Page 170 - Read Online
P. 170
Page 6 of 24 He et al. Microstructures 2023;3:2023037 https://dx.doi.org/10.20517/microstructures.2023.29
Table 1. A brief overview of design based on band structures in infinite meta-structures
Design Algorithm Meta-structure and performance Description Year
type
Complete MLP RBF is suitable for single parameter prediction, while MLP 2019
band RBF-NN can meet multi-parameter prediction [92]
structures
CNN Construct digital structure genomes through forward 2021
prediction. Thus, the target property structures can be
[93]
quickly extracted from the genomes
GAN Generate optimal structure based on customized dispersion 2022
CNN and accelerate design processes [94]
GAN Generate and screen structures with excellent attenuation 2022
CNN performance. The dataset is generated through secondary
[95]
mirroring, which lacks flexibility
Compared to MLP, TNN can solve the problem of data 2019
Tailoring MLP inconsistency and is suitable for multi-parameter inverse
bandgaps TNN [96]
design
GA The model is insufficient to provide accurate predictions 2020
MLP beyond the training data range and only performs well
within local data points [97]
The model can obtain the target modular metamaterial but 2020
GA [98]
MLP cannot find the configuration beyond the dataset
AE Can accurately process data beyond the dataset. Only a 2020
MLP relatively small region of the design space in RVE is
[99]
explored using a nine-parameter analytical function