Ain Shams Engineering Journal (Feb 2025)
Thin wire and circular disk resonator MXene solar absorber optimized using Machine learning for renewable energy applications
Abstract
Investigated a range of solar absorber configurations made from different materials in an effort to maximize energy output and reduce energy losses in the context of renewable energy sources, especially solar energy. In this study, two-dimensional MXene was used as a resonator based on the Tungsten (W) substrate in thin wire and circular disk integration resonator MXene-based solar absorber (TWCDIRMSA). The design of the resonator combines a thin wire with a circular disk. The TWCDIRMSA has 90 % absorption with a 1750 nm wideband and 96 % absorption with a 930 nm wideband. The solar absorber structure analyzed AM 1.5 irradiation and 91.9 % achieved absorption. This solar absorber achieved absorption rates of 88.35 % at UV, 92.66 % at VIS, 89.56 % at NIR, 96.29 % at MIR, and 97.76 % at FIR spectra respectively. A variation in electric and magnetic intensity with colour-changing was also investigated and 94.37 % absorption by design in the 200–3000 nm wavelength region. The implementation of machine learning (ML) aims to reduce simulation time by one-fourth when compared to traditional methods while also reducing other simulation requirements, with the investigation demonstrating the ML model’s highest achieved R2 value of 0.94594 and a mean squared error of 1.294 × 10-4 under parameter variation. This proposed solar absorber polarization is insensitive to an incident angle of up to 70 degrees and a comparison with previous work is also presented in the table. With these remarkable abilities, this TWCDIRMSA has particular uses in solar inductions.