Lead Performer: National Laboratory of the Rockies - Golden, CO
Partners:
-- Lawrence Berkeley National Laboratory - Berkeley, CO
-- Pacific Northwest National Laboratory - Portland, OR
-- National Institute of Standards and Technology - Gaithersburg, MD
-- Colorado School of Mines - Golden, CO
-- Cornell University - Ithaca, NY
-- Semantic Interoperability Consultants - San Francisco, CA
DOE FY23 Funding: $1,300,000
Funding Type: core lab call, merit review 2022
Project Term: October 1, 2023 - September 30, 2025
Related Projects: Brick - Skewering the Silos, Project Haystack - Metadata for the Masses, Spawn, Open Building Control, VOLTTRON, BOPTEST
Project Overview
High-performance supervisory building operation—advanced control, fault-detection and diagnostics, grid response, and other applications—is a "no regret" strategy for building energy reduction, demand management, and decarbonization. These strategies are effective, savings of 30% have been demonstrated for a range of systems. They are needed to take full advantage of all electric systems whose performance is sensitive to ambient conditions and of hybrid configurations that support a decarbonization transition. And they can be implemented with low up-front cost and carbon relative to envelope and HVAC equipment upgrades.
One barrier to deploying high-performance supervisory strategies at scale is the challenge of configuring and customizing them for specific individual buildings and systems. Not only are buildings and their HVAC systems bespoke—no one is exactly the same as another—but so are building automation system installations. This necessarily makes the installation of building software a manual process, usually by someone who knows either the software to be installed or the existing installation and its conventions, but never both.
Automating the installation of building operation software requires that the building management system have a standard Semantic Model of the building system in question. A Semantic Model is a structured queryable description of the system, its components, their relationships, and their sensing and actuation points. Queryability allows building operation software to systematically "discover" the building and its systems and automatically configure itself to them. A second important use of queryability is that it allows the Semantic Model itself to be checked for soundness and completeness relative to the needs of certain classes of applications. This gives building operators a way of specifying and procuring Semantic Models for their buildings.
ASHRAE Standard 223P "Semantic Modeling for Building Monitoring and Control Applications" is a new standard that is being developed to address this gap. The standard is based on W3C Semantic Web standards like RDF (Resource Descriptor Format), a metadata schema for describing objects, properties, and relationships; SPARQL (SPARQL Protocol and RDF Query Language) query language—computer scientists love recursive acronyms—and SHACL (SHApe Constraint Language). In addition to supporting the development of the standard, the project team is creating templates and tools for building standard-compliant models, checking models for compliance and completeness relative to a set of target applications, and for translating to and from other metadata schema. The team is also collaborating with other BTO project teams to support Standard 223P in other BTO-supported control platforms and tools.
Contacts
- DOE Technology Manager: Amir Roth
- Principal Investigators: Avijit Saha, NLR; Marco Pritoni, LBNL; Steven Bushby, NIST