<![CDATA[Global Horizontal Radiation (GHR) and Global Horizontal Illumination (GHI) are critical environmental parameters that play a vital role in solar energy development, precision agriculture, and sustainable urban planning. However, their prediction remains challenging due to the high variability caused by atmospheric conditions. This study evaluates the performance of various machine learning models in predicting GHR and GHI using a comprehensive dataset comprising 29 environmental features. The models tested include Linear Regression, Random Forest Regressor, XGBoost Regressor, LightGBM Regressor, Support Vector Regressor (SVR), and Artificial Neural Network (ANN). The results consistently show that ensemble-based models, particularly LightGBM Regressor, provide the best predictive performance for both target variables, achieving very high R-squared values (approaching 0.999). XGBoost and Random Forest also demonstrate highly competitive performance. ANN performs well, while Linear Regression and SVR show lower accuracy. These findings underscore the significant potential of advanced machine learning models in predicting environmental parameters with high accuracy, which has important implications for renewable energy optimization, smart agriculture, and sustainable urban planning.]]>
Copyrights © 2025
