Bio

Meiyi Li is a Ph.D. candidate in Civil Engineering at The University of Texas at Austin, where she works with Prof. Javad Mohammadi on trustworthy and sustainable AI for energy systems. She previously studied Electrical and Computer Engineering at Carnegie Mellon University and earned B.S./M.S. degrees in Electrical Engineering from Shanghai Jiao Tong University, graduating from the prestigious Outstanding Engineers Honor Class.

Her research spans real-time optimization of power grids, renewable and storage integration, decentralized energy coordination, transportation electrification, and the cybersecurity of AI-driven infrastructures. She focuses on developing theoretically grounded, policy-aware AI frameworks that ensure energy system operations are not only fast and efficient but also reliable, safe, and sustainable.

Her contributions have been recognized with numerous awards, including the Chevron Energy Graduate Fellowship (2025), Highest-Ranked University Team award in the ARPA-E Grid Optimization Competition (2021), and the IEEE PES Best of the Best Paper Award (2019).

Design and Control of Self-Adaptive Physical Systems

Energy systems are being transformed by renewables, storage, electrification, and climate resilience pressures, creating safety-critical infrastructures where failures can trigger blackouts, service disruptions, and economic risks. Operating these systems requires decision tools that are both fast and reliable, yet traditional optimization solvers are too slow for real-time needs. Artificial Intelligence (AI) and Machine Learning (ML) promise rapid intelligence, but speed without guarantees is unacceptable in such contexts, as unreliable decisions can amplify risks. At the same time, data centers and large-scale AI models increasingly demand vast amounts of electricity and water for computation and cooling. This creates a dual challenge: developing trustworthy, context-aware AI for energy system operations while ensuring that increasingly resource-intensive AI can be supplied sustainably by the very energy systems it seeks to improve.

My research addresses this challenge by reimagining AI as a trustworthy decision engine for energy systems. I develop frameworks that integrate optimization theory, machine learning, and deep domain knowledge of physical dynamics, operational constraints, and regulatory requirements. By embedding rigor, interpretability, and policy awareness into AI-driven decision tools, my work advances a paradigm shift from “speed alone” to “speed with trust” . The resulting methods enable fast, reliable, and safe decision-making for real-time energy operations while aligning with environmental and societal imperatives.