Events Calendar

Epstein ISE Department Seminar

Speaker: Victor Zavala, Assistant Computational Mathematician, Mathematics and Computer Science Division, Argonne National Laboratory, Argonne, IL

Talk Title: "A Stochastic Electricity Market Setting with Fair Pricing Properties"

Abstract: We argue that deterministic market clearing settings introduce strong price distortions (difference between day- ahead and expected real-time prices) that lead to arbitrage, biased (unfair) payments, and the need for uplifts. We propose a stochastic market clearing formulation in which deviations between day-ahead and real-time quantities are penalized using $\ell_1$ terms with parameters given by incremental bid prices. We prove that the formulation yields price distortions that are bounded by the incremental bid prices and prove that adding a similar penalty term to transmission flows ensures boundedness throughout the network. We provide conditions under which the stochastic formulation yields day-ahead quantities and flows that converge to the medians of real-time counterparts. This result implies that day-ahead variables only converge to expected value quantities when the distributions are symmetric. We demonstrate that convergence to expected value quantities can be induced using squared $\ell_2$ penalty terms. Our arguments against deterministic settings suggest that comparisons between deterministic and stochastic settings based solely on social surplus are insufficient to fully appreciate the benefits of stochastic settings. We thus present a new set of metrics to perform benchmarks. This is joint work with John Birge and Mihai Anitescu.

MONDAY, MARCH 10, 2014
ANDRUS GERONTOLOGY BLDG (GER) ROOM 206
3:30 - 4:30 PM

Biography: Victor M. Zavala is an assistant computational mathematician in the Mathematics and Computer Science Division at Argonne National Laboratory and he is a fellow in the Computation Institute at the University of Chicago. He received his B.Sc. degree from Universidad Iberoamericana (2003) and his Ph.D. degree from Carnegie Mellon University (2008), both in chemical engineering. He is currently a recipient of the DOE Office of Science Early Career Award under which he develops scalable algorithms for optimization under uncertainty. He also leads an advanced grid modeling project funded by DOE Office of Electricity to develop and test large-scale power grid models and he participates in the Multifaceted Mathematics for Complex Energy Systems project funded by DOE Office of Science. His research interests are in the areas of mathematical modeling of energy and power systems, uncertainty modeling, stochastic optimization, and real-time operations.


Host: Daniel J. Epstein Department of Industrial and Systems Engineering

More Information: Seminar-Zavala.doc

Location: Ethel Percy Andrus Gerontology Center (GER) - Room 206

Audiences: Everyone Is Invited

Contact: Georgia Lum