Austin Lin awarded IEEE PES Prize Conference Paper for work quantifying HVAC load-shifting efficiency

Austin Lin, PhD student in Electrical and Computer Engineering (ECE), received a Prize Conference Paper award for his presentation Experimental Investigation of Building HVAC Load-Shifting Efficiency at the 2024 IEEE Power & Energy Society General Meeting in Seattle. His work aims to control the power consumption of buildings to provide flexibility to the power grid.

As the electrical grid incorporates renewable energy sources into its infrastructure, it is becoming increasingly challenging to ensure that the power supply and demand are matched at all times. The process of scheduling power generation to meet fluctuating demands is called balancing the grid––and Lin’s research investigates the grid-balancing potential of commercial buildings, through the flexible use of their heating, ventilation, and air conditioning (HVAC) systems. These buildings account for about 30% of all U.S. electrical consumption.

“Buildings represent a huge load on the grid,” he explained, “But they also represent a very flexible load. When you plug your phone in, you want it to be charging right away, but with building HVAC systems you usually don’t care when it is running, as long as the temperature in the building is okay.”

This means that HVAC power could be strategically adjusted, allowing buildings to actively participate in grid balancing, so long as the room temperature changes remain modest. This could help balance out the fluctuations of renewable generation, at times when the wind is gusty or clouds are passing over solar panels. Lin thinks of the buildings as virtual batteries––where over-consuming power charges the “battery” and makes the building colder, while under-consuming power discharges the “battery” and makes the building warmer. Although the building always consumes power, the “battery” is created by the relative power consumption, compared to what the building would have otherwise consumed. Within this context, Lin focuses on load shifting, or moving around the timing of power usage without changing the total energy consumption of the building.

“What we’re specifically doing is seeing if we can utilize HVAC fans for short timescale balancing, where we’re looking at shifting load around within a single hour of the day,” said Lin. “There’s a lot of resources that work on that timescale. A lot of the faster dynamics aren’t necessarily shown in simulations and models, and haven’t been identified.”

To understand more about these dynamics, Lin analyzed an existing dataset of measurements from the HVAC systems of 14 buildings on the U-M campus to quantify their efficiency. He was interested in quantifying how effective storing energy in building “batteries” could be as a grid resource.

The presented study found that, although the buildings had some inefficiencies in returning to their normal operating states after load shifting events, they were more effective at shifting their usage on sub-hourly timescales than previous studies had indicated. The analysis found that buildings with quicker recovery times after the events tended to be more efficient at load shifting, and that events with more shifted power were less efficient. Lin’s findings indicate that the tuning of building controllers and strategic planning of events may be able to improve load shifting efficiency.

The next steps for this research will identify the building systems and controls causing the observed inefficiencies, so that building managers can better plan for load shifting events. Lin will continue this work with his PhD advisor, Johanna Mathieu, associate professor of ECE.

“At the end of the day, what excites me is that there’s no one answer,” he said, “I see a lot in the news about climate change and decarbonization being huge issues, but so many people are focused on solutions that add new resources. They’re not thinking about the flexibility that we already have available. To me, there’s always the part of using what we have, and just seeing how far we can go with it.”

The paper was selected as a Prize Conference Paper from thousands of paper submissions at the IEEE PES meeting. Lin co-authored the paper with Mathieu and Shunbo Lei, a postdoctoral fellow alum of Mathieu’s research group and current assistant professor at the Chinese University of Hong Kong – Shenzhen.