EUVIMEDEUVIMED entdecken
Crossref

Research on Rapid Design Method of Cross-polarization Metasurface Based on Deep Learning

Hang Zhang, Jiao Wang, Yannan Jiang

Advances in Engineering Technology Research · 12.03.2026 · 16 (1) · 89

DOI: 10.56028/aetr.16.1.89.2026 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Engineering Technology Research, online erschienen am 12.03.2026. Verfasst von Hang Zhang und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study presents a deep learning-driven framework that integrates pseudo-ellipticity angle parametrization with digital coding metasurfaces, aiming to address the inefficiency limitations inherent in conventional manual design approaches for reflective cross-polarization conversion metasurfaces. Specifically, the pseudo-ellipticity angle is employed to characterize cross-polarization conversion performance, and a Conditional Generative Adversarial Network is adopted for training, which enables the rapid design of cross-polarization conversion metasurfaces. The proposed rapid design method successfully realizes the design of cross-polarization conversion metasurface with diversified configurations, thereby validating the effectiveness of this approach.

Zitieren

APA

Zhang, H., Wang, J., & Jiang, Y. (2026). Research on Rapid Design Method of Cross-polarization Metasurface Based on Deep Learning. Advances in Engineering Technology Research, 16(1), 89. https://doi.org/10.56028/aetr.16.1.89.2026

Vancouver

Zhang H, Wang J, Jiang Y. Research on Rapid Design Method of Cross-polarization Metasurface Based on Deep Learning. Advances in Engineering Technology Research. 2026;16(1):89. doi:10.56028/aetr.16.1.89.2026

BibTeX
@article{Zhang2026,
  author = {Zhang, Hang and Wang, Jiao and Jiang, Yannan},
  title = {Research on Rapid Design Method of Cross-polarization Metasurface Based on Deep Learning},
  journal = {Advances in Engineering Technology Research},
  year = {2026},
  volume = {16},
  number = {1},
  pages = {89},
  doi = {10.56028/aetr.16.1.89.2026}
}
Kostenlos starten →