Coverage assessment for core design characteristics in a LFR
Description
The accurate prediction of reactor physics observables is essential for the design, licensing, and safe operation of advanced nuclear reactors. For Lead-cooled Fast Reactors (LFRs), confidence in simulation results relies on a sound validation basis supported by experimental evidence, benchmark calculations, and an understanding of the uncertainties associated with numerical predictions.
This thesis will investigate a limited number of reactor physics observables that are particularly relevant for core design. The exact set of observables will be defined at the start of the project in consultation with the supervisors, but representative examples include core power distributions, burnup reactivity swing, and control rod worth.
For each selected observable, the student will assess the current state of validation by reviewing available experimental data, benchmark exercises, and published studies. In parallel, reactor physics simulations will be performed using the Serpent 2 Monte Carlo code to reproduce the selected quantities for representative LFR models.
A key aspect of the work will be the assessment of the role of nuclear data in the prediction of these observables. Using available sensitivity and uncertainty analysis methodologies, the student will identify the design options and model inputs contributing most significantly to each observable and quantify the resulting uncertainties. The work will provide insight into the confidence that can be placed in current reactor physics predictions and identify areas where additional validation or improved nuclear data would have the greatest impact.
The outcome of the thesis will be a structured assessment of the validation status of the selected reactor physics observables, highlighting their supporting evidence, associated uncertainties, and remaining knowledge gaps.
Activities
Familiarization
- Become familiar with Lead-cooled Fast Reactor technology and reactor physics fundamentals.
- Learn the basics of Monte Carlo reactor physics calculations using Serpent 2.
- Become acquainted with the reference LFR core model and the computational workflow used within the research group.
Literature Review
- Review reactor physics observables relevant to LFR core design.
- Review experimental programmes, benchmark exercises, and published validation studies related to the selected observables.
- Review the principles of nuclear data sensitivity and uncertainty analysis and their application to reactor physics.
Reactor Physics Simulations and Sensitivity Analysis
- Perform Serpent calculations for the selected observables using representative LFR core models.
- Analyse the sensitivity of the observables to nuclear data using available sensitivity analysis techniques.
- Identify the nuclear data (e.g. reaction cross sections, fission spectra, prompt neutron multiplicity) with the largest impact on the calculated observables.
- Investigate the dependence of the results on relevant modelling assumptions where appropriate.
Uncertainty Quantification and Validation Assessment
- Quantify the nuclear-data-induced uncertainties associated with the selected observables.
- Compare calculated results and uncertainty estimates with available benchmark and experimental evidence.
- Assess the level of confidence in the prediction of each observable.
- Identify knowledge gaps and prioritize areas where improved validation data or nuclear data would most effectively reduce prediction uncertainties.
Analysis and Thesis Writing
- Consolidate and interpret the simulation, sensitivity, and uncertainty results.
- Discuss the implications for LFR core design and validation.
- Prepare the MSc thesis and present the results during the final defence.
Expected Outcomes
By the end of the project, the student will have:
- developed proficiency in Monte Carlo reactor physics calculations with Serpent 2;
- gained experience in sensitivity and uncertainty analysis;
- identified the nuclear data that most strongly influence selected LFR core design observables;
- quantified the associated prediction uncertainties;
- assessed the current validation basis of the selected observables and identified priorities for future validation efforts.