Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning
With the rapid advancement of infrared detection technology, the development of infrared stealth materials has become a pressing need. The study of optical micro/nano infrared stealth materials, which possess selective infrared radiation properties and precise structural features, is of significant...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Photonics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2304-6732/12/1/67 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587642873577472 |
|---|---|
| author | Kunyuan Zhang Delian Liu Shuo Yang |
| author_facet | Kunyuan Zhang Delian Liu Shuo Yang |
| author_sort | Kunyuan Zhang |
| collection | DOAJ |
| description | With the rapid advancement of infrared detection technology, the development of infrared stealth materials has become a pressing need. The study of optical micro/nano infrared stealth materials, which possess selective infrared radiation properties and precise structural features, is of significant importance. By integrating deep reinforcement learning with a multilayer perceptron, we have framed the design of radiation-selective films as a reinforcement learning problem. This approach led to the creation of a Ge/Ag/Ge/Ag multilayer micro/nano optical film that exhibits infrared stealth characteristics. During the design process, the agent continuously adjusts the thickness parameters of the optical film, exploring and learning within the defined design space. Upon completion of the training, the agent outputs the optimized thickness parameters. The results demonstrate that the film structure, optimized by the agent, exhibits a low average emissivities of 0.086 and 0.147 in the 3∼5 µm and 8∼14 µm atmospheric windows, respectively, meeting the infrared stealth requirements in terms of radiation characteristics. Additionally, the film demonstrates a high average emissivity of 0.75 in the 5∼8 µm range, making it effective for thermal radiation management. Furthermore, we coated the Si surface with the designed thin film and conducted experimental validation. The results show that the coated material exhibits excellent infrared stealth properties. |
| format | Article |
| id | doaj-art-5e528943c53649ceb2723d9d77d67838 |
| institution | Kabale University |
| issn | 2304-6732 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Photonics |
| spelling | doaj-art-5e528943c53649ceb2723d9d77d678382025-01-24T13:46:24ZengMDPI AGPhotonics2304-67322025-01-011216710.3390/photonics12010067Design Method of Infrared Stealth Film Based on Deep Reinforcement LearningKunyuan Zhang0Delian Liu1Shuo Yang2School of Optoelectronic Engineering, Xidian University, Xi’an 710071, ChinaSchool of Optoelectronic Engineering, Xidian University, Xi’an 710071, ChinaXi’an Institute of Applied Optics, Xi’an 710065, ChinaWith the rapid advancement of infrared detection technology, the development of infrared stealth materials has become a pressing need. The study of optical micro/nano infrared stealth materials, which possess selective infrared radiation properties and precise structural features, is of significant importance. By integrating deep reinforcement learning with a multilayer perceptron, we have framed the design of radiation-selective films as a reinforcement learning problem. This approach led to the creation of a Ge/Ag/Ge/Ag multilayer micro/nano optical film that exhibits infrared stealth characteristics. During the design process, the agent continuously adjusts the thickness parameters of the optical film, exploring and learning within the defined design space. Upon completion of the training, the agent outputs the optimized thickness parameters. The results demonstrate that the film structure, optimized by the agent, exhibits a low average emissivities of 0.086 and 0.147 in the 3∼5 µm and 8∼14 µm atmospheric windows, respectively, meeting the infrared stealth requirements in terms of radiation characteristics. Additionally, the film demonstrates a high average emissivity of 0.75 in the 5∼8 µm range, making it effective for thermal radiation management. Furthermore, we coated the Si surface with the designed thin film and conducted experimental validation. The results show that the coated material exhibits excellent infrared stealth properties.https://www.mdpi.com/2304-6732/12/1/67infrared stealthmultilayer filmsreinforcement learning |
| spellingShingle | Kunyuan Zhang Delian Liu Shuo Yang Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning Photonics infrared stealth multilayer films reinforcement learning |
| title | Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning |
| title_full | Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning |
| title_fullStr | Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning |
| title_full_unstemmed | Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning |
| title_short | Design Method of Infrared Stealth Film Based on Deep Reinforcement Learning |
| title_sort | design method of infrared stealth film based on deep reinforcement learning |
| topic | infrared stealth multilayer films reinforcement learning |
| url | https://www.mdpi.com/2304-6732/12/1/67 |
| work_keys_str_mv | AT kunyuanzhang designmethodofinfraredstealthfilmbasedondeepreinforcementlearning AT delianliu designmethodofinfraredstealthfilmbasedondeepreinforcementlearning AT shuoyang designmethodofinfraredstealthfilmbasedondeepreinforcementlearning |