Energy generation forecasting: elevating performance with machine and deep learning

Abstract Distribution System Operators (DSOs) and Aggregators benefit from novel Energy Generation Forecasting (EGF) approaches. Improved forecasting accuracy may make it easier to deal with energy imbalances between production consumption. It also aids operations such as Demand Response (DR) management in Smart Grid architecture. This work aims develop test a new solution for EGF. combines various methodologies running EGF tests on historical data buildings. The experimentation yields different resolutions (15 min, one hour, day, etc.) while reporting errors. optimal technique should be relevant variety of applications trial-and-error manner, utilizing strategies, ensemble approaches, algorithms. final evaluation incorporates performance metrics coefficient determination ( $${R^{2}}$$ R 2 ), Mean Absolute Error (MAE), Squared (MSE) Root (RMSE), presenting comparative analysis results.