Postdoctoral Researcher Position In Artificial Intelligence, Hybrid Modelling And Digital Twins

Postdoctoral Researcher position in Artificial Intelligence, Hybrid Modelling and Digital Twins for Small Modular Reactors (SMRs)

Application Form

IMDEA Energy Institute is a Severo Ochoa Centre of Excellence dedicated to conducting research and development (R&D) in the fields of renewable energy and clean energy technologies, while fostering strong partnerships with leading companies in the energy sector ((Use the "Apply for this Job" box below).).

IMDEA Energy is opening a Postdoctoral Researcher Position for an interdisciplinary project between the Artificial Intelligence Unit and the Nuclear Technologies Unit, looking for talented and motivated candidates to start a new research line on applying AI techniques to model, simulate, predict, optimize and evaluate SMR systems.

We are looking for a solid profile in scientific modeling of physical systems and Artificial Intelligence experience. The project will start defining the scientific and technological basis, with simplified models and simulated data. Later, the research will evolve with more complex models and experimental data coming from research and industrial collaborators in the nuclear sector.

This researcher will work in close collaboration with researchers from the Nuclear Technologies Unit for the definition, development and validation of the physical models used as the basis of the hybrid models and digital twins as the main focus of this position.

• Conduct a state-of-the-art review about physical, data-based and hybrid models, and digital twins applied to nuclear energy systems (particularly SMRs), contributing to the scientific roadmap of the new research line.
• Identify and select the most suitable AI methodologies and base physical models (neutronics, thermal-hydraulics) for modeling problems such as an SMR digital twin.
• Design hybrid models combining physical knowledge, behavioral equations, and AI techniques, and develop methodologies to estimate variables that cannot be measured directly.
• Develop machine learning models and surrogate models to accelerate physical simulations in SMR components.
• Design the digital twin architecture, defining the data architecture and the interface between the physical model and the AI learning layer.
• Develop an initial digital twin demonstrator for a SMR system or component.
• Apply early anomaly detection, diagnostic, and predictive maintenance techniques, as SMR digital twin applications, when available data.
• Incorporate methods for uncertainty quantification, robustness, explainability, and validation of AI models.
• Generate, organize, and analyze databases from simulated and experimental data, and validate digital models against them.
• Contribute to national/international project proposals and deliverables
• Publish and present results at scientific conferences (e.g., Spanish Nuclear Society Annual Meeting, SNETP, CEIDEN, AI forums).
• Co-supervise undergraduate/graduate staff.

• PhD in Artificial Intelligence, Computer Science, Data Science, Mathematics, Physics, Nuclear Engineering, or equivalent engineering with high computational modeling.
• Proven experience (pointing specific publications in the motivation letter) in scientific modeling, simulation and/or control of [BG4.1]complex nonlinear dynamical physical systems.
• Proven experience (pointing specific publications in the motivation letter) in different data-based modeling and AI methodologies.
• Demonstrated solid programming skills in Python [BG5.1](add link to github or portfolio in CV).

• Strong track record of scientific publications in AI.
• Knowledge of thermo-hydraulics, heat transfer, nuclear energy, or reactor safety.
• Experience with physical modeling and simulation of physical, thermal, energy, or industrial systems (e.g., MOOSE, RELAP, TRACE).
• Experience with surrogate models.
  
  Experience with digital twins.
• Experience with hybrid physical-data architectures (e.g., PINNs).
• Experience with time-series analysis, anomaly detection, or predictive maintenance.
• Experience with reinforcement learning applied to optimization and control.
• Experience developing, assessing, and validating predictive models.
• Experience with uncertainty quantification, explainable AI, and robustness evaluation.
• Experience with Open Modelica, Modelica, OpenFOAM, OpenMC, ANSYS, COMSOL, or other simulation tools.
• Experience with cloud platforms and GPU computing (e.g., HPC, AWS).
• Good software engineering practice for scientific code (version control, testing, documentation).
• Proactive and solution-oriented mindset, with strong multidisciplinary motivation to learn physics and computation and apply AI to the nuclear sector.
• Practical experience delivering outputs in international scientific or industrial projects.
• Strong organizational and documentation skills.
• Experience supervising students.
• Excellent spoken and written communication skills in English.

• Adequacy of the candidate's training and experience to the…