Abstract:
Operational decision-making in modern power systems is increasingly shaped by uncertain market signals, such as electricity prices and curtailment levels. In this talk, I will present our research group’s recent efforts to develop data- driven methods for forecasting these signals, and further leveraging them to inform asset-level decision-making. First, I will present a multivariate statistical approach for electricity price forecasting designed to capture system, market, and temporal dependencies that are prevalent in electricity price signals. The proposed approach is evaluated on two years of electricity prices from the California Independent System Operator, showing significant improvements in both point and probabilistic forecast metrics relative to well-established statistical and emerging deep learning methods. Independent validation against industry-adopted forecasting systems further demonstrates the approach’s competitive performance and practical relevance. I then turn to how variability in market signals (naturally viewed as a challenge for asset management) can, counter-intuitively, be turned into an opportunity for improved decision-making. In particular, I will present a grid- informed maintenance optimization framework for wind energy assets that incorporates grid- level information, such as electricity prices and curtailment, to support condition-based maintenance decisions. Together, these results highlight how market signals can be accurately predicted, and further leveraged to inform asset management, bridging forecasting and optimization in modern power systems.

Biography:
Aziz Ezzat is an Assistant Professor of Industrial & Systems Engineering at Rutgers University, where he leads the Renewables & Industrial Analytics (RIA) Research Group. He received his PhD degree in Texas A&M University, and his BSc. Degree from Alexandria, Egypt, both in Industrial & Systems Engineering. Aziz’s research develops data science, AI, and machine learning methods for energy, environmental, and industrial systems, with support from the National Science Foundation, U.S. Department of Energy, the state of New Jersey, and industry partners. His work has appeared in leading journals such as Technometrics, Annals of Applied Statistics, IISE Transactions, and IEEE Transactions on Sustainable Energy. Aziz is a recipient of the A. Walter Tyson Early Career Award, the IIF-SAS ® Research Methodology Award, and the Excellence in Teaching Awards from the Operations Research and Data Analytics Divisions of the Institute of Industrial & Systems Engineers (IISE). He served as the 2023-2024 President of the Energy Systems Division of IISE, where he introduced numerous initiatives to advance the broader impacts of data and decision sciences, including the inaugural PG&E Energy Analytics Challenge—an industry-sponsored, national-scale energy forecasting competition. He is a professional member of INFORMS, IISE, IEEE, and IIF. More about his research and teaching can be found at: https://sites.rutgers.edu/azizezzat/.