A new study by the Technical University of Cluj-Napoca introduces an innovative load disaggregation architecture designed to improve both system responsiveness and energy analysis efficiency
Understanding how energy is actually used inside buildings is key to making homes and energy systems more efficient. A new scientific publication from the Energy Transition Research Center (EnTReC) of the Technical University of Cluj-Napoca, partner of the REN+HOMES project, introduces an innovative solution that makes this process faster and smarter.
The study, presented at the Energy and Sustainability Conference 2025 and published on E3S Web of Conferences, presents a new architecture that combines artificial intelligence with fog computing, a technology that processes data closer to where it is generated, instead of relying only on distant cloud servers. This means quicker responses, lower data transfer needs, and more efficient energy monitoring.
The system analyses visual representations of energy consumption data to classify different load profiles. Two distinct consumption patterns were successfully identified using Random Forest and XGBoost classifiers, demonstrating accurate load disaggregation even in complex operational contexts.
This new method opens the door to more responsive energy management solutions for homes and buildings, helping users better understand their energy consumption and supporting the transition towards smarter, more sustainable energy systems.
Cover photo credits: Conny Schneider on Unsplash