arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1906.03387physics.opticscs.LG
收藏

基于深度神经网络的全介质超颖表面建模新方法

A Novel Modeling Approach for All-Dielectric Metasurfaces Using Deep Neural Networks

University of Massachusetts Lowell(马萨诸塞大学洛厄尔分校);Massachusetts Institute of Technology(麻省理工学院);East China Normal University(华东师范大学);Lockheed Martin Corporation(洛克希德·马丁公司);University of Central Florida(中佛罗里达大学)
Sensong An, Clayton Fowler, Bowen Zheng, Mikhail Y. Shalaginov, Hong Tang, Hang Li, Li Zhou, Jun Ding, Anuradha Murthy Agarwal, Clara Rivero-Baleine, Kathleen A… 展开作者

Sensong An, Clayton Fowler, Bowen Zheng, Mikhail Y. Shalaginov, Hong Tang, Hang Li, Li Zhou, Jun Ding, Anuradha Murthy Agarwal, Clara Rivero-Baleine, Kathleen A. Richardson, Tian Gu, Juejun Hu, Hualiang Zhang

摘要:

超颖表面已成为一种有前途的手段,用于操纵光学波前的平面和高性能的光学器件。传统的元表面器件设计依赖于试错法来获得目标电磁(EM)响应,这种方法需要大量的努力来研究大量可能的元原子结构。在本文中,介绍了一种深度神经网络方法,与目前用于组装基于元表面的设备的技术相比,该方法在速度和准确性方面都有显着提高。我们的神经网络方法克服了三个关键的挑战,限制了以前的基于神经网络的设计方案:输入/输出矢量尺寸失配,准确的EM波相位预测,以及适应3-D介电结构,并可以普遍适用于各种各样的超颖表面器件设计在整个电磁波谱。使用这种新的方法,神经网络的例子,能够产生按需设计的元原子,元表面过滤器,相变可重构元表面进行了演示。

英文摘要:

Metasurfaces have become a promising means for manipulating optical wavefronts in flat and high-performance optical devices. Conventional metasurface device design relies on trial-and-error methods to obtain target electromagnetic (EM) response, an approach that demands significant efforts to investigate the enormous number of possible meta-atom structures. In this paper, a deep neural network approach is introduced that significantly improves on both speed and accuracy compared to techniques currently used to assemble metasurface-based devices. Our neural network approach overcomes three key challenges that have limited previous neural-network-based design schemes: input/output vector dimensional mismatch, accurate EM-wave phase prediction, as well as adaptation to 3-D dielectric structures, and can be generically applied to a wide variety of metasurface device designs across the entire electromagnetic spectrum. Using this new methodology, examples of neural networks capable of producing on-demand designs for meta-atoms, metasurface filters, and phase-change reconfigurable metasurfaces are demonstrated.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑