Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces
Metasurfaces generate desired electromagnetic wavefronts using sub-wavelength structures that are much thinner than conventional optical tools.However, their typical design method is based on trial and error, which is adversely inefficient in terms of the consumed time and computational power. This...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/173142 |
| _version_ | 1811697166127726592 |
|---|---|
| author | Kanmaz, Tevfik Bulent Ozturk, Efe Demir, Hilmi Volkan Gunduz-Demir, Cigdem |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Kanmaz, Tevfik Bulent Ozturk, Efe Demir, Hilmi Volkan Gunduz-Demir, Cigdem |
| author_sort | Kanmaz, Tevfik Bulent |
| collection | NTU |
| description | Metasurfaces generate desired electromagnetic wavefronts using sub-wavelength structures that are much thinner than conventional optical tools.However, their typical design method is based on trial and error, which is adversely inefficient in terms of the consumed time and computational power. This paper proposes and demonstrates deep-learning-enabled rapid prediction of the full electromagnetic near-field response and inverse prediction of the metasurfaces from desired wavefronts to obtain direct and rapid designs. The proposed encoder-decoder neural network was tested for different metasurface design configurations. This approach overcomes the common issue of predicting only the transmission spectra, a critical limitation of the previous reports of deep-learning-based solutions. Our deep-learning-empowered near-field model can conveniently be used as a rapid simulation tool for metasurface analyses as well as for their direct rapid design. |
| first_indexed | 2024-10-01T07:50:56Z |
| format | Journal Article |
| id | ntu-10356/173142 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:50:56Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1731422024-01-19T15:41:41Z Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces Kanmaz, Tevfik Bulent Ozturk, Efe Demir, Hilmi Volkan Gunduz-Demir, Cigdem School of Electrical and Electronic Engineering School of Physical and Mathematical Sciences School of Materials Science and Engineering LUMINOUS! Centre of Excellence for Semiconductor Lighting & Displays Science::Physics Deep Learning Electromagnetic Near Fields Metasurfaces generate desired electromagnetic wavefronts using sub-wavelength structures that are much thinner than conventional optical tools.However, their typical design method is based on trial and error, which is adversely inefficient in terms of the consumed time and computational power. This paper proposes and demonstrates deep-learning-enabled rapid prediction of the full electromagnetic near-field response and inverse prediction of the metasurfaces from desired wavefronts to obtain direct and rapid designs. The proposed encoder-decoder neural network was tested for different metasurface design configurations. This approach overcomes the common issue of predicting only the transmission spectra, a critical limitation of the previous reports of deep-learning-based solutions. Our deep-learning-empowered near-field model can conveniently be used as a rapid simulation tool for metasurface analyses as well as for their direct rapid design. Submitted/Accepted version This research is supported by the Scientific and Technological Research Council of Turkey413 (TÜBİTAK), under its 1004 - Center of Excellence Support Program grant no:20AG001. H.V.D.414 also acknowledges the support from TUBA. 2024-01-17T04:14:05Z 2024-01-17T04:14:05Z 2023 Journal Article Kanmaz, T. B., Ozturk, E., Demir, H. V. & Gunduz-Demir, C. (2023). Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces. Optica, 10(10), 1373-1382. https://dx.doi.org/10.1364/OPTICA.498211 2334-2536 https://hdl.handle.net/10356/173142 10.1364/OPTICA.498211 2-s2.0-85175494664 10 10 1373 1382 en Optica © 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement (https://opg.optica.org/library/license_v2.cfm#VOR-OA). application/pdf |
| spellingShingle | Science::Physics Deep Learning Electromagnetic Near Fields Kanmaz, Tevfik Bulent Ozturk, Efe Demir, Hilmi Volkan Gunduz-Demir, Cigdem Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces |
| title | Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces |
| title_full | Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces |
| title_fullStr | Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces |
| title_full_unstemmed | Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces |
| title_short | Deep-learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces |
| title_sort | deep learning enabled electromagnetic near field prediction and inverse design of metasurfaces |
| topic | Science::Physics Deep Learning Electromagnetic Near Fields |
| url | https://hdl.handle.net/10356/173142 |
| work_keys_str_mv | AT kanmaztevfikbulent deeplearningenabledelectromagneticnearfieldpredictionandinversedesignofmetasurfaces AT ozturkefe deeplearningenabledelectromagneticnearfieldpredictionandinversedesignofmetasurfaces AT demirhilmivolkan deeplearningenabledelectromagneticnearfieldpredictionandinversedesignofmetasurfaces AT gunduzdemircigdem deeplearningenabledelectromagneticnearfieldpredictionandinversedesignofmetasurfaces |