Analysis of graphene coatings on various metallic/oxide crystal/composite material substrates using machine learning for enhanced solar thermal energy conversion.

Because energy interest demands clean and sustainability in the last ten years. Solar thermal energy conversion, where sunlight can be absorbed to convert it into heat can stand as an alternative for this purpose. Graphene dispersed with different substrates enables us to get torsion control over light absorption and heat transport. This work discusses the optothermal properties of graphene-based coatings on different substrates such as CuO, MAPBI3, Fe, etc. The optothermal properties of such CuO-graphene, MAPBI3-graphene, and Fe-graphene combinations display the highest average absorptance of 96.8% across the solar spectrum between 0.2 and 2.5 μm followed by 86.7% by MAPBI3-graphene. However, Fe-graphene depicts a significantly lower value of 24.3%. A critical inspection of these optothermal properties would enrich one with critical knowledge of design optimisation in graphene-coated solar absorbers. Thus, the data collection time is greatly reduced using ML compared to running simulations which have a step size of about 8 h per change. Where the machine learning efficacy is 98% for the thickness optimization of Fe, CuO, and MAPBI3 with 25% test data. Of much potential interest are the solar absorbers developed using these materials in fields such as solar thermal energy harvesting, air/water heaters, and industrial heating systems.