Implementation of model predictive control for an HVAC system in a mid-size commercial building

Abstract

The article presents field experiment results from the implementation of a model predictive controller which optimizes the operation of a variable volume, dual-duct, multi-zone HVAC unit serving an existing mid-size commercial building. This full-scale proof-of-concept study was used to estimate the benefits of implementing advanced building control technologies during a retrofit. The control approach uses dynamic estimates and predictions of zone loads and temperatures, outdoor weather conditions, and HVAC system models to minimize energy consumption while meeting equipment and thermal comfort constraints. The article describes the on-line implementation of the hierarchical control system, including communication of the optimal control scheme with the building automation system, the controlled set-points and the component-level feedback loops, as well as the measured energy and indoor comfort performance benefits from the demonstration. The building-scale experiments and the receding-horizon control algorithm implementation results are described. Site measurements show this algorithm, when implemented in state-of-the-art direct digital control systems, consistently yields energy savings and reduces peak power while improving the indoor thermal comfort. The demonstration results show energy savings of 20% on average during the transition season, 70% on average during heating season, and 10% or more peak power reduction, all relative to pre-configured, rule-based schedules implemented in the retrofitted direct digital control system.

Acknowledgment

The authors are grateful for the financial support and guidance provided by the Strategic Environmental Research and Development Program (SERDP) and Environmental Security Technology Certification Program (ESTCP) Office under the leadership of Drs. Jeff Marqusee and Jim Galvin for this project, EW-0938. Engineer Research and Development Center–Construction Engineering Research Laboratory (ERDC-CERL) and the University of Illinois provided demonstration site support, particularly David Schwenk, Andrew Friedl, Joe Bush, and Clint Wilson. CERL researchers authored the original sequence of controls. Mike Boogemans, Alpha Controls and Services, retrofitted the systems and implemented the DDC mode. Mr. Teja Kuruganti and his team at Oak Ridge National Laboratory (ORNL) supported the sensor network installation and commissioning. Views, opinions, and/or findings contained in this report are those of the authors and should not be construed as an official Department of Defense position or decision unless so designated by other official documentation. This material is based upon work supported by the U.S. Army Corps of Engineers, Humphreys Engineer Center Support Activity, under Contract No. W912HQ-09-C-0056.