Related Experiment Video
Updated: Jul 31, 2025
07:14
Microfabrication of Implantable Optics Integrated in a Microstructured Imaging Window for Advanced In Vivo Imaging
Published on: April 11, 2025
691
Deep learning-based inverse design of microstructured materials for optical optimization and thermal radiation
Jonathan Sullivan1, Arman Mirhashemi2, Jaeho Lee3
1Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA, USA.
Scientific Reports
|May 6, 2023
Summary
This study introduces an aggregated neural network for optimizing microstructures in thermal management. The AI framework accelerates material design by predicting optimal properties and even discovering new materials for aerospace applications.
Area of Science:
- Materials Science
- Optical Engineering
- Artificial Intelligence
Background:
- Engineered microstructures are crucial for thermal management in aerospace and space.
- Traditional material optimization is slow and limited by design variables.
Purpose of the Study:
- To develop an efficient inverse design process for optical microstructures.
- To overcome limitations of traditional material optimization methods.
Main Methods:
- An aggregated neural network combining a surrogate optical network and an inverse neural network was developed.
- The surrogate network emulates finite-difference time-domain (FDTD) simulations.
- A self-learning loop was created using FDTD evaluation and retraining.
Main Results:
- The framework predicts microstructure design properties for desired optical spectra.
- It can identify new material properties for optimization or match existing materials.
- The approach allows for complex, user-constrained optimization.
Conclusions:
- The deep learning-derived approach enables efficient inverse design of optical microstructures.
- This framework is applicable to thermal radiation control in aerospace and space systems.
