DC6 — Flexibility Modeling and Forecasting for EDC Fleets

Host: Technical University of Denmark (DTU), Lyngby, Denmark — Dynamical Systems section (DynSys), DTU Compute.
Work package: WP2 (Carbon-Aware Intelligence and Control)
Duration and start: 36 months, full-time; start between February and September 2027 (latest start September 2027).

The project

Carbon-aware control can only plan ahead if it has good forecasts of future grid conditions. In this project you build the forecasting engine for a fleet of edge datacenters.

You will extend a Smart Energy operating system developed at DTU with forecasts of carbon intensity, energy prices, and grid congestion across many time horizons, together with the uncertainty estimates that risk-aware scheduling needs. You will also combine the limits of many sites into fleet-wide flexibility functions.

What you will work on:

• Produce multi-horizon forecasts for carbon, prices, and congestion, from five minutes to a day ahead.
• Develop flexibility-function representations that compress a site’s constraints into a compact form.
• Design a hierarchical method that adds up flexibility across the whole fleet.

What you will deliver:

• A forecasting suite for grid signals with measured accuracy and calibrated uncertainty bounds.
• An aggregation framework, with a report on its economic value in simulated market scenarios.

Secondments

You will spend two research visits with partners in the network:

• Energy Cool (EC Group) (Denmark) — 3 months from around October 2027 (industry), with Henrik Thorsen: model how datacenter cooling systems can take part in demand-response schemes.
• EPRI Europe (Dublin, Ireland) — 3 months from around May 2028 (academic), with Eamonn Lannoye: test multi-horizon forecasts for carbon intensity, energy prices, and grid congestion using EPRI’s market-analytics platforms.

The host

The Dynamical Systems section at DTU Compute works on modeling, optimization, forecasting, and control of energy systems, with a focus on smart energy.

Supervision

Main supervisor: Henrik Madsen (ORCID) — Professor in the Dynamical Systems section, DTU Compute. He is a leading expert in smart-energy systems and stochastic dynamical systems.

Co-supervisors:

• Henrik Thorsen — Energy Cool (EC Group)
• Eamonn Lannoye — EPRI Europe

Profile we are looking for

• Time-series forecasting, statistics, or stochastic modeling
• Programming and data analysis
• Knowledge of energy systems is a plus
• MSc (completed or near completion) in data science, applied mathematics, electrical engineering, or a related field
• Curiosity, initiative, and good spoken and written English

You must also meet the MSCA eligibility rules — no PhD yet, the mobility rule, and open to all nationalities.

How to apply

Official vacancy at the host institution: link coming soon.

Until then, apply through the central AEGIS application portal: AEGIS application portal (reference position DC6), and read what to include in your application.

Questions about this position? Contact the coordinator, Prof. Paul Pop, aegis@compute.dtu.dk.