Climate downscaling
How can statistical and generative downscaling methods represent the aspects of weather and climate that matter for energy applications, including statistical fidelity, physical consistency, and power-system extremes?
Ziqi Wei
Ph.D. Student in MIT's CSE–CEE Interdisciplinary Ph.D. Program
Hi, I'm Ziqi! I'm a Ph.D. student at MIT, working in the Howland Lab with Prof. Michael Howland and closely with Prof. Saurabh Amin. My research focuses on representing weather and climate uncertainty in low-carbon energy systems, with interests spanning climate downscaling, climate–energy interactions, adaptive power-system decision-making, and, more recently, LLM-based multi-agent negotiation.
Before starting my Ph.D., I graduated from Weiyang College, Tsinghua University, with a B.S. in Mathematics and Physics and a B.Eng. in Environmental Engineering. My earlier research included climate and building-energy modeling advised by Prof. Michael T. Craig and Prof. Xi Lu, precipitation uncertainty modeling at Stanford advised by Prof. Sarah Fletcher, and renewable-resource drought analysis at Cornell advised by Prof. K. Max Zhang.
If you'd like to talk about research, climate, energy, or brainstorm something fun together, I'd love to hear from you.
Research interests
Climate-energy interactions
How do uncertainties in weather and climate propagate into renewable power generation and electricity demand?
Power-system decision-making under uncertainty
How can power-system planning and operations be dynamically optimized under weather and climate uncertainty to support adaptive, resilient, and low-carbon decisions?
LLM-based multi-agent negotiation
Can LLM-based multi-agent systems reproduce the strategic interactions, coalition dynamics, and negotiated outcomes observed in real-world multilateral negotiations?