Graphene-Based Machine Learning Optimized Surface Plasmon Resonance Solar Absorber Design for Renewable Energy Applications

Sustainable energy solutions are required since conventional energy sources, such as fossil fuels, cause environmental degradation and resource depletion. In the present study, we have investigated the graphene-based metamaterial solar absorber (GBMSA) for designing a wide range of solar energy harvesting systems. In the 200–4000 nm range, the average absorption of GBMSA is 91.09%, and its reflection is 8.9% with 8.15 × 10⁻⁶ transmission. UV has the lowest absorption (88.41%) and NIR the highest (92.32%). In the MIR and VIS regions, the average absorption approaches 90%, with GBMSA reflecting the remaining energy. Transmission is nearly zero across the entire solar spectrum. With an average R² value of 90% and a mean squared error of 2.23 × 10⁻⁶, the machine learning method for predicting the performance of the GBMSA cuts modeling time from 56 to 7 h. The GBMSA is polarization-insensitive to TM and TE waves, maintaining over 50% absorptance up to a 70° incident angle, making it appropriate for various light sources. With these findings, the GBMSA discovers effective applications in solar air and water heating, industrial heating, and solar induction systems.