Lead Performer: Community Energy Labs – Portland, Oregon

Partner(s): ENERlite Consulting – Sacramento, CA

DOE Total Funding: $206,000

FY21 DOE Funding: $206,000

Project Term: July 2021 – May 2022

Funding Type: SBIR phase I release 2 – 2021

Project Objective

Model predictive control (MPC) to leverage building thermal mass has been shown to be effective reducing at HVAC energy use and peak demand. Academic buildings have the size and thermal mass to be able to leverage MPC for HVAC energy and demand reduction, but they rarely have the capital budgets to procure building automation systems or the operational expertise to implement these strategies.

Community Energy Labs, LLC (CEL) will cater to educational buildings and other buildings in the MUSH sector by develop a protocol and specification for configuring MPC on low-cost hardware. The protocol will target hybrid grey-box MPC models—specifically LBNL’s MCPy and PNNL’s neuroMANCER—as these have fewer configuration parameters than full detail models. It will be optimized to collect the most impactful parameters, with consideration for manual effort, and will be tested with and packaged for staff that currently operate and maintain these buildings. This system will allow facility managers to implement smart, grid-responsive HVAC control strategies without requiring either expertise in building climate control or expensive fully automated control systems.

Project Impact

CEL has identified over 13,000 candidate academic buildings in California, Oregon, and Washington. MPC could not only help these buildings save energy, but also allow them to avoid peak demand charges which will come into force within the next five years.

Contacts

DOE Technology Manager: Amir Roth

Principal Investigator: Jennifer Worrall