Modeling VVER-1200 Reactor Pressure Vessel by using Computational Fluid

Dynamics Tools
Huseyin Ayhan, Sule Ergun
Hacettepe University, Department of Nuclear Engineering, Beytepe, Ankara 06800, Turkey

Abstract
In this study, the Reactor Pressure Vessel (RPV) of the VVER-1200 (AES-2006) was modeled and thermal-hydraulic
behavior of coolant in the RPV was investigated. The model set for the study includes 3 dimensional geometry
description of the RPV. While the geometry is defined, the console elements (which help keeping RPV in its original
position) are ignored since they are small and their effects on thermal or hydraulic behavior on the flow is negligible.
The outer sections of the RPV were modeled in detail. For the modeling of inner sections, many of the core supporting
structures locating at the lower plenum are modeled almost as exactly as they are. For the parts and components
that are not modeled in detail and are simplified, porous media approach was adopted in order to perform realistic
simulations. For the modeling, a detailed description of the VVER-1000 (V-392) RPV was used as the base model.
The design difference in fuel assembly and RPV height for the VVER-1200 RPV is taken into consideration. As
the result of using the described 3 dimensional modeling for the simulations of the thermal-hydraulic behavior of the
VVER-1200 reactor core, the temperature distribution and pressure drop are calculated. The calculated values are in
good agreement with the values reported in the literature.
Keywords: VVER-1000, VVER-1200, CFD, Reactor Pressure Vessel, Porous Media Approach

1. Introduction
It is an essential aim of CFD investigations to simulate 3D flow phenomena in highly complex technical systems.
In case of nuclear reactor pressure vessels (RPV) the scales of interest are spread over several orders of magnitude.
The height of a typical RPV is longer than 10 m and more with an inner diameter of several meters. The smallest
geometrical details, which have significant influence on mixing phenomena, are in the range of 1 mm. To resolve this
technical system in all details would require a grid far beyond present computational capabilities.
Because of this large scale of geometrical parameters, once the system is modeled by using Finite Volume Method
(FVM), the mesh structure must be set such that even the smallest structure must be involved in the model established. Mesh and grid structure established as described requires defining approximately 108 -109 grids only for the
core region. Having this many grids for the calculations make not only high performance computers but also long
computation time necessary.
Therefore, instead of modeling the system by taking every single detail into account, the system is modeled as a
flow volume and the details of the design are defined by using proper parameters and coefficients and by defining these
parameters and coefficients to the analyzer program. This helps user to have shorter computing time while affordable
computers are used. In this study, this is achieved by applying porous media approach.
For the model described in this study, 107 grids were defined. Porosity which is defined as the ratio of fluid volume
to total (fluid and solid) volume, total solid surface, hydraulic diameter and loss coefficients which lead to pressure
drop are used as modeling parameters as required by the selected solver. The coefficients required to apply porous
media approach are calculated by taking the complicated parts and sections of the system into account.
Corresponding author. Tel.: +90 312 297 73 00; fax: +90 312 299 21 22.
E-mail address: huseyinayhan@hacettepe.edu.tr (H. Ayhan).

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
Phase 2 of the OECD VVER-1000 coolant transient benchmark was defined (Kolev et al., 2004). This reference
problem is adopted in this study, which is a pure thermal-hydraulic problem with given boundary conditions and
geometrical dimensions with some changes (VVER-1200 reactor core is a little bit larger than VVER-1000 reactor
core).
2. Porous Medium Approach for Complex Reactor Core
In the porous media procedure, the region of interest is divided into a network of volumes or lumped regions, each
of which will be characterized by the volume-averaged parameters. This approach has been widely used for analysis
of large regions containing only fluid (continuum) and for flows in regions with immersed solids.
Typical geometries of interest are a fuel rod array in a nuclear reactor core or the shell side of a steam generator
in which an array of tubes carrying the primary coolant is immersed. Normally these arrays are periodic with an
anisotropy characterized by the nominal array geometry. In specific circumstances rod distortions or foreign material
blockages can exist. The porous media approach is a practical method of analyzing both the reactor core and the steam
generator situations.
In the core section, when it is compared with the RPV, there are a lot of small sections and structures. While the
RPV of VVERs has the height of about 12.5 m, the fuel rod has the diameter of 9,1 mm and the spacer grids have
the thickness of 1-2 mm. These small components are often found in the core section and the orientation of them is
regular and periodic. When the fuel rod and fuel rod+spacer portions are examined separately, there is almost
no anisotropy in these systems. Since, in the core region the reactor geometric complexity requires the adoption of
a smaller scale segmentation (finer mesh), high computational effort is required for CFD calculations. The porous
media approach is applicable for this problem and it reduces computational time and effort.
3. Computational Model
3.1. Conservation equations
Turbulent flows are characterized by fluctuating velocity fields. Since these fluctuations can be of small scale
and high frequency, they are too computationally expensive to simulate directly in practical engineering calculations.
Instead, the instantaneous (exact) governing equations can be time-averaged, ensemble-averaged, or otherwise manipulated to remove the resolution of small scales, resulting in a modified set of equations that are computationally less
expensive to solve. However, the modified equations contain additional unknown variables, and turbulence models
are needed to determine these variables in terms of known quantities.
The equation for conservation of mass, or continuity equation, can be written as follows (Inc. ANSYS, 2009):

+ (~u) = 0
(1)
t
where and u represent fluid density and velocity, respectively.
Conservation of momentum in an inertial (non-accelerating) reference frame is described by (Inc. ANSYS, 2009)
"
!#

2
T
(~u) + (~u~u) = p + (~u + ~u ) ~uI + ~g + F~
(2)
t
3
where p is the static pressure, ~g is the gravitational body force, is the molecular viscosity, I is the unit tensor and
F~ is external body force, respectively. F~ also contains other model-dependent source terms such as porous-media and
user-defined sources.
The energy equation can be written in the following form (Inc. ANSYS, 2009):

(E) + (~u(E + p)) = eff T

h j J~j + (eff ~u) + S h
(3)
t
j
where J~j is the diffusion flux of species j. The first three terms on the right-hand side of Eq. 3 represent energy transfer
due to conduction, species diffusion, and viscous dissipation, respectively. S h includes user defined heat source. In
Eq. 3, eff is the effective conductivity and for the realizable k- model, the effective thermal conductivity is given by
2

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
c p t
Prt
where and Prt represent the thermal conductivity and the turbulent Prandtl number, respectively. t represents the
turbulent viscosity (eddy viscosity) and for the realizable k- model, it is computed from
eff = +

k2
(4)

where C is a function of the mean strain and rotation rates, the angular velocity of the system rotation, and the
turbulence fields ( k and ). In Eq. 4, k and  represent turbulent kinetic energy and turbulent dissipation rate and the
modeled transport equations for k and  in the realizable k- model are (Inc. ANSYS, 2009):
"
!
#

t k

(k) +
(ku j ) =
+
+ Gk + Gb  Y M + S k
(5)
t
x j
x j
k x j
"
!
#

t 
2

() +
(6)
(u j ) =
+
+ C1 S  C2
+ C1 C3 Gb + S 
t
x j
x j
 x j
k
k + 
t = C

where

,
C1 = max 0.43,
+5
"

k
=S ,


S =

q
2S i j S i j

In these equations, Gk represents the generation of turbulence kinetic energy due to the mean velocity gradients,
Gb is the generation of turbulence kinetic energy due to buoyancy, Y M represents the contribution of the fluctuating
dilatation in compressible turbulence to the overall dissipation rate. C2 and C1 are constants, k and  are the
turbulent Prandtl numbers for k and , respectively. S k and S  are user-defined source terms. The model constants are
C1 = 1.44, C2 = 1.9, k = 1.0,  = 1.2, C3 = tanh |u |
where u is the ratio of the component of the flow velocity parallel to the gravitational vector over the component of
the flow velocity perpendicular to the gravitational vector.
3.2. Porous media approach
The presence of solid objects in the flow domain has two significant effects: the geometric effect of displacing
fluid, and the physical effect of altering the momentum and the energy exchanges within the flow domain. In the
porous media approach solid material equivalent to the real solid material is uniformly dispersed in the flow domain.
By equivalent, it is traditionally meant that the same volume of material yields the same overall shear and heat transfer
effects between the fluid and solid surface. In this way the characteristics of porosity, flow resistance, and heat source
(sink) of each cell replicate the volume-averaged characteristics of their real domain counterparts. In porous media
model, volume porosity is defined as the ratio of fluid volume to total volume. Thus:
V =

Vf
Vs + V f

(7)

where V f and V s represent fluid volume and solid volume, respectively.

Porous media are modeled by the addition of a momentum source term to the standard fluid flow equations. The
source term is composed of two parts: a viscous loss term and an inertial loss term. If a perforated plate or tube
bank are modeled, the viscous loss term can be eliminated and the inertial loss term can be used alone. Remembering
~ a key concept associated with the porous media approach, is the
the term, F~ in Eq. 2, the distributed resistance R,
resistance force per unit volume of fluid exerted on the fluid by the dispersed solid. It is defined by the following
relation (Todreas and Kazimi, 2001):
Z
~ = 1
hRi
(p~n + ~n) dA
(8)
V f Af s
where A f s represents interface area between fluid and solid regions. In Eq. 8, hi represents volume average operator.
3

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
R
R
The term A p~n dA is the form drag, and the term A ~n dA is the friction drag. For pure axial flow (flow along
fs
fs
the rod axis) without any spacers or baffles , there is no form drag. In transverse flows , due to the presence of boundary
layer separation phenomena, both form and friction drag forces are important. For flow at an arbitrary direction with
respect to the rod axis, separation effects, which are responsible for the transverse form drag, will also affect the axial
friction drag. However, due to lack of general constitutive relations accounting for simultaneous action of form drag
and friction phenomena for flow at an arbitrary direction , simple combinations of axial and transverse flow resistance
~ is composed of the transverse components: hR
~ x i and
correlations are typically used. Hence the resistance vector hRi
~ y i which are equal and the axial component, hR
~ z i.
hR
For axial flow, applying a force balance, we obtain:
Z
1~
p
~
k
(9)
hRz i =
(p~n + ~n) dA =
Vf
z
Af s
Here pressure drop term includes friction and form loses. By using the definition of pressure drop, we obtain:
p = ( f

1
1
z
+ K) hihvz i2 = Cz hihvz i2
De
2
2

(10)

where K represents form loss coefficient, De represents hydraulic diameter of selected unit cell and calculated by,
De =

4A f
Pw

where Pw represents wetted perimeter in unit cell. In Eq.10, f is the friction factor and is available in terms of circular
tube friction factor fct , i.e., for turbulent flow in a triangular array (Rehme, 1972):
f = fct [1.045 + 0.071(P/D 1)] : Re = 104
f = fct [1.036 + 0.054(P/D 1)] : Re = 105
where P and D represent fuel lattice pitch and fuel rod diameter, respectively.
The coefficient C in Eq.10 represents inertial resistance and its unit is 1/m. The coefficient C used in this calculation and the porosity values for different regions are tabulated in Table 1. These values are calculated for each region
seperately using the detailed geometry reported in (Kolev et al., 2004).
Table 1: Computational parameters for porous media model.

Perforated Plates Region

Core Support Plates Region
Assembly Support Region
Fuel Rod Region
Fuel Rod and Spacer Region

Cx
152100
25100
26000
2342
47850

C (1/m)
Cy
152100
25100
26000
2342
47850

Cz
152.1
25.1
26
2.342
47.85

porosity
0.3400
0.4286
0.4280
0.5540
0.0912

User-defined source terms (S k and S  ) are taken to be zero for present study. Secondary flows are so small and
axial flow is dominant in reactor core. In reactor core, coolant flows through the narrow channel so there is almost no
changes in the turbulent kinetic energy except spacer top and bottom regions. The contribution of additional turbulent
kinetic energy on pressure drop is negligible in this study.
4. Computational Domain
4.1. Design and geometrical parameters of the VVER-1000 reactor
The overall height of the pressure vessel of VVER-1000 is about 12.5 m and its inner diameter amounts to 3.6
m. The location of main inlet and outlet nozzles of the reactor vessel is non-uniform in azimuthal direction and
4

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
asymmetric with respect to the core symmetry axes. The diameter of these nozzles are 0.85 m. The core contains
163 hexagonal fuel assemblies; each one consists of 312 fuel pins, 18 guide channels for control rods and a central
instrumentation tube. The diameters of fuel pins, guide channels and central rods are 9.1 mm, 12.6 mm and 11.2 mm,
respectively. The fuel pins are arranged in triangular grid. The main design details are shown in Fig. 1.

Figure 1: Computational domain of the VVER-1000 RPV (left) and 1/6 symmetry of fuel assembly (right) (Kolev et al., 2004).

The coolant enters the vessel by the inlets, flows downwards through the downcomer and enters the lower plenum
by passing a perforated elliptical bottom plate. It flows through channels formed by the lower parts of the core support
columns. The upper part of the support columns serve as flow distributors at core entrance. In this part the columns
are perforated and hollow. Crossing the core bottom plate the water enters the core. In the core region, the coolant is
heated up by the core during it passes through the fuel rods and also spacer structures and leave the core region. A
more detailed description of the components and geometry can be found in OECD benchmark specifications (Kolev
et al., 2004).
4.2. Design differences in the VVER-1200 reactor core
There are several significant design differences between VVER-1000 reactor core and VVER-1200 reactor core.
Two of these differences are adopted in this study. One of them is core height and the other is number of spacer grids.
VVER-1200 reactor core is about 5% longer than VVER-1000 core and has one more spacer element.
4.3. Mesh structure
GAMBIT, a geometry and mesh generation code is used to create the mesh structure of domain. 3D geometry is
created for this study. The outer sections of the RPV are created in detail. For the inner sections, many of the core
supporting structures locating at the lower plenum are created almost as exactly as they are. For the parts and core
region geometry created as a blocks. There are five different blocks, which represent perforated plates region, core
support plates region, assembly support plates region, fuel regions and spacer regions, respectively. Mesh structure
is created meticulously. In section interfaces, mesh growth ratios are selected at reasonable levels. For the whole
geometry, 10.5 million mesh elements are created.
5

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
4.4. Boundary conditions
Calculations are made for normal operating conditions and for different inlet temperature condition. VVER-1200
reactors are operating at nominal conditions with a thermal power of 3200 MW. The system pressure amounts 16.2
MPa with a core inlet coolant temperature of approximately 571 K. The mean coolant mass flow rate is at 19 tonnes
per second. Piecewise-linear profile is used for thermo-physical properties of working fluid. Cosine profile is adopted
for heat generation.
5. Results
5.1. Calculations for normal operation
Normal operating condition parameters are indicated in Section 4.4. Heat generation profile is recalculated for the
porous media model. In porous media model, since computational domain is fluid, heat sources can only be applied
for fluid regions. To make correction in heat source, the value of heat source, which is available for solid region, is
divided by the porosity value.
5.1.1. Temperature distribution in reactor core
In case of normal operation, temperature distribution of reactor core is presented in Fig. 2. Reactor inlet temperature is set as 571 K and outlet temperature is calculated as 601 K. Nominal core temperature rise is reported as 30-31
K in literature. So porous media approach gives acceptable result for core temperature distribution. The result of the
temperature rise agree with the operational parameters.

Figure 2: Temperature distribution at centerline of reactor core.

In nuclear reactors, axial flow is the dominant flow. Lateral flows are negligible in core region (especially through
the spacer region). This behavior can be seen in Fig. 3. Therefore, porous media approach for core region is a good
approximation.

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria

Figure 3: Flow pathlines through the core region.

5.1.2. Pressure drop in RPV

According to the OECD Benchmark report (Kolev et al., 2004), pressure drop between reactor inlet nozzle (P0 in
Fig. 1) and core support region (P2 in Fig. 1) is about 140 kPa, and pressure drop between core support region and
top section of the fuel assemblies (P4 in Fig. 1) is about 200 kPa.

Figure 4: Pressure distribution in reactor pressure vessel.

Pressure drop behavior is presented in Fig. 4. It can be seen in Fig 4, result obtained with porous media approach
is in a good agreement with reported values. Detailed pressure drop values can be seen in Kolev et al. (2004).

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
5.2. Effects of cold leg temperature variations on system
Variations at cold leg temperature is a significant issue for nuclear reactor safety. For instance, trip in one of the
reactor coolant pump results in significant temperature increase at inlet nozzle. Furthermore, as mentioned before,
main flow direction in the core section is axial one. Although theres mixing through the core, any increase in inlet
temperature affects the core outlet temperature. This may result in reaching high cladding temperature in the core,
especially in the hot channel.
5.2.1. Temperature behavior
One of the cold inlets is changed to 575 K (the others constant at 571 K) in this part of the analysis. Then the
assembly outlet temperature (in related outlet section) increases to 605 K, which is 4 K higher than nominal condition.
This behavior can be seen in Fig. 5. Similar results, which is about the coolant mixing in the core region, are reported
in (Bottcher, 2008; Bottcher and Krumann, 2010; Bieder et al., 2007).

Figure 5: Temperature distribution in RPV in case of presense temperature variation at one of the cold inlets

5.2.2. Pressure behavior

Temperature variation (4 K) does not effect the pressure drop characteristic in the core. As in nominal condition,
pressure drop between the inlet nozzle and the assembly outlet is about 340 kPa.
6. Conclusion
In this study, CFD calculations for 3D VVER RPV are performed using ANSYS Fluent commercial CFD tool,
and temperature and pressure behaviors are investigated. In place of detailed modeling of VVER core, porous media
approach is adopted in the calculations. Porous media parameters used in modeling are generated for support regions,
fuel regions and spacer regions seperately. Mainly the geometric details are taken from geometry of VVER-1000
reactor. Then the design differences are added. For this calculation, porous media approach gives the advantage for
computational requirements and time due to the good performance even in coarse mesh structure.
8

Ayhan and Ergun / BgNS conference 18-21 September 2013, Sunny beach, Bulgaria
Study shows that the porous media approach is an efficient tool for the complex reactor core section, since it can
give reasonable result for temperature behavior and pressure characteristics.
Acknowledgements
This study was financially supported by Turkish Atomic Energy Authority (TAEK, PN: 2011TAEKA5.H1.P1.04).
References
Bieder, U., Fauchet, G., Betin, S., Kolev, N., Popov, D., 2007. Simulation of mixing effects in a VVER-1000 reactor. Nuclear Engineering and
Design 237, 17181728.
Bottcher, M., 2008. Detailed CFX-5 study of the coolant mixing within the reactor pressure vessel of a VVER-1000 reactor during a nonsymmetrical heat-up test. Nuclear Engineering and Design 238, 445452.
Bottcher, M., Krumann, R., 2010. Primary loop study of a VVER-1000 reactor with special focus on coolant mixing. Nuclear Engineering and
Design 240, 22442253.
Inc. ANSYS, 2009. ANSYS Fluent 12 Users Guide.
Kolev, N., Aniel, S., Royer, E., Bieder, U., Popov, D., Topalov, T., 2004. VVER-1000 coolant transient benchmark (V1000CT)-Volume II:
Specifications of the VVER-1000 Vessel Mixing Problems. Commissariat a lEnergie Atomique and OECD Nuclear Energy Agency .
Rehme, K., 1972. Pressure drop performance of rod bundles in hexagonal arrangements. International Journal of Heat and Mass Transfer 15, 2499
2517.
Todreas, N.E., Kazimi, M.S., 2001. Nuclear Systems II: Elements of Thermal Hydraulic Design. Taylor & Francis, ISBN-10: 1560320796. 1st
edition.