Post-tensioned

Non-linear finite element concrete

analysis of post-tensioned structures

concrete structures
509
Z̆eljana Nikolić and Ante Mihanovic´
Faculty of Civil Engineering, University of Split, Croatia Received January 1996
Accepted November
1996

Introduction
Numerical modelling of non-linear prestressed structures has been studied by
many authors[1-6]. The model presented in this paper is applied to non-linear
analysis of post-tensioned concrete structures, partially or fully prestressed.
The stress in prestressed tendons is decreased as a result of various
influences which directly affect the bearing capacity of the structure.
Consequently, all changes should be computed as precisely as possible. The
model developed in this paper makes it possible to compute friction losses,
losses in the anchorage zone caused by concrete deformation and standard
losses caused by the short-term linear and non-linear deformation of concrete.
The prestressed tendons and reinforcing bars are embedded into the
concrete[7-9]. They were modelled by one-dimensional isoparametric three-
node elements. These elements make it possible to model generally curved
prestressing tendons and reinforcing bars. The influence of the prestressed
tendon on the concrete is modelled by distributed normal and tangential load
along the tendon and the two forces concentrated on the anchors.
The computation for post-tensioned structures is organized in phases. In the
phase which precedes the prestressing of the tendons the structure is computed
taking into account the dead load and one part of the permanent load. In the
prestressing phase the tendons are tensioned individually. The prestress force
can be applied at once or incrementally. In the third phase, which follows the
tensioning of all tendons, the structure is computed taking into account the
remaining part of the dead load and the live load.

Material modelling
A brief description of the material models adopted in this analysis is presented
herein. Further details may be found in [8,9].

Compressive behaviour of concrete

In the present analysis an elasto-viscoplastic approach[10] was employed to
model the compressive behaviour of the concrete. The yield condition was Engineering Computations,
Vol. 14 No. 5, 1997, pp. 509-528.
formulated in terms of the first two stress invariants[11-13]. © MCB University Press, 0264-4401
EC Beyond the yield surface, a viscoplastic response[10,14] was assumed for
14,5 concrete in compression. The standard associated flow rule[11-15] was assumed
valid for yielded concrete.
Concrete crushing is a strain controlled phenomenon which is monitored by
a fracture surface similar to a yield surface. The crushing condition can be
obtained simply by converting the yield criterion, described in terms of stresses,
510 directly into the strains[11-13,16].

Tensile behaviour of concrete

The response of concrete under tensile stresses was assumed to be linear elastic
until the fracture surface was reached. A smeared crack approach[11-13] was
used to model the behaviour of cracked concrete with opening and closing of
tensile cracks, taking into account the decrease in the tensile strength[11-13].
Cracks were assumed to form in planes perpendicular to the direction of the
maximum principal tensile stress as soon as this stress reaches the concrete
tensile strength. The shear stiffness in the cracked zones was modelled by the
reduction of the shear modulus[11-13].

Reinforcing and prestressing steel

Reinforcing bars and prestressing tendons exibit uniaxial behaviour in the
longitudinal direction and they were both modelled by a uniaxial elasto-
viscoplastic model[11,14].

Determination of tendon geometry

Prestressed tendons can occupy a general position within the concrete element;
they can be either straight or parabolic. They can also consist of straight and
parabolic parts (Figure 1). All tendons can be simulated in this way, whether
straight or parabolic, if the incontinuity of the first derivation is obtained
during the transfer from the region of one curvature into another.
The tendons are discretized with one-dimensional isoparametric three-node
finite elements which can occupy a general position on a two-dimensional
concrete element[7-9]. The tendon position is determined by nodes whose co-
ordinates are defined in a global co-ordinate system. In order to ensure
continuity between tendon elements, the boundary nodes have to be placed at
the intersection point of the tendon and the boundaries[7] of a two-dimensional

Figure 1.
Possibility of tendon
determination
concrete element (Figure 2). The paper uses an iterative procedure for Post-tensioned
interpolation co-ordinates of the central node[8,9]. concrete
structures
The prestress force transfer
The tendon force at any cross-section depends on the applied force and
prestress losses[17]. In this section the losses of the prestress force caused by
friction between concrete and the tendons will be discussed. Generally, it is 511
necessary to determine the prestress influence in the internal points of the
tendon if the external forces at the tendon ends are known.
The friction between the tendon and the concrete causes losses in the
prestress force. Figure 3 shows the forces acting on an infinitesimal element ds
of a curved tendon. The prestress force at distance s from the beginning of the
tendon will be denoted by S(s). This prestressing results in a distributed load

Figure 2.
Tendon position
determined by nodes

Figure 3.
Forces acting on an
infinitesimal element of
curved tendon
EC within the tendon, i.e. normal Pn(s) and tangential pt(s) on the concrete element.
14,5 Normal load at any cross-section of the tendon depends on the curvature radius
r(s) of the tendon and the intensity of the prestress force at that section[4,17]:

(1)
512 while tangential load pt(s) is the frictional force per unit length, given as:

(2)

where µ is the friction coefficient between the tendon and concrete[4,17]. The
force in the tendon at a distance s from beginning of the tendon can be
expressed as:

(3)
where SA is the known prestress force at one end of the tendon, while A is a
point at the beginning of the tendon (s = 0 in point A).
If equations (1) and (2) are introduced into equation (3), after the necessary
transformations, the force in the tendon at any point, at distance s from the
beginning, can be expressed as:

(4)

Possibilities of tendon prestressing

Tendons can be prestressed on one end or at both ends. If the tendon is
prestressed on one end applying force SA, then force SB at the other end of the
tendon can be computed according to expression (4) by integration along the
entire tendon length. The force at any cross-section of the tendon and the
distributed load can be obtained directly according to equations (1), (2) and (4).
If the tendon is prestressed from one end, and if the forces SA and SB at both
tendon ends are known, then the friction coefficient can be computed according
to expression:

(5)
If the tendon is prestressed on both ends, the force decreases, as a result of
friction between the tendon and concrete, if the distance from the end is
increased. In symmetrical prestressing, the problem can be simply solved since
the decrease in force is the greatest at the middle of the beam. For a beam with
length l the force in the tendon at distance l/2 can be calculated according to
equation (4), whereas the value g(s) is obtained by integration from the Post-tensioned
beginning to the middle of the beam. concrete
If the tendon is asymmetric or if the prestressing forces at the tendon ends structures
are not equal (SA≠SB), the procedure is more complex. The minimum force will
occur at the cross-section which has not been previously known. Let us denote
by x the distance of that cross-section from end A, and by l – x the distance from
end B. The force at that cross-section Smin can be computed according to forces 513
SA or SB by using one of the following expressions:

(6)

The forces should be equal regardless of the end where they were computed. By
equating terms in (6) we shall obtain an equation where the integration limit is
an unknown value. This equation is solved numerically. In this model,
asymmetrical prestressing is performed by taking a cross-section with the
greatest decrease in force. The prestressing force in a cross-section assumed
according to (6) is calculated before computing the entire structure. According
to the ratio between forces S Lmin and S Rmin, the assumed cross-section is moved
either to the left or to the right. This procedure is repeated for each tendon
separately in the phase of input data preparation. Subsequently, the structure is
computed and the possible difference between the two forces Smin to the left or
to the right from the selected cross-section can be ignored.

Numerical model for prestressing

Computation of the prestress force effects
After defining the tendon position, it is necessary to determine the influence of
the prestress force on the concrete. The structure is subjected to external
prestress forces on the anchors. The influence of the prestressed tendon on the
concrete is exerted by two forces at the ends, whereas there is a certain influence
which is computed as normal and tangential load per unit length of the tendon.
These values are computed in the co-ordinate system of the tendon and are later
transformed into a global co-ordinate system of the structure.
The tendon nodes are given by their global co-ordinates (xj*,yj*), (j = 1, 2, 3).
The co-ordinate of an arbitrary point can be expressed by the shape function
hj = hj(ζ) of a one-dimensional element, where ζ is an independent normalized
co-ordinate, and by global co-ordinates of the tendon nodes[15]:

(7)
where hj(ζ), (j = 1, 2, 3), are shape functions of a one-dimensional isoparametric
three-node element[8,9,15]. Differential element of the length ds can be presented
as[7-9,15]:
EC
14,5
(8)
while the values dx,dy are given by expressions:
dζ dζ
514

(9)
The prestress force at any point on the tendon can be computed according to (4),
where expression
s ds
g(s) = ∫
A r(s)

is a curvilinear integral which has to be solved by numerical integration. The

integration is performed numerically in the Gaussian points of the linear
element, so that values ds and r(s) have to be computed in those points. The
curvature radius r is defined as a reciprocal value of the tendon curvature k4
according to the expressions:

(10)
The radius of the tendon curvature is always a positive value. The direction of
the normal load depends on the curvature of the tendon[4] so that a sign should
be taken into account. The values
dx dy
x= , y′ =
dζ dζ
are given according to (9), while

Integer g(s) is comput for each finite element as:

(11)
where index i denotes the Gaussian points of one-dimensional element.
Subsequently, it is possible, according to the force at the beginning of the
finite element S1, to compute the force at the end of element S3. The force in the
central node is computed as the average value of the forces in the boundary
nodes of the tendon finite element. The prestress force at any point of 1D Post-tensioned
element can be presented as: concrete
structures
(12)

After computing the prestress force in all nodes, it is necessary to compute the 515
distributed normal and tangential load according to (1) and (2), where the
direction of the normal load depends on the sign of the tendon curvature.

Equivalent nodal forces

The influence of the prestressed tendon on the concrete is exerted by two
compressive forces at the ends of the tendon and the distributed normal and
tangential stress along the tendon. The prestressed tendon is generally
curvilinear, modelled by one-dimensional isoparametric three-node elements
and it can occupy a general position on a two-dimensional concrete element
(Figure 4). It is necessary to compute the equivalent nodal forces of a two-
dimensional finite element.
Let us denote the global co-ordinate of the point at which a concentrated
force is acting with (Xp, Yp). The equivalent force at node i, (i = 1, ..., 8) of two-
dimensional element, as a result of the concentrated force on the anchor, can be
expressed as:

(13)
where Sx and Sy are components of the force in the direction of the axis x and y
in a global co-ordinate system, and Ni (ζp, ηp) is the value of the shape function
of a two-dimensional element at the point where the force is acting. The
respective local co-ordinates ( ζ p , η p ) of the concrete element should be
associated to the point with the known global co-ordinates (X p , Y p ). The
mapping procedure from the global into local co-ordinate system is performed
numerically[7-9].
The total force caused by distributed load along the cable, in the direction of
the axis of the global co-ordinate system, can be computed by the integration
along the cable element according to the expression:

(14)
The integrals in expression (14) are solved numerically in the Gaussian points
of a one-dimensional element. The equivalent nodal forces on a two-dimensional
element caused by distributed normal and tangential load can be determined
according to the expression:
EC
14,5

(15)

516 where Pxi, Pyi are equivalent forces in the the direction of axes x and y in the ith
node of a 2D element, Ni (ζj, ηj) are the values of the ith shape function of a 2D
element in the jth Gaussian point of an 1D element, and is the value of the forces
in the direction of the axis of the global co-ordinate system computed according
to expression (15) in the jth Gauss point of a 1D element. Thus, it is possible to
perform inverse mapping procedures where the respective local co-ordinates of
a 2D element (ζi, ηi) are associated to the Gaussian points of the tendon with a
known normalized co-ordinate.
The total equivalent nodal load of a 2D element caused by tendon
prestressing can be represented in the form:

(16)
where i (i = 1, ...., 8) is the node of a 2D element.

Figure 4.
Concrete element load
caused by prestressing
Losses in the prestress force Post-tensioned
The model developed in this paper makes it possible to compute standard concrete
losses caused by short-term linear and non-linear concrete deformation, losses structures
in the anchorage zone caused by concrete deformation and losses caused by
friction.
Standard losses caused by concrete deformation are directly related to the
applied model for concrete, reinforcement and prestressed tendons described 517
earlier.
Losses in the anchorage zone caused by concrete deformation are obtained
directly. In the zone where the prestress force is introduced into the structure,
local concentration of compressive stresses occurs in the direction of the tendon
and significant tensile stresses occur in the direction perpendicular to the
tendon. At those points, farthest from the place where the force is applied, the
state becomes more stable. Owing to the described concentration of stress, the
prestress force is significantly decreased at the beginning of the tendon. The
described effects occur along the so-called anchorage length[17]. The accuracy
of the obtained results depends on the density of the finite elements mesh in this
zone.
The computation of losses caused by friction was described in detail earlier
in this paper.

Numerical procedure
The computation of post-tensioned structures was performed in phases. The
load can be applied incrementally in each phase.
In this paper the initial stiffness method[14] was applied for solving the
system of non-linear equations within each load increment. Consequently, the
tangential stiffness matrix was formed at the beginning of each increment and
it was kept in subsequent iterations.

Phase I
In the first phase the structure was computed taking into account the load it
carried before the prestressing of tendons (dead load and one part of permanent
load). A concrete or reinforced concrete structure is analysed herein.
According to the known geometry and load we form the global stiffness
matrix KI for phase I according to the expression:

(17)
where KC is the concrete stiffness matrix[8,11] and KR is the reinforcement
stiffness matrix[7-9], which is obtained by the numerical integration along the
reinforcing bar according to the expression:

(18)
EC In equation (18) BR is a strain matrix of the reinforcement element[7-9], ER is the
14,5 tangential modulus of elasticity of the reinforcement, AR is the cross-section
area of the bar, ds is a differential element of the length and ζ is the independent
normalized co-ordinate.
Global load vector (vector of residual forces) FI is determined according to
expression:
518 FI = FC + FR (19)
where FC is a vector of external forces and residual forces on concrete
element[8,11,14], while FR is vector of residual forces due to reinforcement
strain[7,8]:

(20)
In equation (20) σr is the normal stress in reinforcement.

Phase II
In the second phase the tendons are tensioned individually. The prestress force
can be applied at once or incrementally, and, thus, gradual prestressing
procedures can be simulated. The previously applied force can be subsequently
decreased, which is sometimes done in practice, in order to reduce high initial
stress in one part of the tendon. During the prestressing phase the respective
tendon is not treated as a structural element. Its geometry is used actually to
compute the initial influence of prestressing which is modelled as a fictitious
distributed load. In subsequent iteration, the tendon functions as a classical
reinforcement with a given initial stress. During successive prestressing of
tendons, the tendon which is currently being prestressed does not influence the
stiffness of the structure, while the previously prestressed tendons take over the
stresses as a classical reinforcement.
The global stiffness matrix in this phase can be presented in the following
form:

(21)
where: i = tendon index, i.e. index of a group of tendons which are being
prestressed;
KIIi = global stiffness matrix at the moment of prestressing the ith
group of tendons;
KI = global stiffness matrix of Phase I;
KPi = stiffness matrix of one tendon or group of tendons which
started functioning as a classical reinforcement[7-9].
The loading vector can be presented in the following form:

(22)
 where: FIIi = global vector of loading at the moment of prestressing the Post-tensioned
ith group of tendons; concrete
FI = load vector after Phase I; structures
∆FjII = vector of equivalent load which represents the influence of
the prestressing force of a given tendon on the concrete
structure[8]. 519
When the prestressing force is introduced into the structure gradually, vector
∆FjII is applied incrementally and not at once.

Phase III
The prestressing of all tendons is followed by the third phase in which the
structure is computed taking into account the remaining part of the dead load
and the live load. Concrete, reinforcement and all prestressed tendons which
function as a classical reinforcement, contribute to the stiffness of the structure.
The load is applied incrementally until failure. The stiffness matrix in this
phase KIII is

(23)
where n is the number of prestressed tendons, i.e. the number of prestressed
tendon groups.
The loading vector can be presented as

(24)
where ∆FIII is the part of the loading vector which resulted from the load taken
over by the structure after completed prestressing. When the load is applied
incrementally vector ∆FIII is applied in increments.

Examples
Example I
The validity of the model was verified on a prestressed beam PP2R3-0, which
had been tested by experiment, analysed in [18]. The characteristics of the beam
and the position of the prestressed tendon and the reinforcement are presented
in Figure 5.
The concrete strength was 43.815MPa. According to the composition of the
concrete mix described in [18] it was possible to estimate an initial modulus of
elasticity of 25,500MPa. The ultimate strength of the prestressing steel was
1,482MPa, while the yield stress was 1,290MPa. The initial modulus of
elasticity was 207,000MPa and the hardening parameter H’=67,547MPa. The
prestressed tendon was straight with a cross-sectional area of 38.48mm2. The
prestressing was performed from both ends by a force of 36.13 kN which caused
EC an initial stress of 938.4 MPa in the tendon. The adopted models for concrete
14,5 and steel are presented in Figure 6.
The tensile reinforcement had a cross-sectional area of 157.08mm2 with a
yield stress 285.11MPa. In the upper zone the embedded reinforcement area was
56.55mm2 with a yield stress of 276MPa, in order to support the shear stirrups
with a diameter of 6mm at 30cm intervals. The modulus of elasticity for the
520 reinforcement was 200,000MPa.
A centric force P = 2.967kN was applied to the beam which gave a midspan
moment of 2.261kNm. The force was increased gradually until failure.

Comparison of results obtained by this programme with experimental results

Figure 7 presents the load-displacement response, while Figure 8 presents the
relation between the stress increment in the prestressed tendon and the applied
load.
The failure of the structure, according to experiments, was caused by the
opening of cracks in concrete and by the yielding of the reinforcement steel at
the midspan moment of 18.0kNm, and according to this paper, the failure was
caused by cracks in concrete for a moment of 16.8kNm. The stresses in
prestressed steel did not reach the yield stress in either case.

Figure 5.
Geometry of the beam
and the position of the
prestressed tendon

Figure 6.
Adopted models for
concrete and steel
The increase in stress in a prestressed tendon, owing to applied load, shows a Post-tensioned
trend similar to the load-deflection response (Figures 7 and 8). Before cracking, concrete
the increase in stress in prestressing steel was relatively small and it varied, structures
according to experiments, from 3MPa to 20MPa, while the greatest increase,
which was obtained numerically, was 13MPa. The opening of the first cracks
was at the load P = 14.84kN, i.e. at the midspan moment of 11.31kNm. After
cracking, the stress in the prestressing steel tended to increase significantly 521
with applied load.

Figure 7.
Load deflection
response of the beam

Figure 8.
Increase of stress in
prestressing steel with
applied load
EC Example 2
14,5 The analysis was performed on a prestressed concrete beam; its geometry and
load are presented in Figure 9. The concrete strength was fB=30MPa and the
prestressed steel fK=1,500MPa. The prestressing was applied on both ends
with two symmetrical tendons and equal forces at both ends. The computation
was performed in three phases. In the first phase only the dead load of the beam
522 was taken into account. In the second phase the tendons were prestressed
individually, while in the third phase the computation included a distributed
load of p = 47kN/m which was increased incrementally until failure. The beam
was previously dimensioned[19], the position of the tendons was selected and
the required initial prestressing force of S = 1,372kN was obtained in the
tendons.
The example was analysed for three values of the friction coefficient between
the tendons and the concrete:
(1) µ = 0,
(2) µ = 0.25 which corresponds to the case when the tendon is embedded in
a protective tube[19]; and
(3) µ = 0.40 when the tendon is directly connected to the concrete.
Figure 10 presents the force in a prestressed tendon I for different cases of
loading for the friction coefficient µ = 0.40.
Figure 11 presents a diagram of the structure displacement in point C
dependent on the load, and Figure 12 presents a diagram of the dependence of
the force in the tendon on the applied load for both tendons, for friction
coefficients µ = 0 and µ = 0.40.
In case (1) failure occurred at factor 2.8 with a significant increase in the
displacement. The force in the tendons was smaller than the initial prestressing
force up to factor 1.7, since the prestressed losses were greater than the increase
in the force in the tendons, which functioned as a tensile reinforcement. The
force in tendon I was less than in tendon II, and the maximum deviation was 2.3
per cent. This difference appeared during successive prestressing, since the
stress was decreased in the previously prestressed tendons. At loading factor

Figure 9.
Geometry and loading
of the structure with
discretization
 Post-tensioned
concrete
structures

523
Figure 10.
Force in prestressed
tendon I for different
cases of loading

Figure 11.
Load-deflection
response of the beam

2.0 the stress in the tendons began to increase significantly, since they took over
great tensile stresses. Immediately before failure, the force in the tendon in the
middle of the beam was 1,681kN.
For cases (2) and (3) in the prestressing phase the stress losses were caused
by friction. Immediately after completed prestressing this decrease in the
middle of the tendon was 2.6 per cent greater from the initial force in the tendon
for case (2), and 4 per cent for case (3). The structure failure occurred at loading
factor 2.9 in both cases. In this example the influence of the friction between the
tendons and concrete was beneficial to the structure; thus, by increasing the
friction coefficient the loading factor was increased at failure. This can be
explained by the fact that friction causes a decrease in the initial stress in the
tendons, so that tendons have a greater storage capacity in taking over tensile
stresses. In practice similar phenomena are eliminated by selecting the degree
of prestressing.
Example 2 was analysed in detail for the friction coefficient between steel
and concrete µ = 0. The analysis included the stress state after prestressing
tendon I in the cross-section 1'–1', parallel to the beam axis. In the area where
EC
14,5

524

Figure 12.
Change in the tendon
force until failure

the prestress force was introduced into the structure, the local concentration of
compressive stresses occurred in direction x, while the tensile stresses which
occurred in direction y, could not be neglected. In the Gaussian point, near the
place where the prestress force was applied, the stress σxx was significant, but
it decreased farther from that point. Stress σyy, presented in Figure 13, was
compressive in the first Gaussian point, while in the next point it was tensile
and gradually decreased. The described behaviour occurred at a distance of 1
(m) from the support after which the state was stabilized, stresses σxx became
dominant, while stresses σyy, were negligible.

Figure 13.
Stress in the anchorage
zone
Owing to the described concentration of compressive stresses and because of Post-tensioned
the assumption that perfect bonding between concrete and the tendon had been concrete
realized, the force in the prestressing tendon decreased at the beginning of the structures
tendon. This decrease was evident at the first finite element, 0.5 (m) from the
place where the force was applied and amounted to 11.2 per cent of the initial
prestress force. In the third finite element, 1.0-1.5 (m) this decrease was
insignificant. Such a behaviour numerically confirmed the effects[19] along the 525
so-called anchorage length, which is approximately the same as the beam
height. An analysis of the same example with a denser mesh of finite elements
would describe the state in this area more precisely.

Example 3
The prestressed beam from Example 2, with the same geometry, loading and
tendon position, was analysed for prestressing by different forces on both
tendon ends. The initial prestressing force for both tendons at the left end was
SA=1,372kN, and at the right end SB=1,320kN. The friction coefficient between
the tendon and concrete was µ = 0.4.
The position of the cross-section with the minimum initial prestressing force
was not known in advance. Since the prestressing force was smaller at the right
end, the required cross-section was at the right half of the beam. In the first
iteration we assumed the cross-section to be at 7m from bearing A. The
computed force in the tendon at that point, according to equation (6), differed
significantly on the left and the right sides. In the second iteration the cross-
section was moved to a 9m distance from the bearing A, where S min was
computed to be 1,290.47kN on the left side, while on the right side Smin was
1,293.40kN. The difference between those two forces was 0.23 per cent. By
moving the cross-section to the left or to the right, it was not possible to achieve
better accuracy with the selected finite element mesh. Consequently, the
procedure for determining the cross-section with the minimum force in the
tendon was considered to be completed, and the structure was computed until
failure.
Figure 10 presents a diagram of the dependence between the loading and the
structure displacement in the middle of the span. Since the initial prestressing
force in the tendon was smaller on the right side, the structure failure occurred
earlier than in the case of prestressing performed on both ends with equal
forces. The loading factor at failure was 2.5.

Example 4
A post-tensioned non-prismatic continuous beam[1] was considered. Figure 14
shows the beam dimensions, the tendon layout and load.
The load was considered in the three phases:
• Phase I (the phase preceding prestressing): the beam’s own weight was
25kN/m 3 and uniformly distributed dead load was ρ d =
20kN/m.
EC • Phase II (prestressing phase): load considered in Phase I, and
14,5 prestressing forces S = 2,000kN applied at one end of the tendon
while the other end was anchored into concrete.
• Phase III (under service-load condition): load considered in Phase II,
uniformly distributed live load pl = 20kN/m and concentrated
load P = 200kN.
526 Material parameters for concrete and prestressing steel are given in Table I.
Table II presents the results obtained for reactions at support A for linear
and non-linear analysis which are compared with the results for the linear
analysis. The results for the linear analysis agree well with those from [1].
However, the non-linear analysis shows that cracks occur already in the phase
of tendon prestressing owing to the great prestressing load, which causes
deviations in the rate of reactions on the bearing. In addition, the model
presented in this paper takes into account the losses caused by the prestress
force due to non-linear deformation of the concrete and the losses in the
anchorage zone which affected the reaction rate.

Figure 14.
Non-prismatic
prestressed continuous
beam

Concrete Prestressing steel

Young’s modulus, Ec 28000N/mm2 Young’s modulus, Es 224,000N/mm2

Uniaxial compressive strength, fc 30N/mm2 Yield stress, fy 1,500N/mm2
Ultimate compressive strain, εu 0.0035 Area thickness, As 2,000mm2
Table I.
Poisson’s ratio, ν 0.25
Material parameters
for Example 4 Tensile strength, ft 3.0N/mm2

This paper Reference[1]

Non-linear Linear Linear
Load case Fx Fy M Fx Fy M Fx Fy M
Table II.
Phase I 107 180 296 107 180 296 105 180 298
Reactions for support
A (kN) for the beam Phase II –308 250 575 –469 273 655 –479 273 647
shown in Figure 14 Phase III –142 476 1098 –251 485 1093 –260 487 1093
Figure 15 presents the force along the prestressed tendon for several cases of Post-tensioned
load. After prestressing, the force in the tendon decreases owing to non-linear concrete
deformation of concrete and great concentration of compressive stresses in the structures
anchorage zone. In the third phase the prestressed tendon starts functioning as
a reinforcement which influences the changes of the force in the tendon. Load-
displacement response of the beam in point C is shown in Figure 16.
527

Figure 15.
Force in prestressed
tendon for different
cases of loading

Figure 16.
Load-displacement
response of the beam
EC References
1. El-Mezaini, N. and Citipitioglu. E., “Finite element analysis of prestressed and reinforced
14,5 concrete structures”, Journal of Structural Engineering, Vol. 117 No. 10, 1991, pp. 2851-64.
2. Van Greunen, J. and Scordelis, A., “Nonlinear analysis of prestressed concrete slabs”,
Journal of Structural Engineering, Vol. 109 No. 8, 1983, pp. 1742-60.
3. Kang, Y.J. and Scordelis, A.C., “Nonlinear analysis of prestressed concrete frames”, Journal
of Structural Division, Vol. 106 No. 2, 1980, pp. 445-63.
528 4. Hofstetter, G. and Mang, H.A., “Nonlinear finite element analysis of prestressed concrete
shells”, in Taylor, C., Owen, D.R.J., Hinton, E. and Damjanić, F.B. (Eds), Proceedings of the
3rd International Conference on Numerical Methods for Non-linear Problems, Dubrovnik,
September 1986, Pineridge Press Ltd, Swansea, 1986, pp. 344-62.
5. Majorana, C., Natali, A. and Vitaliani, R., “Analysis of three-dimensional prestressed
concrete structures using a non-linear material model”, Engineering Computations, Vol. 7,
1990, pp. 157-66.
6. Seraj, S.M., Kotsovos, M.D. and Pavlovic´, M.N., “Non-linear finite element analysis of
prestressed concrete members”, Proceedings of the Institution of Civil Engineering, 1992,
pp. 403-18.
7. Elwi, A.E. and Hrudey, T.M., “Finite element model for curved embedded reinforcement”,
Journal of Engineering Mechanics, Vol. 115 No. 4, 1989, pp. 740-54.
8. Nikolic´, Z̆., “Development of the numerical model for post-tensioning of plane reinforced
concrete structures”, Ms.Sc. Thesis, Faculty of Civil Engineering, University of Split, 1993.
(in Croatian).
9. Mihanovic´, A. and Nikolić, Z̆., “Numerical model for post-tensioning concrete structures”,
International Journal for Engineering Modelling, Vol. 6 No. 1-4, 1993, pp. 221-8.
10. Perzyna, P., “Fundamental problems in viscoplasticity”, Recent Advances in Applied
Mechanics, Vol. 9, 1996, pp. 243-377.
11. Damjanić, F.B., “Reinforced concrete failure prediction under both static and transient
conditions”, PhD Thesis, C/Ph/71/83, University of Wales, Swansea, 1983.
12. Owen, D.R.J., Figueras, J.A. and Damjanic', F.B., “Finite element analysis of reinforced and
prestressed concrete structures including thermal loading”, Computational Methods in
Applied Mechanical Engineering, Vol. 41, 1983, pp. 323-66.
13. Damjanić, F.B., “A finite element technique for analysis of reinforced and prestressed
concrete structures”, in Bergan, P. et al. (Eds), Finite Element Methods for Nonlinear
Problems, Springer, New York, NY, 1986, pp. 623-37.
14. Owen, D.R.J. and Hinton, E., Finite Element in Plasticity: Theory and Practice, Pineridge
Press Ltd, Swansea, 1980.
15. Marović, P., S̆imunović, S. and Damjanic´, F.B., “Short introduction to the program package
TULISA”, in Bićanić, N., Marović, P., Owen, D.R.J., Jović, V. and Mihanović, A. (Eds)
Proceedings of 4th International Conference on Nonlinear Engineering Computations,
Split/Swansea, Pineridge Press Ltd, Swansea, 1991, pp. 339-46.
16. Chen, W.F., Plasticity in Reinforced Concrete, McGraw-Hill, New York, NY, 1981.
17. Lin, T.Y. and Burns, N.H., Design of Prestressed Concrete Structures, 3rd ed., John Wiley &
Sons., New York, NY, 1980, pp. 295-7.
18. Harajli, M.H. and Kanj, M.Y., “Service load behavior of concrete members prestressed with
unbonded tendons”, Journal of Structural Engineering, Vol. 118 No. 9, 1992, pp. 2569-89.
19. Tomićić, I., Manual for the Computation of Reinforced Concrete Structures, Society of
Croatian Structural Engineers, Zagreb, 1993 (in Croatian).