A Fortune-Telling Model Predictive Control System Could Help Slash Energy Bills Using Waste Heat

Researchers at the Norwegian University of Science and Technology (NTNU) have designed a proof-of-concept heating system which uses model predictive control (MPC) to warm buildings with waste heat from a nearby data center — boosting efficiency over a traditional timer-based control system.

"Making predictions is actually completely different from divination," explains Natasa Nord, professor in the Department of Energy and Process Engineering, of the team's work on boosting energy efficiency at the campus. "Whereas a fortune teller rubs a ball and receives her prediction out of thin air, MPC is research based."

NTNU's Gløshaugen is already extremely energy efficient, thanks to smart reuse of waste heat generated by supercomputers in its on-site data center. Cold water is piped into the data center to cool the computers, and the warmed water piped out to heat the rest of the campus — slashing the amount of heat it needs to pull from the district heating network.

Nord and colleagues, though, believed the campus could be improved still further, by making the control system for the heating system smarter. Enter model predictive control, which pulls in data about the heating system and historical information about everything from regional weather and usage of the supercomputers to make predictions about how the heat should be used.

In simulation, the MPC approach revealed the potential for small but measurable savings of around 1.8 per cent based on a 2021 price model. "We can assume that the effect of the power savings would be even greater today, although I wouldn’t speculate on how much," says Nord.

"Today, cost schemes use different price models for night- and daytime energy use. In any case, our test shows that with MPC we can maximize financial savings for heat consumers like the Gløshaugen campus. The results also show that MPC was more stable and robust than the traditional control system, which is a positive factor for system safety."

The same control system could be used elsewhere too, the researchers argue — anywhere there's surplus heat available. "Acquiring good data that we can build into the programming can be challenging," Nord admits, speaking to potential barriers to the approach's wider adoption.

"Some components have protocols that we struggle to communicate with. They require some work to adapt the programming. But we're continuously working on implementation, and this is something that researchers at NTNU can help other players with."

The team's work is available under open-access terms in the journal E3S Web of Conferences.

Main article image courtesy of Geir Mogen/NTNU.