Precise solar energy forecasting is essential for the dependable and efficient operation of photovoltaic (PV) energy production for smart grid systems due to its erratic nature. Changes in weather, temperature, and solar radiation can impact electricity production, which in turn can have an impact on grid stability and energy management strategies. This research examines a number of supervised machine learning methods for predicting close- term solar energy in smart grid applications. Some of the tested models include Extreme Gradient Boosting (XGBoost), Gradient Boosting (GB), Random Forest (RF), Support Vector Regression (SVR), and Multi-Layer Perceptron (MLP). Weather patterns and past data on solar energy are among the inputs. The Mean Absolute Error (MAE), coefficient of determination (R2), and normalized Root Mean Square Error (nRMSE) are a few of the performance indicators utilized. The findings demonstrate that, in terms of prediction accuracy and robustness, ensemble-based models, particularly Random Forest and XGBoost, outperform other methods. The findings demonstrate how smart grid environments may use ensemble learning techniques to increase the accuracy of their predictions.
Pyatimath et al. (Thu,) studied this question.