IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS

Parameter calculation and verification of concrete plastic damage model

of ABAQUS
To cite this article: Li Qingfu et al 2020 IOP Conf. Ser.: Mater. Sci. Eng. 794 012036

View the article online for updates and enhancements.

This content was downloaded from IP address 110.54.151.244 on 23/07/2020 at 12:38

ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

Parameter calculation and verification of concrete plastic

damage model of ABAQUS

Li Qingfu1*, Guo Wei 1 and Kuang Yihang 1

1
School of Water Conservancy Science and Engineering, Zhengzhou University,
Zhengzhou, China
*
[email protected]

Abstract. The purpose of this paper is to study the method for determining the parameters of
ABAQUS CDP model based on the concrete stress-strain curve provided by the code for
concrete structure design. A finite element model was established to calibrate the value of the
stress-strain curve cutoff and the damage factor of the concrete, and a calculation method of
damage factor was recommended. The results show that the relative errors between first crack
load, ultimate load and deflection at ultimate load with the experimental results were less than
5%. It is accordingly concluded that the finite element analysis results adequately reflected the
experimental results.

1. Introduction
ABAQUS software shows good applicability in solving nonlinear analysis. The concrete damage
plasticity model (CDP model) provided by ABAQUS was first proposed by Lubliner[1], and the concept
of stiffness recovery introduced by Lee[2] improved its model. The CDP model realizes stiffness change
through damage factor, and considers crack development, crack closure and partial stiffness recovery of
concrete under cyclic loading[3-5]. The basic rules include: yield criterion, flow rule and hysteresis rule.
The CDP model has been widely used in seismic analysis. It can be used to simulate the behavior of
concrete structure and composite structure under cyclic loading, and can properly predict the punching
shear response of the slab-column composite structures[6-8]. Rama et al.[9] used the concrete plastic
damage model to evaluate the fracture energy and characteristic length for different grades of concrete
using CDP model, and studied the effect on fracture properties of different input parameters like
dilatation angle and eccentricity. Wosatko et al.[3] studied the development and closure of cantilever
beam cracks in the concrete plastic damage model based on the plasticity theory. The study showed that
the model could be applied to isotropic plastic materials, and the reduction and recovery of stiffness
were considered in both the tensile stage and the compression stage, which could be used to analyze
crack closure. Fang et al.[10] simulated the reinforced concrete under cyclic loading and compared the
influence of concrete uniaxial stress-strain curve on the simulation results. It is recommended to use the
formula provided in the Chinese code as the tensile curve and other parts of the concrete. Reasonable
selection of CDP model parameters is the key to improve the accuracy of numerical simulation. In this
paper, based on the constitutive relationship provided in the concrete structure design code[11] (GB
50010-2010), a test beam was selected to establish a finite element model, and to calibrate the CDP
model parameters.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036


t0 A

E0
B

wt = 1 J
H wt = 0 (1-dt ) E0
G O
(1-dc ) E0 (1-dt )(1 − dc ) E0 C 
D
F wc = 0 wc = 1
E0
Uniaxial compression
stress-strain curve

Figure 1. Schematic diagram of elastic modulus

recovery of concrete under cyclic loading.

2. Concrete damage plasticity model

The concrete damage plasticity model (CDP model) assumes that the concrete material is continuous,
isotropic and uniform, and the elastic modulus of concrete will be lost when the concrete shows inelastic
strain. As the stiffness reduction under cyclic loading can be partially restored, the stiffness recovery
coefficient w is defined to control this phenomenon. Under uniaxial cyclic loading, stiffness reduction
and restoration of concrete under tensile and compressive loading are different. Therefore, the stiffness
recovery factor needs to be defined separately in both directions. The damage factor d which is
defined as the reduction of secant modulus relative to the initial elastic modulus at a certain point, as
presented in the Equations (1) and (2).
E = (1 − d ) E0 (1)
Where the damage factor d is a function of stress state and uniaxial tensile and compressive
damage variables d t and dc . Under uniaxial cyclic loading, ABAQUS assumes:
(1-d ) = (1 − st dc )(1 − sc dt ) (2)
Where st and sc are respectively functions of the stiffness recovery stress state related to stress
inversion, and are determined as:
 st = 1- wt r *  11 0  wt  1
 ( )
 (3)
 sc = 1- wc 1- r  11  0  wc  1

*
( )
1 ( >0 )
( ) ( )
11
r *  11 = H  11 =  (4)
0 ( 11 < 0)
Where wt and wc are the tension and compression stiffness recovery coefficient. Under the
uniaxial cyclic load (tension-compression-tension), the reduction and recovery of concrete stiffness are
shown in Figure 1. When the concrete is in an axially stretched state, the tensile stress increases linearly.
When the peak tensile stress of concrete (point A) is reached, the concrete cracks and then loads to point
B. When the tensile stiffness of concrete decreases, the unloading will be carried out according to the
effective stiffness (1-dt ) E0 , just as the path BC. When the axial pressure is applied to the concrete in
the reverse direction, if w c = 0 (the compression stiffness does not recover after the tensile damage
occurs), it is loaded by the path CD, and if wc = 1 , it is loaded by the path CMF. When point F is reached,
it is unloaded and then reverse loaded and stretched. If wt = 1 , it is loaded according to path GJ. If
wt = 0 , it is loaded by path GH.

2
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

3. Model parameters determination method

3.1Concrete model
The concrete constitutive relationship is not given in the CDP model, and the ABAQUS user needs to
define the constitutive relationship. The most representative concrete constitutive relations are three
kinds: the Chinese code[11] (GB50010-2010), the European code[12] (EN 1992-1-1: 2004) and the full
stress-strain curve equation proposed by American scholar Kent-Park[13]. The European code does not
give the tensile stress-strain curve of concrete. The stress-strain curve in the Chinese code introduces
the damage evolution parameters, which has better convergence in calculation compared with the Kent-
Park model. Therefore, the stress-strain curve provided by the Chinese code is used to determine the
parameters in the CDP model. The uniaxial stress-strain curve in the code is calculated by the following
Equation (5).
 = (1-D ) E0 (5)
Where D is the damage evolution parameter and E0 is the initial elastic modulus. ft ,r and fc,r
use the standard values of axial tensile strength and axial compressive strength.

3.2Damage factor
The CDP model directly uses the damage evolution parameters provided by the Chinese code, which
leads to the difficulty of convergence of the calculation. According to the principle of energy
equivalence, the calculation method of concrete damage factor proposed by Sidoroff[14] can be applied
to the calculation of concrete damage factor under tension and compression. The results are practical
and easy for iterative convergence. The damage factors were determined by referring to Equation (6).

d =1− (6)
E0
Where  and  are determined by equation (5), E0 is the initial modulus of elasticity of the
concrete.

3.3CDP model parameters

The CDP model considers the inelastic behaviour of concrete by defining damage factors in both
compression and tension. It is assumed that the failure modes of concrete mainly include crack and crush.
The evolution of yield surface is mainly controlled by tensile equivalent plastic strain  tpl and
compressive equivalent plastic strain  cpl . The strain input in the model is inelastic strain, and it is
necessary to ensure that the corresponding equivalent plastic strain and  cpl are greater than 0 and
increase with the increase of damage factor. Accordingly, the stress-strain curve section of concrete
between 0.5  c,r (0.5 times the peak compressive strain of concrete corresponding to uniaxial
compressive strength) and 0.6  t ,r (0.6 times the peak tensile strain of concrete corresponding to uniaxial
tensile strength) is selected as the elastic section.
Figure 2 shows the relationship between the compressive strains of the model. The compressive
inelastic strain can be calculated by the following equations:
 cin =  c −  0elc (7)
c
 0elc = (8)
E0
dc  c
 cpl =  cin − (9)
(1 − dc ) E0

3
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

Where  cin is the compressive inelastic strain, and  0elt is the elastic compressive strain
corresponding to the initial elastic modulus,  tpl is compressive plastic strain, d t is the compressive
damage factor.
Figure 3 shows the relationship between the tensile strains of the model. The tensile cracking strain
can be calculated by the following equations:
 tck =  t −  0elt (10)
t
 0tel = (11)
E0
dt  t
 tpl =  tck − (12)
(1 − dt ) E0
Where  cck is the tensile cracking strain, and  cck is the elastic tensile strain corresponding to the
initial elastic modulus,  tpl is tensile plastic strain, d c is the tensile damage factor.
c t

 c ,r t0 ( t ,r ,  t ,r )

 c0 ( in ,  in )
E0
E0
(1 − dc ) E0 E0 (1-dt ) E0 E0

 cin  0clc c  tck  0clt t

 c
pl
 cl
c t pl
 t
el

Figure 2. Uniaxial compression Figure 3. Uniaxial tensile

stress-strain curve of CDP. stress-strain curve of CDP.
Table 1. Some parameters of CDP.

（）  fb 0 / fc 0 K 

30 0.1 1.16 0.6667 0.0005

Table 1 shows the recommended values for other parameters of the CDP model for ordinary concrete.
The expansion angle  and eccentricity  are parameters related to the yield surface flow rule. K
is the parameter that controls the shape of the yield surface. fb 0 is the concrete biaxial compressive
strength, fc 0 is the uniaxial compressive strength, and the ratio of the two is taken as the recommended
value.  is the viscous parameter defined by the CDP model. The larger  is, the easier it is to
converge. The smaller  is, the higher the accuracy is. When  is 0.0005, the accuracy and
convergence can be satisfied.

4. Model parameters calibration

The CDP model does not specify the ultimate inelastic strain of concrete. Therefore, it is necessary to
calibrate the truncation of tension and compression stress of concrete. Taking the test beam STEEL-40S
in the literature [14] as an example, the finite element analysis was carried out in order to test the damage
factors calculated by the formula recommended in this paper. At the same time, the cut-off of the stress-
strain curve of the concrete was preliminarily calibrated.

4
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

SD400 D10 SD400 D10@100

35 130
250
85
100 900 400 900 100 150

2400

Figure 4. Detailed view of the beam and specimen cross-section.

Table 2. Material properties of reinforcement.[15]
Young’s modulus Yield strength Ultimate strength
Diameter(mm) Area(mm2) f y (MPa) fu (MPa)
E (GPa)
9.5 71.3 173 446 577

Table 3. Parameters of concrete mechanical behaviour.

Compressive Compressive Tensile Tensile
a
Compressiv b
Tensile
xc inelastic damage xt cracking strain damage
e stress  c stress  t
strain  c
in
factor dc  tck (×10-5) factor d t
0.5 21.46843 0.000000 0.000000 0.6 1.712712 0 0.000000
0.8 26.56861 0.000460 0.200082 0.8 2.187247 1.6620105 0.104482
1 27.30000 0.000757 0.274751 1 2.410000 3.07461538 0.159226
1.5 22.75000 0.001696 0.459431 2 1.260460 17.1016608 0.570048
2 17.06250 0.002669 0.594573 3 0.809976 28.9777670 0.718586
3 10.50000 0.004468 0.740319 6 0.423590 61.6366447 0.856098
6 4.55000 0.009442 0.879125 10 0.278854 104.041989 0.909560
9 2.86364 0.014285 0.921703 20 0.164949 209.292464 0.950815
15 1.63669 0.023905 0.954150 40 0.099964 419.292418 0.972925
20 1.20477 0.031903 0.965932
xc = c / c , r ,  c is the real compressive strain of concrete, and  c , r is the peak compressive strain
a

of concrete corresponding to uniaxial compressive strength  c, r .

b
xt = t / t , r ,  t is the real tensile strain of concrete, and  t , r is the peak tensile strain of concrete
corresponding to uniaxial tensile strength  t ,r .

4.1Test beam and parameters

Figure 4 shows detailed views of the simply supported beam STEEL-40S. The beam was designed and
fabricated with a length of 2400 mm (span 2200 mm), a cross sectional area of 150 ×250 mm, the depth
of outer reinforcing bar of 215 mm and the depth of inner reinforcing bar of 165 mm. SD400 D10
reinforcement was used for compression reinforcement, tensile reinforcement and stirrup. Table 2 shows

5
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

the material properties of SD400 D10. The specified design strength of concrete applied for forming the
specimens was 40 MPa. The measurements of compressive strength were 40.91 MPa.
The standard value of the compressive strength after conversion is f ck =27.3MPa .The standard
value of concrete compressive strength is used in the calculation of CDP model parameters. Table 3 lists
the calculated damage factors and related parameters.
The model was established according to the geometry, loading mode and constraint position of the
STEEL-40S test beam. The material properties of concrete adopt the parameters in Table 3. The
constitutive model of reinforcement adopted bilinear model. In the model, the concrete elements were
represented using C3D8R hexahedral elements, while the reinforcement is simulated using truss
elements. The reinforcement was embedded in the concrete and the degrees of freedom are the same as
the values of the concrete element’s degrees of freedom at the corresponding position. The element size
for both concrete and steel was 50mm3. This model adopted the method of setting steel plate to avoid
the phenomenon of stress concentration at the loading point and the supporting point and adopted the
displacement method in the loading process.
70
60
60
50
50
Load (kN)
Load (kN)

40
40 Experimental
Experimental Beam yielding C20T40
Beam yielding 30
30 C20T40 C15T40
C20T10 C9T40
C20T6 20 C3T40
20 Cracking Cracking
C20T3
10 10

0 0
0 10 20 30 40 50 60 70 0 20 40 60 80
Deflection (mm) Deflection (mm)

(a) xc max = 20 . (b) xt max = 40 .

Figure 5. Loading-deflection curves at mid-span of different parameters.

4.2Analytical results and discussion

In Figure 5, the test results were compared with the results of the finite element analysis. In order to
study the influence of the truncation of the stress-strain curve and the value of damage factor on the
calculated results, different values were taken for A and B in the finite element analysis. When A and B
were less than 3, the load-mid-span deflection curve of the finite element model was greatly different
from the test value. Therefore, finite element models were established to take 3, 9, 15, and 20 for A, and
take 3, 6, 10 and 40 for B. The cracking load of the test beam STEEL-40S is 13.28kN, the ultimate load
is 56.91kN, and the deflection at the ultimate load is 40.51mm. Table 4 shows the first crack load and
ultimate load in the finite element analysis. When A, the theoretical value of the ultimate load was
60.69kN, and the deflection at the ultimate load was 14.20mm, which did not reflect the actual situation
of the test. When B, the theoretical value of the cracking load was 3.26kN from the true value, there was
no falling section in the load-span deflection curve within deflection=100mm. When A, the values of
tensile damage factor and compression damage factor were both greater than 0.95, and the relative errors
between first crack load, ultimate load and deflection at ultimate load with the experimental results were
less than 5%. It is accordingly concluded that the finite element analysis results adequately reflected the
experimental results.
Table 4. The first crack moment and the experimental and analysis values of the maximum moment.
First crack Ultimate Failure
Specimen xc max xt max
load Pcr (kN) load Pu (kN)
C20T3 20 3 18.92 60.69

6
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

C20T6 20 6 17.96 58.74

C20T10 20 10 17.64 58.05
C20T40 20 40 16.54 56.37
C3T40 3 40 16.54 -
C6T40 6 40 16.54 57.41
C9T40 9 40 16.54 55.74
C15T40 15 40 16.54 55.99

5. Conclusion
This paper investigated the effects of the value of the stress-strain curve cutoff and the damage factor of
the concrete on finite element analysis. Through the above discussions, the following conclusions can
be drawn:
1) According to the damage factor calculation formula recommended in this paper, the concrete
stress-strain curve provided by the code for concrete structure design was applied to the finite element
model. The obtained model parameters have been verified, and the finite element model can be used for
the crack resistance check and bearing capacity check.
2) When the values of tensile damage factor and compression damage factors were both greater than
0.95, the relative errors between first crack load, ultimate load and deflection at ultimate load with the
experimental results were less than 5%. The deformation and damage of the finite element model were
consistent with the experimental phenomena. Therefore, the recommended damage factor is greater than
0.95.

Acknowledgment
The authors would like to acknowledge the financial support provided by the National Natural Science
Foundation of China (Grant No. 51679220).

References
[1] Lubliner, J., Oliver, J., Oller, S., Oñate, E. (1989) A plastic-damage model for concrete
International Journal of Solids and Structures 25:299-326
[2] Lee, J., Fenves, G.L. (1998) Plastic-Damage Model for Cyclic Loading of Concrete Structures.
Journal of Waterway, Port, Coastal and Ocean Engineering 124:892-900
[3] Wosatko, A., Genikomsou, A., Pamin, J., Polak, M.A., Winnicki, A. (2018) Examination of two
regularized damage-plasticity models for concrete with regard to crack closing. ENG FRACT
MECH 194:190-211
[4] Ortiz, M. (1985) A constitutive theory for the inelastic behavior of concrete. MECH MATER 4:67-
93
[5] Chi, Y., Yu, M., Huang, L., Xu, L. (2017) Finite element modeling of steel-polypropylene hybrid
fiber reinforced concrete using modified concrete damaged plasticity. ENG STRUCT 148:23-
35
[6] Sarikaya, A., Erkmen, R.E. (2019) A plastic-damage model for concrete under compression. INT
J MECH SCI 150:584-93
[7] Genikomsou, A.S., Polak, M.A. (2015) Finite element analysis of punching shear of concrete slabs
using damaged plasticity model in ABAQUS. ENG STRUCT 98:38-48
[8] Alfarah, B., Lopez-Almansa, F., Oller, S. (2017) New methodology for calculating damage
variables evolution in Plastic Damage Model for RC. structures ENG STRUCT 132:70-86
[9] Rama, J.R.J.S., Chauhan, D.C.D.R., Sivakumar, M.S.M.V., Vasan, A.V.A., Murthy, A.M.A.R.
(2017) Fracture properties of concrete using damaged plasticity model -A parametric study
STRUCT ENG MECH 64:59-69
[10] Zihu, F., Haijun, Z., Shaoying, L., Qiang, X. (2013) Choose of ABAQUS concrete stress-strain
curve. Building Structure A2:559-61
[11] GB 50010-2010. (2011) Code for concrete structure design.

7
ACHE2019 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 794 (2020) 012036 doi:10.1088/1757-899X/794/1/012036

[12] EN 1992-1-1: 2004. (2004) Eurocode 2: Design of concrete structures.

[13] D.S B (1982) Stress-Strain Behavior of Concrete by Overlapping Hoops at Low and High Strain
Rates ACI Journal 13-27
[14] Kim, S., Kim, S. (2019) Flexural behavior of concrete beams with steel bar and FRP reinforcement
J ASIAN ARCHIT BUILD 18:94-102