Forecasting Global Building Energy Dynamics under Socioeconomic Transitions: An XGBoost-Based Approach
Accurate long-term forecasting of building energy demand is a cornerstone for achieving sustainable climate mitigation targets. However, traditional modeling frameworks often rely on static assumptions of internal heat gains, neglecting the evolving socioeconomic drivers that shape global energy patterns. This research bridges that gap by introducing a dynamic, data-driven approach to forecast internal heat gains using the XGBoost ensemble model. By integrating historical data with future projections aligned with the Shared Socioeconomic Pathways (SSPs), the study provides a comprehensive, globally consistent dataset that enhances predictive accuracy and supports climate-resilient building design strategies. Dynamic Modeling of Internal Heat Gains The study leverages advanced machine learning, specifically XGBoost, to model and forecast key variables influencing internal heat gains such as household size and energy intensity across major end-uses. Unlike conventional regression-bas...