Leveraging Brick Ontology to Enhance Energy Reporting and Compliance in Data Centers

Abstract

This thesis aimed to develop a unified semantic modelling framework utilising Brick and RealEstateCore ontologies to automate energy reporting and ensure compliance with the EU Energy Efficiency Directive (EED) requirements for data centers with an IT capacity exceeding 500 kW. The research followed a four-stage approach: starting with data collection at the data center facility, then moving to semantic modeling with Brick vocabulary for data center infrastructure representation, followed by RDF conversion with SHACL validation for model integrity verification, and finally automated SPARQL-based computation of KPIs. The framework produced a validated RDF model comprising more than 20 entities that describe spatial zones, IT equipment, and sensor data. The system generated KPIs that achieved compliance standards through automated calculations, showing an average PUE of 1.3 and indicating 30% additional energy consumption beyond the needs of IT equipment. The semantic modelling approach reduces manual reporting efforts, improves data accuracy, and enables scalable deployment across multiple data center facilities and was validated using SHACL validation. Overall, the framework establishes a foundation for automated EED compliance reporting while supporting advanced analytics for energy optimization and operational excellence in data center management.