BWR core thermal–hydraulic uncertainty and sensitivity analysis with improved bypass modeling features

Abstract

Boiling Water Reactors (BWR) designs have several special features to be considered while analyzing their thermal–hydraulic performance. One of those are the core regions where the coolant is restricted by physical boundaries of coming in contact with fuel rods. These bypass regions can be classified into core bypass (in the core periphery), bundle bypass (between the assemblies), and internal assembly bypass (water rods/channels). Subchannel thermal–hydraulic analysis usually simplifies the modeling of these regions by not accounting for heat transfer to the bypass coolant flow, aiming to be conservative in predicting safety margins to acceptance criteria.

Since the nuclear industry is embracing economically efficient Best Estimate (BE) simulation methodologies in place of the conservative methodologies, there is a heightened emphasis on the advancements in modeling the BWR core bypass regions in subchannel thermal–hydraulic analyses. Ziyad et al. (2022) have improved the advanced sub-channel code CTF by developing and implementing models for bypass related phenomena in BWRs. For the application of the code in BE analysis, rigorous uncertainty quantification becomes necessary. This involves propagating inherent uncertainties in model inputs for newly developed bypass modeling features in addition to the traditional model inputs. This propagation is important for accurately quantifying uncertainties in System Response Quantities (SRQs) which informs the safety margins and hence has economic incentive.

In this research, uncertainties in the input parameters are propagated in steady-state simulations through an assembly-resolved full core model and a subchannel-resolved single fuel assembly model of the Peach Bottom Unit 1 at End of Cycle 2. The statistical analysis tool Dakota is used as a driver for CTF, and it is employed for conducting the uncertainty propagation. Random Monte Carlo sampling techniques are utilized for input preparation, while the Spearman correlation metric is employed for sensitivity analysis.

The sensitivity analysis of the full core model indicates that bypass flow fraction is a strong function of the void fraction in the active core region. This phenomenon is only possible to be captured by employing the bypass modeling which employs pressure equalization across all subchannels. The void fraction prediction is also affected by other bypass modeling features as established by Ziyad et. al. (2022), hence each of the developed features finds its importance in the analysis. It also has been found that isolated single assembly modeling is inadequate to predict thermal–hydraulic conditions in bypass regions as lateral flow between the assembly gaps cannot be captured.