Sallam, Mahmoud et Kaur, Kuljeet.
2026.
« ProHome: Recurrent proximal policy optimization-driven advanced home energy management ».
Sustainable Energy, Grids and Networks, vol. 47.
Prévisualisation |
PDF
Kaur-K-2026-34296.pdf - Version publiée Licence d'utilisation : Creative Commons CC BY. Télécharger (5MB) | Prévisualisation |
Résumé
Advanced Home Energy Management Systems (HEMS) are essential for integrating Renewable Energy Sources (RES) and Electric Vehicles (EVs) into smart homes; however, many existing solutions overlook the temporal dependencies in energy dynamics and are evaluated under overly simplified or fully known price signals. This work proposes ProHome, a HEMS framework based on Recurrent Proximal Policy Optimization (RPPO), where a Long Short-Term Memory (LSTM)-enabled policy learns memory-dependent control strategies from sequential observations. ProHome coordinates a comprehensive set of household assets—namely, heating, ventilation, and air conditioning (HVAC), electric water heater, deferrable appliances, photovoltaic (PV) generation, an energy storage system, and an EV with bidirectional power exchange—under time-varying electricity prices without assuming access to future prices. A realistic smart home simulation environment driven by historical data was developed for training and testing. Evaluated over a comprehensive 1089-day dataset (roughly 3 years with a 75/25 train-test split), ProHome performs competitively against Deep Reinforcement Learning (DRL) baselines. While operating costs are comparable to standard Proximal Policy Optimization (PPO) (reducing the mean daily cost by a modest 1%), the proposed recurrent agent provides more reliable EV readiness by increasing the state of energy at departure by 2.1% toward the user-defined target, all while maintaining near-zero comfort violations for HVAC and hot-water temperature.
| Type de document: | Article publié dans une revue, révisé par les pairs |
|---|---|
| Chercheur(-euse): | Chercheur(-euse) Kaur, Kuljeet |
| Affiliation: | Génie électrique |
| Date de dépôt: | 04 sept. 2026 20:42 |
| Dernière modification: | 26 sept. 2026 19:15 |
| URI: | https://espace2.etsmtl.ca/id/eprint/34296 |
Actions (Authentification requise)
 |
Dernière vérification avant le dépôt |