Semantic rule-based approach for automated energy resource management in buildings

The widespread use of renewable energy sources leads to the adoption of more sophisticated and intelligent real-time energy management solutions. Automated energy management systems allow consumers to play an active role in their flexibility management while ensuring their energy needs are met. However, a lack of trust in autonomous decision-making poses a significant challenge to their adoption. This work presents a novel semantic-based framework for automated energy management in buildings, integrating semantic rules and ontologies, expert knowledge, and machine learning models to enhance decision transparency and adaptability. By leveraging a semantic-based model, the proposed framework improves real-time decision-making, facilitates interoperability between data sources, and provides context-aware explanations, fostering user trust and system reliability. The framework has been tested and validated within a cyber-physical infrastructure, ensuring its robustness in real-world scenarios. A case study on the management of lighting and air conditioning demonstrates the advantages of this approach. The results confirm that the framework effectively adapts to evolving conditions, ensures reliable decision-making, and fosters user trust by providing interpretable justifications for automated actions. This facilitates a more efficient use of energy resources, reduces costs, and supports the transition toward a more sustainable and renewable-based power sector. • Novel semantic-based framework for automated energy management in buildings. • It leverages semantic technologies, expert knowledge and ML models. • Improves decision-making and provides context-aware explanations, fostering trust. • The framework has been tested and validated within a cyber-physical infrastructure. • The framework adapts to evolving conditions, ensuring reliable decision-making.