SIE Seminar: Harsha Gangammanavar
Thursday, November 16, 2023 - 2:00 p.m. (MST)
Harsha Gangammanavar
Associate Professor
Operations Research & Engineering Management
Southern Methodist University
"Towards a Sustainable Power Grid: Stochastic Hierarchical Planning and Pricing Mechanisms"
ENGR 301
Abstract: Including variable energy resources (such as wind and solar) on a large scale imposes several operational and pricing challenges in power systems. The current practice of planning operations and pricing that rely on deterministic optimization tools is ill-suited for a future with abundant uncertainty due to these intermittent resources. To address the operational challenges, we present a stochastic hierarchical planning (SHP) framework in the first part of the talk. This framework captures operations at day-ahead, short-term, and hour-ahead timescales, as well as the interactions between decisions and stochastic processes across these timescales. While stochastic counterparts of individual optimization problems (e.g., unit commitment and economic dispatch) have been studied previously, this presentation is built around a comprehensive computational treatment of planning frameworks stitched together in a hierarchical setting. Our extensive computational experiments reveal that, relative to its deterministic counterpart, the SHP framework improves reliability, environmental, and economic metrics of assessing power systems operations. In the second part of the talk, we tackle the challenge of electricity pricing under uncertainty. To this effect, we present new market-clearing formulations that are based on the principles of stochastic programming, specifically the notion of nonanticipativity. We design new pricing mechanisms by developing suitable dual optimization problems for these models. Through our analysis, we illustrate the ability of these new pricing mechanisms to provide desirable features such as long-run revenue adequacy for the system operators and cost recovery and price distortion under every scenario for all market participants.
Joint work with S. Atakan (Amazon), S. Ariyarathane (SMU), and S. Sen (USC).
Biography: Harsha Gangammanavar is an associate professor in operations research and engineering management and an affiliated Data Science Institute faculty member at Southern Methodist University. His research focuses on optimization under uncertainty, including developing models for large-scale infrastructure systems, designing algorithms for stochastic programming and distributionally robust optimization that combine decomposition and sampling, and developing scalable and general-purpose optimization solvers. His research is supported by grants from the Air Force Office of Scientific Research, Office of Naval Research, and the Office of Science at the Department of Energy. He received his Ph.D. in operations research and M.S. in electrical engineering from the Ohio State University and a B.E. in electronics and communication engineering from Visvesvaraya Technological University, India.