Analysis of the impact of irradiance and temperature on photovoltaic production: A statistical and machine learning approach

This study explores the influence of solar irradiance (Ir) and ambient temperature (T) on photovoltaic (PV) production (P) by combining statistical analysis and deep learning techniques. A strong positive correlation was found between irradiance and PV output with Pearson coefficient (R=0.9211), while temperature exhibited a moderate effect (R=0.4477). A two-way analyze of variance (ANOVA) confirmed the statistical significance of these environmental factors. Furthermore, an autoencoder-based model was developed to capture complex nonlinear relationships and outperformed classical regression models in terms of accuracy and generalization. These results highlight the potential of machine learning methods for improving the understanding and optimization of PV systems in variable climatic conditions. Combines statistical analysis and deep learning to assess PV production factors Employs an autoencoder model to capture nonlinear relationships more effectively The model integrates a new interaction term <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow><mml:msub><mml:mi>I</mml:mi> <mml:mi>r</mml:mi></mml:msub> <mml:mo>.</mml:mo> <mml:mi>T</mml:mi></mml:mrow> </mml:math> that increases sensitivity to joint environmental variations.