Documente Academic
Documente Profesional
Documente Cultură
Received 25 January 2005; received in revised form 2 August 2005; accepted 4 August 2005
Abstract
The Ignalina NPP has a pressure suppression type of confinement, which is referred to as the accident localization system
(ALS). The ALS prevents the release of the radioactive material from the NPP to the environment during a loss-of-coolant
accident (LOCA). Ten water pools are located in the two ALS towers (five pools in each tower), which separate the dry well from
the wet well. These water pools condense the accident-generated steam and prevent high overpressures in the compartments.
The steam distribution device (SDD), with the vertical vent pipes (nozzles) that are inserted under the water of the condensing
pools, connects the dry well and the wet well. In case of an accident, these components must be capable of withstanding the
dynamic loads generated by a LOCA for successful pressure suppression function.
This paper presents the transient analysis of the SDD and their connections to the vertical steam corridors following a LOCA.
A thermo-hydraulic analysis of the SDD was performed using the state-of-the-art COCOSYS code to determine pressure and
temperature histories resulting from a LOCA. The finite element code NEPTUNE was used to evaluate the structural integrity of
the SDD and its supporting reinforced concrete wall. Results show that, although portions of the SDD undergo plastic response
and the outside surface of the vertical steam corridor reinforced concrete wall cracks, the structural integrity of the SDD and
wall are maintained during a LOCA.
2005 Elsevier B.V. All rights reserved.
Abbreviations: ALS, accident localization system; BRU-B, valve for steam discharge to ALS tower (Russian abbreviation); BSRC, bottom
steam reception chamber; CTCS, condenser tray cooling circuit; ECCS, emergency core cooling system; GDH, group distribution header; HCC,
hot condensate chamber; LOCA, loss-of-coolant accident; MCC, main circulation circuit; MDBA, maximum design basis accident; MSV, main
safety valve; NPP, nuclear power plant; RBMK, Russian acronym for channelized large power reactor; SDD, steam distribution device; SDH,
steam distribution header; VATESI, Lithuanian state atomic energy safety inspection
Corresponding author. Tel.: +370 37 401918; fax: +370 37 351271.
E-mail address: gintas@mail.lei.lt (G. Dundulis).
0029-5493/$ see front matter 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.nucengdes.2005.08.001
NED-4318;
No. of Pages 10
Nomenclature
A
F1
F2
m
vw
Greek letters
p
atmosphere density inside the pipe
(kg/m3 )
w
condensing pool water density (kg/m3 )
1. Introduction
The nuclear reactors of the Ignalina NPP are
RBMK-1500 type reactors. These reactors do not
possess the conventional Western style containment.
Instead, the Ignalina NPP has a pressure suppression
type of confinement, which is referred to as the acci-
Fig. 1. The scheme of Ignalina NPP ALS and location of main components of MCC: (1) fuel channel, (2) main circulation pumps, (3) suction
header, (4) pressure header, (5) GDH, (6) ECCS headers, (7) hot condensate chamber, (8) CTCS pumps and heat, (9) discharge pipes section,
(10) pipes from MSV and BRU-B, (11) pipes from reactor cavity, (12) condensing pools, (13) steam distribution headers, (14) BSRC sprays,
(15) water seals between HCC and BSRC, (16) BSRC vacuum breakers, (17) air removal corridor sprays, (18) air venting channel, (19) gas
delay chamber tank, (20) gas delay chamber, (21) reinforced compartments, (22) down hatches, (27) MSV and BRU-B, (28) drum separators,
(29) BSRC, and (30) reactor.
ing pools, connect the dry well and the wet well. These
components should be capable of withstanding the
dynamic loads for successful pressure suppression in
case of an accident. The Lithuanian nuclear regulator, VATESI, recommended performing this analysis in
order to verify that the design and strength of the SDD
and their connections to the vertical steam corridors are
sufficient to withstand the thermal and dynamic pressure loading in case of a LOCA. To determine pressure
and temperature loadings, the calculations were performed employing the state-of-the art computer code
COCOSYS (COCOSYS, 2001).
The structural integrity analysis was performed for
the SDD in the first condensing pool using the dynamic
loading results (temperature and pressure histories)
from the thermo-hydraulic analysis. The NEPTUNE
(Kulak and Fiala, 1988) computer code was used for
the transient structural analysis. The code is based upon
the central difference explicit integrator. Thus, the code
does not employ stiffness or flexibility matrices but
is based upon a nonlinear internal nodal force vector.
This approach is ideal for transient, nonlinear analyses in which metals are deforming in an elastoplastic
mode, concrete is cracking/crushing and contact impact
is taking place. When individual elements reach a failed
state, their contributions to the internal nodal force vector is reduced to zero and there is no change required
to the solution algorithm. Validation of the NEPTUNE
Regarding material properties, the model to be analyzed is constructed from two materials: the SDH
pipelines and vertical steam nozzles are made from
12X18H10T steel and the compartment walls are made
from reinforced heavy concrete M300. The mechanical properties (Norms and Rules, 1984) of the con-
Table 1
Material properties of reinforced concrete
Material
Youngs modulus
(GPa)
Poissons ratio
Tensile strength
(MPa)
Compressive
strength (MPa)
Yield stress
(MPa)
Tensile strength
(MPa)
Concrete
Rebars
27
205
0.2
0.3
1.5
NA
17
NA
NA
392
NA
590
NA = not applicable.
Table 2
Material properties of steel 12X18H10T
Characteristic
Temperature ( C)
20
100
200
300
510
216
35
55
205
461
206
30
55
200
421
187
27
54
190
412
177
26
52
180
Fig. 4. Finite element model of the SDD 19 900 1600 and support
wall: (1) SDD pipe, (2) concrete support wall, (3) vertical vent pipes,
(4) longitudinal connectors, (5) SDD support constrains, and (6) cross
connectors.
The pressure F1 (force per unit area of flow) represents the loading of the vent pipe due to steam/gas
mixture flow. The loading to the vent pipe wall is calculated from Eq. (1). This loading was calculated for
all three types of SDDs installed in Ignalina NPP ALS:
F1 =
m2
(Pa)
p A2
(1)
The equation for F1 is derived from the formula provided in (ANSI/ANS-58.2-1988) by assuming
p 0. Such an assumption is reasonable because the
steam/gas mixture is discharged under the water and the
water column compensates the pressure difference in
the vent pipe and the gas space above the water layer
in the pool. The results of the dynamic loading, force
F1 , of the vertical vent pipes due to steam/gas mixture
flow are presented in Fig. 7.
The loading applied to the SDDs due to pool swell,
which is briefly described in Section 1, is defined as
F2 . It represents the specific pressure of water impact
on the SDD header. This loading appears only in the
case when the pool swell phenomenon occurs and the
water surface level reaches the SDD headers. This is a
short-term loading, but may be important due to large
mass of the water lifted. This pressure is calculated by
Eq. (2):
F2 = w v2w (Pa)
(2)
3. FE analysis results
The aim of the structural integrity analysis of the
SDD subjected to a MDBA was to evaluate the following:
structural integrity of the SDD header;
structural integrity of the SDD supporting wall;
structural integrity of the connection between the
SDD header and the vertical vent pipes.
Stress analysis results are presented as stress snapshots at the time step of maximum stress. The von
Mises stresses in the SDD header near the connection
of the SDD to the concrete support wall are presented in
Fig. 10. The stresses had a maximum value of 236 MPa.
The displacement field of the SDD resulting from
a MDBA are presented in Fig. 11. The maximum
displacement was located in the connection of the
SDD header pipe and wall with a maximum value of
13.73 mm.
To show more detail, the stress analysis results are
presented as stresstime curves in the elements adjacent to the connection of the SDD header and the
supporting wall. The temporal variation of the normal
and shear stresses are presented in Fig. 12. It is seen
that the normal stress, xx , is the largest with a maximum value of 242 MPa at 2.00 s. The other stresses are
smaller. It was determined that the pipe material yield
limit (177 MPa) was exceeded at this location of the
Fig. 10. von Misses stress (Pa) distribution in the SDD 19 900
1600 near the connection of the header to the support wall.
Fig. 12. Normal and shear stress histories in the SDD 19 900 1600
adjacent to the connection of the SDH and vertical vent pipes.
Fig. 14. Displacement (m) field in the SDD 19 900 1600 headersupport wall.
Fig. 15. Temporal evolution of normal stresses in the concrete adjacent to the connection of the SDH and support wall.
The normal stress, xx , for compression in the first integration point (layer 1) is the largest with a value of
1.21 MPa. The concrete limit for tension is 1.5 MPa
and for compression it is 17 MPa. It was determined
that the concrete limit for tension was exceeded at
the fifth integration point, but the limit for compression was not exceeded. This indicates that the limit in
tension is reached and the concrete cracking begins;
tensile failure occurred only at one of the integration
points.
The temporal variation of the axial stress, xx , in the
reinforcing bars in the SDD header support wall adjacent to the connection of the SDD header to the support
wall are presented in Fig. 16. Four layers of rebars are
used in the quadrilateral plate element to represent the
four layers of rebars in the reinforced concrete wall. It
is seen that the axial stress in the first rebar layer, which
10
Acknowledgments
The authors would like to acknowledge the support
and access to the latest NEPTUNE code provided by
the US Department of Energy and Argonne National
Laboratory. The authors also want to express gratitude
to the administration and technical staff at the Ignalina
NPP for providing information regarding operational