Vous êtes sur la page 1sur 9

Transaction B: Mechanical Engineering

Vol. 16, No. 2, pp. 159{167

c Sharif University of Technology, April 2009

Simulation of Turbulent Flow Through

Porous Media Employing a v2f Model
R. Bahoosh Kazerooni1 and S. Kazemzadeh Hannani1;
Abstract. In this article, a v2f model is employed to conduct a series of computations of incompress-
ible ow in a periodic array of square cylinders simulating a porous media. A Galerkin/least-squares nite
element formulation employing equal order velocity-pressure elements is used to discretize the governing
equations. The Reynolds number is varied from 1000 to 84,000 and di erent values of porosities are
considered in the calculations. Results are compared to the available data in the literature. The v2f model
exhibits superior accuracy with respect to k " results and is closer to LES calculations. The macroscopic
pressure gradients for all porosities studied showed a good agreement with Forchheimer-extended Darcy's
law in the range of large Reynolds numbers.
Keywords: Porous media; Turbulent ow; Volume averaging; v2f model.

INTRODUCTION develop near the solid boundaries of the pores. The

inertial ow regime is the second type of ow, which
Porous media can be found in a wide variety of is initiated at Rep = 1 10, when the boundary
engineering applications and natural systems. For the layers become more pronounced and an \inertial core"
design of engineering systems, it is generally desir- appears. The developing of the \core" ows outside
able to calculate the pressure drop across the porous the boundary layers is the reason for the non-linear
medium and to predict the ow eld characteristics. relationship between the pressure drop and ow rate.
Most porous materials used in traditional engineering As the Rep increases, the \core" ows enlarge in size
applications have very small pores and small perme- and their in uence becomes more and more signi cant
ability, therefore, the uid speed is relatively small. to the overall ow picture. This steady non-linear
In these porous materials, the dominant ow regime laminar ow regime persists up to Rep  150. The
is the laminar ow. High speed uid ow through third ow regime detected by [1] is an unsteady laminar
porous media with considerable permeability can lead ow eld in the Reynolds number ranges of 150 to 300.
to turbulent ow within the pores. At Rep  250, the rst evidence of unsteady ow is
Dybbs and Edwards [1] have studied experimen- observed in the form of laminar wake oscillations in
tally the characteristics of turbulent ow through the pores. These oscillations take the form of traveling
porous medium using laser anemometry and ow waves characterized by distinct periods, amplitudes
visualization techniques. They reported that four and growth rates. In this ow regime, these oscillations
ow regimes may occur through porous media. Fol- exhibit preferred frequencies that seem to correspond
lowing [1], the rst type of ow is the Darcy or to growth rates. Vortices form at Rep  250 and persist
creeping ow regime, when the ow is dominated by up to Rep  300. Finally, based on ow visualization
viscous forces and which occurs at Rep < 1 (based on techniques, a highly unsteady and chaotic ow regime,
the average particle dimension and the average pore which occurs for Rep > 300 and which qualitatively
velocity). At Rep  O(1), boundary layers begin to resembles turbulent ow, is categorized as a fourth ow
1. Center of Excellence in Energy Conversion, Department of
regime by [1].
Mechanical Engineering, Sharif University of Technology, Modeling turbulent ow through porous media
Tehran, P.O. Box 11155-9567, Iran. is important in engineering applications, such as sim-
*. Corresponding autor. E-mail: hannani@sharif.edu ulating compact heat exchangers, composite seawalls
Received 5 March 2007; received in revised form 28 August 2007; and composite break-waters, to mention just a few [2-
accepted 23 September 2007 5]. More traditional applications can be found in the
160 R. Bahoosh Kazerooni and S. Kazemzadeh Hannani

design of uidized bed combustors and in enhancing details). During the last few years, the v2f (DNS based)
petroleum extraction from reservoirs. Other practical turbulence model, originally introduced by Durbin [15],
phenomena of natural origin and environmental im- has become increasingly popular, due to its ability to
portance are the contaminants transported by air ow correctly account for near-wall damping without the
through forests and corps. use of ad-hoc damping functions. Wall e ects through
Masuoka and Takatsu [6] derived and solved a porous media and the strong blockage of ow in the
zero equation turbulence model for ow through porous stagnation and impingement region of pores are very
media by use of the local volume-averaging technique. crucial, therefore, it seems that the v2f model is a good
They modeled the e ective eddy di usivity as the candidate.
algebraic sum of the eddy di usivities estimated from Summarizing the introduction, turbulent ow is
two types of vortices: The pseudo vortex and the ubiquitous, occurring in macroscopic scales of the
interstitial vortex. They studied turbulent ow and universe and galaxies, down to the interior of porous
heat transfer through stacked spheres. Alvarez et media. A turbulent simulation of porous media en-
al. [7] proposed and solved a one equation `macroscopic' counters two challenges. The rst problem consists
turbulence model. They derived a k equation, starting of presenting an accurate turbulence model mimicking
from a k l version of conventional RANS (Reynolds the complex ow features in the pores. The second
Averaged Navier-Stokes) models. challenge is to develop a suitable `macroscopic' tur-
Two-equation `macroscopic' turbulence models bulence model, substituting the complex topology of
are also proposed in the literature. Antohe and the pore structures, using the concept of space-time or
Lage [8] derived a two-equation macroscopic turbulence time-space averaging (so-called black box model). The
model by applying the time averaging operator to present work is focused on the rst challenge.
the generalized model for ow through porous media
(known as the Brinkman-Forchheimer-Extended Darcy GOVERNING EQUATIONS AND V2F
model). Getachew et al. [9] extended Antohe's work by MODEL
taking into account the Forchheimer term into higher
order terms. Following another approach, Kuwahara et The ow is assumed incompressible. The mean
al. [10] and Nakayama and Kuwahara [11] proposed a ow variables satisfy the following Reynolds Averaged
two-equation macroscopic turbulence model obtained Navier-Stokes equations (RANS):
by spatially averaging the Reynolds-averaged Navier-
Stokes equations and k " model transport equations. r:U = 0; (1)
For more information on `macroscopic' modeling, see
Pedras and Lemos [12] and the references cited by these DU
authors. = rp + r:[( + t )(rU + rUT )]: (2)
Kuwahara et al. [10], Nakayama and Kuwa-
hara [11] and Pedras and Lemos [12,13] conducted The v2f model is applied for evaluating t . This is a
numerical experiments for turbulent ows through a general model that is valid up to the wall, circumvent-
periodic array of cylinders, using the conventional ing the use of wall functions (see below).
`microscopic' two-equation turbulence model, based on The v2f model of Durbin [15-17] and Behnia et
RANS equations. Kuwahara et al. [14] performed the al. [18] consists of solving two additional equations,
Large-Eddy Simulation (LES) to study the turbulent with respect to the standard k and " model, the wall-
ow through porous media using a `microscopic' point normal stress, v2 , transport equation and an elliptic
of view and compared the results with a two equation relaxation function, f , equation. This model was de-
k " model. Direct Numerical Simulations (DNS) are veloped for improving the modeling of wall e ects and,
the best choice for validating the result of any turbu- more precisely, uses the contribution of wall normal
lence model through porous media, which, however, stress to obtain the eddy viscosity. The Boussinesq
will not be available for engineering applications in the approximation is used for the stress-strain relationship
near future, due to the large memory and CPU time as follows:
2 @Ui @Uj
In this paper, to study turbulent ow through ui uj ij k = t + : (3)
3 @xj @xi
porous media, we have employed a v2f turbulence
model using a `microscopic approach'. The v2f model The turbulent time scale, T , in Durbin's v2f model
is a higher order turbulence model than conventional is [18]:
two-equation models. This model circumvents the
" #
use of wall functions and introduces a more real- k
istic expression for eddy viscosity and predicts the T = min max ; CT ; p : (4)
impingement region more accurately (see below for " " 2 3C v2 S
Simulation of Turbulent Flow Through Porous Media 161

The strain-rate magnitude, S , in Equation 4 is de ned With compensated f~, Equation 10 is converted into:
2 1 P
Sij Sij : (5) f~ = L2 r2 f~ = C1 + (5 C1 )v2 =k + C2 k :
3 T k (14)
 is the kinematics viscosity, Sij is the strain rate According to this modi cation, the dissipation rate of
tensor, and the turbulent length scale, L, is [18]: Equation 9 is changed into:
" " # #
" #
k 3= 2 k 3 =2  3 =4 
L = CL max min ;p ; C 1=4 : (6) @v2
+ Uj
@v2 @
+ t 6
v2 ~
+ kf:
" 3C  v S2 " @t @xj @xj k @xj T (15)
C , CL , CT and C , which are model constants, will
be de ned in Equations 20. Using this modi cation, the upper limit of the realiz-
The Durbin's v2f model can be summarized by ability condition for stagnation ow is altered for the
the following four transport equations: v2f model (for more details see [19]).
The modi ed turbulent time scale, T , is written
@k @k @  @k as [19]:
+ Uj = + t + Pk "; (7)
@t @xj @xj k @xj "  r  #
k  0:6k
   T = min max ; CT ;p ; (16)
@" @" @  @" C P C " " 6C v2 S
+U = + t + "1 k "2 ;
@t j @xj @xj k @xj T (8) and the modi ed turbulent length, L, scale is [19]:
"  # " " # #
@v2 @v2 @  @v2 " k 3 =2 k 3 =2  3= 4
+ Uj = + t v2 + kf; L = CL max min ;p ; C :
@t @xj @xj k @xj k (9) " 6C  v 2 S

"1=4 (17)

2=3 v2 =k P Sij is the mean strain rate tensor:

f L2 r2 f = (C1 1) + C2 k : (10)
T k
Sij = (@Ui =@xj + @Uj =@xi )=2: (18)
Pk is the production of turbulent kinetic energy and is
de ned below (see Equation 19). Pk , in Equations 7, 8 and 10, is the production of the
The term kf in Equation 9 represents the re- turbulent kinetic energy and de ned as:
distribution of turbulence energy from the streamwise
component. This term is non-locally obtained by Pk = ui uj @Ui =@xj : (19)
solving an elliptic relaxation equation for f [18].
The eddy viscosity, vt , is given by: The coecients of the modi ed model are as fol-
lows [19]:
vt = C v2 T: (11) C = 0:22; CL = 0:23; C = 85;
The no slip boundary conditions, (y ! 0), are approx-
imated as: CT = 6:; C1 = 0:4; C2 = 0:3;
2k C"2 = 1:9; k = 1:; " = 1:3;
kw = 0; "w = lim 2 ; (12a)
y !0 y " #
" # k
v2 C"1 = 1:4 1 + 0:045 2 : (20)
vw2 = 0; fw = 20v2 lim : (12b) v
y!0 "w y 4
It is known that the v2f model has a poor numerical
Variable y denotes the coordinate normal to the wall. stability. Sveningsson and Davidson [20] and Davidson
Lien & Kalitzin [19] proposed a modi cation of et al. [21] proposed a simple modi cation for improving
the v2f model that allows a simple explicit boundary stability. The source term, kf~, in the v2 transport
condition at walls for elliptic relaxation. They intro- (Equation 15), includes the modeled pressure strain
duced a new variable for f . This new variable is called term, which is damped near walls as compensated f~
compensated f~ and is de ned by: goes to zero. Since v2 represents the wall-normal stress,
it should be the smallest normal stress, i.e. v2  u2
f~ = f 5
: (13) and v2  w2 and, thus, v2 should be smaller than
k2 2k=3 [20,21]. In the homogeneous region far away from
162 R. Bahoosh Kazerooni and S. Kazemzadeh Hannani

the wall, the Laplace term is assumed to be negligible

2 f~
i.e. @x@j @x j
! 0, then Equation 14 reduces to:
2 1 P
f~ = C1 + (5 C1 )v2 =k + C2 k : (21)
3 T k
On the region far away from the wall, the Laplace term
is not negligible and, as a consequence, v2 gets too
large, so that v2  2k=3.
A modi cation is to use the right-hand side of
Equation 14 as an upper bound on the source term,
kf~, in the v2 -equation, i.e.:
The Source Term of the v2 transport equation = Figure 2. Structural unit chosen for the computation
~ 2 C1 k + (5 C1 )v2 1 + C2 Pk : (22)
min kf;
3 T are selected as:  = 0:3, 0.4, 0.5, 0.64, 0.75, 0.84 and
This modi cation ensures that v2  2k=3 (for more The boundary conditions related to periodicity
details see [20,21]). are as follows:
PROBLEM DEFINITION U(x = 0:; y) = U(x = 2H; y); (23)
The problem considered is shown schematically in U(x; y = 0) = U(x; y = H ): (24)
Figure 1. In this gure a periodic array of square
cylinders is shown. Based on the geometrical pe- The pressure gradient, p, in a periodically fully devel-
riodicity, only one structural unit is chosen for the oped ow, can be expressed by [22]:
computational domain. The selected area is shown @p @P
in Figure 2. Therefore, the ow was solved only (x; y) = + (x; y); (25)
through the domain that is shown in Figure 2. By this @x @x
technique, eddies larger than the scale of the porous where is a constant and represents the overall pres-
structure are intentionally neglected, since such large sure drop over the gradient imposed on the ow [22].
eddies cannot be detected through a simulated porous The function, P (x; y), behaves in a periodic manner,
medium. so that:
The Reynolds number, ReH = uD H= , is based
on the center-to-center distance, H . The porosity of P (x; y) = P (x + 2H; y + H ): (26)
this domain is calculated from  = 1 (D=H )2 . The The ow rate through the periodic cell (Figure 2) was
porosities used in the calculation of the present study set equal to unity by normalizing the variables and
tuning the value of .
The boundary conditions for turbulent kinetic
energy and its dissipation are as follows:
k(x = 0:; y) = k(x = 2H; y);
k(x; y = 0:) = k(x; y = H ); (27)
"(x = 0:; y) = "(x = 2H; y);
k(x; y = 0:) = k(x; y = H ): (28)
The boundary conditions for Reynolds stress v2 and
elliptic relaxation can be written as:
v2 (x = 0:; y) = v2 (x = 2H; y);
Figure 1. Periodic array of square cylinders. v2 (x; y = 0:) = v2 (x; y = H ); (29)
Simulation of Turbulent Flow Through Porous Media 163

f~(x = 0:; y) = f~(x = 2H; y); 17928 elements, therefore, mesh systems of the order
of 12000 seem to be suciently accurate and all
f~(x; y = 0:) = f~(x; y = H ): (30) calculations presented below are obtained using this
mesh system.
Boundary conditions for no slip boundaries were men- Regarding numerical convergence, the normalized
tioned previously by Equations 12a and 12b. residuals for all variables were brought down to 10 5 .
For Reynolds numbers lower than 5  104 , the under
RESULTS relaxation coecient of U and p is 0:8, k and " is 0.6,
v2 and f~ is 0.5 and, for Reynolds numbers equal to or
The governing equations are numerically solved inside greater than 5  104 , the under relaxation coecient of
the domain and the results are shown in Figures 3 to 13. U and p is 0:7, k and " is 0.4, v2 and f~ is 0:3.
The governing equations are discretized using a In Figure 4, streamlines are shown at ReD =
GLS (Galerkin/Least-Squares) nite element method, 40000 and for porosity = 0.84. Pressure contour
employing equal order interpolation for the velocity solutions at ReD = 40000 and porosity = 0.84 are
and pressure in conjunction with a pressure stabilizing shown in Figure 5. In Figure 6, turbulent kinetic energy
method (see for more details [23,24]). contours are depicted at ReD = 40000 for porosity
The Reynolds number, ReH = uD H= , is varied = 0.84, employing the v2f model. The turbulent
from 1000 to 500000. uD = hU if is a Darcian kinetic energy results are converted into macroscopic
velocity and H is set equal to unity. In this work, turbulent kinetic energy by applying the following
the value of D is varied as 0.2236, 0.4, 0.5, 0.6, 0.7072, volume average operator:
0.7746 and 0.83667. The respected porosities are 0.95,
0.84, 0.75, 0.64, 0.5, 0.4 and 0.3. The Reynolds
number based on H is converted to a Reynolds number
based on D (ReD = ReH  D) for comparing the
present computations with the recent results of the
literature [14]. D is the particle dimension and, in this
study, ReD is varied between 1000 to 84000.
For checking mesh-independency, the ow
through the porosity of 0.84 has been solved at
ReH = 1000, 5000, 10000, 50000 and 100000 using
7508, 11948 and 17928 elements. The microscopic
turbulent kinetic energy normalized by the square of
Darcian velocity, k=u2D , of these solutions are shown
at x=H = 1. The Reynolds number of Figure 3 is
ReD = 40000. The mesh stretching near the wall
in these three solutions is similar. There is a good
agreement between the solution data of 11948 and Figure 4. Streamlines for porosity = 0.84 and
ReD = 40000.

Figure 3. Normalized microscopic turbulent kinetic Figure 5. Pressure contours for porosity = 0.84 and
energy at ReD = 40000 and x=H = 1. ReD = 40000.
164 R. Bahoosh Kazerooni and S. Kazemzadeh Hannani

Figure 6. Normalized turbulent kinetic energy contours

for porosity = 0.84 and ReD = 40000. Figure 8. Normalized turbulent kinetic energy for
Z porosity = 0.64.
hkif = V1 kdV; (31)

Vf = V: (32)
The macroscopic turbulent kinetic data are normalized
by the square of Darcian velocity, hkif =u2D . These
data are shown versus Reynolds number based on D, in
Figures 7 to 10. These curves are sketched for various
porosities of 0.84, 0.64, 0.5 and 0.3. The results of
present calculations employing the v2f model have been
compared with the results of [14] obtained using LES.
The normalized macroscopic turbulent kinetic energy
remained almost constant at large Reynolds numbers.
It seems that the v2f model and LES both predict
a Reynolds number independent normalized solution
of k for large ReD number ows. For the range of
ReD < 1000, no logical correlation can be concluded
between LES and v2f results. It is well known that Figure 9. Normalized turbulent kinetic energy for
turbulence model constants exhibit appreciable non- porosity = 0.5.

Figure 7. Normalized turbulent kinetic energy for Figure 10. Normalized turbulent kinetic energy for
porosity = 0.84. porosity = 0.3.
Simulation of Turbulent Flow Through Porous Media 165

universalities in the `very low Reynolds number' limits

(ReD  1000) (see also [24,25]). We conjecture that for
`very low Reynolds number' turbulent ows, neither the
present v2f model nor LES calculations of [14], which
use an RNG k " near wall model, are suitable for
prediction with acceptable accuracy.
Macroscopic turbulent kinetic energies of all
porosities in this study are depicted in Figure 11.
The results of the present study are compared with
the result of the LES solution of [14]. The slope
of the present study data is smaller than the slope
of LES and low Reynolds k " model data. At
lower porosities, the normalized macroscopic turbulent
kinetic energy of LES [14] is larger than the v2f
predictions. The di erence can be attributed to the
use of a renormalized group (RNG) subgrid scale model Figure 12. Normalized Dissipation rate of turbulent
in LES [14], leading to overestimation of turbulent kinetic energy versus (1 )=1=2 .
kinetic energy near stagnation or impingement regions.
Figure 12 presents the normalized dissipation rate of
turbulent kinetic energy for all porosities used in the simpli ed into Equation 34.
present study. The present data are compared with the  
data of Nakayama and Kuwahara [11]. The dissipation dhpif D 1 
2 = 1:75 3 ; (ReD > 3000):
rate of the turbulent kinetic energy of the present study dx uD  (34)
is in good agreement with the results of [11].
The Forchheimer equation is known as a popular The dimensionless pressure gradient, dhpif =dx
equation for estimating the pressure drop through 2
[D=uD ], is depicted in Figure 13. The line in this
porous media. Ergun's empirical equation, accounting gure indicates the relation given by Equation 34.
for the Forchheimer drag in packed beds of particle The pressure results of the present study are in good
diameter D, is given by [14]: agreement with Forchheimer-extended Darcy's law.
dhpif D 150(1 )2  1  CONCLUSION
2 = +1:75 3 :
dx uD  uD D  (33)
A series of computations of incompressible ow in a
periodic array of square cylinders simulating a porous
Following Kuwahara et al. [14], the rst term on the media for various Reynolds numbers and di erent
right hand side of Equation 33 could be neglected porosities, employing the v2f turbulence model, are
at high Reynolds number, therefore, Equation 33 is carried out. The macroscopic turbulent kinetic energies

Figure 11. Normalized turbulent kinetic energy versus Figure 13. Dimensionless macroscopic pressure gradient
(1 )=1=2 . versus (1 )=3 .
166 R. Bahoosh Kazerooni and S. Kazemzadeh Hannani

of this study are compared to the results available in U time averaged velocity vector
the literature. Good agreement for Reynolds number u time uctuation velocity vector
is observed between LES and the present calcula- uD Darcian velocity
tions. However, for a low Reynolds number ow of
the order 1000, discrepancies between the results are V total volume of porous media
noted. Further study is needed to elucidate the main Vf volume of uid in porous media
shortcoming of simulations, both in the context of v2 normal to the wall component of
LES and turbulence modeling. The dissipation rate Reynolds stress
of the turbulent kinetic of the present study showed
good agreement with the available results in the liter- xi coordinate components
ature. The macroscopic pressure gradient results for
all porosities are in concordance with the Forchheimer Greek Letters
law. Our next step is to derive a `macroscopic' v2f overall pressure drop gradient
model for porous media. The v2f model could predict " dissipation rate of turbulent kinetic
a better correlation for turbulent kinetic energy that energy
is important for driving a macroscopic model for ow ' porosity
through porous media. It can be argued that in the
macroscopic models, the inherent advantages of using  kinematics viscosity
a v2f model are not obvious and should be proved t turbulent viscosity
mathematically. However, various applications can  density
be stated where porous and non-porous media cal-
culations must be performed simultaneously. Indeed,
the implementation of a single more accurate model REFERENCES
would be very useful in these cases. The derivation, 1. Dybbs and Edwards, R.V. \A new look at porous
computation and validation of the macroscopic model media uid mechanism - Darcy to turbulent", in Fun-
will be conducted subsequently. damentals of Transport Phenomena in Porous Media
(Book), J. Bear and V. Corapcioglu, M. Nijho , Eds.,
NOMENCLATURE the Netherlands, pp. 199-254 (1984).
2. Missirlis, D., Yakinthos, K., Palikaras, A., Katheder,
C ; CL ; C ; C2 ; C"1 ; K. and Goulas, A. \Experimental and numerical in-
C"2 ; CT ; k ; " turbulence model constants vestigation of the ow eld through a heat exchanger
D particle diameter for aero-engine applications", Int. J. of Heat and Fluid
Flow, 26, pp. 440-458 (2005).
f elliptic relaxation function
3. Yang, Y.T. and Hwang, C.Z. \Calculation of turbulent
f~ compensated elliptic relaxation ow and heat transfer in a porous-baed channel", Int.
function J. of Heat and Mass Transfer, 46, pp. 771-780 (2003).
H distance of two particles from each
other 4. Garcia, N., Lara, J.L. and Losada, I.J. \2-D numerical
analysis of near- eld ow at low-crested permeable
k turbulent kinetic energy breakwaters", Coastal Engineering, 51, pp. 991-1020
hki f
intrinsic volume average of turbulent (2004).
kinetic energy 5. Karim, M.F. and Tingsanchalil, T. \A coupled nu-
L turbulent length scale merical model for simulation of wave breaking and
p pressure hydraulic performances of a composite seawall", Ocean
Engineering, 33, pp. 1-15 (2005).
P periodic component of pressure
variable 6. Masuoka, T. and Takatsu, Y. \Turbulence model
Pk production of the turbulent kinetic for ow through porous media", Int. J. Heat Mass
energy Transfer, 39, pp. 2803-2809 (1996).
ReH Reynolds number based on H 7. Alvarez, G., Bournet, P.E. and Flick, D. \Two di-
ReD Reynolds number based on D mensional simulation of turbulent ow and transfer
through stacked spheres", Int. J. Heat Mass Transfer,
Rep Reynolds number based on particle 46, pp. 2459-2469 (2003).
Sij strain rate tensor 8. Antohe, B.V. and Lage, J.L. \A general two-equation
macroscopic turbulence model for incompressible ow
t time in porous media", Int. J. Heat Mass Transfer, 40, pp.
T turbulent time scale 3013-3024 (1997).
Simulation of Turbulent Flow Through Porous Media 167

9. Getachew, D., Minkowycz, W.J. and Lage, J.L. \A impinging on a at plate", Int. J. Heat Mass Transfer,
modi ed form of the k " model for turbulent ows of 41, pp. 1845-1855 (1998).
an incompressible uid in porous media", Int. J. Heat
Mass Transfer, 43, pp. 2909-2915 (2000). 19. Lien, F.S. and Kalitzin, G. \Computations of transonic
ow with the v2 f turbulence model", Int. J. of Heat
10. Kuwahara, F., Kameyama, Y., Yamashita, S. and and Fluid Flow, 22, pp. 53-61 (2001).
Nakayama, A. \Numerical modeling of turbulent ow
in porous media using a spatially periodic array", J. 20. Sveningsson, A. and Davidson, L. \Assessment of
Porous Media, 1, pp. 47-55 (1998). realizability constraints in v2 f turbulence models",
11. Nakayama, A. and Kuwahara, F. \A macroscopic Int. J. of Heat and Fluid Flow, 25, pp. 785-794 (2004).
turbulence model for ow in a porous medium", ASME 21. Davidson, L., Nielsen, P.V. and Sveningsson, A. \Mod-
J. Fluids Eng., 121, pp. 427-433 (1999). i cation of the v2 f model for computing the ow in
12. Pedras, H.J. and Lemos, J.S. \Macroscopic turbu- a 3D wall jet", in 4th Int. Symp. on Turbulence Heat
lence modeling for incompressible ow through unde- and Mass Transfer, Antalya, Turkey (2003).
formable porous media", Int. J. Heat Mass Transfer,
44, pp. 1081-1093 (2001). 22. Kelkar, K.M. and Patankar, S.V. \Numerical predic-
tion of ow and heat transfer in a parallel plate channel
13. Pedras, H.J. and Lemos, J.S. \On the mathematical with staggered ns", J. of Heat Transfer, 109, pp. 25-
description and simulation of turbulent ow in a 29 (1987).
porous medium formed by elliptic rods", ASME J.
Fluids Eng., 123, pp. 941-947 (2001). 23. Hannani, S.K., Stanislas, M. and Dupont, P. \Incom-
14. Kuwahara, F., Yamane, T. and Nakayama, A. \Large pressible Navier-Stokes computations with SUPG and
eddy simulation of turbulent ow in porous media", GLS formulations - a comparison study", Computer
Int. Com. in Heat and Mass Transfer, 33, pp. 411-418 Methods in Applied Mechanics and Engineering, 124,
(2006). pp. 153-170 (1995).
15. Durbin, P.A. \Near-wall turbulence closure modeling 24. Hannani, S.K. and Stanislas, M. \Incompressible tur-
without damping functions", Theoretical and Compu- bulent ow simulation using a Galerkin/least-squares
tational Fluid Dynamics, Chap. 3, pp. 1-13 (1991). formulation and a low Reynolds k " model", Comput.
16. Durbin, P.A. \On the k " stagnation point anomaly", Methods Appl. Mech. Engineering, 181, pp. 107-116
Int. J. of Heat and Fluid Flow, 17, pp. 89-90 (1995). (2000).
17. Durbin, P. \Separated ow computations with the k 25. Hannani, S.K. and Stanislas, M. \Finite element simu-
" v2 model", AIAA Journal, 33 pp. 659-664 (1995). lation of turbulent Coutte-Poiseuille ows using a low
18. Behnia, M., Parneix, S. and Durbin, P.A. \Prediction Reynolds k " model", Int. J. Num. Methods in Fluids,
of heat transfer in an axisymmetric turbulent jet 30, pp. 83-103 (1999).