基于加速精英遗传算法的非对称超构表面最优化设计OA

A transparent metasurface design method is proposed based on geometric coding and an Accelerated Elitist-Preserving Genetic Algorithm(AEGA).To verify this method,a Transparently Stacked Metasurface(TSMS)designed by the AEGA is utilized to achieve asymmetric polarization decomposition.Simulation results indicate that in 8.5～9.7 GHz,the metasurface can decompose linearly polarized incident waves into reflected waves and transmitted waves.Meanwhile,for linearly polarized waves incident from the opposite direction,the designed metasurface can decompose them into transmitted Left-Hand Circular Polarization(LHCP)waves and reflected Right-Hand Circular Polarization(RHCP)waves.Furthermore,for circularly polarized waves,within the band of 7.57～9.72 GHz,the metasurface is capable of converting LHCP waves into transmitted RHCP waves and RHCP waves into reflected LHCP waves.At the same time,for waves incident from the opposite direction,the metasurface enables full transmission of LHCP waves and complete reflection of RHCP waves.Unlike existing optimization designs,this method not only enables the independent optimization of the metal structures in multilayer metasurfaces,but also allows for the optimization of the unit cell size and thickness.This not only increases the probability of finding the global optimum,but also presents an advantage in the design of electrically small unit cells.