Fluoride-salt-cooled high-temperature test reactor thermal-hydraulic licensing and uncertainty propagation analysis

Abstract

An important Fluoride-salt-cooled High-temperature Reactor (FHR) development step is to design, build, and operate a test reactor. Through a literature review, liquid-salt coolant thermophysical properties have been recommended along with their uncertainties of 2-20%. This study tackles determining the effects of these high uncertainties by proposing a newly developed methodology to incorporate uncertainty propagation in a thermal-hydraulic safety analysis for test reactor licensing. A hot channel model, Monte Carlo statistical sampling uncertainty propagation, and limiting safety systems settings (LSSS) approach are uniquely combined to ensure sufficient margin to fuel and material thermal limits during steady-state operation and to incorporate margin for high uncertainty inputs. The method calculates LSSS parameters to define safe operation. The methodology has been applied to two test reactors currently considered, the Chinese TMSR-SF1 pebble bed design and MIT's Transportable FHR prismatic core design; two candidate coolants, flibe (LiF-BeF2) and nafzirf (NaF-ZrF4); and forced flow and natural circulation conditions to compare operating regions and LSSS power (maximum power not exceeding any thermal limits). The calculated operating region accounts for uncertainty (2 [sigma]) with LSSS power (MW) for forced flow of 25.37±0.72, 22.56±1.15, 21.28±1.48, and 11.32±1.35 for pebble flibe, pebble nafzirf, prismatic flibe, and prismatic nafzirf, respectively. The pebble bed has superior heat transfer with an operating region reduced ~10% less when switching coolants and ~50% smaller uncertainty than the prismatic. The maximum fuel temperature constrains the pebble bed while the maximum coolant temperature constrains the prismatic due to different dominant heat transfer modes. Sensitivity analysis revealed 1) thermal conductivity and thus conductive heat transfer dominates in the prismatic design while convection is superior in the pebble bed, and 2) the impact of thermophysical property uncertainties are ranked in the following order: thermal conductivity, heat capacity, density, and lastly viscosity. Broadly, the methodology developed incorporates uncertainty propagation that can be used to evaluate parametric uncertainties to satisfy guidelines for non-power reactor licensing applications, and method application shows the pebble bed is more attractive for thermal-hydraulic safety. Although the method was developed and evaluated for coolant property uncertainties for FHR, it is readily applicable for any parameters of interest.