freedom. Based on GAs, transgenic technology is introduced into metasurface design to improve both performance and flexibility. We describe the generation of a transgenic factor (TF), how it operates on metasurface design, and explain the feasibility of enhancing design freedom using a TGA. To target the broadband regime and high efficiency, an electromagnetic polarization converter and an absorber are then designed by TGA, respectively. As a result, polarization conversion of more than −1 dB and absorption of over 90% were achieved from 8.09 GHz to 24.90 GHz and 7.71 GHz to 20.01 GHz, respectively. Both simulated and measured results are highly consistent, which validates the stable performance achieved by the proposed TGA design. What is more, a performance comparison between different TFs illustrates the subtle influence of TFs, revealing that the performance is improved to some extent by an increase of surface-design freedom. Significantly, this effort provides more freedom in the design method, leading to the realization of many expected improvements.