Speaker Series - Christopher DeMarco, PhD

Date: 
Tuesday, November 27, 2012 - 11:30am

Christopher DeMarco, PhD
Grainger Professor of Power Engineering
University of Wisconsin-Madison

Enhancing Power Grid Stability through Analytics: Information is Power, and Power is Information

Many challenges in the North American power grid revolve around coordinating and controlling a commodity that moves near speed-of-light, across large geographic distances.  Trends of the last 20 years of U.S. utility restructuring have accentuated these challenges, as the formation of regional transmission organizations and wholesale power markets have encouraged the shipment of more power, over longer distances.  These challenges are likely to grow if the U.S. wants to fully exploit its renewable resources, which often lie distant from major load centers.

Among the recommendations to emerge from studies following the 2003 northeastern U.S. blackout was a call for improved measurement technology, to enhance "situational awareness" in the grid.  With major investment in recent years for Smart Grid technologies, dramatically improved sensors in the form of "phasor measurement units" (PMUs) have indeed been widely deployed across North America.  Among the opportunities for grid engineering today is that of using voluminous data PMUs provide to improve real time monitoring and control decisions, to make the power system less vulnerable to disturbances, more stable, and more reliable.

This work will examine a near-real-time stability monitor for grid operations, based solely on phasor measurement unit data, without need for detailed network model parameters or topology that underlie traditional methods.  Inspired in part by large-scale analytics in such diverse applications as Netflix's Cinematch, this work will argue that the class of "big data" algorithms widely applied in consumer behavior prediction can be adapted to more physically based problems, such as grid stability.  In particular, we propose a singular value decomposition (SVD) based algorithm for PMU data to improve prediction of power grid vulnerability to voltage instability, a phenomena that has contributed to a number of major blackouts and near miss events.

Christopher DeMarco holds the Grainger Professorship in Power Engineering at the University of Wisconsin-Madison, where he been a member of the faculty of Electrical and Computer Engineering (ECE) since 1985.  He has served as ECE Department Chair (2002-2005), and is UW-Madison Site Director for the National Science Foundation IUCRC Power Systems Engineering Research Center (2004-present).  He was recipient of the UW-Madison Chancellor's Distinguished Teaching Award in 2000.  Dr. DeMarco received his PhD degree at the University of California, Berkeley in 1985, and his B.S. degree from the Massachusetts Institute of Technology in 1980, both in Electrical Engineering and Computer Sciences.  His research and teaching interests center on dynamics, control and optimization of electrical energy systems.