Accepted Manuscript

Modeling of anisotropic hardening of sheet metals including description of the

Bauschinger effect
Fusahito Yoshida, Hiroshi Hamasaki, Takeshi Uemori
PII:

S0749-6419(15)00033-9

DOI:

10.1016/j.ijplas.2015.02.004

Reference:

INTPLA 1882

To appear in:

International Journal of Plasticity

Received Date: 29 August 2014

Revised Date:

19 January 2015

Accepted Date: 3 February 2015

Please cite this article as: Yoshida, F., Hamasaki, H., Uemori, T., Modeling of anisotropic hardening of
sheet metals including description of the Bauschinger effect, International Journal of Plasticity (2015),
doi: 10.1016/j.ijplas.2015.02.004.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Modeling of anisotropic hardening of sheet metals

including description of the Bauschinger effect
Fusahito Yoshidaa*, Hiroshi Hamasakib and Takeshi Uemoric*
a

RI
PT

Department of Mechanical Science and Engineering, Hiroshima University,

1-4-1 Kagamiyama, Higashi Hiroshima 739-8527, Japan
b
Department of Mechanical Science and Engineering, Hiroshima University,
1-4-1 Kagamiyama, Higashi Hiroshima 739-8527, Japan
c
Department of Advanced Mechanics and Manufacturing, Okayama University,
1-1-1, Tsushima-naka, Kita-ku, Okayama 700-0082, Japan

TE
D

M
AN
U

SC

ABSTRACT
The present paper proposes a framework for constitutive modeling of plasticity to describe the evolution of
anisotropy and the Bauschinger effect in sheet metals. An anisotropic yield function, which varies continuously with
increasing plastic strain, is defined as an interpolation between two yield functions at two discrete levels of plastic
strain. Several types of nonlinear functions of the effective plastic strain are proposed for the interpolation. A
framework for a combined anisotropic-kinematic hardening model of large-strain cyclic plasticity with small elastic
strain is presented, and the details of modeling which is based on the YoshidaUemori model (Yoshida and Uemori
et al., 2002a, 2002b, 2003) are described. The shape of the yield surface and that of the bounding surface are
assumed in the model to change simultaneously. The model was validated by comparing the calculated results of
stressstrain responses with experimental data on r-value and stress-directionality changes in an aluminum sheet
(Hu, 2007) and a stainless steel sheet (Stoughton and Yoon, 2009), as well as the variation of the yield surface of an
aluminum sheet (Yanaga et al., 2014). Furthermore, anisotropic cyclic behavior was examined by performing
experiments of uniaxial tension and cyclic straining in three sheet directions on a 780-MPa advanced high-strength
steel sheet.
Keywords: Anisotropic hardening; B. constitutive behaviour; B. anisotropic material; B. metallic material

* Corresponding author. Tel: +81 82 424 7541; fax : +81 82 424 7541

Introduction

AC
C

1.

EP

E-mail address: fyoshida@hiroshima-u.ac.jp (F. Yoshida)

The use of plasticity models that properly describe material behavior is of vital importance for accurate
numerical simulation of sheet metal forming. Conventional plasticity models assume that the shape of the yield
surface does not change during a plastic deformation, and consequently, r-values and flow stress directionality
calculated with the models remain constant throughout the deformation. However, some metallic sheets exhibit
significant changes in planar r-value anisotropy and flow stress directionality (e.g., Hu, 2007; Stoughton and Yoon,
2009; An et al., 2013; Safaei et al., 2014) and the shape of yield surface (e.g., Tozawa, 1978; Kuwabara et al., 1998;
Yanaga et al., 2014; Yoon et al., 2014) as plastic strain increases.
The anisotropy of a solid is expressed by an anisotropic yield function that involves a set of material parameters,
Ck , k = 1, 2, ..., N . To describe the evolution of anisotropy, the material parameters should be given as functions of
the effective plastic strain , i.e., as Ck ( ) . The material parameters in yield functions for sheet metals are usually
determined using experimental data on uniaxial tension stresses (e.g., 0 , 45 , 90 ) and r-values (e.g., r0 , r45 , r90 ) in
several sheet directions and some biaxial stresses (e.g., equi-balanced stress b ). If these stresses and r-values are

ACCEPTED MANUSCRIPT

formulated as functions of the plastic strain, i.e.,

0 ( ), 45 ( ), 90 ( ), b ( ), r0 ( ), r45 ( ), r90 ( ) ,

the

material parameters Ck can be calculated from these for any effective plastic strain values.
Stoughton and Yoon (2009) employed Hills (1948) quadratic yield function in the framework of a nonassociated flow rule to describe anisotropic hardening of several types of aluminum and stainless steel sheets. Hu
(2007) used a fourth-order polynomial yield function and succeeded in reproducing both the stress and r-value
anisotropies of an aluminum sheet. In their models, the material parameters of a yield function are calculated for

r0 ( ), r45 ( ), r90 ( ) ).

0 ( ), 45 ( ), 90 ( ), b ( ),

RI
PT

every time step using a given set of stress values and r-values (

SC

However, direct calculation of material parameters for every time step of sheet metal
forming simulation is only applicable to yield functions whose material parameters are calculated explicitly as
functions of 0 , 45 , 90 , b , r0 , r45 , r90 , etc.; it is not applicable to more complicated models whose material
parameters are determined using only iterative numerical approaches such as the N-R method (e.g., Barlat et al.,
2003, 2005, 2007; Banabic et al. 2005; Yoshida et al. 2014).
An alternative approach is approximation equation building for material parameters, such as Ck ( ), k = 1, 2, ..., N .
For that purpose, M sets of material parameters, Ck( i ) , i = 1, 2, .., M , determined at M discrete levels of the effective

( , )

M
AN
U

plastic strains, i , i = 1, 2, .., M , are employed (e.g., Aretz, 2008; Safaei et al., 2014; Yanaga et al., 2014). An
advantage of this approach over the aforementioned direct calculation approach is that only limited experimental
data (M sets) corresponding to i , i = 1, 2, .., M are needed. However, in most of cases, the approximation equations
involve many coefficients that need to be determined from analysis of experimental data. Instead of using
approximation equations for material parameters, Plunkett et al. (2006) expressed the current yield function
directly as a linear interpolation of two yield functions,

i ( , i )

and

i +1 ( , i +1 ) , within the plastic

AC
C

EP

TE
D

strain range i i +1 .
Another important issue in material modeling is describing the Bauschinger effect, especially for accurate
numerical simulation of springback of high-strength steel (HSS) sheets (e.g., Yoshida and Uemori, 2003; Eggertsen
and Mattiason, 2009, 2010; Wagoner et al, 2013). Although many types of kinematic hardening models have been
proposed to describe the Bauschinger effect and cyclic plasticity behavior (e.g., Armstrong and Frederick, 1966;
Ohno, N., 1982; Chaboche and Rousselier, 1983; Ohno and Wang, 1993; Geng and Wagoner, 2002; Yoshida et al.,
2000, 2002a, 2002b, 2003, 2008; Haddadi et al. 2006; Taleb, 2013; for more details, refer to Chaboches review,
2008), very few models have been developed to describe stress and r-value anisotropy evolution simultaneously. Socalled distortion yield function modeling is another type of formulation used to represent the Bauschinger effect and
stress-strain responses under non-proportional cyclic loading (e.g., Shiratori et al, 1979; Voyadijis and Froozesh,
1990; Kurtyka and Zyczkowski, 1996; Francois, 2001; Feigenbaum and Dafalias, 2007; Barlat et al., 2011, 2013,
2014). However, to the best of the present authors knowledge, among these works, only Barlat et al.s
homogeneous anisotropic hardening (HAH) model (2011, 2013, 2014) reproduces the Bauschinger effect well
together with the anisotropy evolution of sheet metals.
The aim of the present paper is to propose a framework for constitutive modeling of plasticity, based on the
associated flow rule, to describe the evolution of anisotropy and the Bauschinger effect in sheet metals. In the
anisotropic hardening model presented in this paper, an anisotropic yield function, ( , ) , which varies
continuously with increasing plastic strain within the range A B , is defined as an interpolation between two
yield functions,

A ( )

and

B ( ) ,

for two discrete levels of effective plastic strains, A and B . A full

i ( ) ,

i = 1, 2, .., M ,
determined at a few discrete levels of plastic strains, i , i = 1, 2, .., M . The proposed model was validated by
comparing the calculated stressstrain responses with experimental data on r-value and stress changes in an
aluminum sheet (Hu, 2007) and a stainless sheet (Stoughton and Yoon, 2009), as well as experimental data on the
variation of yield surface of an aluminum sheet (Yanaga et al., 2014). This anisotropy evolution model is able to
describe the Bauschinger effect by incorporating it into the kinematic hardening model. This paper describes a
combined anisotropickinematic hardening model constructed in the framework of the YoshidaUemori two-surface
modeling approach (Yoshida and Uemori, 2002a, 2002b, 2003), wherein the shape of the yield surface and that of
the bounding surface change simultaneously. The model was validated by comparing the numerical prediction of

description of anisotropy evolution requires only a limited number of yield functions,

ACCEPTED MANUSCRIPT

sheet-direction-dependent cyclic plasticity behavior to experimental results for a 780-MPa advanced HSS (AHSS)
sheet.

2.

Modeling of anisotropic hardening

2.1. Framework of modeling

elastic and plastic parts,

RI
PT

In the present paper, a model is constructed within the framework of the associated flow rule of plasticity. With
the assumption of small elastic and large plastic deformation, the rate of deformation D is decomposed into its

De and De , respectively, as follows:

D = De + D p .

(1)

The decomposition of the continuum spin W is given as follows:

W = +W p ,
(2)
p
where W denotes the plastic spin and is the spin of substructures (Dafalias, 1985). The constitutive equation
o

= &

SC

of elasticity is expressed as follows:

= C : De

and

are the Cauchy stress and its objective rate, respectively,

is the spin tensor; and C is the

M
AN
U

where

(3)

elasticity modulus tensor.

The initial yield criterion is expressed by the following equation:
f = 0 ( ) Y = ( ) Y = 0 ,

(4)

where Y is the initial yield stress and is the effective stress. In general, the evolution of anisotropy is expressed by
the evolutionary change of the yield function as a function of the effective plastic strain . Let us begin our
discussion with a simple case that excludes the description of the Bauschinger effect. The subsequent yield function
after plastic deformation is expressed by the following equation:

TE
D

f = ( , ) F ( ) = ( , ) F ( ) = ( , ) (Y + H F ( ) ) =0,

(5)

where F and H F are the flow stress and its workhardening component, respectively. Note that ( ,0) = 0 ( ) .
Based on the work conjugate concept, the effective plastic strain is defined as follows:

& = : D p , = & dt .

(6)

The associated flow rule is written as follows

where

& ( 0)

f & &
=
,

EP

Dp =

(7)

is the plastic multiplier. For a homogeneous yield function of degree one,

, the work conjugate

AC
C

type definition of the effective plastic strain rate (see Eq.(6)) reduces to & = & . Thus, the elasto-plasticity
constitutive equation can be written as follows:
o

= C ep : D ,

ep
C:
C :
C =

+ H
:C :

where

H = H F =

F dH F
=
.

(8)

if & = 0
if & > 0

(9)

(10)

ACCEPTED MANUSCRIPT

In the present paper, the anisotropic hardening of the yield surface is expressed as follows:

( , ) = ( )A ( ) + (1 ( ) ) B ( ) for A B

B ( ) are two different yield functions defined at the effective plastic strains A and B ,
respectively, and ( ) an interpolation function of the effective plastic strain, where
1 = ( A ) ( ) ( B ) = 0 .
(12)
Note that the types of these two yield functions, A ( ) and B ( ) , do not need to be the same. For example Fig.
1 shows a case of A ( ) being a Tresca-type function and B ( ) being a von Mises-type function. An advantage
of this modeling framework is that, if the two yield functions A ( ) and B ( ) are convex, ( , ) always
and

RI
PT

Here,

A ( )

(11)

satisfies the convexity. The derivatives are expressed as follows:

EP

y / 0

TE
D

(13)

M
AN
U

1.5

0.5

SC

( )
( )
,
= ( ) A
+ (1 ( ) ) B

( )
= (A ( ) B ( ) )
.

AC
C

-0.5

-1

-1.5
-1.5

Tresca
( = 1.0)
Tresca-von Mises ( = 0.3)
von Mises
( = 0)
-1

-0.5

0.5

1.5

x / 0

Fig. 1 Change of yield function from Tresca (

= 1 ) to von Mises ( = 0 ).

ACCEPTED MANUSCRIPT

( ) . Assuming that A = 0 (at

that A ( ) = 0 ( ) and B ( ) = ( ) , Eq.(11)

Several linear and nonlinear functions can be used for the interpolation function
initial yielding) and

B = (at

infinitely large strain) and

reduces to the following

( , ) = ( )0 ( ) + (1 ( ) ) ( ) , 0 .

(14)

RI
PT

Some examples of forms of interpolation functions are as follows:

( ) = exp( ) ,

(15)

( ) = a exp( 1 ) + (1 a )exp(2 ) ,
where

(16)

, a, 1 , and 2 are material constants.

corresponding to M discrete plastic strain points,

SC

If we have M sets of experimental data ( 0 , 45 , 90 , b , r0 , r45 , r90 , etc.) for material parameter identification

1 ( = 0), 2 ,..., i , i +1,..., M , we can determine M sets of yield

M
AN
U

functions 1 ( ) , 2 ( ) , ., i ( ), i +1 ( ), , M ( ) . Using an interpolation function ( ) , the yield function

( , ) can be defined by the following equation:

( , ) = ( )i ( ) + (1 ( )) i +1 ( ) for i i +1 .

(17)

The following nonlinear equation is proposed for use as the interpolation function:
pi

i
( ) = 1
for i i +1 ,
i +1 i
where pi (i = 1, 2, ..., M 1) are material constants.

TE
D

2.2. Modeling of polynomial-type yield function

(18)

Among the various types of anisotropic yield functions available, stress polynomial-type models (e.g., Hill 1948,
Gotoh 1977, Soare et al. 2008, Yoshida et al. 2013) are suitable for use in modeling anisotropy evolution. A
polynomial-type yield criterion is given by the following equation:
f = ( m ) ( ) Fm = m Fm = 0 ,

(19)

EP

where ( m ) ( ) denotes the m-th order stress polynomial-type yield function. For example, when m = 6 (Yoshida et
al. 2013) under plane stress condition,

(6) = C1 x6 3C2 x5 y + 6C3 x4 y2 7C4 x3 y3 + 6C5 x2 y4 3C6 x y5 + C7 y6

AC
C

+9 ( C8 x4 2C9 x3 y + 3C10 x2 y2 2C11 x y3 + C12 y4 ) xy2

(20)

+27 ( C13 x2 C14 x y + C15 y2 ) xy4 + 27C16 xy6 .

In the same manner as Eq.(11), when the following equation is assumed

(m) ( , ) = ( )A( m) ( ) + (1 ( )) B(m) ( ) ,

(21)

it reduces to an interpolation for material parameters Ck , k = 1, 2,..., N , as follows

Ck = ( )Ck (A) + (1 ( )) Ck (B) .

(22)

ACCEPTED MANUSCRIPT

Here, C k (A) and C k (B) are material parameters determined at the effective plastic strains,
Assuming that

A = 0 (at

initial yielding) and

B = (at

B , respectively.

and

infinitely large strain) and that

Ck ( A) = Ck (0) and

Ck (B) = Ck ( ) , Eq.(22) reduces to the following:

Ck = ( )Ck (0) + (1 ( ) ) Ck ( ) , 0 .

RI
PT

(23)

In discretization form:

Ck = ( )Ck (i) + (1 ( ) ) Ck (i +1) , i = 1, 2, ..., M 1 .

(24)

2.3. Validation of the model

SC

To validate the model, calculated stress-strain responses to r-value and stress changes were compared with the
corresponding experimental data for an aluminum sheet (after Hu, 2007) and a stainless steel sheet (after Stoughton
and Yoon, 2009), as well as the variation of yield surface of an aluminum sheet (after Yanaga et al., 2014) .
Hu (2009) presented interesting experimental data on the drastic changes in r-values ( r0 , r45 , r90 ) and flow stresses

M
AN
U

( 0 , 45 , 90 , b ) that occurred in a 3003-O aluminum sheet with increasing plastic strain. The flow stresses and rvalues calculated using the proposed model were compared with Hus experimental data. A sixth-order polynomial
model, Eq. (20), was employed for the yield function. Using this yield function, flow stresses 0 , 45 , 90 , b and
r-values r0 , r45 , r90 were given as functions of material parameters C1 ~ C16 in explicit forms (see Yoshida et al.,
2013).
1/ 6

1/ 6

1/ 6

(25)

r0 =

S 3T + 9U + 27C16
C6
C2
, r45 =
, r90 =
2C1 C2
2S + 12T + 18U
2C7 C6

(26)

TE
D

90

C1
C1
= 1 0 , 45 = 2
0, b = 0
S
C7
S + 9T + 27U + 27C16

S = C1 3C2 + 6C3 7C4 + 6C5 3C6 + C7 ,

T = C8 2C9 + 3C10 2C11 + C12 , U = C13 C14 + 3C15

EP

Figure 2 shows experimental data on the change in the flow stresses ( 0 , 45 , 90 , b ) and the corresponding
values predicted using Eq. (23). Interpolation equation Eq. (16) was employed using material constants a = 0.5 ,

1 = 70 and 2 = 10 . Material parameters of the yield function, Ck (0) and Ck ( ) , determined at = 0 and ,

AC
C

are listed in Table 1. The calculated stresses agree fairly well with the experimental results, although the predicted
b is slightly lower than the experimental value at relatively small strains. The comparisons of r-values are
illustrated in Fig. 3. The predicted values of r0 , r45 , and r90 agree well with the experimental results.

ACCEPTED MANUSCRIPT

160

100

RI
PT

120

0 (experiment = calculation)
45 (experiment)
45 (calculation)

80
60

90 (experiment)
90 (calculation)

40

SC

0, 45, 90, b (MPa)

140

b (experiment)

20

b (calculation)

0
0.05

0.1

0.15

0.2

M
AN
U

0.25

0.3

Fig. 2 Flow stresses of 3003-O aluminum sheet and values predicted using the anisotropy evolution model (see Eqs.
(23) and (16)).

0.7

r45 (experiment)
r45 (calculation)
r90 (experiment)
r90 (calculation)

EP

45

r ,r ,r

90

0.8

r0 (experiment)
r0 (calculation)

TE
D

0.9

AC
C

0.6

0.5

0.4

0.05

0.1

0.15

0.2

0.25

0.3

Fig. 3 R-values of 3003-O aluminum sheet and values predicted using the anisotropy evolution model (see Eqs. (23) and (16)).

ACCEPTED MANUSCRIPT

Ck (0) , k = 1~16, for 3003-O

aluminum sheet (experimental data are from Hu, 2007), determined at initial yielding ( = 0 ) and Ck ( ) at
infinitely large strain ( = ).
Table 1 Material parameters in the sixth-order polynomial-type yield function,

Ck ( )

at =

C2
0.826
C10
1.189
C2
0.767
C10
0.830

C3
0.815
C11
1.310
C3
0.742
C11
0.822

C4
0.854
C12
1.558
C4
0.660
C12
1.194

C5
0.988
C13
1.271
C5
0.705
C13
1.017

160

100
80
60

20
0

C7
1.510
C15
1.455
C7
1.378
C15
0.950

C8
1.290
C16
1.369
C8
1.043
C16
0.866

M
AN
U

120

TE
D

0, 45, 90, b (MPa)

140

40

C6
1.205
C14
0.533
C6
0.827
C14
0.401

RI
PT

at = 0

C1
1.000
C9
1.132
C1
1.000
C9
0.795

SC

Ck (0)

0.05

EP

0.1

0 (experiment = calculation)
45 (experiment)
45 (calculation)
90 (experiment)
90 (calculation)
b (experiment)
b (calculation)

0.15

0.2

0.25

0.3

AC
C

Fig. 4 Flow stresses of 3003-O aluminum sheet and values predicted using the anisotropy evolution model (see Eqs (24) and
(18)).

ACCEPTED MANUSCRIPT

r0 (experiment)
r0 (calculation)

0.9

r45 (experiment)
r45 (calculation)

RI
PT

r90 (experiment)
r90 (calculation)

0.7

45

r ,r ,r

90

0.8

SC

0.6

0.4
0

0.05

M
AN
U

0.5

0.1

0.15

0.2

0.25

0.3

Fig. 5 R-valuesof 3003-O aluminum sheet and values predicted using the anisotropy evolution model (see Eqs (24) and (18)).

Table 2 Material parameters in the sixth-order polynomial-type yield function,

Ck (i) , k = 1~16; i = 1~3, for 3003-O

= 0, 0.05, 0.30 .
at = 0
Ck (2)

at = 0.05

C2
0.826
C10
1.189
C2
0.789
C10
0.908
C2
0.770
C10
0.835

AC
C

Ck (3)

C1
1.000
C9
1.132
C1
1.000
C9
0.919
C1
1.000
C9
0.807

at = 0.30

C3
0.815
C11
1.310
C3
0.696
C11
1.037
C3
0.737
C11
0.839

EP

Ck (1)

TE
D

aluminum sheet (experimental data are from Hu, 2007), determined at three levels of effective plastic strain:

C4
0.854
C12
1.558
C4
0.623
C12
1.368
C4
0.656
C12
1.199

C5
0.988
C13
1.271
C5
0.670
C13
1.063
C5
0.701
C13
1.028

C6
1.205
C14
0.533
C6
0.927
C14
0.136
C6
0.839
C14
0.404

C7
1.510
C15
1.455
C7
1.386
C15
0.944
C7
1.378
C15
0.986

C8
1.290
C16
1.369
C8
1.119
C16
0.901
C8
1.053
C16
0.871

The flow stresses and r-values calculated using Eq. (17) and the interpolation equation of Eq. (18) for M = 3
were compared to the experimental data, and the results are shown in Figs 4 and 5, respectively. Three discrete
plastic strain points,

i = 0,

0.05, and 0.3 were selected to define 1 ( ) , 2 ( ) , and 3 ( ) . The material

parameters of the yield function,

Ck (1) , Ck (2) , and Ck (3) , determined at = 0 , 0.05, and 0.3, respectively, are

listed in Table 2. The material constants used in Eq. (18) were p1 = 0.25 for 0 0.05 and p2 = 0.60 for

0.05 0.3 . The calculated results agree well overall with experimental results for both the stresses and r-

ACCEPTED MANUSCRIPT

values even though we used just three sets of experimental data for stresses and r-values at
The yield surfaces at the plastic strain vlues of

= 0, 0.05, and 0.3 are illustrated in Fig. 6.

i = 0, 0.05, and 0.3.

200
Experiment

RI
PT

150
100

SC

y (MPa)

50

M
AN
U

-50
-100
-150

-200
-200 -150 -100

-50

50

100

150

200

TE
D

x (MPa)

Fig. 6 Yield surfaces at

i = 0, 0.05, and 0.3.

Plunkett et al. (2006) presented an anisotropic hardening model using a linear interpolation, i.e., corresponding to
the case of

pi = 1 in Eq. (18), based on Eq. (17). In their model, the flow stress, F , is also expressed by linear

EP

interpolation, together with the yield function, ( ) , as follows:

F = ( ) F ( i ) + (1 ( ) ) F ( i +1) , i = 1, 2, ..., M 1 , k = 1, 2, ..., N .

(27)

AC
C

Consequently, stressstrain curves calculated using the model are always piecewise linear, as illustrated in Fig. 7 for
the case of M = 4 for a 3003-O aluminum sheet. The change in the r-values with increasing plastic strain, calculated
using the model are illustrated in Fig. 8. These results show that the nonlinear interpolation model produces more
realistic stress-strain responses and r-value changes (see Figs. 7 and 8) than those calculated using the linear
interpolation model, although the former uses fewer sets of experimental data (M = 3) than the latter (M = 4).

ACCEPTED MANUSCRIPT

160
140

100

RI
PT

0, 45, 90 (MPa)

120

80

0 (experiment)
45 (experiment)
90 (experiment)
0 (calculation)
45 (calculation)

60

20

SC

40

90 (calculation)

0
0.05

0.1

0.15

0.2

0.25

M
AN
U

0.3

Fig. 7 Calculation of flow stresses for 3003-O aluminum sheet using linear interpolation equation.

r0 (calculation)
r45 (experiment)
r45 (calculation)
r90 (experiment)
r90 (calculation)

0.7

0.6

AC
C

r0, r45, r90

EP

0.8

r0 (experiment)

TE
D

0.9

0.5

0.4
0

0.05

0.1

0.15

0.2

0.25

0.3

Fig. 8 Calculation of r-values for 3003-O aluminum sheet using linear interpolation equation.

ACCEPTED MANUSCRIPT

The second example considered was the anisotropic hardening of a type 719-B stainless steel sheet (after
Stoughton and Yoon, 2009). In this sheet, r-value planar anisotropy remains fixed throughout the plastic
deformation. Figure 9 shows the r-values for various sheet directions, calculated by the yield function, together with
the corresponding experimental data. The flow stresses, 0 , 45 , 90 , b , calculated using Eq. (24) and Eq. (18),
with M = 3, are compared to the experimental data in Fig. 10. Three discrete plastic strain points,

i = 0, 0.1 and 0.5

were selected to define the sixth-order polynomial-type yield functions 1 ( ) , 2 ( ) and 3 ( ) . Material

Ck (1) , Ck (2) and Ck (3) determined at = 0 , 0.1 and 0.5, respectively, are listed

RI
PT

parameters of the yield function,

in Table 3. The material constants are p1 = 0.3 for 0 0.1 and p2 = 0.5 for 0.1 0.5 . The calculated

result agree well overall with the experimental result for the stresses. The yield surfaces at the plastic strains,
0.1 and 0.5, are shown in Fig. 11.

SC

3.5
Experiment
Calculation

M
AN
U

2.5

1.5

TE
D

r-value r

= 0,

15

EP

30

45

60

75

90

Tension axis angle from R.D. (degree)

AC
C

Fig. 9 R-value for type 790-B stainless steel sheet; experimental data (after Stoughton and Yoon, 2009) and
predictions obtained using sixth-order polynomial-type yield function (Yoshida et al., 2013).

ACCEPTED MANUSCRIPT

800

RI
PT

600
500

0 (experiment=calculation)

400

45 (experiment)
45 (calculation)

SC

0, 45, 90, b (MPa)

700

300

90 (experiment)

200

90 (calculation)

M
AN
U

b (experiment)

100

b (calculation)

0
0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

TE
D

Fig. 10 Flow stresses for type 790-B stainless steel sheet; experimental data (after Stoughton and Yoon,
2009) and predictions obtained using the proposed model (see Eqs (24) and (18)).

1000

Experiment

AC
C

y (MPa)

EP

500

-500

-1000
-1000

-500

500

1000

x (MPa)
Fig. 11 Yield surfaces for type 790-B stainless steel sheet and experimental data (from Stoughton and Yoon, 2009)
for

i = 0, 0.1, and 0.5.

ACCEPTED MANUSCRIPT

Table 3 Material parameters in the sixth-order polynomial-type yield function,

Ck (i) , k = 1~16; i = 1~3, for 790-B

stainless steel sheet (experimental data are from Stoughton and Yoon, 2009), determined at three levels of effective

at = 0
Ck (2)

at = 0.10
Ck (3)

C2
1.368
C10
1.462
C2
1.368
C10
1.446
C2
1.368
C10
1.471

C3
1.509
C11
1.526
C3
1.523
C11
0.969
C3
1.522
C11
1.629

C4
1.667
C12
1.063
C4
1.668
C12
1.041
C4
1.732
C12
1.149

C5
1.571
C13
0.996
C5
1.562
C13
0.882
C5
1.663
C13
0.939

C6
1.464
C14
1.278
C6
1.446
C14
1.351
C6
1.584
C14
1.311

M
AN
U

at = 0.50

C1
1.000
C9
1.517
C1
1.000
C9
1.562
C1
1.000
C9
1.485

C7
0.982
C15
0.882
C7
0.969
C15
1.120
C7
1.062
C15
1.198

C8
1.181
C16
0.927
C8
1.041
C16
0.953
C8
1.130
C16
1.009

RI
PT

Ck (1)

= 0, 0.10, 0.50 .

SC

plastic strain:

The third example considered was the variation of the yield surface of a sheet metal during plastic deformation.
Yanaga et al. (2014) obtained very interesting experimental data for a 6016-T4 aluminum sheet by performing a
specially designed biaxial test, in which the shape of the yield surface changed significantly with increasing plastic
strain. In the present study, the following Yld2000-2d yield function (Barlat et al., 2003),
with Eqs (14) and (18) for 0.002 0.16 (M = 2,

yld2000 ( s) , was

1 = 0.002 and 2 = 0.16) to model this scenario:

1/ m

TE
D

m
m
m
1
yld2000 ( s ) = yld2000 ( S%1' , S%2' , S%1" , S%2" ) = S%1' - S%2' + 2S%2'' + S%1'' + 2S%1" + S%2"
2

In Eq. (28), s is the stress deviator and

which are defined as follows:

AC
C

.
(28)

S%1' , S%2' and S%1" , S%2" are the principal values of s' and s" , respectively,

sxx
s xx
4 5 3 2 6 2 4

s , s = s = 1 2 2 4
yy
5
4
6
3
yy

s 3
s xy
0
0

xy

EP

sxx 1 0 0

s' = syy = 0 2 0

s 0 0
7
xy

used

0
0
38

sxx
s
yy
sxy

(29)

The Yld2000-2d parameters, 1 ~ 8 , m , for 1 = 0.002 and 2 = 0.16 are listed in Table 4. In Fig. 12, the yield
surfaces (equi-plastic work surfaces) for the 6016-T4 aluminum sheet under proportional loading, calculated using
the proposed model (Eqs (24) and (18)), are compared with the corresponding experimental results. For the
calculation only two sets of Yld2000-2d parameters determined at

1 = 0.002 and 2 = 0.16, as listed in Table 4,

with an interpolation parameter, value p1 = 0.5 for Eq. (18) were used. The calculated results for the yield surfaces
agree well with the experimental data. In contrast to this approach, Yanaga et al. gave the following approximation
equations for 1 ~ 8 and m:

k = Ak Bk exp ( Ck ) + ak , k = 1, 2, ..., 8

(30)

ACCEPTED MANUSCRIPT

m=

A1 A2
.
1 + exp {( x0 ) / x}

(31)

yield surface data sets at

1 = 0.002

and

2 = 0.16

and one more data set at a plastic strain between

RI
PT

In Eqs (30) and (31), Ak , Bk , Ck , ak , x0 and x are coefficients (34 coefficients in total) whose values need to be
determined experimentally. Yanaga et al. demonstrated that thus formulated yield function reproduces the
experimentally obtained yield surfaces very well. However, this approach to approximation equation building for
material parameters might lead to a difficulty in the determination of many coefficient values. From this perspective,
the present approach offers advantages in that it requires the determination of many fewer parameter values and it
requires fewer sets of experimental data for parameter identification. In the case of the 6016-T4 sheet, only two
were needed to determine the 1 ( ) and 2 ( ) parameters,

1 and 2 ,

e.g., at

0.06 was needed to determine the

interpolation parameter, p1 . Furthermore, the proposed approach always guarantees the convexity of the yield

( , ) ,

for any type of yield function, when

i ( )

and

i +1 ( )

are convex. In contrast, with the

SC

surface

M
AN
U

approach that uses approximation equations for material parameters, convexity is theoretically guaranteed only for
yield functions that satisfy the convexity for any material parameters (which is true for Yld2000-2d, for example,
but not for high-order polynomial-type yield functions).

400
350

250

TE
D

y (MPa)

300

200
150

AC
C

50

= 0.02
= 0.04
= 0.06
= 0.10
= 0.16

50

100

150

Calculation

= 0.002

EP

100

Experiment

200

250

300

350

400

x (MPa)

Fig. 12 Calculation of the change in yield surfaces according to the present model for a 6014-T4 aluminum sheet
(experimental data are from Yanaga et al., 2014)

ACCEPTED MANUSCRIPT

k , k = 1~8, and m for 6016-T4 aluminum sheet,

= 0.002 and 0.160 (after Yanaga et al., 2014).

Table 4 Material parameters in Yld2000-2d yield function,

determined at two levels of effective plastic strain:

3.

0.907

0.993

0.922

1.045

1.032

0.988

0.830

1.429

6.7

0.974

1.022

0.972

1.010

1.003

0.992

1.001

1.204

38.7

Combined anisotropic-kinematic hardening model

RI
PT

k and m
at = 0.002
k and m
at = 0.160

M
AN
U

f = ( , ) Y = (% , ) Y = 0,
% = ,

SC

3.1. Framework of modeling

Kinematic hardening modeling is the most popular approach used to describe the Bauschinger effect and the
cyclic plasticity behavior of metals. The proposed anisotropic hardening model is easily combined with kinematic
hardening models. In kinematic hardening modeling, the yield function f is given by the following equation:
(32)

where denotes the backstress. Based on the following definitions of the effective plastic strain and its rate

& = % : D p , = & dt ,
the associated flow rule is written as follows:

Dp =

f & &
=
,
%
%

(33)

(34)

TE
D

where & = & .

Most kinematic hardening models assume the following form of the evolution equation of the back stress:

A
A
( ) x & = % x & .
Y

(35)

Here (o) denotes the objective rate. For example, in the linear kinematic hardening model:

H LK
H
( ) & = LK % & ,
Y
Y
A = HLK , x = 0 .

EP

(36)

In the ArmstrongFrederick model (1966):

1
( ) 2 & = 1 % 2 & ,
Y

AC
C

(37)

A = 1, x = 2 .

The YoshidaUemori kinematic hardening law has the same form (for details, see Sec. 3.2.). The constitutive
equation is given in the same form as Eqs. (8) and (9), where H = H F = F / is replaced with H K as follows:

H = H K = A

:x.
%

(38)

3.2. Anisotropic hardening based on the Yoshida-Uemori model

The YoshidaUemori model (2002a, 2002b, 2003) was constructed within the framework of two-surface
modeling, wherein the yield surface moves kinematically within a bounding surface, as schematically illustrated in

ACCEPTED MANUSCRIPT

Fig. 13. To describe anisotropic hardening (i.e., expansion of the surface with shape change) and also kinematic
hardening, the bounding surface F is expressed by the equation:

F = ( , ) ( B + R) = 0 ,

(39)

TE
D

M
AN
U

SC

RI
PT

where denotes the center of the bounding surface, and B and R are the initial size of the surface and its
workhardening component, respectively. To include the description of anisotropic hardening in the model, it is
assumed that the shapes of both the yield and bounding surfaces vary simultaneously as a function of the effective
plastic strain, as schematically illustrated in Fig. 14.

AC
C

EP

Fig 13 Schematic illustration of the YoshidaUemori two-surface model

Initial state

After plastic deformation

Fig. 14 Schematic illustration of the changes of the yield surface and the bounding surface.

ACCEPTED MANUSCRIPT

The kinematic hardening of the yield surface describes the transient Bauschinger deformation, which is
characterized by early re-yielding and subsequent rapid change in workhardening rate. The relative kinematic
motion of the yield surface with respect to the bounding surface is expressed by the following equation:

* = .

(40)

a
* = C ( )

Ca
a & ,

% C
* *
Y

* & =

* = (* ) , a = B + R Y .

RI
PT

The evolution of * is given by the following equation:

(41)

(42)

o
b

kb

= k ( ) & = % k & .

Thus, in Eq. (35),

A = Ca + kb,

x=C

and in Eq. (9),

H = H K = Ca + kb

M
AN
U

SC

An ArmstrongFrederick-type evolution equation is used to express the kinematic hardening of .

k ,
*

a
a
: C
k .
C
*
*

(43)

(44)

(45)

written as

TE
D

With respect to the expansion of the bounding surface, i.e., the evolution of R, in the first version of the Yoshida
Uemori model (2002b, 2003), the following equation based on the Voce hardening law was proposed:
(46)
R = RVoce = Rsat {1 exp(k )} ,

R& = R&Voce = k ( Rsat RVoce )& ,.

(47)

However, it is not necessary to use the Voce-type formulation. For example, based on the Swift law:
(48)
R = RSwift = K ( 0 + )n 0n ,

EP

the following evolution equation can be obtained

AC
C

( n 1)/ n
R& = R& Swift = nK 1/ n ( RSwift + K 0n )
& .

(49)

Furthermore, a combination of the above two hardening laws, Eqs (48) and (49), is also possible and can be
expressed as follows:

R& = R&Swift + (1 ) R&Voce , 0 1 ,

(50)

where is a weighting coefficient. This model has high flexibility in describing various levels of workhardening at
large strain levels.
One of the features of the YoshidaUemori model is that it is able to describe the workhardening stagnation that
appears in a reverse stress-strain curve for a certain range of reverse deformation (see Hasegawa and Yakou, 1975;
Christodoulou et al., 1986). This phenomenon is closely related to the strain-range and mean-strain dependency of
cyclic hardening. Specifically, the larger the cyclic strain range is, the larger the saturated stress amplitudes are. This
dependency is expressed by the stagnation of the expansion of the bounding surface for a certain range of reverse

& > 0 ) and non-hardening ( R& = 0 ) of the bounding surface are determined
deformation. The states of hardening ( R

ACCEPTED MANUSCRIPT

for a so-called non-IH (isotropic hardening) surface, g , defined in the stress space as follows and schematically
illustrated in Figs 15(a) and (b):
(51)
g = ( q, r ) r = 0 ,
where q and r denote the center and size of the non-IH surface, respectively. It is assumed that the center of the

only when the center point of the bounding surface, q, lies on the surface

R& > 0 when

g ( q, r ) = ( q, r ) r = 0 and =

g ( q, r ) o
: >0

(see Fig. 15(b)), i.e.,

(52a)

SC

R& = 0 otherwise.

RI
PT

bounding surface q exists either on or inside of the surface g . The expansion of the bounding surface takes place

(52b)

In an analysis of some experimental data, the plastic strain region of workhardening stagnation was found to

q=

M
AN
U

increase with the accumulated plastic strain. To describe this phenomenon, it was assumed that the surface g
moves kinematically as it expands. The governing equations of the kinematic motion and expansion of the surface
are given by Eqs (53) and (54), respectively.

(1 h )
( q) ,
r

r& = h .

(53)
(54)

Here, h is a parameter that controls the strength of the workhardening stagnation characteristic. A larger value of h
corresponds to a larger strain region within which workhardening stagnation occurs, and as a result, a larger value of

or even

= von

TE
D

h leads to weaker cyclic hardening of a material. We may assume that the shape of the surface
Mises type, throughout the deformation, because the shape of

g , is fixed = 0

g has not been measured

experimentally yet, and its effect on the stress-strain calculation would be rather minor.

EP

In the proposed model, the size of the yield surface is held constant. However, if we carefully observe the stressstrain response during unloading after plastic deformation, we find that the stressstrain curve is no longer linear but
rather is slightly curved due to very early re-yielding and the Bauschinger effect. To describe this phenomenon, in
the model, the following equation for plastic-strain-dependent Youngs modulus is introduced:
(55)
E = Eo ( Eo Ea ) {1 exp( p )} ,

AC
C

where Eo and Ea are Youngs modulus for virgin and infinitely large pre-strained materials, respectively, and is a
material constant.

Fig. 15 Schematic illustration of the non-IH surface

g defined in the stress space, when expansion of the bounding

M
AN
U

surface (a) stops, and (b) takes place.

3.3. Validation of the model

SC

RI
PT

ACCEPTED MANUSCRIPT

To validate the model, cyclic stress-strain responses for a 780R AHSS sheet were examined. Cyclic strain tests
and uniaxial tension tests were conducted using specimens cut from the sheet in three directions: 0o (rolling
direction), 45o and 90o. The flow stresses and r-values of the three specimens were significantly different,
depending on the sheet direction. The uniaxial tension test results showed that the stress directionality, measured
with respect to normalized stresses,

45 / 0

and

90 / 0 , varies with increasing plastic strain, whereas r-values

TE
D

do not change much throughout the deformation range ( r0 = 0.46 0.04 , r45 = 1.12 0.10 , r90 = 0.62 0.06 ).
A sixth-order polynomial yield function (Yoshida et al., 2013, see Eq. (20)) was employed for the calculations. For
the evolution law for material parameters, Ck , k = 1, 2,...,16 , Eq. (23), with the interpolation equation Eq. (15), was
employed, with = 50.0 . The material parameters of the yield function,

Ck (0) and Ck ( ) , are listed in Table 5.

AC
C

EP

The YoshidaUemori model parameters are listed in Table 6. It should be emphasized that, in the framework of this
anisotropic-kinematic hardening model, a set of kinematic hardening parameters are determined from uniaxial
experiments (uniaxial tension and cyclic plasticity) only in the rolling (0o) direction and remain fixed throughout the
plastic deformation. The calculated stress-strain curves in the three sheet directions in uniaxial tension were
compared to the corresponding experimental results, as shown in Figs 16(a) and (b). The calculated results agree
well with the experimental results for plastic-strain-dependent stress directionality. The experimental and calculated
results for stress-strain responses under cyclic straining are illustrated in Figs 17(a) and (b), respectively. The
calculated results reflect the sheet-direction-dependent cyclic hardening characteristic well.

ACCEPTED MANUSCRIPT

Ck (0) , k = 1~16, for 780R AHSS

at infinitely large strain ( = ).

Table 5 Material parameters in the sixth-order polynomial-type yield function,

Ck (0)

at = 0
Ck ( )

at =

C1
1.0000
C9
0.6219
C1
1.0000
C9
0.6651

C2
0.6315
C10
0.3354
C2
0.6306
C10
0.3976

C3
0.3813
C11
0.4604
C3
0.4069
C11
0.5905

C4
0.2635
C12
0.7029
C4
0.2608
C12
0.8057

Ck ( )

C5
0.3632
C13
1.2282
C5
0.3798
C13
1.2639

C6
0.4843
C14
0.7522
C6
0.6406
C14
0.4497

B (MPa)

b (MPa)

550

750

100

C2

800

250

12.0

0.2

(b) parameters for plastic-strain-dependent Youngs modulus

Ea (GPa)

200.0

160.0

120.0

EP

TE
D

E0 (GPa)

C8
1.0678
C16
1.0707
C8
1.1349
C16
1.2436

C = C1 for 0 0.005 ; C = C2 for > 0.005 .

AC
C

C1

Rsat (MPa)
140

M
AN
U

Y (MPa)

SC

Table 6 YoshidaUemori parameters

(a) plasticity parameters

C7
0.6357
C15
0.8913
C7
.8381
C15
0.9255

RI
PT

sheet, determined at initial yielding ( = 0 ) and

ACCEPTED MANUSCRIPT

RI
PT

1000

600

0 experiment
calculation

400

45o experiment
calculation

SC

True stress (MPa)

800

90o experiment
calculation

0.02

M
AN
U

200

0.04

0.06

0.08

0.1

True strain

AC
C

EP

TE
D

Fig. 16 Stress-strain curves for 780R AHSS under uniaxial tension in three sheet directions.

ACCEPTED MANUSCRIPT

0o
45o

1200

90o

RI
PT

400
0

SC

True stress (MPa)

800

-400

M
AN
U

-800
-1200
-0.1

-0.05

0.05

0.1

0.05

0.1

True strain

(a) Experiment

1200

45

90

EP

400
0

AC
C

True stress (MPa)

800

TE
D

-400
-800

-1200
-0.1

-0.05

0
True strain
(b) Simulation

Fig. 17 Stress-strain curves for 780R AHSS under cyclic loading in three sheet directions.

ACCEPTED MANUSCRIPT

4.

Concluding remarks

The present paper proposes a framework for constitutive modeling of plasticity to describe the evolution of
anisotropy and the Bauschinger effect in sheet metals. The highlights are summarized as follows.
An anisotropic yield function, which varies continuously with plastic strain, is defined by a nonlinear
interpolation function of the effective plastic strain using a limited number of yield functions determined at a
few discrete points of plastic strain (see Eqs (11) and (14)). In this modeling framework, it is possible to use
any type of yield function; the convexity of the yield surface is always guaranteed.

When using a polynomial type of yield function, this model reduces to the interpolation of its material
parameters (see Eqs (22) -(24)).

This approach, which requires only one interpolation equation, offers a great advantage over other approaches
in that it involves fewer material parameters; some other approaches use many approximation equations for
each material parameter.

SC

RI
PT

Several types of nonlinear equations (see Eqs (15), (16) and (18)) that involve very few material parameters can
be used as interpolation functions.

M
AN
U

The description of the Bauschinger effect and the evolution of anisotropy is made possible by incorporating the
proposed anisotropic hardening model in a kinematic hardening model. A set of kinematic hardening
parameters can be identified from experiments independent of anisotropic hardening parameters, and their
values remain fixed throughout the plastic deformation (see Appendix: scheme of anisotropic-kinematic
hardening model).

The proposed anisotropic hardening model was validated by comparing calculated results of stressstrain
responses, r-value variations, and changes in the shape the yield surface, with experimental data for several
types of aluminum and stainless steel sheets. For cyclic plasticity, details of the modeling approach, which is
based on the YoshidaUemori model, were described. The model was validated by comparison with the results
of uniaxial tension and cyclic tensioncompression tests for several sheet directions for an AHSS sheet.

TE
D

Acknowledgement

EP

The authors would like to thank Professor T. Kuwabara of Tokyo University of Agriculture and Technology for
allowing us to use his published data on the evolutionary change of the yield surfaces on 6016-T4 sheet. We are
grateful to Mr. H. Hasegawa of Hiroshima University for his help with experiment on an AHSS sheet.
Appendix. Scheme of material parameter identification in the anisotropic-kinematic hardening model

AC
C

In this framework of anisotropic-kinematic hardening modeling based on the Yoshida-Uemori model, parameters
in an anisotropic yield function and in the kinematic hardening law are determined independently, as follows:
(a) Anisotropic parameters
The stress-strain responses under monotonic proportional loadings predicted by the Yoshida-Uemori model,
using an anisotropic yield function, are almost the same as those calculated by the isotropic hardening model
(refer to Yoshida and Uemori, 2003). Thus the anisotropic parameters in a yield function can be identified in
just the same way as that in the isotropic hardening modeling, where we use experimental data of flow stresses
(e.g., 0 , 45 , 90 ) and r-values (e.g., r0 , r45 , r90 ) in several sheet directions and some biaxial stresses (e.g.,
equi-balanced stress b ). Note that, to describe the anisotropy evolution, only a few sets of these experimental
data at discrete levels of plastic strains, i , i = 1, 2, .., M are needed, e.g., for 3003-O aluminum in Section 2.3
(case of M = 3), 45 , 90 , b and r0 , r45 , r90 data at = 0 (initial yielding), 0.05 and 0.30 were used together with the
continuous stress-strain data of

0 0p .

ACCEPTED MANUSCRIPT

RI
PT

(b) The Yoshida-Uemori model parameters

This model contains 7-8 plasticity (kinematic hardening) parameters (e.g., see Table 6(a)) and 3 elasticity
parameters (e.g., see Table 6(b)). The plasticity parameters are identified from uniaxial tension and cyclic
straining experiments, 0 0 , in rolling direction (refer to Yoshida and Uemori, 2002b). The elasticity
parameters are determined from the sequential tension-unloading experiment (refer to Yoshida et al., 2002a).

TE
D

M
AN
U

SC

References
An, Y. G., Vegter, H., Melzer, S., Triguero, P. R., 2013, Evolution of the plastic anisotropy with straining and its
implication on formability for sheet metals. Journal of Materials Processing Technology, 213, 1419-1425.
Aretz, H., 2008, A simple isotropic-distortional hardening model and its application in elastic-plastic analysis of
localized necking in orthotropic sheet metals. Int. J. Plast. 24, 1457-1480.
Armstrong, P. J., Frederick, C. O., 1966. A mathematical representation of the multiaxial Bauschinger effect. GEGB
report RD/B/N731. Berkley Nuclear Laboratories.
Banabic, D., Aretz, H., Comsa, D. S., Paraianu, L., 2005. An improved analytical description of orthotropy in
metallic sheets. Int. J. Plast. 21, 493-512.
Barlat, F., Brem, J. C., Yoon, J. W., Chung, K., Dick, R. E., Lege, D. J., Pourgoghrat, F., Choi, S. H., Chu, E., 2003.
Plane stress yield function for aluminum alloy sheets - part 1: theory. Int. J. Plast. 19, 1297-1319.
Barlat, F., Aretz, H., Yoon, J. W., Karabin, M. E., Brem, J. C., Dick, R. E., 2005. Linear transfomation-based
anisotropic yield functions. Int. J. Plast. 21(5), 1009-1039.
Barlat, F., Yoon, J. W., Cazacu, O., 2007. On linear transformations of stress tensors for the description of plastic
anisotropy. Int. J. Plast. 23(5), 876-896.
Barlat, F., Gracio, J. J., Rauch, E. F., Vincze, G., 2011. An alternative to kinematic hardening in classical plasticity,
Int. J. Plast. 27, 1309-1327.
Barlat, F., Ha, J., Gracio, J. J., Lee, M.-G., Rauch, E. F., Vincze, G., 2013. Extension of homogeneous anisotropic
hardening model to cross-loading with latent effects, Int. J. Plast. 46, 130-142.
Barlat, F., Vincze, G., Gracio, J. J., Lee, M.-G., Rauch, E. F., Tome, C. N., 2014. Enhancements of homogeneous
anisotropic hardening model and application to mild and dual-phase steels, Int. J. Plast. 58, 201-218.
Chaboche J. L, Rousselier G., 1983. On the plastic and viscoplastic constitutive equations, Part I and II.

Transactions of ASME, Journal of Pressure Vessel Technology 105,153-164.

AC
C

EP

Chaboche, J. L., 2008. A review of some plasticity and viscoplasticity constitutive theories. Int. J. Plast. 24, 16421693.
Christodoulou, N., Woo, O.T., MacEwen, S.R., 1986. Effect of stress reversals on the work hardening behaviour of
polycrystalline copper, Acta Metall. 34, 15531562.
Dafalias, Y. F., 1985.The plastic spin, Trans. ASME, J. Appl. Mech., 52, 865-871.
Eggertsen, P.-A., Mattiasson, K., 2009. On the modelling of the bending-unbending behaviour for accurate
springback predictions, Int. J. Mech. Sci. 51, 547-563.
Eggertsen, P.-A., Mattiasson, K., 2010. On constitutive modeling of springback analysis, Int. J. Mech. Sci. 52, 804818.
Feigenbaum, H. P., Dafalias, Y. F., 2007. Directional distortional hardening in metal plasticity with
thermodynamics, Int. J. Solids and Structures 44, 7526-7542.
Francois, M., 2001. A plasticity model with yield surface distortion for non-proportional loading, Int. J. Plast. 17,
703-717.
Geng, L. M., Wagoner, R. H., 2002. Role of plastic anisotropy and its evolution on springback, Int. J. Mech. Sci. 44,
123-148.
Gotoh, M., 1977. Theory of plastic anisotropy based on a yield function of 4th-order (plane stress state) .1. Int. J.
Mech. Sci. 19, 505-512.
Haddadi, H., Bouvier, S., Banu, M., Maier, C., Teodosiu, C., 2006. Towards an accurate description of the
anisotropic behavior of sheet metals under large plastic deformation: modelling, numerical analysis and
identification, Int. J. Plast. 22, 2226-2271.
Hasegawa, T., Yakou, T., 1975. Deformation behaviour and dislocation structures upon stress reversal in
polycrystalline aluminium. Mater. Sci. Eng. 20, 267276.
Hill, R., 1948. A theory of the yielding and plastic flow of anisotropic metals. Proc. Roy. Soc. Lond. A 193, 281-297.

ACCEPTED MANUSCRIPT

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

Hu, W. L., 2007. Constitutive modeling of orthotropic sheet metals by presenting hardening-induced anisotropy. Int.
J. Plast. 23, 620-639.
Kurtyka, T., Zyczkowski, M., 1996. Evolution equations for distorsion al plastic hardening, Int. J. Plast. 23, 191-213.
Kuwabara, T., Ikeda, S., Kuroda, K., 1998, Measurement and analysis of differential work hardening in cold rolled
steel sheet under biaxial tension, Journal of Materials Processing Technology 80-81, 517-523.
Ohno, N., 1982. A constitutive model of cyclic plasticity with a non-hasrdening strain range. ASME J. Applied
Mechanics 49, 721-727.
Ohno N, Wang J-D, 1993. Nonlinear kinematic hardening rule with critical state of dynamic recovery, Part I:
Formulation and basic features for ratchetting behavior, International Journal of Plasticity 9, 375-390.
Plunkett, B., Lebensohn, R. A., Cazacu , O., Barlat, F., 2006. Anisotropic yield function of hexagonal materials
taking into account texture development and anisotropic hardening. Acta Materialia 54, 4159-4169.
Safaei, M., Lee, M.-G., Zang, S., Waele, W. D., 2014, An evolutionary anisotropic model for sheet metals based on
non-associated flow rule approach, Computational Materials Science, 81, 15-29.
Shiratori, E., Ikegami, K., Yoshida, F., 1979. Analysis of stress-strain relations by use of anisotropic hardening
plastic potential, J. Mech. Phys. Solids 27, 213-229.
Soare, S., Yoon, J. W., Cazacu, O., 2008. On the use of homogeneous polynomials to develop anisotropic yield
functions with applications to sheet forming. Int. J. Plast. 24, 915-944.
Stoughton, T. B, Yoon, J. W., 2009, Anisotropic hardening and non-associated flow in proportional loading of sheet
metals, Int. J. Plast. 25, 1777-1817.
Taleb, L., 2013, About cyclic strain accumulation of the inelastic strain observed in metals subjected to cyclic stress
control, Int. J. Plast. 43, 1-19.
Tozawa, Y., 1978, Plastic deformation behavior under conditions of combined stress. In: Koistinen, D. P., Wang,
N.-M. (Eds.), Mechanics of Sheet Metal Forming, Plenum Press, pp.81-110.
Voyiadjis, G. Z., Foroozesh, M., 1990, Anisotropic distortional yield model, Trans. ASME, J. Appl. Mech. 57, 537547.
Wagoner, R. H., Lim, H., Lee, M.-G., Advanced issue in springback. Int. J. Plast. 45, 3-20.
Yanaga, D., Takizawa, H., Kuwabara, T., 2014, Formulation of differential work hardening of 6000 series aluminum
alloy sheet and application to finite element analysis, Trans JSTP 55, 55-61. (in Japanese)
Yoon, J. W., Lou, Y., Yoon, J., Glazoff, M. V., 2014. Asymmetric yield function based on the stress invariants for
pressure sensitive metals, Int. J. Plast. 56, 184-202.
Yoshida, F., 2000. A constitutive model of cyclic plasticity. Int. J. Plast. 16, 359-380.
Yoshida, F., Uemori, T., Fujiwara, K., 2002a. Elastic-plastic behavior of steel sheets under in-plane cyclic tensioncompression at large strain. Int. J. Plast. 18, 633-659.
Yoshida, F., Uemori, T., 2002b. A model of large-strain cyclic plasticity describing the Bauschinger effect and
workhardening stagnation. Int. J. Plast. 18, 661-686.
Yoshida, F., Uemori, T., 2003. A model of large-strain cyclic plasticity and its application to springback simulation.
Int. J. Mech. Sci. 45, 1687-1702.
Yoshida, F., Kaneda, Y., Yamamoto, S., 2008. A plasticity model describing yield-point phenomena of steels and its
application to FE simulation of temper rolling. Int. J. Plast. 24, 1792-1818.
Yoshida, F., Hamasaki, H., Uemori, T., 2013. A user-friendly 3D yield function to describe anisotropy of steel
sheets, Int. J. Plast. 45, 119-139.