A Review of Artificial Intelligence-Based Techniques to Estimate Atmospheric Parameters Influencing the Performance of Concentrating Photovoltaic/Thermal Systems

Abstract

Concentrating photovoltaic/thermal (CPV/T) technology is regarded as the most auspicious part of renewable energy capable of reducing reliance on fossil fuels due to its superior performance and hybrid output nature. CPV/T technology aims to reduce the cost of the renewable systems by replacing the costly solar cell material with relatively cheap optical devices that concentrate the light collected from the sun to a small solar PV cell and simultaneously generating useful heat energy for process heat applications. However, the electrical and thermal performances of systems utilizing the methodology mentioned above get strongly affected by atmospheric parameters like solar radiation, ambient temperature, and the solar spectrum. In recent years, due to the advantages tendered by Artificial Intelligence tools to solve ambiguous and non-linear problems, many authors have used intelligent system-based techniques for the prediction of the above-mentioned atmospheric parameters. This paper presents a review of artificial intelligence-based techniques, including Artificial Neural Network, Genetic Algorithm, and their composite models for the estimation of atmospheric parameters that significantly influence the working of hybrid concentrating PV/thermal systems. The review demonstrates the feasibility and accuracy of artificial intelligence-based tools for precise solar insolation and ambient air temperature prediction.