|
| References | Year | Methodology | Parameters/metrics | Merits/limitation |
|
| Patel et al. [21] | 2022a | Graphene and LSTM | Resonator and substrate thickness | (i) Precisely predict absorption levels, shorten simulation time, and use fewer resources |
| (ii) Training takes more time and memory |
| (iii) High computation burden |
|
| Patel et al. [22] | 2022b | SiO2 and regression models | Resonator thickness, SiO2 width, SiO2 height, and chemical potential of graphene | (i) More than 0.99 prediction accuracy |
| (ii) It was unable to match complicated datasets accurately |
|
| Zhang et al. [23] | 2022 | Titanium and gallium arsenide metal | Absorber absorption efficiency | (i) The building was effectively angle-insensitive and capable of maintaining strong absorption |
| (ii) Yet effective validation is not achieved in this research |
|
| Parmar et al. [24] | 2022 | CNN-regression | Incidence angle, substrate, and resonator thickness | (i) Precision at a high level |
| (ii) Lack of spatial invariance with respect to the supplied data |
|
| Patel et al. [25] | 2022c | General regression neural network | Different structural parameters | (i) Decreased simulation time and increased accuracy |
| (ii) It may be quite large, which would make calculation difficult. There is not a perfect way to make things better |
|
| Donda et al. [26] | 2021 | CNN | Absorption spectrum response and low frequency | (i) It appeals to applications that need quick on-demand design |
| (ii) However, the performance of the model is very poor |
|
| Patel et al. [27] | 2022d | Polynomial regression models and MgF2 substrate | Substrate and resonator thickness, thickness, needlepoint, and plus shape width | (i) Ability to absorb with great accuracy |
| (ii) The probability of prediction error is higher than in other methods |
|
| Patel et al. [28] | 2022e | KNN and Ge2Sb2Te5 (GST) substrate | Thickness: metasurface, substrate, and ground plane | (i) Superior prediction accuracy |
| (ii) The prediction phase is very slow due to the higher amount of data |
|
