Optimization of Photonic Nanocrystals for Invisibility Using Artificial Intelligence
Photonic crystals are structures that can control the propagation of light by creating periodic changes in the refractive index. These structures facilitate directing electromagnetic waves in specific directions and making objects invisible by creating a bandgap. Photonic crystals, as microscopic st...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | fas |
| Published: |
Isfahan University of Technology
2024-12-01
|
| Series: | Journal of Advanced Materials in Engineering |
| Subjects: | |
| Online Access: | https://jame.iut.ac.ir/article_3558_569b92c979f712d03fa49495ea8fe79c.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Photonic crystals are structures that can control the propagation of light by creating periodic changes in the refractive index. These structures facilitate directing electromagnetic waves in specific directions and making objects invisible by creating a bandgap. Photonic crystals, as microscopic structures with unique optical properties, are used in the design of invisibility systems. However, the design and optimization of these structures, especially with new methods like deep learning, have not been thoroughly investigated. Utilizing deep learning techniques can be highly beneficial in this area. Therefore, this paper employs the deep neural network architecture ResNet to optimize photonic crystals. ResNet can assist designers in selecting suitable materials and determining the optimal dimensions and arrangements of photonic nanostructured crystals for invisibility by extracting complex and nonlinear features from input data. The phenomenon of negative refraction in photonic crystals and the way light propagates in the proposed invisibility were studied using FDTD simulation. The results indicated that appropriate design of photonic crystals using deep learning could contribute to the creation of efficient structures. |
|---|---|
| ISSN: | 2251-600X 2423-5733 |