Page 310 - Thermal Hydraulics Aspects of Liquid Metal Cooled Nuclear Reactors

280 Thermal Hydraulics Aspects of Liquid Metal Cooled Nuclear Reactors

Fig. 6.2.1.4 (Left) k-ε-Pr t and (right) TMBF compared with TEFLU forced convection
4
experimental results, Re jet ¼10 , d¼orifice diameter, and X¼distance to orifice
(Carteciano and Gr€ otzbach, 2003).

TMBF and its respective improvement needs to be performed. In this regard, within
the framework of the SESAME project, ASCOMP has implemented this model in
their code TransAT. As a next step, this model is foreseen to be extensively validated
against the variety of reference databases that will be generated within the SESAME
project.
It must be noted that due to the mathematical difficulties, that is, in the form of a
large number of equations and the respective model coefficients, the second-moment
closure is less attractive for complex flow configurations. In addition, due to large
number of differential equations, it requires high computational resources. Apart from
that, the resulting large number of model coefficients would require tuning, which is a
cumbersome task, to make the model work for most of the applications.

6.2.1.2.2 Algebraic heat flux models

A full differential second-moment closure model could represent a good platform for
deriving simpler AHFM. The resulting AHFM form a class of heat flux models, which
is simple compared with the second-order closure and yet sophisticated enough to
provide a good physical meaning. A number of AHFMs have been developed
with an explicit or implicit formulation, for example, by Abe and Suga (2001),
Hanjalic (2002), Launder et al. (1975), and Kenjeres et al. (2005).

6.2.1.2.2.1 Explicit AHFM
In an explicit formulation, the turbulent heat flux is calculated by a gradient hypoth-
esis with the eddy diffusivity concept. This is defined by using appropriate turbulence
timescales. This leads to four-parameter turbulence models, like presented by Hwang
and Lin (1999), Nagano and Shimada (1996), Abe et al. (1995), and Nagano et al.
(1999). Because of the isotropic nature of the model, these four-parameter models