Timezone: »

 
Poster
Adversarially robust learning for security-constrained optimal power flow
Priya Donti · Aayushya Agarwal · Neeraj Vijay Bedmutha · Larry Pileggi · J. Zico Kolter

Tue Dec 07 08:30 AM -- 10:00 AM (PST) @
in Poster Session 1 »
In recent years, the ML community has seen surges of interest in both adversarially robust learning and implicit layers, but connections between these two areas have seldom been explored. In this work, we combine innovations from these areas to tackle the problem of N-k security-constrained optimal power flow (SCOPF). N-k SCOPF is a core problem for the operation of electrical grids, and aims to schedule power generation in a manner that is robust to potentially $k$ simultaneous equipment outages. Inspired by methods in adversarially robust training, we frame N-k SCOPF as a minimax optimization problem -- viewing power generation settings as adjustable parameters and equipment outages as (adversarial) attacks -- and solve this problem via gradient-based techniques. The loss function of this minimax problem involves resolving implicit equations representing grid physics and operational decisions, which we differentiate through via the implicit function theorem. We demonstrate the efficacy of our framework in solving N-3 SCOPF, which has traditionally been considered as prohibitively expensive to solve given that the problem size depends combinatorially on the number of potential outages.
Keywords: Optimization · Robustness · Adversarial Robustness and Security

Author Information

Priya Donti (Carnegie Mellon University)
Aayushya Agarwal (Carnegie Mellon University)
Neeraj Vijay Bedmutha (Carnegie Mellon University)
Larry Pileggi (Carnegie Mellon University)

Lawrence Pileggi is the Tanoto professor and Head of electrical and computer engineering at Carnegie Mellon University, and has previously held positions at Westinghouse Research and Development and the University of Texas at Austin. He received his Ph.D. in Electrical and Computer Engineering from Carnegie Mellon University in 1989. He has consulted for various semiconductor and EDA companies, and was co-founder of Fabbrix Inc., Extreme DA, and Pearl Street Technologies. His research interests include various aspects of digital and analog integrated circuit design, and simulation, optimization and modeling of electric power systems. He has received various awards, including Westinghouse corporation’s highest engineering achievement award, a Presidential Young Inves¬tigator award from the National Science Foundation, Semiconductor Research Corporation (SRC) Technical Excellence Awards in 1991 and 1999, the FCRP inaugural Richard A. Newton GSRC Industrial Impact Award, the SRC Aristotle award in 2008, the 2010 IEEE Circuits and Systems Society Mac Van Valkenburg Award, the ACM/IEEE A. Richard Newton Technical Impact Award in Electronic Design Automation in 2011, the Carnegie Institute of Technology B.R. Teare Teaching Award for 2013, and the 2015 Semiconductor Industry Association (SIA) University Researcher Award. He is a co-author of "Electronic Circuit and System Simulation Methods," McGraw-Hill, 1995 and "IC Interconnect Analysis," Kluwer, 2002. He has published almost 400 conference and journal papers and holds 40 U.S. patents. He is a fellow of IEEE.

J. Zico Kolter (Carnegie Mellon University / Bosch Center for A)

More from the Same Authors