Project will develop human-in-the-loop intelligent systems to optimize energy use in buildings
Source: [Columbia Engineering]
The National Science Foundation (NSF) has awarded Xiaofan (Fred) Jiang, assistant professor of electrical engineering, with a CAREER Award, the most prestigious recognition the NSF gives to young researchers. Jiang will receive $500,000 to fund his project, “A Scalable Occupant-Driven Energy Optimization System for Commercial Buildings.”
Buildings are responsible for 40% of the nation’s energy spend. Jiang plans to use his expertise in sensors, wearables, cyber-physical systems, and data analysis to help building occupants track their personal energy use in real-time and receive recommendations for actionable steps to reduce their “energy footprint.”
“The system we are proposing could move us toward a more environmentally responsible and sustainable future,” said Jiang. “By providing real-time visibility into the energy cost of each action, people will have at their fingertips the information they need to reduce their personal energy usage and reduce building energy waste.”
The project will include elements of real-time sensing and actuation, building energy monitoring, indoor localization, large-scale time-series analytics, and recommender systems. To realize the proposed system, Jiang will focus on three research thrusts. First, a digital twin of a commercial building, together with a simulation environment, will be created to model how human actions can drive energy savings. It will also help determine algorithms for computing each occupant’s energy footprint in shared spaces. Second, Jiang will design a reinforcement learning-based recommender system that can uncover which actions have the most potential to realize energy savings while adapting to user preferences. As a third component, the project will discover the most effective incentives and feedback methods to encourage energy-saving behaviors and their adoption rate, as well as increase recommendation quality.
In addition to providing insights for occupants and building managers, Jiang will also make research products from his work publicly available for those who seek to translate findings to their own system designs and to educate the public on how individual actions impact energy. The project will also develop online course modules for undergraduate and graduate students on embedded systems, mobile computing, Internet-of-Things, and deep reinforcement learning.
Working with colleagues in psychology and civil engineering, Jiang plans to create a living testbed on the Columbia campus to study the effectiveness of the recommendation and incentive system and evaluate its scalability. Eventually, the campus deployment site will be made open to the public as an ongoing exhibition and as a means to collect more data for the system.
“Most people have little or no idea that their actions and choices while at work could potentially have a significant impact on the energy consumption of their workspaces,” said Jiang. “This experience will help reinforce that awareness and pave the way towards a more energy conscious future, where people are active participants in large self-optimizing building systems.”
Jiang, who also co-chairs the Smart Cities Center at the Data Science Institute, joined Columbia Engineering in 2015. His research lies at the intersection of systems and data, with projects spanning wearable systems for pedestrian safety, the intelligent built environment, air quality monitoring, and connected health. His research has been published in top venues with over four thousand citations and featured in many popular media outlets, including The Economist, New York Post, Mashable, Gizmodo, The Telegraph, and Fast Company.
—by Allison Elliot