Third-party cyber evaluations involving OpenAI models
openai.com·13h ago
AI Research
arXiv CS.AI·Jun 10·5 min read
A Survey on Semantic Modeling for Building Energy Management
TL;DR
Building Energy Management (BEM) faces challenges due to inconsistent data models and limited semantic interoperability. This survey reviews 60 semantic models and introduces Ontology Evidence Completeness (OEC) to assess how well operational concepts are mapped to ontology classes.
✦ Why It Matters
Engineers can leverage insights on semantic modeling to enhance interoperability in BEM applications.
Key Takeaways
- Assess your BEM systems for coverage of abstract operational concepts.
- Utilize Ontology Evidence Completeness to evaluate ontology effectiveness.
- Consider ontology reuse and adaptation for improved interoperability.
- Focus on enhancing context-awareness in BEM applications.
Full Summary
Building Energy Management (BEM) plays a crucial role in minimizing energy consumption and CO2 emissions in buildings. Despite advancements in Internet of Things (IoT) technologies that generate vast amounts of operational data, challenges remain due to inconsistent data models and device descriptions.
This survey analyzes 60 semantic models and over 20 ontology-based BEM use cases, introducing Ontology Instantiation Rates (OIR) and Ontology Evidence Completeness (OEC) as new metrics for evaluating ontology effectiveness. Findings indicate that while physical structures and technical systems are well-represented, abstract operational concepts like performance indicators and control logic are often overlooked.
Consequently, many BEM applications rely on reusing and adapting existing ontologies to fill these gaps. The survey clarifies the strengths and weaknesses of current semantic models, guiding future developments towards more interoperable and context-aware BEM systems.