Akademik Veri Yönetim Sistemi
APPLIED SCIENCES-BASEL, cilt.15, sa.17, 2025 (SCI-Expanded, Scopus)
Featured Application This work introduces an optimized stacked ensemble model that significantly enhances the accuracy of hourly energy consumption forecasting. This model is particularly valuable for Digital Twin simulations in smart buildings. Its ability to accurately predict energy usage and robustly handle various real-world data challenges (like structural changes, sudden spikes, data quality issues, and appliance malfunctions) makes it an ideal tool for active energy management. Specifically, it can be applied to optimize building energy efficiency by enabling precise adjustments to HVAC systems and lighting, facilitate predictive maintenance of energy-consuming assets by allowing for timely interventions before failures intensify, support sustainable energy management strategies through accurately forecasting savings from efficiency upgrades and tracking their long-term impact, and enhance grid stability via improved demand-side management and real-time anomaly detection. By providing highly reliable and adaptive energy consumption forecasts, this model directly contributes to smarter, more efficient, and sustainable energy management.Abstract Modern energy grids, with their regional diversity and complex consumption patterns, require accurate short-term forecasting for operational efficiency and reliability. This study introduces a Stacking Ensemble Forecasting (SEF) framework for multi-region household energy demand, utilizing an optimized stacking ensemble model tuned via Bayesian Optimization to achieve superior predictive accuracy. The framework significantly improved accuracy across Diyarbak & imath;r, Istanbul, and Odemis, with a final model demonstrating up to 16.47% RMSE reduction compared to the best baseline models. The final model's real-world performance was validated through a Simulated Digital Twin (SDT) environment, where scenario-based testing demonstrated its robustness against behavioral changes, data quality issues, and device failures. The proposed SEF-SDT framework offers a generalizable solution for managing diverse regions and consumption profiles, contributing to efficient and sustainable energy management.