首页|Customizable multifunctional metasurface absorber based on bidirectional deep neural networks covering the quasi-entire terahertz band

Customizable multifunctional metasurface absorber based on bidirectional deep neural networks covering the quasi-entire terahertz band

扫码查看
In this work,we propose a novel approach that combines a bidirectional deep neural network[BDNN]with a multifunctional metasurface absorber[MMA]for inverse design,which can effectively address the challenge of on-demand customization for absorbers.The inverse design of absorption peak frequencies can be achieved from 0.5 to 10 terahertz[THz],covering the quasi-entire THz band.Based on this,the BDNN is extended to broadband absorption,and the inverse design yields an MMA at the desired frequency.This work provides a broadly applicable approach to the custom design of multifunctional devices that can facilitate the evaluation and design of metasurfaces in electromagnetic absorption.

terahertzinverse designbidirectional deep neural networkmetasurface

丁志鹏、苏巍、叶李鹏安、周远航、李文龙、Riaz Ali、唐斌、姚红兵

展开 >

College of Mechanics and Engineering Science,Hohai University,Nanjing 211100,China

School of Microelectronics and Control Engineering,Changzhou University,Changzhou 213164,China

2024

中国光学快报(英文版)
中国光学学会 中国科学院上海光学精密机械研究所

中国光学快报(英文版)

CSTPCD
影响因子:1.305
ISSN:1671-7694
年,卷(期):2024.22(6)