Smart data-driven building management framework and demonstration

Abstract

The building sector holds a significant impact over global energy usage and carbon emissions, making effective building energy management vital for ensuring worldwide sustainability and meeting climate goals. In line with this objective, this study aims to develop and demonstrate an innovative smart data-driven framework for building energy management. The framework includes semantic multi-source data integration schema, AI-empowered data-driven optimization and predictive maintenance strategies, and digital twin for informative and interactive human-equipment-information building management platform. A case study was conducted in a typical chiller plant on a campus located in Hong Kong, China. The results show that the deployment of the proposed smart data-driven framework achieves chiller sequencing control in a more robust and energy-efficient manner. Specifically, the proposed control strategy achieves energy savings of 5.9% to 12.2% compared to the conventional strategy. This research represents an important step forward in the development of smarter and more sustainable building management practices, which will become increasingly critical as we strive to reduce our environmental impact and combat climate change.