Page 378 - Thermal Hydraulics Aspects of Liquid Metal Cooled Nuclear Reactors

(U)RANS pool thermal 6.2.4

hydraulics

L. Koloszar, V. Moreau
von Karman Institute for Fluid Dynamics, Sint-Genesius-Rode, Belgium; Department of
Energy and Environment, Center for Advanced Studies, Research and Development in
Sardinia (CRS4), Pula, CA, Italy

Reactor pool thermal hydraulics is targeting to reproduce the flow and thermal field in
the primary heat-exchanging loop of a nuclear reactor. Numerical simulation of a reac-
tor pool is still a very challenging problem despite the ever increasing computational
power and available numerical techniques. The main difficulty in the application of
CFD codes to such problems are due to

l the complex multiphysical environment coexisting in such applications and the lack of gen-
eral knowledge of their modeling in an economic way,
l the difficulties arising while building the numerical model, including the definition of the
conditions to be simulated.
In the chapter title, the term pool is a short name for what should rather be “thermal
loop in pool configuration.” To give hints about what the argument is, we will first
work by differentiation. The first big difference is that about 99% of the CFD calcu-
lations involve a fluid coming from outside the computational domain through one or
more inlets. The fluid performs or withstands some action in the domain, such as pres-
sure loss, heat exchange, phase change, and chemical reaction. Then, the fluid leaves
the computational domain through one or more outlets.
In pool configuration, however, the fluid stays always inside the domain, performing
some action while circulating. Not much of a difference one would say. In practice, but
not in principle, errors, in general, and slowly emerging issues, in particular, end up
being trapped inside the CFD domain, not leaving it along with the flow through the
outlet. This causes some very unusual/peculiar CFD issues to arise, which will not
be encountered in regular CFD studies. Two relevant examples come readily in mind:
l The mass conservation must be strictly enforced; otherwise, the pool will slowly overflow or
otherwise slowly but eventually empty. Corrective actions must be taken to compensate the
possible accumulation of numerical errors rising due to the temperature-dependent density.
l In the presence of a free surface, a slight unphysical mixing of the phases across the surface
will not be evacuated in the outlet before being macroscopically noticeable. Instead, it will
slowly accumulate in time (or in iteration) eventually vanishing the relevance of the simu-
lation. Therefore, a strict control of the phase separation must be enforced.
The second big difference is that about 95% of CFD models are characterized by a
very restricted number of features (one or two at most): jet flows, backward-facing

Thermal Hydraulics Aspects of Liquid Metal Cooled Nuclear Reactors. https://doi.org/10.1016/B978-0-08-101980-1.00021-1
Copyright © 2019 Elsevier Ltd. All rights reserved.