Related Experiment Video
Updated: Aug 13, 2025
13:44
Simulation, Fabrication and Characterization of THz Metamaterial Absorbers
Published on: December 27, 2012
15.4K
Lightweight Machine-Learning Model for Efficient Design of Graphene-Based Microwave Metasurfaces for Versatile
Nengfu Chen1, Chong He1, Weiren Zhu1
1Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai 200240, China.
Nanomaterials (Basel, Switzerland)
|January 21, 2023
Summary
This study introduces a machine learning network for designing graphene-based microwave absorbers, simplifying the process of creating broadband absorbers with tunable conductivity for advanced microwave applications.
Area of Science:
- Materials Science
- Electrical Engineering
- Artificial Intelligence
Background:
- Graphene enables optically transparent microwave metasurfaces with broadband absorption.
- Current design methods for graphene metasurfaces are complex, requiring extensive parameter sweeping and expert knowledge.
Purpose of the Study:
- To develop a machine learning network for the forward prediction and inverse design of versatile microwave absorbers.
- To leverage graphene's tunable conductivity for intelligent metasurface design.
Main Methods:
- A lightweight and efficient machine learning network incorporating normalization and transposed convolution layers.
- An optimization-based inverse design system for versatile microwave absorber creation.
Main Results:
- Demonstrated representative cases with highly promising performance in meeting diverse absorption requirements.
- The proposed network effectively predicts reflection spectra and facilitates inverse design.
Conclusions:
- The developed machine learning network significantly streamlines the intelligent design of graphene-based metasurfaces.
- This approach holds substantial potential for various microwave applications requiring tailored absorption properties.
