Next-generation segmented Halo-T metamaterial ultra-wideband design using MXene-TiO2-Cu materials performance prediction and optimization with machine learning assisted solar absorber for industrial heating systems

This is the authors' abstract. We don't add key points for this paper.

Solar energy is a great choice for sustainable energy utilization due to its pollution-free and reliable nature. Solar energy is utilized with a solar thermal absorber, which efficiently converts solar radiation to heat, and this absorber is made of various types of designs and materials, such as Metamaterials, Two-dimensional (2D) materials, etc. And also, an advanced solar absorber is utilized with Machine Learning (ML) to optimized structure and predict the absorptance. In this study, we investigate the ML-driven assisted optimization of MXene-based Segmented Halo-T Metamaterial Absorber (MSHTMA), which has a tri-layered (MXene-TiO 2 -Cu) metasurface structure with a novel T-shaped resonator. This MSHTMA resonator is made of metallic conductive 2D MXene material, and this 2D MXene-based metasurface resonator is placed on the TiO 2 substrate. And MSHTMA have high thermal stability due to the 1084.62 °C melting point of Cu-based back layer, which reduced the transmittance of the radiation. This highly thermal stable and polarization-insensitive MSHTMA achieved 95.66% absorptance between the 200-3000 nm range. And this MSHTMA achieved ultra-broadband absorptance of the 400 and 2110 nm bandwidth. This MSHTMA was analyzed with Locally Weighted Linear Regression (LWLR) and Neural Network Regression (NNR) algorithms and compared both predictions in terms of R 2 and Mean Squared Error (MSE). This highly ML optimized MSHTMA is utilized for the industrial thermal heating application and mining applications.