Smart Home Energy Management System Empowered with Digital Twin

The growing use of renewable energy in modern power systems has increased the demand for efficient, real-time energy management in smart homes. Smart homes are equipped with distributed generation, energy storage system, EVs, and flexible loads. These homes face challenges from renewable variability, dynamic pricing, and user behavior uncertainties — limitations for conventional HEMS. This paper proposes a Digital Twin-assisted, edge-based real-time HEMS (DT-HEMS) that combines a live digital model of the home’s energy management system with Long Short Term Memory (LSTM)-based forecasting block and edge computing. The system predicts PV generation, EV availability, and appliance demand, continuously refining forecasts and control actions using real-time data. Simulations with 15-minute scheduling intervals show the efficiency of DT-HEMS in improving forecasting accuracy, responsiveness, and operational cost savings — achieving up to 10% savings at a 15-minute resolution. The results highlight the potential of integrating Digital Twins, edge computing, and machine learning for scalable, privacy-preserving, and resilient smart home energy management.