RESEARCH AND APPLICATION OF THE STACKING ENSEMBLE MODEL IN SHORT-TERM LOAD FORECASTING

Abstract

This study presents the development and implementation of a Stacking Ensemble model for short-term load forecasting, a critical power system management and operations task. The Stacking Ensemble model integrated multiple machine learning algorithms, including XGBoost, LightGBM, Ridge Regression, Random Forest, and Support Vector Regression, to enhance forecasting accuracy and reduce standard error metrics such as Mean Absolute Error, Mean Absolute Percentage Error, Mean Squared Error, and Root Mean Squared Error. The findings demonstrate that the Stacking Ensemble model outperforms individual models when handling complex and volatile datasets. A comprehensive comparison between the Stacking Ensemble and individual models such as XGBoost, LightGBM, Random Forest, and Support Vector Regression reveals that Stacking improves forecasting accuracy and offers excellent stability across different scenarios. This research validates the Stacking Ensemble as a robust and effective approach for optimizing short-term load forecasting, contributing to enhanced energy management and more efficient power system operations.