Smart thermostats have revolutionized the way households heat and cool their homes, utilizing machine learning to adapt to patterns of occupancy and personal preferences, resulting in reduced energy consumption. While this technology has primarily been applied to individual dwellings, researchers at MIT are now investigating its potential on a larger scale – managing the heating and cooling of an entire campus using artificial intelligence (AI).
The Challenge of Campus Energy Management
Campuses like MIT face unique challenges when it comes to energy management. Existing building management systems often struggle to quickly respond to fluctuations in occupancy or external factors such as weather forecasts or grid carbon intensity. As a result, spaces are heated and cooled inefficiently, consuming more energy than necessary. To address this issue, researchers have turned to AI to develop a framework that can analyze and predict optimal temperature set points for individual rooms, taking into account various factors. By doing so, they aim to optimize the existing systems without the need for manual intervention.
Understanding the Complexity
“Heating and cooling a campus is not that different from what homeowners do,” explains Les Norford, an architecture professor at MIT involved in the research. However, the challenges faced on campus are more intricate. Factors such as classroom usage duration, weather predictions, heat from sunlight, and adjacent room conditions all play a role. Norford and his team, comprising experts from various departments, are focused on tackling these complexities.
Developing Smarter Building Controls
The development of smarter building controls begins with a physics-based model that utilizes differential equations to understand how objects heat up, cool down, and store heat. External data, including weather, grid carbon intensity, and classroom schedules, are also taken into consideration. Using this information, the AI system determines optimal thermostat set points every hour, striking a balance between occupants’ thermal comfort and energy efficiency. Real-life testing validates the effectiveness of these set points, with feedback from building occupants shaping the AI algorithms for energy and carbon emission savings.
Pilot Programs and Future Prospects
The initial pilot programs have focused on testing thermostat set points in specific buildings and classrooms on the MIT campus. As the research progresses, the team aims to expand these pilots to cover an entire building. While energy savings have been estimated for individual classrooms, the potential for substantial energy savings across an entire building is promising. If successful, this AI-driven approach could revolutionize energy management on campuses, paving the way for a more sustainable future.
Q: What is the goal of the research at MIT?
A: The goal is to reduce on-campus energy consumption using artificial intelligence.
Q: How does the AI system determine optimal thermostat set points?
A: The AI system considers factors such as weather, grid carbon intensity, and classroom schedules to determine the best trade-off between occupants’ comfort and energy use.
Q: How will the research benefit campuses?
A: The research aims to optimize heating and cooling systems on campuses, leading to substantial energy savings and a more sustainable approach to energy management.
Q: Are there plans to expand the pilot programs beyond individual classrooms?
A: Yes, the team plans to expand the pilot programs to cover entire buildings and lab spaces in the future.