Special Seminar: Management and Learning for Future Energy Systems

Vassilis Kekatos, Postdoctoral Associate, ECE Deptartment, University of Minnesota

Monday, April 6, 2015, 4:00–5:00pm

Room 200, Cummings Hall

Current power infrastructure faces major challenges related to uncertainty, security, and environmental impact. While new technologies with cyber capabilities are installed throughout the grid, effective ways for processing the resultant data are now demanded. Advances in large-scale algorithms, low-rank and sparse modeling, and online optimization, are definitely pertinent in power engineering problems. In this context, this talk presents contemporary data-driven methods for enhancing our energy grids. As a control example, we will first demonstrate how online distribution-free schemes coupled with recent convex relaxations can be effective for reactive power control and voltage regulation in microgrids with renewables. Studying the potential of revealing the topology of the underlying physical grid by simply processing real-time energy market data is the second theme of this talk.

About the Speaker

Dr. Vassilis Kekatos obtained his PhD in Computer Engineering and Science from the Un. of Patras, Greece. He is a recipient of a Marie Curie Fellowship, while he received the postdoctoral career development award (honorable mention) by the University of Minnesota in 2014. During the summer of 2012, he worked for Windlogics Inc. During 2014, he stayed with the University of Texas at Austin and the Ohio State University as a visiting researcher. His research focus is on monitoring, control, and optimization for future energy systems.

For more information, contact Haley Tucker at haley.tucker@dartmouth.edu.