This study aims to design energy demand forecasting models for energy management in hybrid microgrid systems using optimized machine learning techniques. By incorporating temperature, humidity, season, hour of the day, and irradiance, the complex relationship between these input parameters and the yield of photovoltaics, generator, and grid energy sources is examined. Five different machine learning models including linear regression, random forest (RF), support vector regression, artificial neural network, and extreme gradient boosting models are adopted in this study. Evaluation of model performance shows that the RF model is the best candidate for the dataset, with a mean-squared error of 0.2023, mean absolute error of 0.0831, root-mean-squared error of 0.4498, and R² score of 0.9992. Shapley additive explanations analysis identified key predictors such as hour, irradiation, and season while highlighting the negative impact of humidity and day of the week on energy demand.

The global energy crisis and ozone layer depletion as a result of carbon emissions have increased the awareness and acceptance of renewable energy sources as an alternative form of electric power, resulting in the sizing of renewable energy sources. However, in order to properly size an energy power system, the information being addressed, such as the load demand, is critical. The Load demand data of Nile University campus is obtained from one of its power stations (PS-1) for a period of eight month. The data was measured from the bus bar of the power station using smart meters on a weekly basis. To power the university campus, the diesel generators are synchronized using Genset controllers with suitable communications interfaces and a SMA hybrid controller, which continually checks the power output of the power sources as well as the working condition of all loads in the busbar. The diesel generators are synchronized using SMA hybrid controllers and combined with the other source of the energy at a common bus bar and used to power the university campus. Additionally, carbon emission data were obtained from the PV solar system reading.