Energy efficient control of buildings became a hot topic when artificial intelligence entered the realm. A paper on advanced building controls, co-authored by Pacific Northwest National Laboratory (PNNL) data scientist Ján Drgoňa, was recently named one of 2022’s best papers in the journal Building and Environment.
The paper, which was written by 23 authors representing 17 institutions from around the globe, describes how building occupants’ behaviors can be modeled into building controls. Drgoňa specializes in learning-based model predictive control—an advanced control method that uses models to predict the behavior of a controlled system—and its application toward controlling heating, ventilation, and air conditioning systems in buildings.
“For decades, there has been very little innovation in the way we control HVAC systems,” said Drgoňa. “Now, however, energy efficiency and grid responsiveness are considered to be the top priorities in buildings control research.”
At PNNL, Drgoňa works closely with Chief Data Scientist Draguna Vrabie to further integrate deep learning into model predictive control. The two were recently part of an international team that authored a “most-cited” paper on model predictive control published in Annual Reviews in Control. Together with data scientist Aaron Tuor and other researchers from PNNL, the team also developed Neuromancer, an open-source software framework for data-driven modeling and advanced control of dynamical systems with downstream applications in energy efficient buildings.
Drgoňa received his PhD in control engineering from the Slovak University of Technology in Bratislava, Slovakia. He has received multiple awards for his work, including another best paper award and was recently quoted in New Scientist magazine in an article about AI-powered control of commercial cooling systems in buildings.
Drgoňa's research is supported by the Data-Model Convergence initiative and the Mathematics for Artificial Reasoning in Science initiative via the Laboratory Directed Research and Development (LDRD) investments at PNNL. Drgoňa is also supported by the Department of Energy, Office of Advanced Scientific Computing Research's “Data-Driven Decision Control for Complex Systems” project where he serves as a task lead, and through the Energy Efficiency and Renewable Energy, Building Technologies Office (BTO) under the “Dynamic Decarbonization through Autonomous Physics-Centric Deep Learning and Optimization of Building Operations” and the “Advancing Market-Ready Building Energy Management by Cost-Effective Differentiable Predictive Control” projects, where he serves as a co-PI.