Exploring Human Thermal Comfort Vs. Smart Thermostats
Assistant Professor Michael Kane, civil and environmental engineering, was awarded a $763K NSF CAREER Award for “Human-Centric Automation in the Built Environment.” The project will enable the design of automation in the built environment that models human physiological and behavioral responses to changing environmental temperature conditions to satisfy competing objectives of energy management and occupant comfort.
Kane joined the Department of Civil and Environmental Engineering in Spring 2017 as an assistant professor. He earned his PhD in Civil and Environmental Engineering – Systems from the University of Michigan. His research focus is in the areas of occupant-centric building controls, community resilience, model predictive control, and hybrid systems.
A Smart Home that Warms the Hearth and Not the Planet
In the Automation in the Built and Living Environment (ABLE) Lab at Northeastern, Assistant Professor Michael Kane, civil and environmental engineering, and his students look at how people and automation interact in man-made environments such as buildings and transportation. The ABLE Lab’s latest project—of which he received a prestigious $763K National Science Foundation CAREER Award—is focused on improving the automation of the home thermostat to not only use less energy, but eventually to anticipate comfort needs.
“The ‘brains’ of thermostats as we know them today are simple rules that help your home hold a temperature,” says Kane. “People don’t really know how much energy they’re consuming until about a month later when their utility bill arrives, making it hard to know how to save energy. We want to shorten that feedback loop to real-time, and to use human-centric automation to make it something you don’t have to worry about.”
The science behind Kane’s work is to understand all the physiological and psychological factors that affect how people perceive their home’s temperature in that moment and change it to make them comfortable. This data could help build artificial intelligence (AI) that predicts how comfortable they are feeling—even before they know it themself—and it helps them understand how to work with the AI to make their energy consumption friendlier to an electric grid with variations in available renewable energy.
Kane and his team are partnering with Professor Misha Pavel from the Bouvé College of Health Sciences at Northeastern, an expert on health behavior and using smart technology, as well as ecobee, a company that creates smart thermostats and other home technologies.
“Human comfort is a very personal and subjective thing,” says Kane. “Sixty-five degrees Fahrenheit in March in Boston feels very different than 65 degrees in August in San Diego. In order to get accurate data, we can’t study thermal comfort only in a lab setting.”
With the funding from the NSF, Kane seeks to overcome this limitation by recruiting 100 homes across the U.S. for a pilot study. Over a period of several years, they will track their home heating and cooling data with ecobee thermostats. Real-time algorithms will monitor irregular changes in the environment or user behavior and prompt participants to take a second to answer a question with a swipe on their smart watch.
“With this info, we can probe into all of the physical and psychological factors that contribute to a person’s comfort” says Kane. “The ultimate goal is to understand these personal internal processes using data from the large cohort of participants to homes comfortable to the people living there while using energy when the grid is clean and cheap, and holding off on heating and cooling for short periods when the grid is dirty and expensive.”
Kane is aiming to collect three years of data from this pilot study, which will then be combined with an existing data set from 100,000 ecobee users who have opted in to anonymously donate their temperature data to science. He hopes to start experimenting with different control algorithms within one to two years.
The project will also foster a collaboration between some of Kane’s students and those in the new heating, ventilation, and air conditioning (HVAC) program at Madison Park Technical Vocational High School in Roxbury. Participants in Kane’s Energy Systems class will complete hands-on demonstrator exercises in this new technology to train the next generation of HVAC technicians, installers, and operators. Kane hopes that connecting the two groups can be a catalyst for building critical relationships between innovative designer and skilled worker.
Award Source: NSF