Short - Term Load Forecasting for IEEE 33 Bus Test System using SARIMAX

Load forecasting plays a crucial role in energy management, providing accurate predictions of electricity demand to optimize resource allocation, grid stability, and economic efficiency. This study investigates the use of the Seasonal Auto-Regressive Integrated Moving Average with Exogenous Variables (SARIMAX) model for load forecasting. The SARIMAX model combines the conventional SARIMA model with exogenous variables, allowing the addition of external factors like weather, holidays, and economic indices that affect power usage. In comparison to conventional SARIMA models, the SARIMAX model improves forecasting accuracy by taking these factors into account, particularly in scenarios with large external impacts. The SARIMAX model for load forecasting is implemented and evaluated in this study. The SARIMAX model is trained and optimized using historical load data and pertinent exogenous variables. The best model parameters are found using a variety of methods, including grid search and cross-validation.