Comput.m & Structures Vol. 32, No. 314, pp. 563-590, 1989 Co457949/89 $3.00 + 0.

00
Printed in Great Britain. 0 1989 Maxwell Pqamon Macmillan plc

NONLINEAR ANALYSIS OF CONCRETE STRUCTURESt

KLAUS-J~RGENBATH&$@JAN WALCWK, llfi ALAN WELC.!E@~~ and NAGIN MrsrRvl/tt
~Ma~chu~t~ Institute of Technology, Cambridge, MA 02139, U.S.A.
JADINA R & D, Inc., Watertown, MA 02172, U.S.A.
TTaylor Woodrow Services, Ltd, Middlesex, U.K.

Abstract-Solution capabilities for two- and three-dimensional nonlinear finite element analysis of
concrete structures are presented. The concrete material is modeled including the triaxial nonlinear
stress-strain behavior, tensile cracking, axon-stiff~~g, compression crushing and str~n-sof~ng. The
results of various sample analyses are given to demonstrate the constitutive model and the solution
strategies. These analyses include the response predictions of a test beam and of small scale models of
two reactor vessels.

1. INTRODUCT’tON The objective in this paper is to present the three-

dimensional concrete model available in ADINA and
During recent years, interest in nonlinear analysis of show some applications. The concrete model was in
concrete structures has increased steadily, because of many respects already proposed in [2]. However, our
the wide use of plain, reinforced and prestressed recent improvements on the model have been very
concrete as a structural material, and because of the significant and promise to make the model much
development of relatively powerful finite element more attractive for practical nonlinear concrete
procedures [l]. If a realistic nonlinear analysis of a analyses.
concrete structure can be carried out, the safety of the ~thou~ we could have chosen to present only our
structure is increased and the cost can frequently be recent improvements on the concrete model, we
reduced. believe that it is of more interest to have the complete
Concrete exhibits a complex structural response model description in this paper with applications.
with various important nonlinearities; namely, a non- The basic aim in the development of the model was
linear stressstrain behavior, tensile cracking and to implement in ADINA a model that with the
compression crushing material failures and tempera- present constitutive descriptions, numerical methods
ture-dependent creep strains [2-91, All these concrete and computing equipment available would satisfy the
nonlinearities depend strongly on the triaxial state of following two criteria. Firstly, the model should be as
stress, and in addition the nonlinearities introduced simple as possible, but reproduce the important non-
by reinforcing and prestressing steel should in general linear and strength characteristics consistent with
be taken into account [l&13]. experimental results. Secondly, the model should be
There are a number of factors that have prevented theoretically sound and numerically stable, so that
the wider acceptability of nonlinear finite element reliable analysis results are obtained.
analysis procedures in the analysis of concrete struc- The material model is a hypoelastic model based
tures. A first important consideration is that the on a uniaxial stress-strain relation that is generalized
~nstitutive properties of concrete have not as yet to take biaxial and triaxial stress conditions into
been identified completely, and there is still no gener- account. Tensile cracking and compression crushing
ally accepted material law available to model concrete conditions are identified using failure surfaces, and
behavior. A second important factor is that nonlinear strain-softening effects are included in the compres-
finite element analysis of concrete structures can be sion and the tensile regions. A particularly valuable
very costly and may require considerable user exper- property of the model is that it is defined by a number
tise. The high cost of nonlinear analysis of concrete of input parameters that provide considerable
structures is largely due to the difficulties encountered flexibility in its use (and indeed the model may also
in the stability and accuracy of the solutions. These be employed to model sand or rock structures).
difficulties, however, are a direct consequence of the In the following sections we first present the con-
specitic numerical implemen~tion of the concrete crete model in detail. We then give some small-size
nonlinearities. applications, merely to demonstrate the details of the
model behavior when it is subjected to various
stress-strain paths. Finally, in Part II of tbe paper we
t Original figures were generated with a colour producing
terminal and submitted in colour.
present the results of the analyses of the Sandia
BCo-author of Part I. pressure vessel, an alkali-silica reacted beam and a
tt Co-author of Part II. PWR prestressed concrete containment. The solution
563
564 KLAUS-JBRGEN BATHE et al.

COMPRESSIVE STRAIN. in Iln

-0032 -0.001

KUPFER ET AL 2.6
(EXPERIMCNTA

PRESENT S
---- ------.

----- I.0

Fig. I. Concrete model uniaxial stress-strain law.

results presented in this section are particularly valu- 2.1. Stress-strain relations
able because they can be compared with and inter- The general multiaxial stress-strain relations are
preted against laboratory experimental results. derived from a uniaxial stress-strain relation ‘6 versus
An important issue in nonlinear analysis is the ‘e”. In this section, we describe the uniaxial and
visualization of the obtained response predictions. In multiaxial stress-strain relations employed in the
this paper we use our standard post-processing pro- model and concentrate on the behavior prior to
gram ADINA-PLOT to present all of our solution tensile cracking or compression crushing.
results. In the following discussion, all uniaxial parameters
are identified by a tilde (“) placed over them (i.e. all
2. PART I: THE CONCRETE MATERIAL MODEL parameters that have been obtained from Fig. I carry
a tilde).
The model implemented employs three basic fea-
2.1.1. Uniaxial conditions. A typical uniaxial stress,
tures to describe the material behavior, namely (i) a
‘5, to uniaxial strain, ‘6, relation (assuming continu-
nonlinear stress-strain relation including strain-
ous loading of the material) is shown in Fig. 1 (61.
softening to allow for the weakening of the material
This stress-strain relation shows that there are basi-
in compression, (ii) failure envelopes that define
cally four strain phases; namely, corresponding to
cracking in tension and crushing in compression, and
.C,>‘e’>O, &,a’~?>~,,, O>‘e’>& and &c”e’>&,
(iii) a strategy to model the post-cracking and crush-
where c?,is the strain corresponding to the uniaxial
ing behavior of the material. In the solution, the
cut-off tensile stress d,, ?,,, is equal to t; . F,, where 5
material can be subjected to cyclic loading conditions,
is an input parameter, CCis the strain corresponding
i.e. the numerical solution allows for unloading and
to the minimum (crushing) stress, d,, that can be
reloading including deactivation of tensile failures.
reached, and E?,,is the ultimate compressive strain. If
However, as will become apparent, the cyclic loading
C,3 % > 0, i.e. the material is in tension, the
conditions are only modeled realistically in situations
stress-strain relation is linear until tensile failure at
of essentially proportional loading.
the stress d,, and a constant Young’s modulus & is
In the following, the material model is described
employed,
for infinitesimal displacement conditions using the ‘5 = &‘e^ (1)
engineering stresses ‘triiand engineering strains ‘eii. In
order to analyze problems with large rotation condi-
tions, the total Lagrangian stress and strain variables
must be substituted for the engineering variables [ 11,
and then the model is directly applicable.7 For ‘5 < 0, we assume the following relation,

‘a/a, =
b%/Jw~l~c) (3)
t We use the notation of Refs [I,21 in this paper. 1 + A@/&) + B(‘P/&)2 + C(9/&,))
 Nonlinear analysis of concrete structures 565

and hence, mined,

$[l - B(‘e”/S$ - 2C(‘e’/&)3] !f<f- (6)

(E =
[1 + A (‘6?/frc)+ B(‘t?/Q + c(‘Z/ZJr]* (4) where f,, is the maximum value of the loading
function that has been reached during the complete
where
solution. In unloading, the initial Young’s modulus,
&, is used to form the incremental stress-strain
A = &l/J% + (P3 - 2P2)&/$ - (2P3 - 3P2 + 1) matrix, both for stiffness and stress calculations.
(P2 - 2P + l)P To obtain the stress-strain relations in loading
conditions, the principal stresses are calculated and
B=(2$/gJ-3)-2A, C=(2-&/$)+A
for each principal stress direction a uniaxial tangent
Young’s modulus, ‘&, corresponding to the strain in
and the strength parameters &, d,, Cc,& = 6c/t?c,d,,
the principal stress direction, ‘epi, is evaluated using
&, p = &,I& and & = a./$,, are obtained from uni-
eqns (2) and (4). When using eqn (4), the current
axial tests. It may be difficult to determine somewhat
strain ‘epiis employed and to account for multiaxial
accurate values of d, and CM;, in which case reasonable
stress conditions the material variables g,, 6”;, & and
values should be assumed by the analyst.
I& are replaced by the variables ai, a:, g: and Z:
The stress-strain relation in eqn (3) assumes
defined in eqns (13) and (14).
monotonic loading conditions. For unloading condi-
Let lap,, ‘ap2and ‘up3be the principal stresses at time
tions and loading back to the stress state from which
t, with ‘upsG Iup2< ‘up, and ‘&, , ‘l?p:p2 and ‘i$,, the
unloading occurred, the initial Young’s modulus i$ is
corresponding uniaxial Young’s moduli. Hence, the
used. For strain states beyond & in compression, it is
material is considered as orthotropic with the direc-
usually assumed that the material cannot resist any
tions of orthotropy defined by the principal stresses.
load and the stresses are set to zero. However, an
The stress-strain matrix corresponding to these direc-
option for a gradual stress unloading beyond 5” is
tions is, considering three-dimensional stress condi-
also included.
tions,

- (1 -v) 5?,,:pl v ‘E,* v ‘E,, 0 0 0

(1 - v)‘J$~ v ‘E23 0 0 0

1 (1 -v)‘J$3 0 0 0

c=(l+“)(l-2”) $1 - 2v) ‘E,* 0 0

symmetric gl - 2v) ‘E,3 0
$1 - 2v) ‘E23

(7)
2.1.2. Multiaxial conditions. The behavior of con-
where v is the Poisson ratio, and the shear modulus
crete under multiaxial stress conditions is very com-
in a coordinate plane is calculated from the weighted
plex and has not been assessed experimentally in a
Young’s modulus corresponding to that plane,
complete manner. Various material models with con-
siderable simplifying assumptions have been pro-
posed in the literature. Considering the variability of ‘Eij 1 1‘Upi1‘$i + 1'Upj1 ‘Epj
‘G, = = ~
concrete materials that need be described in practice, 2(1+V) 2(1+V) I’UpiISI’UpjI ’ (*)

the objective in this work was to develop an effective

but simple model that provides sufficient flexibility to The above stress-strain relations for material load-
the analyst to fit the basic material behavior. ing conditions are only employed in the calculation of
The stress-strain relations are evaluated differently the stiffness matrix at time t. Considering the evalua-
depending on whether the material is loading or tion of the stress increment, u, from time t to time
unloading. t + At, the stress integration is approximated in the
To characterize loading and unloading conditions following manner,
we define a loading function ‘J
u = ee (9)
If= ‘f (5)
where e is the strain increment. The stress-strain
where ‘9 = (4 ‘s~‘s~)“2, ‘Sij= ‘dii - ‘a,&, ‘a, = l/3 ‘u,,, matrix C is as C in eqn (7) but the uniaxial Young’s
and 6, is the Kronecker delta. The material is loading moduli ‘$ are replaced by the moduli ‘&. These are
except when the unloading conditions are deter- evaluated using the uniaxial stress-strain relationship
566 KLAUS-JORGEN
BATHEet al.

(a

ENStLE FAILURE

Fig. 2. Triaxial failure envelopes reduced to two-dimen-

sional conditions.

Fig. 3. Triaxial tensile failure envelope.

in Fig. 1, namely, for ‘GoB 0 we have

qi = $ (10) 2.2. Material failure emelopes

To model the failure of the material in tension and
and if IcPic 0, we define
compression in two- and three-dimensional analysis
r+Ar f
and to account for multiaxial conditions in the
ap1=
uniaxial stress-strain law of Fig. 1, failure envelopes
are employed. Figure 2 shows an example of the
failure envelopes used corresponding to two princi-
pal stress directions. Regarding the tensile failure it is
noted that considering one principal stress direction
and then the tensile strength of the material in this direction
does not change with the introduction of a tensile
stress in the other principal stress direction, but a
(12) compressive stress decreases this tensile strength.
The failure envelopes shown in Fig. 2 are a special
Here, ‘+“ePi and ‘ePiare the strain components at time case of the triaxial failure envelopes for tensile and
t + At and t in the directions of the principal stresses. compressive failures shown in Figs 3 and 4.

- INPUT TO ADINA
---KHAN AND SlWGY

Fig. 4. Triaxial compression failure envelope.

 Nonlinear analysis of concrete structures 567

The triaxial tensile failure envelope is a straightfor-

ward generalization of the envelope for two-
dimensional analysis. However, the compressive
failure envelope is more complex and requires more
data input. The envelope can be used to represent a
large number of different envelopes like the biaxial
envelope of Liu et al. [7] and the triaxial failure
surface of Khan and Saugy[8]. The shape of the
compressive failure surface used is largely based on
the experimental results reported by Kupfer et al. [6]
and Launay and Gachon [9].
The compressive failure envelope is input using 24
discrete stress values. Firstly, the values crb,/dc are
input. These values define at what stress magnitudes
‘a,, the discrete two-dimensional failure envelopes for COMPRESSlC+4
additional stresses ‘upzand ‘cprare input. These failure
STRAIN
envelopes are defined by the failure stress values
bp$/c?~(i = 1, . . . ,6;j = 1,2,3) that correspond to the Fig. 5. Increase of strength parameters for model under
stress magnitudes ‘up1= ‘tag,, ‘ap2= /I ‘ups(B is a con- multiaxial conditions.
stant) and ‘up2= ‘up,.
The failure envelopes are employed to establish the
uniaxial stress-strain law accounting for multiaxial state in the failure envelopes. In the following we
stress conditions, and to identify whether tensile or consider how one single plane of tensile failure devel-
crushing failure of the material has occurred. Having ops and how the material fails in compression crush-
calculated the current principal stresses, to establish ing.
the uniaxial stress-strain law it is assumed that ‘up, Tensile failure occurs if the tensile stress in a
and ‘up2 are held constant and the minimum stress principal stress direction exceeds the tensile failure
that would have to be reached in the third principal stress. In this case it is assumed that a plane of failure
stress direction to cause crushing of the material is develops perpendicular to the principal stress direc-
calculated using the failure envelopes (see Fig. 4). Let tion. The effect of this material failure is that the
this stress be 8: and y, = a:/~?,, then we also use normal and shear stiffnesses across the plane of
failure, and the corresponding normal stress and
5; =y,d, (13) shear stresses, are reduced.
Prior to tensile failure the stress-strain material law
e’: = (C, r: + C,Yl)% p:=(C,Y:+C,Yl)& (14) is given by eqns (7)-(12). Assuming that ‘upi is larger
than the tensile failure stress, the new material
where C, and C, are constants defined by the user; stress-strain matrix is in the stiffness matrix calcula-
tion,

1 1
0 0 0 0 0

-[I -v2
1 ‘I!& v‘$3
‘E2, 0
0 0 0

i.2
0%
2(1 + v) 0 0 ’

symm.

normally, C, = 1.4 and C, = -0.4. The constants where the ‘Epi are the uniaxial Young’s moduli
8:) Cl, Pi, 5: are employed instead of the unprimed evaluated in the principal stress directions using eqns
variables in order to establish, using eqn (4), the (2) and (4) and the 'E,, are evaluated using eqn (8).
uniaxial stress-strain law under multiaxial conditions The constant rl. is a small value, normally lo-‘, to
(see Fig. 5). reduce the stiffness normal to the tensile failure plane,
To identify whether the material has failed, the and the constant b, normally 0.5, reduces the shear
principal stresses are used to locate the current stress stiffness in the tensile failure plane.
 568 KLAUS-JORGEN
BATHEet al.

For the stress calculation, the following stress-

strain matrices are used.
For the tensile stress normal to the tensile failure
plane and the shear stresses in this plane, we use the
total strains to calculate the total stresses with

C=IyIrn. Gp/, G;] (‘6)

STRAIN NWIMAL

RILURE PLANE
where Ef and G$, G{, are evaluated as shown in
Fig. 6. In this figure 5 is a variable, input by the user,
that defines the amount of tension stiffening. Figure
6 shows that G$ is a fraction of the initial shear
modulus, &,/2(1 + v), corresponding to the total
strain normal to the tensile failure plane. Also, Fig. 6
assumes loading from zero stress directly into the
tensile region. If the tensile stress is reached by
unloading from a compressive stress, the strain nor-
/ 1
I

mal to the tensile failure plane is measured from the St t0 % FAILURE PLANE

strain at which the stress is zero (see the example of Fig. 6. Material moduli for stress calculation under tensile
Fig. 7). failure. (a) Calculation of Young’s modulus E,, normal to
tensile failure plane. (b) Calculation of shear modulus in
For the remaining stress components, the incre- tensile failure plane.
ment in the stresses is evaluated from the incremental
strains using failure is still active. The failure is considered to be
inactive provided the normal strain across the plane
’ ‘Q 0
(17) becomes negative and less than the strain at which the
symm. (1 - v’)‘G,, ‘last’ failure occurred and is active otherwise. There-
fore, a tensile failure plane may repeatedly be active
where the r$i are the uniaxial Young’s moduli eval- and inactive.
uated using eqn (12) and ‘E2, and TG23are evaluated If a tensile failure plane has developed, which may
using eqn (8) but with r&i instead of ‘J!&. or may not be active, the material stress-strain
Considering eqns (15), (16) and (17), we note that relations are always established as described above
by use of eqn (16) the tensile stress normal to the but corresponding to the principal stress directions in
tensile failure plane and the shear stresses in this the failure plane and the direction perpendicular to
plane are gradually released. Also, the equations this plane. Hence, instead of using the principal
show that plane stress conditions are assumed to exist stresses and corresponding directions as done for the
at the plane of tensile failure. The factor tf, is not set unfailed material, the stress conditions along and
exactly equal to zero in order to avoid the possibility normal to the material tensile failure plane are used
of a singular stiffness matrix. The value to be em- to evaluate the stress-strain matrix. Also, when a
ployed for nS must depend on a number of physical failure plane is or was active, a subsequent failure
factors, and in the numerical solution it is at this plane is assumed to form perpendicular to the direc-
time best to leave nJ as a variable that is input by the tion of the one that developed first, once a normal
analyst. stress along the original failure plane has reached the
For visualization, the ‘plane of tensile failure’ is tensile failure stress. It follows that at any integration
referred to as a ‘crack’, but we should interpret this point, the direction of the third tensile failure plane
terminology judiciously because a physical crack does is fixed once failure has occurred in two directions.
not actually develop at the element integration point. It may also happen that after tensile failure of the
Instead, the material has failed in one principal stress material (in one or two directions) the material fails
direction. in compression crushing, which is identified, as usual,
Equations (15), (16) and (17) describe the solution by entering the compression crushing envelope in
when tensile failure occurs. To identify compression Fig. 4 with the principal stress(es) that act(s) along
failure, the largest principal stress IuP, is employed to the tensile failure plane(s).
establish from Fig. 4 by interpolation, the biaxial If the material has crushed in compression, it is
failure envelope on ‘up2 and ‘ups. The material has assumed that the material strain-softens into all
crushed if the stress state corresponding to ‘rr,,*and directions.
‘a,, lies on or outside the biaxial failure envelope.
2.4. Post-compression crushing, strain-softening
2.3. Post-tensile cracking behavior behavior
Once a tensile plane of failure has formed, it is Consider first uniaxial stress conditions. As shown
checked in each subsequent solution step whether the in Fig. 1, for a uniaxial strain smaller than &, the
 Nonlinear analysis of concrete structures 569

material has crushed and softens with increasing 2.7. Example solutions
compressive strain, i.e. ‘1 is negative. In the following, we present some simple, small-size
Under multiaxial stress conditions the compression applications to demonstrate the details of the model
crushing is identified using the multiaxial failure behavior when it is subjected to various load condi-
envelope, and once the material has crushed isotropic tions. In’ all these examples a constant Poisson ratio
conditions are assumed using the uniaxial v is used.
stress-strain law with the constants a:, Pi and so on Figure 7 summarizes the response obtained when
(see Fig. 5) corresponding to the multiaxial condi- subjecting a single plane stress element to biaxial
tions at crushing. The Young’s modulus for the strain-controlled loading. This solution includes ini-
isotropic material law is evaluated using tiation, closing and reopening of a crack; loading,
unloading and reloading with large load steps; and
crushing of the material with strain-softening behav-
ior. In this figure, the numbers indicate load steps
where ‘epj and eps are the strain component and in the analysis. Note that the material model can
incremental strain component at time t measured in be used with relatively large load step increments.
the direction of the principal stress ‘rrp,. Figure 8 shows the material failure for different load
Note that, as shown in Fig. 1, when ‘epsbecomes steps indicated in Fig. 7.
equal to or less than P:, the stresses are set to zero. In the next example solution a fully constrained
However, an option is available for a gradual stress concrete specimen modeled by four 4-node plane
unloading in which case the constant Young’s mod- stress elements is subjected to a uniform temperature
ulus & = -5:/c: is employed, i.e. in eqn (18) we then loading ‘8. The initial temperature is zero degrees and
use IS=&. the temperature is raised to 450°C leading to material
If unloading of the crushed material in the strain- crushing, see Fig. 9. Note the increase in the crushing
softening region occurs characterized by e,r > 0, the strength of the material is due to the biaxial stress
initial Young’s modulus $ is used. In all these conditions.
calculations the principal stresses are checked individ- Next, in Fig. 10, a simple plane stress concrete
ually whether a positive value has been reached, and element subjected to biaxial compression loading is
if so, the stress in the corresponding direction is set considered. After crushing, the material relation of
to zero. Sec. 2.4 is used, i.e. all Young’s moduli are set to &.
In a further example solution, a single 2D plane
stress element is loaded to trace the post-cracking
2.5. Poisson’s ratio in the compressive region
behavior of the concrete material model under biaxial
It has been observed in experiments that the ratio loading in comparison with uniaxial loading (see Fig.
of lateral strain to principal compressive strain re- 11). The shift in the stress-strain 1D response de-
mains constant until approximately 80% of the max- pends on the crp,, up2relation (positive/positive, posi-
imum compressive stress d,. tive/negative) as well as on the magnitude of the
Usually we assume in our analyses that the Poisson Poisson ratio. It should be noted that strain-softening
ratio is constant; however, as an option a value, vr, post-cracking models (r > 1.0) are to be used with
can be used when the material dilates under compres- caution since such models may lead to a nonunique
sion. The value of vr is given by [lo], solution [14] (see also Sec. 3.2).
Extensive cracking and complex post-cracking be-
vs = v whenY~=:QY.
‘Q havior can be observed in the analysis of concrete
pressure vessels. In this part of the paper we present
detailed solutions of two ‘reduced’ models. Analyses
v~=v,-(v~-v)/~ when y2>yo of the complete pressure vessel structures are pre-
sented in Part II of the paper.
Figure 12 shows a simple finite element model of
(19)
a part of the Sandia pressure vessel. This model repre-
where v is the initial Poisson’s ratio, v, is the maxi- sents a slice of the vessel at the elevation 5.969 m. The
mum value of Poisson’s ratio at failure and y. is slice is subjected to boundary and loading conditions
usually set to 0.8. corresponding to the whole structural behavior when
the vessel is under internal pressure. Three eight-node
axisymmetric elements were used to model the con-
2.6. Temperature eflects
crete of the pressure vessel. The l-node axisymmetric
In some analyses temperature strains, ‘e”, need to and 2-node truss elements were used to model the
be included. These strains are taken into account by steel reinforcements. Figure 13 shows crack distribu-
replacing the total incremental strain, e, in the gov- tions for different load levels. In Fig. 14 a comparison
erning incremental equations by the strain e - elh, with experimental data is presented. Note that in
where e” is the thermal incremental strain and is Fig. 13 cracks are plotted at the integration points,
calculated from the temperature conditions. the size of a cracking symbol depends on the con-
 KLAUS-JORGENBATHE et al.

s
zz-SS3MlS II-SS3YlS

zz-SS3LllS AA-SS3YIS
 Nonlinear analysis of concrete structures 571

C.A.S. 32,3.4-F
512 KLAUS-JURGEN
BATHEet al.
 Nonlinear analysis of concrete structures 513

.I0 .I5 4

TMREE S-NODE AXISVMTRIC

CONCRETE ELEMENTS
8
2 5

.I3 ! .I@

SEVEN 2-NODE (LINES)

AND
ELEVEN I-NODE (RINGS)
PRESTRESSING TRUSS
ELEMENTS

Fig. 12. A finite element model for a slice of the Sandia pressure vessel at elevation 5.969 m (see also Fig.
19). Top: 2D axisymmetric concrete elements. Bottom: prestressing truss elements.

tributing volume at the integration point, circles model. The model can be employed in a number of
represent hoop cracks and two parallel lines a short important concrete analyses, notably for collapse
distance apart represent cracks in the Y-Z plane. A analyses of two- and three-dimensional concrete
smaller distance between the parallel lines is used structures. One important aspect is the generality of
when the crack is closed. the model with respect to the stress-strain law and
A simple model of a part of the PWR pressure failure surfaces used. Also, the model is deemed
vessel (this vessel is also considered in Part II of the computationally efficient because relatively large in-
paper, see Fig. 30) is shown in Fig. 15. As for the cremental load steps can be employed (see Figs 14
Sandia vessel, a slice of the PWR vessel is considered. and 17).
This slice now corresponds to the height 22.30 m. Although a powerful model already, further im-
Two 8-node axisymmetric finite elements were used provements to the concrete model are surely desir-
to model the concrete of the vessel. The steel hoop, able. Some features that would appropriately gain
longitudinal and radial reinforcement is modeled further research attention are the modeling of time-
using l-node axisymmetric and 2-node truss ele- dependent behavior, a more accurate representation
ments. The model is subjected to boundary and of the post-crushing response and further refinements
loading conditions that correspond to the response of in modeling the tensile-stiffening behavior, the effects
the complete structure. The crack distributions for of Poisson’s ratio and cyclic loading conditions.
different solution times are shown in Fig. 16. Note
that three cracks have been formed at some of the
3. PART n: APPLICATIONS OF ADINA 5.0 TO THREE
integration points at the final load level. In Fig. 17 the TEST STRUCFURE5
solution results are compared with the results of an
analysis using a simple force-balancing method. Taylor Woodrow plc (TW), a worldwide engineer-
ing group, has been involved in the design and
2.8. Concluding remarks construction of nuclear power structures for over 30
The objective of Part I of this paper was to years. During this period much knowledge has been
summarize the current status of our concrete material gained on the short- and long-term behavior of
 (bl
LY

4 I t* I I<,1 t I<, STRESSES CRACKlNG AN0 STRESSES

AT INTECRATlON POINTS AT INTEGRATION POlNTS
FOR PRESSURE FOR PRESSURE I
3.17E*OS N/II**2 3.66E*05 N/H**2

CRACKlNG AN0 STRESSES CRACKING AN0 STRESSES

AT INlECRATlON POINTS AT INTEGRATION POINTS
FOR PRESSURE FOR PRESSURE
6.6261+05 Nltl*+Z I .S.BE’GS N/t+--2
I 1 I

Fig. 13. ~elopment of cracks in the Sandia model of Fig. 12. In 3D analysis a crack at an integration point is represented by two disks a small distance apart,
the crack lying in the plane between the disks. Hence, in this figure parallel lines represent cracks in the Y-Z plane; circles represent hoop cracks. The size of the
two disks corresponds to the contributing volume at the integration point.
 Nonlinear analysis of concrete structures

STEP 12
OFF GRAPH

ELEVATION lS.Q6QM
I

0 ADINA. I2 STEP SOLUTION

+ EXPERImNlAL RESULTS

lIO5
PRESSURE lN/M**2)

Fig. 14. Displacement-pressure response of the Sandia pressure vessel.

concrete and a comprehensive degree of expertise has within some of the improvements to the concrete
been accumulated in the design, analysis and testing material model that are found in the latest ADINA
of a wide range of concrete structures. TW has 5.0 code.
extensive in-house testing facilities which at first were The work presented in this section was originally
used in support of the nuclear industry and later have carried out using the modified ADINA 8 1 system but
been involved in a wider range of concrete structures has subsequently been rerun using the latest ADINA
including the offshore, civil and conventional 5.0 version, these later results being presented now.
building areas. This test work has supplemented Three applications are given of which two represent
the analytical work which itself has augmented the situations where the original analysis was carried out
testing. prior to the test results being available. The third
To increase its analytical capabilities it was decided analysis supplemented the design but will eventually
to acquire a nonlinear finite element system which be used in conjunction with a physical model at
included a concrete material model. After reviewing present under construction in the Taylor Woodrow
a number of alternatives TW obtained the ADINA 8 1 Laboratories.
code in 1983. In using the package on a wide range
of reinforced concrete structures a number of poten- 3.1. One-sixth scale model of a reinforced concrete
tial improvements to the concrete model could be containmeht
envisaged, and TW decided to add in extra features The U.S. Nuclear Regulatory Commission (NRC)
and change existing ones (see [lo, IS]). The form of has initiated the construction and testing of a series
these changes and their influence were shared with of scale models of containment buildings. This
ADINA R & D, Inc. and have been incorporated formed part of a containment integrity program, one

i.

Fig. IS. A finite element model for a slice of the PWR pressure vesselat elevation 22.30 m (seealso Fig. 32).
576 KLAUS-JGRGEN
BATHEet ai.

-
-

- - -
~
- -
I
 Nonlinear analysis of concrete structures 511

t FORCE-BALANCING METHOD

Fig. 17. Loadclisplacement response for the PWR model of Fig. 15.

_i

1
WING LINL -

._

Fig. 18. General arrangement of the Sandia pressure vessel. It is a 1/6th scale model.
578 KLAUS-J@RGEN
BATHEet al.

IDINA ORIGINAL XVMN 0.000 7

XWIAX 3.810
05700 Ywll% 0.000 i
‘1V”AX t t *x3

Fig. 19. Finite element mesh of the Sandia pressure vessel.

I I

Fig. 20. Compressive failure envelope used for analysis of the Sandia pressure vesse1. Comparison of
experimental values, ADINA input values used in the analysis and ADINA default values.
 Nonlinear analysis of concrete structures 519

item of which was the construction of a one-sixth The physical model contained a steel liner.
scale model of a reinforced concrete containment. ADINA does not as yet contain an axisymmetric
This structure was to be analyzed prior to testing by shell element, therefore the liner was modeled using
a number of participating organizations, four from hoop and meridional truss elements with due al-
the U.S. and six from Europe. One of the groups lowance for the Poisson’s ratio effect that would be
submitting an analysis was sponsored by the Central in the plate but which is missing in the equivalent set
Electricity Generating Board (CEGB) of the United of truss elements.
Kingdom. Work was carried out for the CEGB by the Finally the soil was modeled as a set of spring
Nuclear Design Associates (NDA), a joint venture of elements. Due to the relatively large thickness of the
Taylor Woodrow and Sir Robert McAlpine. bottom cap the choice of spring stiffnesses was not
The analysis formed part of a ‘Round Robin’ expected to significantly influence the final ultimate
pretest series of analyses, these being documented in load.
a NRC document [16] and formed part of the NDA The model was constructed by Sandia National
validation work for the use of the ADINA code [ 17) Laboratories and tested in July 1987, by which time
Details of the test structure are shown in Fig. 18. It all analysis work had been carried out, supplied to the
can be seen that whilst in general being an axisym- NRC and disseminated in a report of May 1987 [ 161.
metric structure, some degree of divergence from this The test consisted basically of an initial low pressure
occurs at major penetrations, and with the disposi- cycle up to 1.15 times design pressure. With design
tion of reinforcement. The ADINA analysis carried pressure at 0.317 MN/m2 this represents an initial
out was assumed to be axisymmetric, the local threc- loading of 0.365 MN/m2. After returning to zero
dimensional effects being investigated by separate pressure a high pressure test was performed up to the
linear analyses. ultimate pressure sustainable by the structure. In the
The finite element mesh produced for the ADINA analysis the initial cycle was not included, the loading
analysis is shown in Fig. 19. The concrete properties consisting of a gradual application of pressure until
used in the analysis were as follows, failure was indicated.
The following commentary will help to describe
both the ADINA analysis findings and the complex
initial E value ($) = 24,800 MN/m2
behavior predicted.
crushing stress limit (8,) = -41 MN/m2
crushing strain (&) = -0.00186
0.138 MN/m’-Cracking occurs at the wall base
ultimate stress (6”) = -30 MN/m2
junction due to bending. This represents less than
ultimate strain (&) = -0.0032
44% of design pressure, but subsequent cracking in
tensile limit (5,) = 4.1 MN/m2
this area progresses across the section in a controlled
Poisson’s ratio (v) = 0.2.
manner. The outcome is that a ‘plastic’ hinge forms
here.
Compressive failure envelopes were based on pub- 0.345 MN/m2-Vertical cracking (that is cracks
lished results [18], these being shown in Fig. 20 in due to hoop stresses) commence at the mid-height
order to indicate how they deviate from the ADINA and quickly spread to reach the dome spring line at
default values. In general TW has used these failure 0.365 MN/m2.
curves for most analyses as it is felt they fit a wider 0.400 MN/m2-Horizontal cracking (due to bend-
range of test results than the default values, although ing and axial tension) appears on the outside face of
the deviation is only of concern in the higher triaxial the wall and quickly spreads through the thickness
stress states. As an aside, it is TW’s policy when using and along the wall towards the dome and wall base
ADINA to try to have a standard approach wherever junction. By about 0.55 MN/m2 the dome and wall
possible, therefore when dealing with conventional are almost completely cracked through, the pressure
reinforced concrete most input parameters are un- being resisted by the liner and reinforcement.
changed from one analysis to the next apart from the 0.552 MN/m2-Cracking occurs at the middle of
compressive limit. This even covers the choice of the underside of the bottom cap. By 0.855 MN/m2
tensile limit where, unless otherwise directed by cir- these cracks have extended across the entire under-
cumstances or direct evidence, the value is assumed side.
to be 10% of the compressive limit. 0.565 MN/m2-Initial yield of the liner at mid-
The reinforcement was treated as a multilinear height in hoop direction, again rapidly extending up
elastic model. The main vertical and hoop reinforce- and down the structure.
ment could be modeled easily but towards the bottom 0.855 MN/m*-Inside hoop reinforcement at mid-
of the wall there existed diagonal seismic reinforce- height reaches first yield.
ment. It was considered sufficiently accurate to model 0.890 MN/m*-All layers of hoop reinforcement at
this as an equivalent extra set of vertical and hoop mid-height have yielded resulting in rapid expansion
bars. Orthogonal bars in the base were likewise of the wall.
replaced by equivalent replacement hoop and radial 1.000 MN/m2-By this pressure most reinforce-
bars. ment has yielded.
580 KLAUS-J~GEN BATHEet al.

3 ] ‘1 i‘ . !^! T 5. 969M

* .
I I
2. 4. 6. a. 10. 12.
#lo5
PRESSURE (N/Mrtx21

Fig. 21. Analysis and experimental displacement response at the mid-height of the radial movement.

1.100 MN/m*-Crushing of outer wall at wall-base At a conference in Denmark in 1988, attendees

junction followed by complete failure at this location. were invited to predict the failure load of one of the
beams that had been tested, but for which the results
It should be pointed out that in most work to date had not yet been disclosed. The results of this test
TW has preferred not to use iterations but to progress were revealed in a subsequent report [19] together
the load at a controlled and sufficiently slow rate with with photographs of crack patterns on similar tested
holds of pressure at chosen pressure levels. This, beams.
particularly with our earlier work, seemed to allow The analysis was carried out on the test beam
the analysis to progress further but care was neces- shown in Fig. 24. The beam deviates from a conven-
sary especially at points of rapid change and near tional test by virtue of the ASR, and from the lack
ultimate. of shear reinforcement in the zone of maximum shear.
A few figures have been produced to illustrate the The mesh used is shown in Fig. 25. The relevant
behavior of the containment. Figure 21 shows the properties chosen for the concrete were based on
comparison between analysis and experiment of the cores removed from the beam. Both vertical and
mid-height radial movement. This is the dominant horizontal cores were taken, the former being 40%
movement in the structure and so a good choice for stronger. Being unable to incorporate this non-
comparison purposes. As can be seen the actual test isotropic behavior a mean compressive strength of
reached 1.0 MN/m2 at which stage excessive leakage 33.0 MN/m* was used together with a tensile strength
occurred, probably through a large liner tear near to of 3.3 MN/m*.
a major penetration. The correlation between the two The analysis reached an ultimate load of 132 kN at
curves is very high, especially as the small deviation each load position. In the experiment the specimen
over the range O-O.4 MN/m* can be accounted for by first failed with a shear failure on one side at
the initial low pressure test carried out. Reference to 127.4 kN.
the table of events indicates that much cracking could Figure 26 shows the load-deflection curve at mid-
have occurred, therefore on repressurization these span and it can be seen that the final failure was fairly
regions would have zero tensile strength. abrupt even though a great deal of cracking had
Two other plots (Figs 22 and 23) show global views occurred in both the flexure and shear zones. This
of the displaced shape and the extent of cracking. could be interpreted as basically indicating a shear
Figure 22 shows global views of the displaced shape failure. Figure 27 shows the extent of cracking at
at various pressure levels, all displacements being various load levels whilst Fig. 28, which plots the
drawn to the same scale. Figure 23 shows the extent largest strains, gives a clear picture of the gradual
of cracking at various pressure levels. flexural damage, initiating on the mid-section bottom
face very early on, and progressing slowly through
3.2. Alkali-silica reacted beam the depth and outwards towards the supports. The
During the period 1984-1989 the Road Directorate subsequent shear damage is seen to be. quite sudden.
of Denmark initiated a project to study the load Comparisons with a similar beam can be made in
carrying capacity of structural members subjected to Fig. 29 which shows the same form of progressive
alkali-silica reaction (ASR). One item of this in- damage with a rapid change on the left hand side at
volved studying the shear strength of concrete beams ultimate load.
subject to ASR. On studying the Danish report it was seen that the
 Nonlinear analysis of concrete structures 581

-...::..+ 1
7 ..:.>.. “.>.
*+.
y..
:..

PRESSURE*O.345 NPA RESSURE- 0.69 WA

________
_-_.-
_._ : __________--.-

1_______e-m-e....
1__________
_...

i” i,

Fig. 22. Global view of the displaced shapes of the Sandia pressure vessel. Displacements magnified by
a factor of 167.
582 KLAUS-JORGENBATHEet al.

i, i”

PRESSURE.0.345 WA PRESSURE-O.366 WA

-i

PRESSURE-O. 5175 WA PRESSURE.0.8549 MPA

r
! c,
Fig. 23. Global view of the extent of cracking in the Sandia model. In the original, yellow-l crack,
blue-2 cracks, red-3 cracks.
 Nonlinear analysis of concrete structures 583

I Heasure
LOAD DIAGRAM I"mm

nchoraqe Zone 5hear Zone Bendlnq Zone Shear Zone Anchoraw Zon

Fig. 24. Alkali-silica reacted beam. Geometry, reinforcement and load diagram.

Fig. 25. Finite element model, 8-node plane stress elements for the beam of Fig. 24.

DISPLACEHENT HEASURED AT POINT A

MID-SPAN DISPLACEHENT IHETRES)

Fig. 26. Load-displacement response as predicted by ADINA for the alkali-silica reacted beam.
584 KLAUS-J~GEN BATHEet al.
 Nonlinear analysis of concrete structures 585

% Load of
Utimate

75

88

100

100

Fig. 29. Experimental results of developments of cracks during loading of an alkali-silica reacted beam.

ASR beams were judged against unaffected speci- Study of the crack dispositions and strains still
mens. On testing these it was found that very early indicates a smearing of cracks instead of the expected
shear failure initiated in the shear zone resulted in dominant single crack. Whilst of not too much
lower strengths even though the unaffected concrete significance in conventionally reinforced structures,
had a higher strength. It was judged that the ASR this phenomenon could be more important in analyz-
caused a great deal of micro-cracking of the concrete ing problems involving mass concrete.
making the beam more ductile, hence redistributing
the stresses more efficiently in the shear zone.
3.3. Size&l ‘B’ PWR prestressed concrete contain-
To investigate this it was decided to rerun the
analysis with a higher concrete strength of ment
58.2 MN/m2, this being the quoted cylinder strength The main difference between this containment and
of the undamaged concrete. Just making this change the Sandia test structure is that the Sizewell structure
resulted, as expected, in an increase in ultimate load, is prestressed by longitudinal, radial and hoop ten-
contrary to the test results. On reflection, the shear dons. Nonlinear analyses have been applied to the
zone is basically unreinforced concrete, therefore full scale structure now under construction at
cracking should be concentrated in fewer cracks and Sizewell, United Kingdom and to a tenth scale model
should propagate more rapidly. This is very similar to which, at the time of writing, is under construction at
the situation of no tensile stiffening after failure and the Taylor Woodrow Laboratories.
a reduced shear reduction factor and hence a run was Figure 30 shows the form of the structure which is
made to use a tensile stiffening factor of 1.2 instead in essence axisymmetric except for the major wall
of 8 and a shear reduction factor of 0.01 instead of penetration (not indicated on this diagram), the
0.5. Whilst the problem showed much more distress tunnel in the base and the disposition of prestress.
in the shear zone it reached higher than expected The latter is provided by hoop tendons that cover
loads. overlapping 240” segments of the wall and the lower
 m I .T

Fig. 30. PWR prestmssed concrete containment section through reactor building.

Fig. 31. PWR containment, tenth scale model: some structural details.
 Nonlinear analysis of concrete structures 587

DINA ORIGINAL XVMlN 0.000

XVMAX 2.512
0.5126 YVHIN 0.000
YVMAX 6.981 i,

Fig. 32. Finite element mesh for the PWR containment model.

portion of the dome together with longitudinal ten- During the pressurization process there are well
dons that cross the dome orthogonally in plane. The defined stages at which the stress distribution could
hoop tendons are anchored on three buttresses and be calculated, these being during the early elastic
the longitudinal tendons at the prestressing gallery period when the load is shared by the concrete,
below the base. reinforcement and prestress, following the total
The model details are shown in Fig. 31 which cracking of the concrete with transfer of the load fully
follows closely the full size structure except for details onto the reinforcement and prestress and finally after
in the bottom cap. The mesh used for the tenth scale yielding of the reinforcement when the load is carried
model analysis is shown in Fig. 32. A similar mesh solely by the prestress. These conditions result in spot
was employed on the full size structure, which was checks and are shown in Fig. 35 which plots the
also being subjected to a more detailed analysis in predicted radial deflection of the actual vessel at
which the element size was roughly halved through- mid-height against the pressure. Close agreement is
out. A comparison of the two sets of results showed indicated with the ADINA analysis, the latter show-
that for this structure the use of two elements through ing the effect of the post-tension stiffening during the
the wall was sufficient. period in which the load is transferred from the
Figures 33 and 34 show progressive deflection concrete onto the steel.
behavior and crack build up at various pressure As in most analyses, there are areas of approxima-
levels. The accuracy of these predictions will only be tion and concern where known behavior has by
apparent after the forthcoming pressure tests but necessity been ignored. In this analysis one such
during the design process certain hand checks were approximation is centered around the prestressing
made. One of these involved the radial displacement system. The three-dimensionality of the prestressing
of the wall at mid-height which is sufficiently removed system both from the hoop and particularly in the
from the wall-base junction and the dome spring line meridional prestressing could be addressed by means
as to make the behavior amenable to solution as a of three-dimensional analyses although due to the
long cylinder. various cyclic symmetries involved this may need to

C.A.S.
32,M-G
588 KLAUS-JORGEN BATHEet al.

PRESSURE-.345 MPA PRESSURE-O.

Fig. 33. Progressive deflection behavior at various pressure levels for the PWR containment model as
predicted by ADINA. Disphcements magnified by a factor of 66.
 Nonlinear analysis of concrete structures

PRESSURE.0.345 HPA PRESSURE-O.592 HPA

PRESSURE-O.69 MPA PRESSURE-O.76 WA

Fig. 34. Progressive crack build up at various pressure levels for the PWR containment model. In the
original, yellow-l crack, blue-2 cracks, red-3 cracks.
590 KLAUS-J~JRGEN
BATHEet al.

consideration are solution strategies, especially with

- RAOIRL DISP. FIT 22.30M new automatic loading methods becoming available,
bond between reinforcement and concrete and non-
orthogonal cracking.
Concrete in all its many forms still remains a very
complex material to model numerically but given the
open exchange of ideas and mutual respect between
code developers and structural analysts each success-
ful application helps to meet tomorrow’s challenges
with an extra degree of confidence.

REFERENCES
I. K. J. Bathe, Finite Element Procedures in Engineering
Analysis. Prentice-Hall, Englewood Cliffs, NJ (1982).
2. K. J. Bathe and S. Ramaswamy, On three-dimensional
nonlinear analysis of concrete structures. Nucl. Engng
Des. 52, 385-409 (1979).
3. W. F. Chen, Plasticity in Reinforced Concrete. Mc-
Graw-Hill (1982).
4. Z. P. Bazant and J. C. Chem, Concrete creep at variable
humidity: constitutive law and mechanisms. Mater.
RD1NQS.C
Struct. RILEM. 18(103), 1-19 (1985).
0 FORCE-BRLANCING METHOD 5. S. Gallegos-Cazares and W. C. Schnobrich, Effects of
creep on the behavior of reinforced concrete gable roof
I--r-r-r-
120. 150. hyperbolic-naraboloids. Civil Ennineerina Studies.
Siructural Besearch Series No. 543,Unive&ty of Illi:
0 ISPLACEMENT (MM1 nois at Urbana-Champaign (1988).
Fig. 35. Radial deflection of the PWR containment at 6. H. Kupfer, H. K. Hilsdorf and H. Rusch, Behavior of
mid-height as a function of applied pressure. concrete under biaxial stresses. J. Am. Concr. Inst. 66,
656666 (1969).
be at the best a 180” analysis and may even require 7. T. C. Liu, A. H. Nilson and F. 0. Slate, Biaxial
a full 360” investigation. The test structure includes stress-strain relations for concrete. J. Struct. Div.,
all the features of the main structure that could ASCE 98, 1025-1034 (1972).
8. M. H. Khan and B. Saugy, Concrete for nuclear
influence this and the need for a full three- reactors. Am. Concr. Inst. Special Publication 34
dimensional analysis should be evident after the (1972).
pressure test. 9. P. Launay and H. Gachon, Concrete for nuclear mat-
A more complicating feature in this instance is the tars. Am. Concr. Inst. Special Publication 34 (1972).
10. J. S. Gedling, N. S. Mistry and A. K. Welch, Evaluation
fact that the prestress is unbonded. This is partly
of material models for reinforced concrete structures.
overcome for the vertical portion of the prestress as Comput. Struct. 24, 225-232 (1986).
only end loads are considered but for the curved 11. J. C. Pyle, Analysis of deep composite steel and concrete
tendons the interaction between displacements and slabs. Comput. Struct. 13, 661-671 (1981).
prestress forces is, whilst not being impossible, a very 12. M. Keuser, B. Kepp, G. Mehlhom and R. Rostasy,
Nonlinear static analysis of end-fittings for GFRP
complex calculation. ADINA contains means of car- prestressing rods. Comput. Struct. 17, 719-730 (1983).
rying this out as loads can be made deflection depen- 13. J. W. Tedesco and W. G. McDougal, Nonlinear dy-
dent and this has been demonstrated to be applicable namic analysis of concrete armor units. Comput. Struct.
to unbonded prestress, but the algorithm to cover the 21, 189-201 (1985).
14. Z. P. Bazant, T. B. Belytschko and Ta-Peng Chang,
general situation is rather formidable. Continuum theory for strain-softening. J. Engig Me&.
Div.. AXE 110. 16661692 (1984).
3.4. Discussion and conclusions 15. N. S. Mistry, Nonlinear analysis of reinforced concrete
The three structures detailed in the previous sec- structures. Finite Element News 4 (1985).
tions have helped to build up a high degree of 16. D. B. Clauss, Round-robin pretest analyses of a 1:6
scale reinforced concrete containment model subject to
confidence in the ADINA system and the concrete static internal pressurisations. U.S. Nuclear Regulatory
model. There is still the need of expertise in its Commission NUREG/CR-4913, SAND87-0891 (1987).
successful application and there are areas that need 17. S. Kloosterman, Validation report for the ADINA-TW
further investigation. In particular the effect of more finite element code. Report PWR/88/83, PMT Docu-
ment SXB-IC-096064 (19881.
discrete cracks as would occur in mass concrete needs
18. Task Committee on Finite ‘Element Analysis of Rein-
more study and from work (not yet published) involv- forced Concrete Structures. A. H. Nilson, Chairman,
ing explosive loadings, there is a need to model the State-of-the-art Report on Finite Element Analysis of
concrete past the ultimate strain state as a com- Reinforced Concrete, ASCE Special Publication (1982).
material. This may simply (Taylor 19. Vejdirektoratet (Road Directorate, Ministry of Trans-
paction
port, Denmark), Load carrying capacity of bridges
Woodrow’s word, not ADINA R & D’s) involve subjected to alkali-silica reactions. The shear strength
marrying the concrete model with the of concrete beams subjected to ASR Interim Report
Drucker-Praeer cao model. Other tonics for future No. 1 (1986).