Postdoc Position: AI-based Load Forecasting for Energy Systems

About the role

Two-year postdoctoral position at Radboud University, Nijmegen, developing AI methods for electricity load forecasting under the European energy transition. The project addresses non-stationarity from EV uptake, distributed solar, and demand response. Funded by a Dutch national grant in partnership with a TSO. Deadline 18 May 2026.

Responsibilities

• Design probabilistic forecasting models for short- and medium-term electricity load.
• Handle distribution shift from rapid electrification (EVs, heat pumps) and weather variability.
• Validate models on real grid data from Dutch operators and publish results.
• Contribute to ongoing collaborations with TenneT and regional DSOs.

Requirements

• PhD in computer science, electrical engineering, applied mathematics, or a closely related field.
• Strong machine learning experience, particularly time-series and probabilistic methods.
• Python proficiency with PyTorch and time-series stacks (PyTorch Forecasting, GluonTS, or similar).
• Fluent English. Dutch is welcomed but not required.

Nice to have

• Prior research on electricity load, demand response, or grid operations.
• Familiarity with probabilistic deep learning (Gaussian processes, normalising flows, deep state-space models).
• Industry collaboration experience with TSOs or DSOs.

How to apply

Open the apply link and submit CV, publication list, research statement, and two reference letters via the Radboud University recruitment portal before 18 May 2026.