Investigation of Machine Learning Optimized Graphene-Based Ultrabroadband Plasmonic Solar Thermal Absorber for Renewable Energy Applications
Abstract
The effective solar absorber (three-layer type) to prevent the greenhouse effect and improve electricity generation has been designed with three different metals gold (Au) for the foundation layer, silver (Ag) for the resonator part construction, and the titanium nitride (TiN) is sandwiched between the two layers of silver and gold. The current three-layer absorber design can study the ultraviolet spectrum (UV), visible area (V), and near and middle infrared regions (NIR and MIR). We have used machine learning algorithm to optimize the structure and ability of solar absorber. To display the highest amounts on the overall observed absorption line, the four wavelength (mu m) values of 0.34, 0.53, 1.01, and 1.91 have been used. With the description of bandwidth and wavelength inclusion, 97.14% can be observed with the wavelength range (mu m) between 1.8 and 2.5 (700-nm bandwidth), and 95.1% by the wavelength values from 0.69 and 2.9 (2210-nm bandwidth), respectively. The overall output is 94.41% for 2800 nm (0.20-3.0 mu m) due to a thin film of graphene addition in the current three-layer design. The design investigated is giving excellent performance for renewable energy applications.