Doktorand i modellering av kärnbränsleprestanda med maskininlärning
Core
PhD research in developing machine learning-based surrogate models for calibrating and simulating nuclear fuel performance to enable faster, more precise predictions with quantified uncertainties.
Role type
PhD researcher (IC)
Builds
Production ML models and computational workflows for nuclear fuel performance simulation
Domain
Nuclear energy / Computational physics / Machine Learning
Deliverable
production ML models
Required skills
Machine learning, statistical calibration, uncertainty quantification, Python programming, numerical methods, physics knowledge
Preferred skills
Bayesian inference, MCMC methods, Gaussian processes, time-series modeling, surrogate modeling, high-performance computing
Technologies
Python, Julia, C++
Responsibilities
Develop and evaluate time-dependent ML models for sequence-to-sequence prediction of fuel behavior; propagate uncertainties between coupled submodels; generate and analyze training/validation data for diverse fuel designs and operating conditions; demonstrate model utility in industrial applications like cladding tangential stress prediction; implement and document computational tools for reproducible workflows.
Seniority
PhD candidate, hands-on research