Impact of the homogenization level, nodal or pin-by-pin, on the uncertainty quantification with core simulators. (April 2018)

• Record Type: Journal Article
• Authors: Castro, E.; Sánchez-Cervera, S.; García-Herranz, N.; Cuervo, D.
• Is Part Of: Progress in nuclear energy. Volume 104(2018)
• Page Start: 218
• Page End: 228
• Publication Date: 2018-04
• Subjects: Uncertainty quantification -- Nodal and pin-by-pin -- Core simulator -- Perturbation theory-based method -- Sampling-based method; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 333.7924
• DOI: 10.1016/j.pnucene.2017.10.001
• Languages: English
• ISSNs: 0149-1970

Abstract

Abstract: This paper assesses the impact of the homogenization level (nodal or pin-by-pin) on the uncertainties predicted by core simulators, highlighting the need of calculations at pin level for a reliable estimate of uncertainties in hot channel factors. In order to perform the analysis, two methodologies for nuclear data uncertainty quantification in stand-alone neutronics calculations have been implemented in the core simulator COBAYA. The first one is based on first-order perturbation theory and allows sensitivity/uncertainty analysis in the multiplication factor. The second one is based on random sampling and allows the uncertainty propagation in all responses computed by the code. Both methodologies are used in conjunction with SCALE system, which provides the capabilities to propagate the nuclear data uncertainties into the few-group constants required for the diffusion calculations. These methodologies are validated using a simplified pin-cell and a fuel assembly, and then applied to a full 3D core in the context of the OECD/NEA UAM benchmark (Uncertainty Analysis in Best-Estimate Modeling for Design, Operation and Safety Analysis of LWRs). Highlights: Uncertainty propagation methodologies are applied in the core simulator COBAYA. Good agreement between deterministic and sampling methods for keff UQ. Power peaking factors do not always follow a normal distribution. Pin-by-pin calculations must be performed to get accurate uncertainty estimates.