Page 175 - Read Online

He et al. Microstructures 2023;3:2023037 https://dx.doi.org/10.20517/microstructures.2023.29 Page 11 of 24

Table 2. A brief overview of design based on wave propagation characteristics in infinite meta-structures
Meta-structure and
Design type Algorithm Description Year
performance
Enhancing noise CNN Predicted the absorption spectra of metasurfaces based on 2021
reduction CNN and conducted experimental verification [106]

CNN Prediction of sound absorption spectra of absorbers based 2022
CGAN on CNN and inverse design based on CGAN [107]

GAN The generated structures can have completely new 2021
configurations and rich local features. They can be in good
agreement with experimental results [108]

TNN Overcoming the data inconsistency caused by the complex 2022
coupling effect between Fabry Perot channels, the
[109]
experimental results are in good agreement

RL Exploring deep subwavelength broadband sound absorption 2022
meta-structures based on RL, replacing the artificial selection
of structural parameters. The accuracy of the design was
verified through sound absorption experiments [110]

RL Employing RL to optimize the huge parameter space with 2023
nine aperture parameters to design broadband sound
absorption meta-structures, and further validated through
[111]
experimentation
CNN Inverse design of the absorber based on the target 2021
absorption spectrum by employing a one-dimensional CNN
model. The difficulty lies in the selection of neural network
[112]
structure and hyperparameters
TNN Inverse design of the absorber based on the target 2023
absorption spectrum by employing TNN. The model uses
fewer hyperparameters and has higher accuracy and
[113]
efficiency than traditional CNN

MLP The inverse design incorporating probability sampling can 2020
gaussian obtain all possible structures. The transmission spectrum
sampling measured in the experiment is highly consistent with the
predicted results, and the accuracy of the report is better
[114]
than models such as ANN and GAN