Dutch Seminar on Digital Energy: Na Li (Harvard University)

This new seminar series aims to bring together mathematics and CS researchers from the Netherlands and internationally, working on analytical techniques for next-generation energy systems. On 12 January, Na Li (Harvard University) will talk about 'Learning and control for residential demand response'.
  • Dutch Seminar on Digital Energy: Na Li (Harvard University)
  • 2022-01-12T15:00:00+01:00
  • 2022-01-12T16:00:00+01:00
  • This new seminar series aims to bring together mathematics and CS researchers from the Netherlands and internationally, working on analytical techniques for next-generation energy systems. On 12 January, Na Li (Harvard University) will talk about 'Learning and control for residential demand response'.
  • What English Intelligent & Autonomous Systems Stochastics Networks & Optimization
  • When 12-01-2022 from 15:00 to 16:00 (Europe/Amsterdam / UTC100)
  • Where via Zoom
  • Contact Name
  • Web Visit external website
  • Add event to calendar iCal

This new seminar series aims to bring together mathematics and CS researchers from the Netherlands and internationally, working on analytical techniques for next-generation energy systems. The seminar has a multi-disciplinary focus, with talks coming from both mathematics (optimization, stochastics and related areas), as well as computer science (artificial intelligence, machine learning etc), with a focus on energy applications, broadly defined. The seminar is organised by CWI, the The Netherlands National Research Centre for Mathematics and Computer Science. One hour talks are held monthly, currently by Zoom.

To participate in the seminars, please register via the mailing list on https://event.cwi.nl/digital-energy/.
--------------------------------------------------------

Wednesday, 12th January 2022, 15:00-16:00 (CET)

Na Li (Harvard University): Learning and control for residential demand response

Residential loads have great potential to enhance the efficiency and reliability of electricity systems via demand response (DR) programs. One major challenge in residential DR is to handle the unknown and uncertain customer behaviors, which are further influenced by time-varying environmental factors. In this talk, we present a set of learning and control methods for regulating loads in residential demand response (DR) by modeling it as a multi-period stochastic optimization problem. Machine learning techniques including both offline and online learning tools are employed to learn the unknown thermal dynamics model and customer opt-out behavior model, respectively. Based on the Thompson sampling framework, we propose an online DR control algorithm to learn customer behaviors and make real-time load control schemes. This algorithm considers the influence of various environmental factors on customer behaviors and is implemented in a distributed fashion to preserve the privacy of customers. This work is based on our collaboration with an industry IoT company, ThinkEco Inc. If time allows, we will briefly present some of our other projects on real-time learning in power systems.
Joint work with Xin Chen, Yingying Li, Yutong Nie, Ran Qin, and Jun Shimada (Founder/CTO of ThinkEco Inc.

Speaker bio: Na Li is a Gordon McKay professor in Electrical Engineering and Applied Mathematics at Harvard University. She received her Bachelor's degree in Mathematics from Zhejiang University in 2007 and a Ph.D. degree in Control and Dynamical systems from California Institute of Technology in 2013. She was a postdoctoral associate at Massachusetts Institute of Technology 2013-2014. Her research lies in the control, learning, and optimization of networked systems, including theory development, algorithm design, and applications to real-world cyber-physical societal systems. She received NSF career award (2016), AFSOR Young Investigator Award (2017), ONR Young Investigator Award(2019), Donald P. Eckman Award (2019), McDonald Mentoring Award (2020), along with some other awards.