Answers are grounded in this paper and cite their sources.
Analyzes environmental impacts on solar panel degradation, suggesting key factors for future resilience strategies.
This study applies machine learning techniques to analyze the degradation of photovoltaic (PV) panels under varying environmental conditions. Using data from nine PV systems in six Portuguese cities, combined with corresponding weather datasets, the research aims to rank environmental factors contributing to performance loss. A Random Forest regression model was employed due to its ability to handle nonlinear relationships and generate interpretable feature importance scores. Through data augmentation and hyperparameter optimization, the model achieved a coefficient of determination R² of 0.981 under the augmentation-inclusive validation scheme. This result is valid in the context of the augmentation approach applied prior to the train-test split, consistent with the exploratory and proof-of-concept scope of the present study. Wind speed, temperature, humidity, and cloud cover emerging as key contributors to degradation. Partial Dependence Plots were used to validate and interpret these rankings. While the model performs effectively on the available data, its generalizability is limited by dataset size, geographic scope, and the absence of operational and material context. As such, the current findings should be viewed as exploratory and not yet ready for industrial application. Nonetheless, this work offers a replicable framework for future studies and contributes toward data-driven strategies for enhancing PV panel performance and environmental resilience.
No takes yet. Share an insight, caveat, or question.
Alladini et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: