Comparative Analysis of Monocrystalline and Polycrystalline Photovoltaic Performance in Arid Climates Using Random Forest for Missing Data Completion

This study employs a Random Forest (RF) algorithm to fill missing data for monocrystalline (m-Si) and polycrystalline (p-Si) photovoltaic modules using satellite data, enabling a complete dataset to support a more accurate performance comparison in an arid climate. Performance was evaluated across different seasons using the Root Mean Square Error (RMSE) and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> metrics. The results reveal that both technologies achieve their best performance during the summer, with p-Si obtaining an RMSE of 98.73 and an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.950, and m-Si achieving an RMSE of 122.6 and an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.946, indicating high predictive accuracy. In contrast, spring and winter exhibit higher RMSE values (e.g., 231.4 for p-Si in spring and 234.84 for m-Si in winter), likely due to higher climatic variability. These findings provide valuable insights for optimizing photovoltaic systems under arid conditions by leveraging satellite data to ensure data continuity.