Intelligent graphene-based metamaterial solar absorber: AI-driven parametric optimization for ultrabroadband absorptance enhancement in renewable energy systems applications and performance analysis

Abstract

The developed global demand for renewable energy technologies has intensified efforts toward developing high-performance solar thermal absorbers as sustainable alternatives to fossil fuels, which continue to deplete natural resources and harm the environment. In the proposed study, a novel multilayer nanostructured broadband solar design is presented to achieve near perfect thermal absorption efficiency over a broad spectral range while maintaining practical applicability for diverse renewable energy systems. The designed structure consists of three carefully selected materials Chromium (Cr), Indium (III) selenide (In₂Se₃), and Tungsten (W) arranged sequentially from the resonator layer to the base layer to promote strong light–matter interaction and effective heat conversion. Furthermore, an ultrathin graphene layer is incorporated to enhance optical absorption, improve thermal stability, and support broadband performance. Numerical results demonstrate that the proposed absorber achieves a near-perfect absorptance of approximately 97% at wavelengths beyond 800 nm, while sustaining about 90% absorption over a wide spectral range at 2800 nm. These broadband absorption characteristics make the absorber highly suitable for efficient solar thermal energy harvesting. To further improve performance, machine learning–based optimization using linear regression was employed, achieving a high coefficient of exploration (R² = 0.98) and low MSE with 0.211187 × 10⁻⁴, confirming the robustness and predictive accuracy of the model. Owing to its versatility and efficiency, the current absorber can be utilized in solar heaters, solar cookers, space heating, crop drying, solar stills, industrial heating, and thermal shielding. This work demonstrates the strong potential of integrating advanced material engineering with machine learning optimization to realize efficient, multifunctional, and environmentally sustainable solar thermal absorber designs for future renewable energy applications.