See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/264176576

Nonlinear time history analysis of a pre-stressed concrete containment vessel

model under Japan's March 11 earthquake

Article  in  Computers and Concrete · January 2014

DOI: 10.12989/cac.2014.13.1.001

CITATIONS READS

5 289

5 authors, including:

An Duan Zuozhou Zhao

Zhejiang University Tsinghua University
7 PUBLICATIONS   62 CITATIONS    13 PUBLICATIONS   14 CITATIONS   

SEE PROFILE SEE PROFILE

Ju Chen Wei Liang Jin

Chinese Academy of Sciences Zhejiang University
48 PUBLICATIONS   816 CITATIONS    180 PUBLICATIONS   1,524 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Durability and life cycle management of concrete structures View project

Muti-environmental & loding similarity theory of concrete structures View project

All content following this page was uploaded by An Duan on 25 February 2015.

The user has requested enhancement of the downloaded file.

Computers and Concrete, Vol. 13, No. 1 (2014) 1-16
DOI: http://dx.doi.org/10.12989/cac.2014.13.1.001 1

Nonlinear time history analysis of a pre-stressed concrete

containment vessel model
under Japan’s March 11 earthquake
An Duan1, Zuo-zhou Zhao2, Ju Chen1, Jia-ru Qian2 and Wei-liang Jin1
1
College of Civil Engineering and Architecture, Zhejiang University, Hangzhou 310058, China
2
Key Laboratory of Civil Engineering Safety and Durability of China Education Ministry,
Tsinghua University, Beijing 100084, China

(Received December 30, 2011, Revised April 3, 2013, Accepted April 31, 2013)

Abstract. To evaluate the behavior of the advanced unbonded pre-stressed concrete containment vessel
(UPCCV) for one typical China nuclear power plant under Japan’s March 11 earthquake, five nonlinear time
history analysis and a nonlinear static analysis of a 1:10 scale UPCCV structure have been carried out with
MSC.MARC finite element program. Comparisons between the analytical and experimental results
demonstrated that the developed finite element model can predict the earthquake behavior of the UPCCV
with fair accuracy. The responses of the 1:10 scale UPCCV subjected to the 11 March 2011 Japan
earthquakes recorded at the MYG003 station with the peak ground acceleration (PGA) of 781 gal and at the
MYG013 station with the PGA of 982 gal were predicted by the dynamic analysis. Finally, a static analysis
was performed to seek the ultimate load carrying capacity for the 1:10 scale UPCCV.

Keywords: earthquake; concrete containment vessel; unbounded tendon; nonlinear time history; finite
element analysis

1. Introduction

As the ultimate barrier against the release of radioactivity, the concrete containment vessel is a
very important structure of nuclear power plant. In 2006, an advanced unbonded pre-stressed
concrete containment vessel (UPCCV) for one typical China nuclear power plant was designed by
Shanghai Nuclear Engineering Research & Design Institute. Since the UPCCV is required to resist
earthquakes, a pseudo-dynamic test of a 1/10 scale UPCCV model structure has been conducted at
Tsinghua University, Beijing China in 2006 (Qian et al. 2007). The test results and a push over
analysis (Duan and Qian 2009) show that the advanced UPCCV would remain elastic under the
safe shutdown earthquake (SSE) of the US criteria.
On March 11, 2011, a devastating earthquake and tsunami hit Japan, and the Fukushima Daiichi
1 nuclear reactor plant was destroyed. This earthquake raised again profound concern regarding
the seismic performance of the nuclear power plants around the world.
In fact, during the past years a large number of experimental and analytical studies focused on

Corresponding author, Associate Professor, E-mail: [email protected]

Copyright © 2014 Techno-Press, Ltd.

http://www.techno-press.org/?journal=cac&subpage=8 ISSN: 1598-8198 (Print), 1598-818X (Online)
2 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

the seismic capacity of containment vessel. In Japan, the Nuclear Power Engineering Corporation
has conducted a series of shaking table tests on a 1/8 scale concrete containment vessel model to
validate its seismic design and reliability under destructive earthquakes (Hirama et al. 2005a, b).
Lee and Song (1999) used a seismic fragility method and a conservative deterministic failure
margin method to evaluate the seismic safety for the containment structure of the Yonggwang
Nuclear Power Plant. In recent years, numerical simulation has become popular with the rapid
development of computer technology. Many studies have been made on the simulations of the
behavior of the nuclear containment under pressure loading far beyond the design basis by using
finite element method (Bash et al. 2003, Prinja et al. 2005). However, researches on the nonlinear
time history analysis of the containment vessel subjected to earthquakes are limited (Kobayashi et
al. 2002, Frano et al. 2010). The nonlinear time history analysis is the most sophisticated and
powerful method in earthquake engineering assessment. It can calculate the displacements,
stresses, strains, damaged positions at any time during the earthquake excitation for the structure.
Frano et al. (2010) carried out a time history analysis for the IRIS containment under an
earthquake excitation with the peak ground acceleration (PGA) of 0.3g, and the acceleration and
displacement time histories were predicted. The PGA of the Japan’s March 11 earthquake is much
larger than 0.3g. Moreover, the detailed responses (like stress and strain distributions, cracking
positions) of the advanced UPCCV subjected to this strong earthquake are demanded.
Hence, this paper presents a nonlinear 3-D dynamic time history analysis carried out with
MSC.MARC finite element (FE) program (MSC 2007a) to predict the global and local response of
the advanced UPCCV for one typical China nuclear power plant under Japan’s March 11
earthquake.

2. The UPCCV test model

Due to loading limitations, the pseudo-dynamic test model had to be geometrically reduced
from the original prototype. Hence, a 1:10 scale model structure of the advanced UPCCV for one
typical China nuclear power plant was constructed. An overall geometry and a photo of the test
model are shown in Fig. 1. The test structure represented all the main structural features of the
advanced UPCCV which is an unbonded pre-stressed concrete cylinder with a hemispherical dome,
two vertical buttresses and a base mat.

(a) Photo of the test model

Fig. 1 Photo and geometry of the test UPCCV model structure
 Nonlinear time history analysis of a pre-stressed concrete containment vessel model 3

(b) Geometry of the test model

Fig. 1 Continued

3. Mathematical modelling

Predicting response of a complex structure like a UPCCV under a strong earthquake needs
complicated nonlinear analysis. The nonlinear analysis was carried out on the 1:10 scale UPCCV
test structure by using MSC.MARC FE program. Examples and benchmarks (Lu et al. 2007, Li et
al. 2011, Miao et al. 2011, Lu et al. 2012) show that this program can precisely simulate the
nonlinear behaviours of reinforced concrete structures. Fig. 2 presents the 3D FE model.

3.1 Modelling of the structural features

The cylinder and the dome were modeled with 4-node, thin shell elements (Elements 139). The
integration through the shell thickness was performed numerically using Simpson's rule (MSC
2007b), and seven integration points were specified. The base mat was modeled with 8-node,
isoparametric, arbitrary hexahedral solid elements (Elements 7). The buttresses were modeled with
beam elements (Elements 98). The reinforcements were modeled with 4-node, isoparametric,
quadrilateral membrane elements (Elements 18) which was inserted into the shell elements,
assuming a perfect bonding between the steel bars and the surrounding concrete. The vertical and
hoop tendons in the cylinder were modeled explicitly with 2-node, truss elements (Elements 9)
which have no flexural stiffness. The FE model was partitioned into 13812 elements, with the total
number of 14585 nodes.
The relative sliding behavior between the unbonded tendons and surrounding concrete could be
simulated by using options provided in MARC, namely, Links and Nodal ties. The truss nodes
(with 3 translational degrees of freedom) for the tendons were defined as tied nodes, while the
shell nodes at the same locations for the concrete cylinder were defined as retained nodes. There
were two ways of tying between the tied nodes and the retained nodes: 1) except for the anchors,
the tied nodes were constrained in the two transverse directions while the longitudinal degree of
4 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

freedom was free and 2) at the location of the anchors, the 3 degrees of freedom of the tied nodes
were tied respectively to those of the retained nodes. In order to achieve the designed hoop and
vertical compressive stresses in concrete cylinder, initial tensile stresses were applied on the
tendons.

Fig. 2 3D FE model for the 1:10 scale UPCCV

Table 1 Scale factors for the 1:10 test model structure

Scale factor
Parameter Symbol Relationship
(model: prototype)
Length Sl Sl 1:10
Mass density Sρ Sρ 1:1
Elastic modulus Se Se 1:1
Time St Sl S ρ / S e 1:10

Frequency Sf S  / S e / Sl 10:1
Acceleration Sa Se /( Sl×Sρ) 10:1
Displacement Su Sl 1:10
Stress Sσ Se 1:1
Strain Sε Se 1:1
Force SF Se×Sl 2 1:100

Table 2 Material properties used in the analysis

Yong’s modulus Yield strength Mass density
Material Poisson’s ratio
/GPa /MPa / (kg·m-3)
Concrete 35.5 0.2 — 2500
Tendons 195 1860
0.3 7850
Rebars 200 500
 Nonlinear time history analysis of a pre-stressed concrete containment vessel model 5

3.2 Modelling of the material

Concrete in compression was treated as an elastic–plastic material using the Buyukozuturk

yield criterion (Buyukozturk 1977) with an associated flow rule and a combined hardening rule. To
define the uniaxial stress-strain relationship the equations suggested by Guo and Zhang (1982) was
used

   
2
 
3

 a   3  2a    a  2  , 0    c
 c  c   c 
   (1)

fc  c
, c    u
  
2

 b  1 
  c  c

in which, σ is the compressive stress, ε is the compressive strain, fc is the compressive strength
which equals to 0.76 fcu, εc is the peak strain and is taken as 0.002, εu is the crushing strain and is
taken as 0.0033, a is the ascending branch parameter and is taken as 1.75, and b is the descending
branch parameter which is taken as 2.59.
The smeared crack approach was adopted to simulate the tensile behavior of cracked concrete.
It is assumed that the concrete behave as a linear elastic material and when the first principal stress
reaches the tensile strength of the concrete (ft = 4.2 MPa) cracks will develop in a direction
normal to the principal stress.
For the steel materials (rebars, and tendons), the elasto-plastic theory based on von Mises yield
criterion was adopted to represent the non-linear behavior.
The material properties of the concrete and the steel used in the analysis of the 1:10 UPCCV
model are presented in Table 2.

3.3 Input of the earthquake response analysis

Three ground motion accelerograms were used in the analysis: 1) an artificial earthquake wave
with the PGA of 0.3g (294 gal), which was also used as the ground excitation in the pseduo-
dynamic test; 2) E-W component of the 11 March 2011 Japan earthquake recorded at the MYG003
station with the PGA of 781 gal (Kyoshin 2011a); and 3) E-W component of the 11 March 2011
Japan earthquake recorded at the MYG013 station with the PGA of 982 gal (Kyoshin 2011b). The
design SSE has a maximum horizontal acceleration of 0.2g (196 gal). Thus, the PGA of the three
ground motions in this investigation are 1.5, 4.0 and 5.0 times those of SSE respectively.
The acceleration time histories and acceleration response spectrums of the three earthquakes for
the prototype are shown in Fig. 3, Figs. 4 and Fig. 5. It should be noted that for the last two
records, the total duration time are 300s. However, the present study focused on the intense phase
of 120s (ranging from 30s~150s). In addition, for the pseduo-dynamic test and the dynamic
analysis of the 1:10 test model structure, these three records were scaled both in time and
amplitude for similitude requirements (see Table 1).
There were five nonlinear time history analysis cases: Case 1 was used to validate the FE
model by comparing the numerical results with the experimental response obtained from the
6 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

(a) Acceleration time history (b) Acceleration response spectrum

Fig. 3 Artificial earthquake

(a) Acceleration time history (b) Acceleration response spectrum

Fig. 4 Ground motion recorded at MYG003 station during the Japan’s March 11, 2011 earthquake

(a) Acceleration time history (b) Acceleration response spectrum

Fig. 5 Ground motion recorded at MYG013 station during the Japan’s March 11, 2011 earthquake
 Nonlinear time history analysis of a pre-stressed concrete containment vessel model 7

Table 3 Main characteristics of the five time history analysis cases

Analysis case Input wave PGA/gal Time duration/s Input direction
1 Artificial wave 2940 2.4 y - axis
2 MYG003 y - axis
7810
3 MYG003 x - axis
12
4 MYG013 y - axis
9820
5 MYG013 x- - axis

pseduo-dynamic test. Case 2 to Case 5 were used to predict the response of the advanced UPCCV
under Japan’s March 11, 2011 earthquake. The main characteristics of the five analysis cases are
listed in Table 3. Take Case 3 for instance, the E-W component of the 11 March 2011 Japan
earthquake recorded at the MYG003 station was used as the input wave, the time duration was
scaled to 12s and the PGA scaled to 7810 gal, and 0°- 180° (x-axis) was chosen as the direction of
the excitation.
Rayleigh damping was assumed with a specified damping ratio (ξ) of 0.05 at both the first and
second vibration modes of the 1:10 test model. Houbolt method (MSC 2007a) was used to
integrate the equations of motion to obtain the dynamic response of the structural system. The
nonlinear algebraic equations were solved iteratively by using Modified Newton-Raphson method
(Ben-Israel 1966). The time interval of the artificial wave and the ground motions recorded at
MYG003 (or MYG013) station are 0.005s and 0.01s respectively. Hence, a time step of 0.0005
was used in Case 1, and 0.001s in other dynamic analysis cases. The chosen time steps didn’t
exceed 10 percent of the period of the highest frequency of interest and were considered to be
adequate for the dynamic analysis.

4. Analysis results and discussions

4.1 Validation of the FE model

Periods, frequencies and directions of the first six modes estimated from the analyses for the
1:10 scale UPCCV structure are given in Table 4. The frequencies in the direction of y-axis and x-
axis for the prototype are 4.343 Hz and 4.377 Hz respectively (Qian et al. 2007). Hence, the
relationship between the frequencies of the FE model and those of the prototype are in accord with
the similitude laws.
As stated before, a pseudo-dynamic test of a 1:10 scale UPCCV structure has been carried out
using the artificial wave (Fig. 3) with the PGA of 2940 gal as the ground motion. 90°-270° (y-axis)
was chosen to be the direction of the excitation. Fig. 6 gives the displacement time history
obtained from the pseudo-dynamic test and from Analysis Case 1. The displacement in the figure
is the horizontal displacement on the top of the cylinder (at the elevation of 4800 mm) relative to
the bottom of the cylinder. As can be seen, the calculated displacement time history showed good
agreement with the experimental result, indicating that the analytical model in this study was
reasonable and has the ability to predict the earthquake response of the UPCCV. Nevertheless, the
peak values predicted by the analysis were about 15 percent lower than those obtained from the
test. The occurrence of the deviance was mainly due to the following reasons: 1) there existed
some construction geometry errors during the building of the test model, and 2) the properties of
the actual concrete may differ a little from those used in the analysis.
8 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

Table 4 First six modes

Mode 1st 2nd 3rd 4th 5th 6th
Period /s 0.0233 0.0229 0.0183 0.0167 0.0164 0.0149
Frequency/Hz 42.93 43.74 54.62 59.76 61.03 67.30
Direction y x Torsion Torsion Torsion Torsion

Table 5 Peak values of displacement and base shear force

Analysis Input Displacement/mm Base shear force/ kN
Direction
Case motion Maximum Minimum Maximum Minimum
Case 2 y 1.56 -1.82 3500.4 -2650.4
MYG003
Case 3 x 1.44 -1.93 3860.1 -2843.3
Case 4 y 1.57 -1.94 4744.5 -2555.4
MYG013
Case 5 x 1.46 -1.99 5017.3 -2546.6

(a) Test results

(b) Analysis results

Fig. 6 Displacement time history of a 1:10 scale UPCCV model under the artificial earthquake
excitation (with the PGA of 2940 gal)
Nonlinear time history analysis of a pre-stressed concrete containment vessel model 9

(a) Case 2

(b) Case 3

(c) Case 4

(d) Case 5
Fig. 7 Displacement time histories predicted by Case 2 to Case 5
10 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

4.2 Simulation results of case 2 to case 5

4.2.1 Displacements and base shear forces

Fig. 7 presents the displacement (as defined in 4.1) time histories predicted by Case 2 to Case 5.
The input ground record and the direction of the excitation are given on the upper right corner of
each graph. Examination of Figs. 7(a) and (b) shows that the response of the 1:10 scale UPCCV
model structure was elastic throughout the duration of the input of ground motion MYG003 and
sustained no permanent displacement at the end of the record. However, when under MYG013
excitation, the UPCCV model structure showed inelastic deformation that caused some residual
displacement. It should be noticed that according to the similitude law, the displacement response
of the prototype should be ten times that of the 1:10 model structure.
The peak values of the base shear force (BSF) of the 1:10 scale UPCCV structure during the
four earthquake excitations are listed in Table 5. When the directions of the excitation were in y-
axis, the maximum calculated BSFs were 3500 kN and 4745 kN for the MYG003 and MYG013
earthquakes, while the maximum BSF calculated using the artificial wave (Case 1) was 2082 kN.

4.2.2 Strains and stresses

When subjected to the earthquake excitation, failure was more likely to occur in the cylinder
than in the dome or the basement, thus emphasis of this research was placed on the behaviour of
the cylinder. For brevity, only the results of the cylinder predicted in Case 4 are displayed.
According to the similitude law (Table 1), the strain and stress would be 1:1 for the 1:10 model
structure and the prototype.

4.2.2.1 Concrete
It is noted that the maximum absolute value of the displacement and the maximum strains
occur at the same time. Thus, the vertical strain distribution in the inner concrete layer of the
cylinder at t = 1.703s (when the minimum displacement is reached) for Case 4 is shown in Fig. 8
(a). In general, the strains are within elastic limit. The lower parts of the cylinder on the 270° side
were in tension, and the largest tensile vertical strain (261.5 × 10-6) appeared at the elevation of
330 mm at 270° azimuth (Position 1). Concrete cracks on inner surface with a crack depth of 40%
wall thickness occurred near the bottom of the cylinder. Fig. 8(b) displays the crack width
distribution and highlights the area of the cracked concrete.
The analysis showed that on the inner surface of the cylinder, flexural cracks developed near
the bottom of the cylinder, while on the outer surface, cracks would appear near the device
opening on the 90° side. The maximum tensile strain in the outer concrete layer occurred at 1.594s
(when the maximum displacement was reached). Fig. 9 presents the strain contour at this moment.
As can be observed, the largest tensile vertical strain (242.9 × 10-6) of the outer layer is
concentrated near the device opening (Position 2), while the largest compressive vertical strain (-
378.6 × 10-6) appears at the elevation of 330 mm at 270°azimuth (Position 1). The Cracks were
noticed near the opening with a crack depth of 50% wall thickness from the outer surface. The
main reasons for concrete cracking at this position are as follows. Firstly, the geometry
discontinuity around the penetration caused stress concentration. Secondly, in the actual structure
(i.e. prototype), a steel liner was used to strengthen the opening area, yet the FE model didn’t take
into account this liner since the 1:10 test structure model which we used for the simulation does
not have this layer of steel liner in it for the sake of simplification of construction.
The vertical strain time histories at the two positions are given in Fig. 10. For position 2, it is
 Nonlinear time history analysis of a pre-stressed concrete containment vessel model 11

(a) Vertical strains

(b) Crack widths (Unit: mm)

(c) Hoop strains

Fig. 8 Strain and crack width distribution in the inner layer of the concrete cylinder at 1.703s for Case 4
12 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

Fig. 9 Strain distribution in the outer layer of the concrete cylinder at 1.594s for Case 4

(a) Position 1

(b) Position 2
Fig. 10 Vertical strain time histories at the two positions for Case 4
 Nonlinear time history analysis of a pre-stressed concrete containment vessel model 13

observed that in the outer layer, progressive cracking near the device opening caused permanent
residual strain at the end of the record. In the middle layer, the maximum strain was smaller than
the cracking strain. Due to prestressing, the strain in the inner layer remained negative throughout
the excitation. Position 1 showed the similar trend with Position 2 except that the cracks initiated
from the inside surface. All the crack depths were within 50% of the wall thickness, and no
through thickness cracking were observed during the excitation of MYG013.
Fig. 8(c) presents the hoop strain contour in the inner concrete layer of the cylinder at t =
1.703s. It is noticed that hoop compression due to prestressing is retained in almost the entire
cylinder region. The maximum compressive hoop strain was at the elevation of 873 mm on the
270° side.

4.2.2.2 Rebar
Analysis results show that the rebar stresses are within elastic limit throughout the excitation of
MYG013. The stresses of rebars in the cylinder at 1.703s for Case 4 are given in Fig. 11. The
stress contours show that the vertical rebars are influenced by the deformation on the 90° and 270°
sides which resist the overturning moment. As expected, the largest compressive and tensile
stresses in the vertical rebars are concentrated at the bottom of 270° side and 90° side respectively.
The stresses in the horizontal rebars of the cylinder are under compression except for the area
around the opening and near the bottom of 270° side.

(b) Horizontal rebars

Fig. 11 Stress distribution in the rebars of the cylinder at 1.703 s for Case 4 (unit: MPa)

Fig. 12 Base shear force –displacement predicted by the static analysis

14 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

4.3 Nonlinear static analysis

To seek the ultimate load carrying capacity of the UPCCV, a nonlinear static analysis of the
1:10 scale model structure was carried out. A monotonically increased lateral force was applied on
the top of the cylinder, and the loading direction was in y-axis.
Fig. 12 presents the relation of BSF versus displacement predicted by the static analysis. The
displacement in the figure is also the horizontal displacement on the top of the cylinder relative to
the bottom of the cylinder. The characteristics of this curve and the main sequence of events
during the lateral loading of the 1:10 scale UPCCV can be summarized as follows. The BSF-
displacement curve is nearly linear up to BSF about 3540 kN (Point A), indicating the structure is
in general within elastic stage, and no cracks are observed in the structure except around the device
opening. Then, flexural cracks appear near the cylinder-base junction of the 90° side, and the
structure begins to soften. At 6791 kN (Point B), the vertical rebar yielding initiated at the bottom
of 90° side follows by cracked zone spreading to a larger region, and the force-displacement curve
bends more sharply. At about 9428 kN (Point C), the vertical tendons at the azimuth of 90° begin
to yield. At 10098 kN (Point D), the crushing strains in concrete are reached at the base of the
cylinder within the range of 270° ± 40°. The BSFs at Point F and E are equal to the maximum
BSFs during the excitations of MYG013 and the artificial wave (1.5 SSE) respectively.
There are two failure modes for the UPCCV: 1) tendon yielding; and 2) concrete crushing.
Based on criterion 1, the ultimate lateral load resisting capacity of the 1:10 scale UPCCV is 9428
kN, 2.0 times the maximum BSF (4745 kN) during the MYG013 earthquake excitation (Case 4),
and 4.5 times the maximum BSF (2082 kN) during the 1.5SSE excitation (Case 1). According to
criterion 2, the ultimate lateral load capacity would be 10098 kN.

5. Conclusions

The main purpose of the present study is to evaluate the behaviour of the advanced UPCCV for
one typical China nuclear power plant under Japan’s March 11 earthquake. Five nonlinear time
history analysis and a nonlinear static analysis of a 1:10 scale UPCCV structure have been carried
out with MARC FE program. From the results of the analysis, the following statements can be
concluded:
1) The calculated displacement time history predicted by Case 1 shows good agreement with
the pseudo-dynamic test result, demonstrating that the analytical model in this study is reasonable
and is capable of predicting the earthquake response of the UPCCV.
2) The 1:10 scale UPCCV model responded elastically throughout the duration of the input of
ground motion MYG003.
3) When subjected to MYG013, the UPCCV model showed inelastic deformation that caused
some residual displacement. The results of Case 4 show that on the inner surface of the cylinder,
flexural cracks would develop near the bottom of the cylinder, while on the outer surface cracks
would appear near the device opening on the 90° side. Besides, all the crack depths were within 50%
of the wall thickness. The hoop compression in concrete due to prestressing was retained in almost
the entire cylinder region. The stresses of the vertical and horizontal rebars were within elastic
limit throughout the excitation of MYG013.
4) According to the tendon yielding failure criterion, the ultimate lateral load capacity of the
 Nonlinear time history analysis of a pre-stressed concrete containment vessel model 15

1:10 scale UPCCV is 9428 kN, 2.0 times the maximum BSF(4745 kN) during the MYG013
excitation, and 4.5 times the maximum BSF (2082 kN) during the 1.5SSE excitation.

Acknowledgements

The financial supports from the National Natural Science Foundation of China (No.51108413),
the Tsinghua Initiative Scientific Research Program (2012THZ02-2), the Fundamental Research
Funds for the Central Universities (No. 2012QNA4016) and the China Scholarship Council
(No.201208330042) are gratefully acknowledged.

References

Bash, S.M., Singh, R.K., Patnaik, R., Kushwaha, H.S. and Venkat Raj, V. (2003), ―Predictions of ultimate
load capacity for pre-stressed concrete containment vessel model with BARC finite element code ULCA‖,
Ann. Nucl. Energy, 30(4), 437-471.
Ben-Israel, A. (1966), ―A newton-raphson method for the solution of systems of equations‖, J. Math. Anal.
Appl., 15(2), 243-252.
Buyukozturk, O. (1977), ―Nonlinear analysis of reinforced concrete structures‖, Comput. Struct., 7(1), 149-
156.
Duan, A. and Qian, J.R. (2009), ―Aseismic safety analysis of a containment vessel model for CNP1000
nuclear power plant‖, Eng. Mech., 26(4), 153-157(in Chinese).
Frano, R.L., Pugliese, G. and Forasassi, G. (2010), ―Preliminary seismic analysis of an innovative near term
reactor: Methodology and application‖, Nucl. Eng. Des., 240(6), 1671-1678.
Guo, Z.H. and Zhang, X.Q. (1982), ―Experimental investigation of the complete stress-strain curve of
concrete‖, J. Build. Struct., 3(1), 1-12 (in Chinese).
Hirama, T., Goto, M., Hasegawa, T., Kanechika, M., Kei, T., Mieda, T., Abe, H., Takiguchi, K. and Akiyama,
H. (2005a), ―Seismic proof test of a reinforced concrete containment vessel (RCCV), Part 1: Test model
and pressure test‖, Nucl. Eng. Des., 235(13), 1335-1348.
Hirama, T., Goto, M., Shiba, K., Kobayashi, T., Tanaka, R., Tsurumaki, S., Takiguchi, K. and Akiyama, H.
(2005b), ―Seismic proof test of a reinforced concrete containment vessel (RCCV), Part 2: Results of
shaking table test‖, Nucl. Eng. Des., 235(13), 1349-1371.
Harris, H.G. and Sabnis, G.M. (1999), Structural Modelling and Experimental Techniques, (2nd Edition),
CRC Press, Boca Raton, FL, USA
Kobayashi, T., Yoshikawa, K., Takaoka, E., Nakazawa, M. and Shikama, Y. (2002), ―Time history nonlinear
earthquake response analysis considering materials and geometrical nonlinearity‖, Nucl. Eng. Des., 212(1-
3), 145-154.
Kyoshin Net. (2011a), ―MYG0031103111446[1]‖, http://www.k-net.bosai.go.jp/. [2011-5-10]
Kyoshin Net. (2011b), ―MYG0131103111446[1]‖, http://www.k-net.bosai.go.jp/. [2011-5-10]
Lee, N.H. and Song, K.B. (1999), ―Seismic capability evaluation of the prestressed reinforced concrete
containment, Yonggwang nuclear power plant Units 5 and 6‖, Nucl. Eng. Des., 192(2-3),189-203.
Li, Y., Lu, X.Z., Guan, H. and Ye, L.P. (2011), ―An improved tie force method for progressive collapse
resistance design of reinforced concrete frame structures‖, Eng. Struct., 33(10), 2931-2942.
Lu, X.Z., Teng, J.G., Ye, L.P. and Jiang, J.J. (2007), ―Intermediate crack debonding in FRP-strengthened RC
beams: FE analysis and strength model‖, J. Compos. Construct., 11(2), 161-174.
Lu, X.Z., Ye, L.P., Ma, Y.H. and Tang, D.Y. (2012), ―Lessons from the collapse of typical RC frames in
Xuankou school during the Great Wenchuan Earthquake‖, Adv. Struct. Eng., 15(1), 139-153.
MSC. Software Corporation (2007a), ―MSC.MARC 2007 (Volume A): Theory and user information‖.
 16 An Duan, Zuo-zhou Zhao, Ju Chen, Jia-ru Qian and Wei-liang Jin

MSC. Software Corporation (2007b), ―MSC.MARC 2007 (Volume B): Element library‖.
Miao, Z.W., Ye L.P., Guan, H. and Lu, X.Z. (2011), ―Evaluation of modal and traditional pushover analyses
in frame-shear-wall structures‖, Adv. Struct. Eng., 14(5), 815-836.
Prinja, N.K., Shepherd, D. and Curley, J. (2005), ―Simulating structural collapse of a PWR containment‖,
Nucl. Eng. Des., 235(17-19), 2033-2043.
Qian, J.R., Zhao, Z.Z., Duan, A., Xia, Z.F. and Wang, M.D. (2007), ―Pseudo-dynamic tests of a 1:10 model
of pre-stressed concrete containment vessel for CNP1000 nuclear power plant‖, China Civil Eng. J., 40(6),
7-13 (in Chinese).

CC

View publication stats