PhD Position: Physics-Informed Generative AI for Synthetic Energy Data

The NWO-funded SHARE project develops realistic, privacy-preserving synthetic energy data for grid planning and decision-making, using operational data from Alliander. The PhD researcher will compare deep generative and probabilistic approaches, incorporate power-flow equations, operational bounds and network topology, and evaluate statistical fidelity, physical plausibility and downstream performance. Collaboration with privacy researchers will integrate differential privacy; results will support an open-source toolbox for distribution-system operators and municipalities.

Supervision is by Yuliya Shapovalova and Tom Heskes in the iCIS Data Science section. The 1.0-FTE appointment begins with 1.5 years and can extend by 2.5 years after a positive assessment; gross monthly salary rises from EUR 3,204 to EUR 4,051. Preferred start: 1 January 2027. Apply through the university’s linked form by 25 October 2026.

MSc by the start date in computer science, AI, data science, applied mathematics, physics, electrical engineering or a related field. Solid machine-learning background, Python and a deep-learning framework, interdisciplinary interest and good English. Energy-system knowledge is not required. The motivation letter should explain a proposed generative approach, a software/data project and the MSc thesis; use the university application form.