|
GPG OUTBRIEF 27
Energy Management Information System with Automated System Optimization
If you missed our latest webinar on a Green Proving Ground (GPG) assessment of the energy management information system with automated system optimization (EMIS with ASO), a recording of the webinar and presentation slides are now available.
Thanks to all the presenters and to the participants for their thoughtful questions, some of which are answered below:
Q: Where can new innovative technologies be tested?
A: Each year, GSA, in collaboration with the U.S. Department of Energy (DOE), releases a request for information (RFI) for emerging technologies. This year’s RFI seeks technologies that will help achieve net-zero carbon buildings and you can find more information on the GPG RFI webpage. The RFI closes this Friday, December 9.
Q: Do you think the superior visualization is a matter of the tool itself or how it was set up? Don’t you have similar capabilities in GSALink, the BMS, or SkySpark?
A: GSA has many sophisticated tools, and our buildings are well-run, but we lack a unified user interface (UUI) that gives us a single integrated view of multiple data streams, access to remote facilities, and both historical and real-time data. Ease of use also makes a difference, and the data visualization capabilities of this EMIS enabled non-technical users to discover long-standing operational issues because of the increased visibility. Automated system optimization helped us run our buildings more efficiently and decreased the burden on system operators.
GSA’s Smart Buildings team will soon release a requirements document for a UUI informed by this EMIS with ASO evaluation.
Q: What can this technology do to reduce peak demand?
A: This evaluation confirmed that the predicted peak demand was within 5% of the measured electrical demand for all four test-bed sites. We did not directly control peak demand, though this capability is being assessed as part of a subsequent GPG study on grid-interactive efficient buildings.
Q: Did energy savings negatively impact comfort?
A: No. In Region 7, building managers and O&M staff regularly spoke with tenants and monitored comfort conditions, and there were no comfort issues.
Q: What were the inputs (e.g., ambient temperature, predicted occupancy) to the peak demand predictive model? What type of machine learning algorithm was used for this?
A: Real-time occupancy and energy usage patterns, weather, past trends, and more are used during a “learning period” where the software learns the building's operations to inform prediction using a proprietary vendor algorithm.
Q: Did your algorithms reduce individual AHU VFD fan speeds based on occupancy? If yes, did you have to disable typical static pressure control sequences?
A: The software does not disable other sequencing, but the occupancy-based mid- and end-of-day ramps do have a higher priority. If a building already has duct static resets, these will function on top of those resets.
Reference to any specific commercial product, process, or service does not constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof.
Webinar Links
|
