Research Online
Faculty of Engineering - Papers (Archive)
2012
Raha Jowkarmeimandi
University Of Wollongong
http://ro.uow.edu.au/engpapers/5251
Publication Details
Aslani, F. & Jowkarmeimandi, R. (2012). Stress-strain model for concrete under cyclic loading. Magazine of Concrete Research, 64
(8), 673-685.
Raha Jowkarmeimandi
A hysteretic stressstrain model is developed for unconned concrete with the intention of providing efcient
modelling for the structural behaviour of concrete in seismic regions. The proposed model is based on the
ndings of previous experimental and analytical studies. The model for concrete subjected to monotonic and
cyclic loading comprises four components in compression and tension an envelope curve (for monotonic and
cyclic loading), an unloading curve, a reloading curve and a transition curve. Formulations for partial unloading
and partial reloading curves are also presented. The reliability of the proposed constitutive model is investigated
for a reinforced concrete member using a non-linear nite-element analysis program. Comparisons with test
results showed that the proposed model provides a good t to a wide range of experimentally established
hysteresis loops.
Notation
Ec
Esec
f c9
n
n1
n2
c
9c
pl
ro
un
9ct
c
f
new
pr
ro
un
Introduction
Experimental programmes in laboratories produce real results
for studying the non-linear behaviour of reinforced concrete
(RC) structures but they are limited to knowledge of particular
cases under restricted structural dimensions, sizes, shapes, loading and boundary conditions. The computational simulation
approach, on the other hand, has no limit to its application
(Maekawa et al., 2003). Interest in materials modelling and
analysis of concrete structures has increased because of the need
to accurately predict the non-linear response of concrete structures under monotonic and cyclic loads. With rapid improvements in computer technology and numerical methods,
microcracks in mortar begins to increase sharply and a continuous pattern of microcracks begins to form. As a result,
undamaged portions that carry the load are reduced and the
stressstrain relationship becomes even more non-linear. The
onset of major microcracking was reported at 7090% of ultimate load (Dabbagh, 2006).
Research signicance
A constitutive model for the description of the response of
concrete under general cyclic loading is presented. The model
has several advantages over previous approaches.
(a) It allows consideration of all the hysteretic characteristics of
the complex behaviour of concrete in a simple and practical
way.
(b) It can be used to simulate the cyclic response of concrete
subject to general load conditions, including partial unloading
or reloading or mixed hysteretic loops involving the transition
from compression to tension stresses or vice versa.
(c) All the required input data can be obtained through
conventional laboratory monotonic compression and tension
tests. This is an important feature issue that determines the
applicability of the model in engineering practice.
The model is veried by comparison with available experimental
results from other research.
1:
c
n(c =9c )
f c9 n 1 (c =9c ) n
2:
:
n n1 [1:02 1:17(Esec =Ec )]0 74 if c < 9c
3:
n n2 n1 (a 28b) if c > 9c
pl un
12:
( un =Ec 9c ) 0:57
Er E c
(un =9c ) 0:57
Reloading curve
6:
Esec f c9 =9c
7:
Ec 3320 ( f c9 )0 5 6900
9:
9c
f c9
Ec
r
r1
f c9
0: 8
17
c
10:
Ec1
ro new
ro un
8:
c ro Ec1 c ro
:
a 3:5(12:4 1:66 3 102 f c9 )0 46
14:
5:
un
Er
11:
13:
where
4:
!1:2
1 [(c un )=(pl un )]
un
1 1:2[(c un )=(pl un )]
:
new un [1 0:09 (un =ro )0 5 ]
c pr
16:
1 [(c un )=(pl un )]
1 1:2[(c un )=(pl un )]
!1:2
3 ( un pr )
17:
f c9 Ec
:
f c9 (9ct =c )0 85
c < 9ct
c . 9ct
pl 0:725 un
Stress ratio
11
10
09
08
07
06
05
04
03
02
01
0
Experiment
Proposed model
05
10
15
Strain ratio
20
25
30
676
f f c9 =10
Stress ratio
Crack-closing model
11
10
09
08
07
06
05
04
03
02
01
0
Experiment
Proposed model
05
10
15
20
Strain ratio
25
30
35
11
10
09
08
07
06
05
04
03
02
01
0
11
Experiment
10
Proposed model
Experiment
09
08
Stress ratio
Stress ratio
07
06
05
04
03
05
10
15
20 25 30
Strain ratio
35
40
45
02
01
0
05
10
15
Strain ratio
20
25
30
11
10
Proposed model
08
Experiment
Proposed model
Stress ratio
Stress ratio
09
11
10
09
08
07
06
05
04
03
02
01
0
07
06
05
04
03
02
01
0
05
10
15
20
Strain ratio
25
30
35
05
10
15
Strain ratio
20
25
30
35
Experiment
30
Proposed model
25
Stress: MPa
20
15
10
05
0
05
40
Experiment
35
Proposed model
Stress: MPa
30
25
20
15
10
05
0
0
00005
0001
00015
Strain
0002
00025
0003
200 mm
1300 mm
A
Top slab
Bottom slab
100 mm
1800 mm
575 mm
Section AA
375 mm
100 mm
400 mm
1000 mm
400 mm
Section BB
200 mm
(a)
1600 mm
A
2 6N16
2 3N12 @ 75 mm
2 3N12 @ 100 mm
R4 @ 100 mm
R4 @ 100 mm
2 6N16
2 6N16
Dv
3 4N16
N12
Dv
Dh
Dh
N12
2 6N12 @ 80 mm
2 6N12 @ 100 mm
1800 mm
Dv
2 4N24
A
Dh
2 6N16
575 mm
2 6N16
2 6N16
1000 mm
1300 mm
400 mm
100 mm
75 mm
400 mm
1600 mm
Wall
B
1000 mm
Edge element
150 mm
3 4N16
2 4N24
575 mm
Section AA
2 R4 @ 100 mm
400 mm
1000 mm
400 mm
Section BB
(b)
679
Reaction pre-load
(2000 kN for zero cyclic lateral load)
Age: days
28
355
Lateral displacement: mm
(a)
SW3
SW4
Longitudinal reinforcement
Dv
Equivalent reinforcement
ratio: %
0.8
0.8
2 3 5W8 @ 160 mm
2 3 5W8 @ 160 mm
Reinforcement
W6
W8
N12
N16
N24
536
498
571
535
524
Ultimate
strength: MPa
597
535
649
638
623
680
42 760
43 670
Figures 1013 show the loaddisplacement response of specimens SW3 and SW4; a summary of the results is given in Table
4. Specimens SW3 and SW4 had longitudinal and transverse
reinforcement ratios of 0.8% with the specimens subjected to a
combination of axial load and lateral reversed cyclic loading. The
ages of SW3 and SW4 at the time of testing were 355 and
358 days respectively. The compressive strength of concrete on
the day of test was 96 MPa.
Transverse reinforcement
2 3 10W6 @ 100 mm
2 3 10W6 @ 100 mm
4.6
7.2
81
96
Specimen SW3 was tested under displacement cycles accompanied by an axial load of 1200 kN. The lateral loading was applied
to the specimen through complete phases of 4, 8 and 12 mm and
Dh
Elastic
modulus:
MPa
Specimen
Spitting tensile
strength: MPa
Axial load
20
Phase 4
16
Phase 3
Phase 2
12
8
Phase 1
4
0
1
4
5
7
2
3
6
8
4 0
8
12
Cycle
16
20
(b)
Compressive
strength:
MPa
Elastic modulus:
GPa
198
179
199
204
195
Equivalent reinforcement
ratio: %
0.8
0.8
Specimen
SW3
SW4
Reinforcement
ratio: %
Concrete
strength:
MPa
Axial load:
kN
Transverse
Longitudinal
0.8
0.8
0.8
0.8
96
96
Peak lateral
load: kN
Downward
1090
683
1200
0
Corresponding
displacement: mm
Upward
1107
753
Downward
Upward
12.12
7.94
12.31
12.32
1400
1200
1000
800
Lateral load: kN
600
400
200
24
20
16
12
0
4
0
200
12
16
20
24
12
16
20
24
400
600
800
1000
1200
1400
Lateral load: kN
24
20
16
12
1400
1200
1000
800
600
400
200
0
4
200 0
400
600
800
1000
1200
1400
681
(a)
(b)
Figures 10(b) and 12(b) show analytical results of SW3 and SW4
compared with the experimental results (Figures 10(a) and 12(a)).
Figures 11 and 13 show the Abaqus simulation using the
proposed stressstrain relationship for concrete for specimens
SW3 and SW4.
Conclusions
A cyclic constitutive model has been developed for unconned
concrete. The following conclusions are drawn from the current
study.
(a) The proposed constitutive model was developed for the
simulation of the response of concrete subjected to cyclic
loadings in both compression and tension.
(b) The model can reproduce the complex behaviour of concrete
under any history of uniaxial cyclic loading (i.e. full loading
and partial loading).
(c) Unloading is assumed to be non-linear and is modelled
1400
1200
1000
800
Lateral load: kN
600
400
200
20
16
12
0
4
0
200
12
16
20
12
16
20
400
600
800
1000
1200
Lateral load: kN
1400
Displacement of top slab: mm
(a)
20
16
12
1400
1200
1000
800
600
400
200
0
4 200 0
4
8
400
600
800
1000
1200
1400
Displacement of top slab: mm
(b)
(g)
(h)
(i)
( j)
(a)
(b)
REFERENCES