Modeling energy systems for a data-center-driven future

Tuesday, January 27-Friday, January 30, 9:00 am-1:00 pm ET each day (4 classes) Location: 4-231

Feel free to drop by. Registration by January 27 appreciated. Email Pablo Duenas (pduenas@mit.edu).

 

For the 17th consecutive year, this 4-session hands-on learning experience continues to evolve, exploring how mathematical modeling can inform and accelerate the transition toward net-zero targets. With a primary focus on electricity systems, the course highlights their central role in a carbon-constrained economy that must deliver reliable, affordable energy while accommodating rapid demand growth, especially from data-center development. Participants will examine critical challenges shaping future power systems, including large-scale carbon-free energy deployment, the expanding potential of demand response, and the accelerating rise of data centers as dominant electricity consumers. Addressing these challenges requires advanced mathematical models to optimize and analyze complex decisions, from grid and generation expansion to flexibility, to ensure the system can reliably meet sustained load growth. In addition to providing theoretical insights, the course offers practical tools that enable participants to perform their own case studies. Real-world applications will illustrate how quantitative modeling can inform key stakeholders, guide public understanding, and support collective action toward a secure, clean, and data-center-ready energy future.

No prior experience is required, although basic familiarity with Python and Julia programming can be helpful. Participants are welcome to attend individual sessions.

Tuesday, January 27

Part 0: Introduction to fundamentals of optimization techniques

Part 1: Covering electricity demand daily

Unit-Commitment (UC): daily dispatch of electricity generation units Managing uncertainty through stochastic optimization of UC

Part 2: Guaranteeing annual electricity production

Medium-term operation planning Managing uncertainty through stochastic hydro-thermal coordination

Wednesday, January 28

Part 3: The network as the backbone of electric systems

Understanding the role of the electricity network Managing network constraints with Locational Marginal Pricing

Part 4: Models for informing utility-scale investments

MACRO: an expansion model for studying low-carbon energy futures

Thursday, January 29

Part 5: Powering AI and data centers

Explaining data centers from an energy perspective Placement and connection of data centers

Part 6: Electricity transmission, storage, and generation expansion planning

openTEPES: informing infrastructure needs for growing demands

Friday, January 30

Part 7: Flexibility and dynamics of data centers

How much flexibility can a data center provide Flexible data centers and the grid

Instructors

Pablo Duenas – Research Scientist at MIT Energy Initiative, pduenas@mit.edu

Deep Deka – Program Manager of Data Center Power Forum at MIT Energy Initiative, deepj87@mit.edu

Andres Ramos – Professor at Universidad Pontificia Comillas, arght@mit.edu

Javier Garcia-Gonzalez – Professor at Universidad Pontificia Comillas, javiergg@mit.edu

Ruaridh McDonald – Energy Systems Research Lead at MIT Energy Initiative, rmacd@mit.edu

Invited speakers

Shaohui Lui – Postdoctoral Associate at MIT Chemical Engineering, shaohuil@mit.edu

Juan Senga – Postdoctoral Associate in the MIT Center for Energy and Environmental Policy Research, jsenga@mit.edu

Yifu Ding – Postdoctoral Associate at MIT Energy Initiative, yifuding@mit.edu