R2 Score Estimation with Various Time Series Models Incorporating 1000 Epochs for Optimization of Accuracy of Renewable Energy Future Trends

The <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$1^{\circ}\mathrm{C}$</tex> rise in global temperature since the pre-industrial era is mainly attributed to the use of fossil fuels and human activities. To mitigate this temperature increase, it is crucial to reduce greenhouse gas emissions. In recent years, renewable energy has gained popularity due to its economic and environmental benefits. This study projects 450 days of renewable energy consumption based on a dataset spanning 5.5 years. Three machine learning algorithms—Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Hist Gradient Boosting (HGBoost)—were utilized to predict renewable energy production and consumption. Based on the margin of error, it can be inferred that producing 2000MW to 5000MW of electricity is a viable option for Romania. The obtained r2 score for Random Forest Regressor is 0.95, whereas it is 0.83 for XGBoost and 0.84 for HGBoost Regressor as a performance parameter for the time series analysis.