Forecasting the Impact of High Solar Module Temperature on Power Output and CO₂ Mitigation: A Machine Learning Approach

The effects of high solar module temperatures on photovoltaic (PV) systems in power output and CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> mitigation were explored. The researchers made the predictions of temperature oscillation impacts on PV system performance and CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions reductions using Linear Regression and Extra Trees Regression models. Results of the research work using Linear Regression gave a higher R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> value in power output prediction, namely, 0.9992, while Extra Trees Regression provided an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> value of 0.9993 for CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions mitigation. Data pre-processing techniques involving imputation, outlier detection, and feature engineering techniques were applied to ensure proper data quality. The empirical study finds that temperature influences both PV system performance as well as the potential reductions in CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>. Machine learning models outperform conventional techniques by providing more dynamic analysis based on data and better precision. These findings emphasize the role of temperature management in the optimization of energy efficiency of PV systems and reducing CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions, hence furthering the effectiveness of solar energy systems in climate change mitigation.