(383i) Decision-Making for Integrated Energy Systems Using Multiscale Optimization and Machine Learning Methods

Research Interests: Process System Engineering, Operations Research, Machine Learning, Energy Markets

As the global energy landscape shifts toward decarbonization, integrating renewable energy sources into the electricity grid has become both a priority and a challenge. Variability and uncertainty from renewables such as wind and solar can strain grid reliability and operational planning. To address these challenges, Integrated Energy Systems (IES)—which combine diverse energy resources and technologies—have emerged as a promising solution to enhance energy efficiency, support grid resilience, and provide greater operational flexibility for market participants.

However, optimizing the design and operation of IES is inherently complex due to multi timescale dynamics and the need to make decisions under uncertainty. Traditional modeling approaches often rely on the price-taker assumption, treating electricity prices as fixed inputs and ignoring the feedback between system operations and market outcomes. To capture the full value and impact of IES, it is essential to adopt more advanced optimization frameworks that consider the IES as an active market participant capable of influencing electricity prices and system conditions.

This work presents a multiscale optimization framework and a machine learning assisted optimization framework to move beyond the price-taker optimization of IES, systematically comparing them with the price-taker optimization. The results show that the price-taker assumption overestimates the economic values of the IES and ignores the uncertainty in the energy markets.

Bio:
Xinhe Chen is a fifth-year Ph.D. candidate in Chemical and Biomolecular Engineering at the University of Notre Dame, advised by Prof. Alexander Dowling. His research lies at the intersection of operations research and energy systems engineering, with a focus on modeling, simulation, and optimization of integrated energy systems within wholesale electricity markets. Supported by the Institute for the Design of Advanced Energy Systems (IDAES), Xinhe develops advanced optimization frameworks that incorporate both market dynamics and system-level uncertainty. He has extensive experience with mathematical programming (linear, nonlinear, and mixed-integer) and machine learning methods for decision-making. Xinhe is seeking full-time opportunities beginning in spring or summer 2026.