Vous êtes sur la page 1sur 113

Theory of Plasticity

Lecture Notes

Spring 2012
Contents

I Theory of Plasticity 1

1 Mechanical Theory of Plasticity 2

1.1 Field Equations for A Mechanical Theory . . . . . . . . . . . . . . . . . . . . 2

1.1.1 Strain-displacement Relations . . . . . . . . . . . . . . . . . . . . . . 3

1.1.2 Equilibrium Equations . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.1.3 Compatibility Conditions . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.1.4 Constitutive Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.2 Uniaxial Inelastic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1.3 Yield Function and Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1.4 General Framework for Plastic Constitutive Relations . . . . . . . . . . . . . . 14

1.5 Uniqueness and Stability Postulates . . . . . . . . . . . . . . . . . . . . . . . 21

1.6 Properties of Yield Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

1.7 Initial Yield Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

1.7.1 von Mises Yield Criterion . . . . . . . . . . . . . . . . . . . . . . . . 40

1.7.2 Tresca Yield Criterion . . . . . . . . . . . . . . . . . . . . . . . . . . 42


ii
1.7.3 Properties of Initial Yield Surfaces . . . . . . . . . . . . . . . . . . . . 43

1.8 Subsequent Yield Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

1.8.1 Isotropic Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

1.8.2 Kinematic Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

1.8.3 Isotropic Hardening vs. Kinematic Hardening . . . . . . . . . . . . . . 53

1.9 Incremental Formulation for Isotropic Hardening . . . . . . . . . . . . . . . . 54

1.10 Incremental Formulation for Kinematic Hardening . . . . . . . . . . . . . . . 66

1.11 Three Dimensional Plasticity . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

1.11.1 Initial Yielding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

1.11.2 Isotropic Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

1.11.3 Kinematic Hardening (3D) . . . . . . . . . . . . . . . . . . . . . . . . 76

1.11.4 Combined Isotropic and Kinematic Hardening . . . . . . . . . . . . . 78

2 Implementation in the Finite Element Method 80

2.1 Principle of Virtual Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

2.2 Principle of Minimum Total Potential Energy . . . . . . . . . . . . . . . . . . 85

2.3 Introduction to Finite Element Method . . . . . . . . . . . . . . . . . . . . . . 86

2.4 An Example of Constant Stress Triangular Elements . . . . . . . . . . . . . . 92

2.4.1 Plane Strain Elements . . . . . . . . . . . . . . . . . . . . . . . . . . 95

2.4.2 Plane Stress Elements . . . . . . . . . . . . . . . . . . . . . . . . . . 96

2.5 Incremental Plasticity Finite Element . . . . . . . . . . . . . . . . . . . . . . . 98

2.6 Numerical Implementation of the Elastoplastic Constitutive Relations . . . . . 102


iii
2.7 Iteration on the Global Equations . . . . . . . . . . . . . . . . . . . . . . . . . 104

3 Limit Analysis 105

3.1 Limit Functions and Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

3.2 Generalized Stresses and Strains . . . . . . . . . . . . . . . . . . . . . . . . . 105

3.3 Limit Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

3.4 Limit Analysis for Beams and Frames . . . . . . . . . . . . . . . . . . . . . . 105

iv
List of Figures

1.1 Generalized deformation of a three-dimensional solid body. . . . . . . . . . . . 3

1.2 Measurement of infinitesimal strains. . . . . . . . . . . . . . . . . . . . . . . . 4

1.3 Uniaxial loading of a prismatic bar. . . . . . . . . . . . . . . . . . . . . . . . . 7

1.4 Rate dependent material. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.5 Time dependent material. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.6 The Baushinger effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.7 Strain hardening. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

1.8 A typical stress-strain curve for metals. . . . . . . . . . . . . . . . . . . . . . 10

1.9 Rigid-perfectly plastic model. . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1.10 Elastic-perfectly plastic model. . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1.11 Elastic-linear hardening model. . . . . . . . . . . . . . . . . . . . . . . . . . . 12

1.12 Multiple stress strain states. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

1.13 Yield function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

1.14 Elastic-perfectly plastic behavior. . . . . . . . . . . . . . . . . . . . . . . . . . 14

1.15 Strain hardening behavior under uniaxial stress state . . . . . . . . . . . . . . . 14


v
1.16 Yield function in terms of the stress tensor and internal variable . . . . . . . . . 16

1.17 Unloading from a yield surface. . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.18 Neutral loading along a yield surface. . . . . . . . . . . . . . . . . . . . . . . 17

1.19 Normality of plastic strains to the yield surface. . . . . . . . . . . . . . . . . . 19

1.20 Two representative material behaviors. . . . . . . . . . . . . . . . . . . . . . . 22

1.21 Monotonic loading. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

1.22 Net work and net complementary work. . . . . . . . . . . . . . . . . . . . . . 24

1.23 Two representative material behaviors both satisfying Postulate 1. . . . . . . . 25

1.24 Moving of yield surfaces accounting for plastic strain increments. . . . . . . . 28

1.25 Representation of the stress tensor as a vector in principal directions. . . . . . . 32

1.26 Stress traction acting on the surface of a general elastoplastic body with outward

unit normal n. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

1.27 Decomposition of the stress traction vector into normal and shear components. 36

1.28 A unit vector defining octahedral planes in the principal space. . . . . . . . . . 40

1.29 von Mises yield surface represented in principal directions of stress tensor. . . . 41

1.30 Projection of von Mises yield surface on the -plane. . . . . . . . . . . . . . . 42

1.31 Simultaneous projection of von Mises and Tresca yield surface on the -plane. 46

1.32 Yield surfaces of both von Mises and Tresca criterion plotted on a single biaxial

stress coordinate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

1.33 The yield surface of a thin-walled tube under combined axial and torsional load-

ing predicted by von Mises yield criterion. . . . . . . . . . . . . . . . . . . . . 48


vi
1.34 The yield surface of a thin-walled tube under combined axial and torsional load-

ing predicted by von Mises and Tresca yield criteria. . . . . . . . . . . . . . . 49

1.35 Initial and subsequent yield surfaces of isotropic modeling from a uniaxial test. 53

1.36 Initial and subsequent yield surfaces of kinematic modeling from a uniaxial test. 54

1.37 The schematic of a thin-walled tube simultaneously subjected to axial and tor-

sional loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

1.38 Two possible loading paths of the tension-torsion test of a thin-walled tube. . . 60

1.39 Characterization of loading path 1 of the tension-torsion test of a thin-walled tube. 61

1.40 The schematic of a thick-walled tube simultaneously subjected to axial and tor-

sional loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

1.41 Elastic-linear hardening behavior of a thick-walled tube simultaneously sub-

jected to axial and torsional loads. . . . . . . . . . . . . . . . . . . . . . . . . 64

1.42 A differential element of a thick-walled tube simultaneously subjected to axial

and torsional loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

1.43 Movement of the yield surface due to kinematic hardening. . . . . . . . . . . . 67

1.44 An elastic and linear-hardening material. . . . . . . . . . . . . . . . . . . . . . 69

1.45 Two possible loading paths of an incremental kinematic hardening model. . . . 70

2.1 Tractions and displacements specified boundaries of a solid body. . . . . . . . 81

2.2 A cantilever beam subjected to a few generalized loads. . . . . . . . . . . . . . 83

2.3 A constant-stress triangular element. . . . . . . . . . . . . . . . . . . . . . . . 88

2.4 A rectangular plate element. . . . . . . . . . . . . . . . . . . . . . . . . . . . 89


vii
List of Tables

viii
Part I

Theory of Plasticity

1
Chapter 1

Mechanical Theory of Plasticity

Notation: direct tensor notation as well as Cartesian index will be used throughout the manuscript.

1.1 Field Equations for A Mechanical Theory

Here we review the governing equations for the infinitesimal deformations (small strain) of a

continuous medium. We restrict ourselves to purely mechanical theory and hence neglect the

first and second law of thermodynamics.

Consider a body B as shown:

As we deform the body particle Q, originally with position vector x measured in an arbitrary

coordinate system, moves to a new point q. The distance vector traveled is denoted by the

displacement vector u(x).


2
Figure 1.1: Generalized deformation of a three-dimensional solid body.

1.1.1 Strain-displacement Relations

1
 = [(u) + (u)] , (1.1)
2

or

" #
1 ui uj
ij = + . (1.2)
2 xj xi

For example, 11 = u1 /x1 gives a measure of the change in length per unit length of a dif-

ferential vector oriented in x1 direction and 12 = 21 [u1 /x2 + u2 /x1 ] of the angle change

between x1 and x2 axes. Note that the engineering strain tensor  is not valid for finite deforma-

tions.
3
Figure 1.2: Measurement of infinitesimal strains.

1.1.2 Equilibrium Equations

t + b = 0. (1.3)

where t is the Cauchy stress tensor, b the body force intensity (usually g), and the density of

the solid body under study.

In terms of index notation, the above equation can be rewritten as

tij,i + bj = 0. (1.4)

1.1.3 Compatibility Conditions

The compatibility equations are restrictions placed on the strain field, (x), to ensure the exis-

tence of continuous and single valued displacements. Using index notation they take the from

2 ij 2 kl 2 ik 2 jl


+ = 0. (1.5)
xk xl xi xj xj xl xi xk

4
Remark: The above expression represents 81 equations although only six of them are unique.

1.1.4 Constitutive Laws

The constitutive law is unique from previous three sets of equations in that it is material depen-

dent. For example, for a linear isotropic material we can write

t = tr()1 + 2, (1.6)

or

tij = kk ij + 2ij , (1.7)

where and are Lame constants.

The inverted form of the above is

1+
 = t tr(t)1. (1.8)
E E

where E is Youngs modulus and is Poissons ratio.

More generally a constitutive law can be written symbolically as

(t) C-law t(t). (1.9)

Mathematically, we can write

Z
t(t) = [(t s)ds] . (1.10)
0

Examples of (classes of) history dependent material include


5
1) Viscoelasticity

2) Viscoplasticity

3) Plasticity (subject of most of this part)

1.2 Uniaxial Inelastic Properties

Consider the axial loading of a bar made from an isotropic material. The Cauchy stress tensor

is given by

t11 P

= A
0 0



t =
0 0 0
(1.11)



0 0 0

where A is the initial cross sectional area. For small strains 11 = u1 /x1 . When we first

load the bar we have linear elastic behavior so that t11 = E11 . Here we assume E is the same

in tension and compression. This is true for most non-granular materials.


The material remains elastic between two points t+ +
11 and t11 , where |t11 | may or may not be

equal to |t
11 |.

In elastic region we have

1) rate independence

2) time independence

3) path independence
6
Collectively, these imply that there is a unique relationship between stress and strain.

Again consider uniaxial loading of a bar

Figure 1.3: Uniaxial loading of a prismatic bar.


Definition: t+
11 and t11 are referred to as the yield stress. Now consider increasing t11 beyond

the yield stress t+


11 , we observe the stress-strain curve becomes nonlinear. Beyond the

yield stress it is possible to have

1) rate dependence

2) time dependence

3) path dependence

Both rate and time dependence are assumed to be negligible in classical plasticity theory.

This excludes

1) Typical polymers
7
Figure 1.4: Rate dependent material.

Figure 1.5: Time dependent material.

2) Metals at high temperature

However, we do have path dependence! For instance, lets continue to load to t+2
11 and then

unload.

2
We observe a new region of elastic behavior between t+2
11 and t11 . In this region, the constitutive

law is governed by t11 = E11 where t11 and 11 represent stress and strain increments.
8
Figure 1.6: The Baushinger effect.

There are three major observations to be made from the nonlinear loading

1) The lower yield stress t2 1 2 1


11 is typically smaller in magnitude than t11 , i.e. |t11 | < t11

(Baushinger Effect).

2) The material remains linear elastic up to its most recent yield stress.

3) Regardless of whether the material has yielded or not, the behavior in the elastic range is

governed by E.

A typical metal will behave as follows

There are several idealizations to this behavior


9
Figure 1.7: Strain hardening.

Figure 1.8: A typical stress-strain curve for metals.

1) Rigid-perfectly plastic.

2) Elastic-perfectly plastic.

3) Elastic-linear hardening.

10
Figure 1.9: Rigid-perfectly plastic model.

Figure 1.10: Elastic-perfectly plastic model.

1.3 Yield Function and Surfaces

Here we consider yielding of a material subject to multiaxial stress states. We have in general

tij 6= 0 and we will indicate such stress states with 2-D sketches.
11
Figure 1.11: Elastic-linear hardening model.

Figure 1.12: Multiple stress strain states.

For multiaxial stress states we still assume:

1) Rate independence

2) Time independence
12
3) Path dependence for inelastic strains

Definition: The yield surface is a scaled valued function (t) of the 6-D stress space which

separates the elastic region from the inelastic region.

Figure 1.13: Yield function.

Typically (t) < 0 = elastic.

For an isotropic material in the elastic region, the constitutive law is governed by

t = tr()1 + 2. (1.12)

Definition: Unloading is the process of inducing stress increments which cause only elastic

strains when the stress state is on the yield surface.

Definition: Loading is the process of inducing stress increments which cause plastic strains

when the stress states is on the yield surface.

13
1.4 General Framework for Plastic Constitutive Relations

In what follows, we restrict the development to materials which exhibit a hardening behavior,

i.e. materials for which plastic strain increments can only occur with an increment in stress.

This excludes the following behavior.

Figure 1.14: Elastic-perfectly plastic behavior.

Consider a uniaxial tension test where we load the material beyond yield and unload it.

Figure 1.15: Strain hardening behavior under uniaxial stress state

Hence, we can write the total strain as

11 = e11 + p11 , (1.13)


14
where e11 = 11 /E.

Now we extend this idea to 3-D theory and write

 = e + p , (1.14)

Here, the elastic strains are governed by Hookes law

1  
e = t trt 1. (1.15)
2G E

Upon complete unloading, t = 0, and hence

1) e = 0.

2)  = p .

The plastic strains depend on the stress history. We need some means of accounting for

this history. To do this, we introduce an internal state vector consisting of a collection of scalar

variables, i.e. = (1 , 2 , 3 , , n ).

We assume these variables characterize the stress history in some sense.

As an alternate viewpoint the variables may be assumed to characterize the current mi-

crostructural state of the material.

Definition: The yield function is a scalar valued function of the stress, t, and the state vector,

which separate regions of elastic and plastic behavior.


,

< 0: elastic.
(t, )

= 0: right on the yield surface.


(t, )
15
Figure 1.16: Yield function in terms of the stress tensor and internal variable

> 0: inaccessible states.


(t, )

We do not allow the yield function to be greater than zero. Rather, we require the yield func-

tion to move to acknowledge loading. This requirement is motivated by physical observation,

i.e., unloading induces only elastic strains!

We can summarize the behavior of the yield function as follows

< 0 and ends at (t + dt, = const) < 0.


Elastic range: starts from (t, )

= 0 and ends at (t + dt, = const) < 0. Alternatively,


Unloading: starts from (t, )

unloading on a yield surface requires d |=const = t
: dt < 0.

= 0 and ends at (t + dt, )


Neutral loading: starts from (t, ) = 0. Alternatively, neutral


loading along a yield surface requires d |=const = t
: dt = 0.

= 0 and requires
Loading: starts from (t, )
: dt > 0. This condition implies that
t

d =
t
: dt +

d = 0 or

d =
t
: dt. Hence, the internal state vector must

change to accommodate loading and the yield function ends at (t + dt, + d)


= 0.

16
Figure 1.17: Unloading from a yield surface.

Figure 1.18: Neutral loading along a yield surface.

The fact that plastic strains only occur if the state vector changes allows us to write

dpij = hkij dk , (1.16)

where hkij is an array of second order tensors.


17
Furthermore, for a work hardening material, the state variables are allowed to change only

if the stress changes. Consequently, the plastic strain may be expressed functionally as

dt).
dp = dp (t, , (1.17)

To proceed further we need a math definition.

Definition: A function F (x, y, z, u, v, w) is said to be homogeneous of degree n in variables

u, v and w if for arbitrary n

F (x, y, z, hu, hv, hw) = hn F (x, y, z, u, v, w).

Now make the following observations:

1) Plastic strains cannot occur if t remains constant. Hence, the expression on the right hand

side of equation (1.17) cannot contain terms that are independent of dtij .

2) The constitutive law should be rate independent. Hence, we can write equation (1.17) in

t).
a rate form as p = p (t, ,

In order to satisfy rate independence we require p to be homogeneous of degree 1 in the

stress rate

p = p (t, , = ap (t, ,


at)
t). (1.18)

These two physical observations collectively imply that equation (1.17) must be homoge-

neous of degree 1 in dt.

adt) = adp (t, ,


dp = dp (t, , dt). (1.19)

To proceed further we shall assume that for changes in the state vector
18
1) The magnitude of d is a function of ,
t and dt;

2) The direction of d is determined by and t.

These assumptions are based on experimental observation where the internal state vector

was taken to be the plastic strains themselves, i.e. = {P11 , P22 , P33 , P12 , P23 , P13 }. Hence, dPij is

always normal to the yield function.

Figure 1.19: Normality of plastic strains to the yield surface.

They further allow us to write

dk = bTk , (1.20)

dt) is a scalar valued function and Tk = Tk (t, ).


where b = b(t, ,

Now let us solve for b. Earlier we had for the case of loading


d = : dt + d = 0. (1.21)
t
19
Substituting for d produces


: dt = bT. (1.22)
t

Hence,


: dt
b =
t
, (1.23)

T

: dt
dk = t
Tk . (1.24)

T

Moreover, the plastic strains are given by


: dt
dPij = hkij dk = t
hkij Tk . (1.25)

T

Now let us simplify this expression by writing

: dt,
dPij = fij (t, ) (1.26)
t

where

hkij Tk
fij (t = . (1.27)
T

We further assume that fij maybe represented as


fij (t, ) g(t, ) ,
= G(t, ) (1.28)
tij

is called the scalar hardening function and g(t, )


where G(t, ) the plastic potential.

Remark: We will rigorously justify this later in the text.


20
Therefore, the constitutive law for plastic strain increments assumes the form


!
dPij g(t, )
= G(t, )

: dt . (1.29)
tij t

To complete the formulation we must find 1) the yield function ; 2) the scalar hardening

function G and 3) the plastic potential g.

A brief summary to Section 1.4:

1) The total strain at a given stress state may be decomposed as  = p + e ;

1
2) For the elastic component, the constitutive law may be written as e = 2
t E tr(t)1 or

1
in an incremental form as de = 2
dt E tr(dt)1;

0 represents all physically accessible states


3) The yield function (t, )

< 0: elastic behavior


a. (t, )


b. t
: dt < 0:unloading


c. t
: dt = 0: neutral loading
 
d. g(t,)
: dt > 0: loading with plastic strain increments dPij = G(t, )
: dt .
t tij t

1.5 Uniqueness and Stability Postulates

Before proceed further, we would like to restrict the materials in a specific manner. As shown

in Figure 1.20, the type of material we will study is typically given by curve A. By not allowing

type B, we will exclude material types 1) ice and 2) concrete to some extent.
21
Figure 1.20: Two representative material behaviors.

Curve A has the representative characteristic dtij dij 0, which curve B does not. This can

be recast in an integrated form. Consider the following loading path

tij = tA B A
ij + (tij tij )(t). (1.30)

We restrict this to be a monotonically increasing stress by delimiting the increasing function of

time (t)

0 (t) 1. (1.31)

It now becomes obvious that, equation (1.30) represents a monotonic loading for tB A
ij tij

and can also be referred to as a proportional loading. Its differential form is given by

dtij = (tB A
ij tij )d(t). (1.32)
22
Figure 1.21: Monotonic loading.

Hence, we can write for curve A

dtij dij = (tB A


ij tij )d(t)dij 0. (1.33)

On the other hand, it is obvious that

(tB A A
ij tij )dij (tij tij )dij = dtij dij 0. (1.34)

Integrating equation (1.34) with respect to ij gives

Z B 
ij

(tB
ij tA B
ij )(ij A
ij ) tij tA
ij dij 0. (1.35)
A
ij

R B  
Definition: The term A
ij
tij tA
ij dij is called Net Work, and it represents the work done
ij

by stresses in excess of the initial stress tA


ij .

Equation (1.35) shows that the net work is positive semi-definite.

Remark: Using integration by parts, the following identity can be shown.


23
Figure 1.22: Net work and net complementary work.

Z B  Z tB 
ij ij
    
tij tA
ij dij + ij A A
ij dtij = tij tij ij A
ij . (1.36)
A
ij tA
ij

In view of the first inequality of equation (1.34) we can show


Z tB 
ij

ij A
ij dtij 0. (1.37)
tA
ij

Definition: The integral (1.37) is referred to as net complementary work.

Therefore we have arrived the following postulate.

Postulate 1: For an inelastic material subjected to a monotonically proportional load, tij =

tA B A
ij + (tij tij )(t), the net stress work and the net complimentary stress work are both

positive semi-definite for stress bounds tB A


ij tij and the increasing function of time

0 (t) 1.

Now let us consider the following two representative material behaviors. As shown in Figure

1.23, both curves A and C satisfy Postulate 1. Although curve C represents a real material and
24
Figure 1.23: Two representative material behaviors both satisfying Postulate 1.

is typical of porous material such as snow and porous aluminum, we wish to eliminate it. To

do this consider the complimentary work for a complete stress cycle, loading from tA B
ij to tij and

unloading back to tA
ij at a stress level within the strain hardening scope. The complimentary

work is given by

I Z tB Z tA
ij ij
ij dtij = ij dtij + ij dtij
tA
ij tB
ij
Z tB Z tB
ij ij
= ij dtij ij dtij
tA
ij tA
ij
I  
(C1) (C2)
= ij ij dtij , (1.38)

(C1) (C2)
where ij and ij are strains on the loading curve C1 and the unloading curve C2, respec-
(C1) (C2)
tively. For such a loading-unloading stress cycle, ij ij and hence

I
ij dtij 0. (1.39)
25
If we perform the same analysis for curve C we would get the opposite sign.

Postulate 2: For an inelastic material subjected to a stress cycle, the complimentary stress work

is negative semi-definite.

A mathematic statement of these two postulates are

Postulate 1:
Z B  Z tB 
ij ij
 
tij tA
ij dij 0, ij A
ij dtij 0, (1.40)
A
ij tA
ij

for a monotonically proportional load.

Postulate 2:
I
ij dtij 0, (1.41)

for a complete stress cycle.

Let us consider the second postulate in more detail. For a stress path tA to tB , we can

integrate the complementary work as


Z tB Z B
ij ij
ij dtij = tij dij + tB B A A
ij ij tij ij . (1.42)
tA
ij A
ij

We now apply the above equation to a full stress cycle

I Z A0
ij 0
 
ij dtij = tij dij + tA A A
ij ij ij
A
ij
Z A0 
ij

= tij tA
ij dij
A
ij

0. (1.43)
26
This requires

Z A0 
ij

tij tA
ij dij 0, (1.44)
A
ij

which is true for any loading-unloading cycle at stress levels not exceeding the ultimate strength.

As a result, the net stress work is positive semi-definite for any load cycle.

1.6 Properties of Yield Surfaces

In what follows we use the stability postulates proposed in Section 1.5 to develop restrictions

on

1) the shape of the yield surface;

2) the direction of the plastic strain.

Consider a stress cycle where we

1) start inside the yield surface at tA ;

2) follow a load path to the yield surface at t;

3) deform the material plastically to t + dt;

4) return elastically to tA ;

Let 1) pA and dp denote the plastic strains prior to the cycle and any additional plastic

strain increments when we have arrived at t + dt, repectively;


27
Figure 1.24: Moving of yield surfaces accounting for plastic strain increments.

1) From tA to t, we have only elastic loading

: t + pA ;
 = e + pA = C (1.45)

2) from t to t + dt, we have

: t + pA + dp ;
 = e + pA + dp = C (1.46)

3) from t + dt to tA , we have

: t + pA + dp .
 = C (1.47)

Now consider the second postulate

I
0 ij dtij
28
Z tij  
= eij + pA
ij dtij
tA
ij
Z tij +dtij  
+ eij + pA p
ij + dij dtij
tij
Z tA
ij
 
+ eij + pA p
ij + dij dtij
tij +dtij
I I Z tij +dtij Z tA
ij
= eij dtij + pA
ij dtij + dpij dtij + dpij dtij . (1.48)
tij tij +dtij

In the above equation

H e
1)  ij dtij = 0 since elastic complementary strain energy is recoverable;

H pA
2) ij dtij = 0 because pA
ij is a constant;

3) By the use of mean value theorem one can reason that

Z tij +dtij  
dpij dtij O dpij dtij , (1.49)
tij

which is a second order differential quantity.

4) Finally, since dpij is independent of the unloading path from tij + dtij to tA
ij

Z tA
ij
 
dpij dtij = tA p
ij tij dtij dij
tij +dtij
 
p p
= tA
ij tij dij dtij dij . (1.50)

Noting the above facts and neglecting the second order differential quantities in (1.49) and

(1.50), equation (1.48) can be rewritten as

 
p
tA
ij tij dij 0, (1.51)
29
or

 
p
tij tA
ij dij 0. (1.52)

Equation (1.51) or (1.52) has the following implications

1) The yield surface is convex. By convex, we mean that any straight line from tA contained

inside the yield surface to another point t inside or on the yield surface must remain inside

or on the yield surface.

2) The plastic strain increments must be perpendicular to the yield surface.

The second implication allows us to write


dpij = , (1.53)
tij

which implies that the gradient of a level function is a vector perpendicular to the level surface

defined by = 0. It can also be shown that 0.

Summary: From the second postulate we have


dpij = . (1.54)
tij

Furthermore, we have previously developed the equation (1.29)


!
dPij g(t, )
= G(t, )

: dt . (1.55)
tij t

30
1.7 Initial Yield Surfaces

Up until now we have been working in a 6-D stress space. Now let us change to the 3-D

principal space. Recall, t = tij (ei ej ). From eigenvalue theory

tn = n, (1.56)
 
t 1 n = 0. (1.57)

(1.58)

It can be shown that the necessary and sufficient condition for nonzero eigenvectors of

equation (1.56) or (1.57) is given by

 
det t 1 = 0. (1.59)

The above equation can always be solved to produce the following representation of the

stress tensor

t = tij (ei ej ), (1.60)

where

tI 0 0


tij

= 0

tII 0 ,
(1.61)



0 0 tIII

31
Figure 1.25: Representation of the stress tensor as a vector in principal directions.

and ei are the principal directions of t. This principal representation of the stress tensor allows

us to put the stress tensor in a vector form, i.e. t = tI e1 + tII e2 + tIII e3 , as shown in Figure

1.25.

To go further with plasticity theory we need to rely on physical observations. The most

classical observation for materials is that the hydrostatic stress states have nothing to do with

plastic behavior. Let v be a unit vector which makes equal angles with the three principal

directions

1
v = (e1 , e2 , e3 ) . (1.62)
3

Planes with unit normal vector defined by v are referred to as the octahedral planes. It now

becomes obvious that any stress vector with the form t = tv = t


3
(e1 , e2 , e3 ) represents a

hydrostatic state of stress since each of the three components is equal tI = tII = tIII = t .
3

Here, the scalar t denotes the magnitude of this hydrostatic stress state. Transforming the stress

represented by tv to any other coordinate systems will also produce a hydrostatic stress state.
32
Now assume we are on the yield surface defined by = 0 and induce a stress change

dt = dtv. This is a hydrostatic stress increment. From experimental observation we know the

corresponding yield function increment d = 0 and remains constant.

Hence, we can write


d = dtI + dtII + dtIII . (1.63)
tI tII tIII

We cannot have d > 0. However, d < 0 is also not possible since changing the sign of

dt would make > 0. Note that

dt
dt = dtv = (e1 + e2 + e3 ) , (1.64)
3
dt
dtI = dtII = dtIII = . (1.65)
3

This allows us to write

!
dt
d = + + = 0. (1.66)
3 tI tII tIII

This is equivalent to require


+ + = 0. (1.67)
tI tII tIII

Hence, stress increments which are hydrostatic must represent neutral loading. Our plas-

ticity theory is independent of volumetric effects. Recall that v = 1


3
(e1 + e2 + e3 ). The

initial yield surface is an open ended cylinder oriented along the vector v which is normal to

octahedral planes. Clearly, shear stress must play an important role in yielding.
33
Figure 1.26: Stress traction acting on the surface of a general elastoplastic body with outward

unit normal n.

Consider the outer boundary of an elastoplastic body with outward unit normal n and stress

traction t(n) as shown in Figure 1.26

The stress traction is related to the stress vector by Cauchys relation

t(n) = n t; (1.68)

(n)
ti = nj tji . (1.69)

Relative to a principal set of axes, we express


34
t = tij (ei ej ), (1.70)

tI 0 0


tij =

. (1.71)
0 tII 0





0 0 tIII

The components of the stress traction then become

(n)
t1 = n1 tI ,

(n)
t2 = n2 tII ,

(n)
t3 = n3 tIII . (1.72)

To proceed further, let us denote the normal and shear component of the stress traction with

N and , respectively, as illustrated in Figure 1.27. The normal stress is related to the stress

traction by

N = t(n) n = tI n21 + tII n22 + tIII n23 . (1.73)

Then, the square of the shear stress is given by

2 2
2 = |t(n) | |N|

2
= (n1 tI )2 + (n2 tII )2 + (n3 tIII )2 (tI n21 + tII n22 + tIII n23 ) . (1.74)
35
Figure 1.27: Decomposition of the stress traction vector into normal and shear components.

One can show that 2 is independent of the value of the hydrostatic stress given by tm =

1
(t
3 I
+ tII + tIII ). This can be shown by incrementing all three components of the principal

stress vector by t, i.e.

tI tI + t, tII tII + t, tIII tIII + t. (1.75)

Substituting the above into equation (1.74), no changes will be found.

Now consider the octahedral plane defined earlier by the unit normal v. This case implies

that

1
n1 = n2 = n3 = . (1.76)
3

Definition: The shear stress on the octahedral plane is referred to as the octahedral shear stress

0 .
36
Substituting the surface normal of octahedral planes (1.76) into the shear stress formulae

(1.74), we obtain

1 2  1
02 = tI + t2II + t2III (tI + tII + tIII )2 (1.77)
3 9
1  
= (tI tII )2 + (tII tIII )2 + (tIII tI )2 . (1.78)
9

Now let us explore some alternate forms for octahedral shear stress 0 . Consider the devia-

toric stress tensor defined by

1
 
s = t tm 1, tm = tr(t) . (1.79)
3

In component form we have



t11 tm t12 t13



sij =
t12 t22 tm t23 .
(1.80)



t13 t23 t33 tm

A few facts about the deviatoric stress tensor should be noted.

1) The trace of the deviatoric stress tensor is zero, i.e.

 
trs = tr t tm 1 = trt tm tr1 = 0. (1.81)

2) The deviatoric stress tensor has the same principal directions as t. Proof: let

tn = tn, (1.82)

sn0 = sn0 . (1.83)


37
Now substituting the definition of deviatoric stress tensor (1.79) into its characteristic

equation (1.83)

(t tm 1)n0 = sn0 ,

tn0 = (s + tm )n0 . (1.84)

Compare equation (1.82) and (1.84), we see that

n0 = n,

t = s + tm . (1.85)

Hence, we can write

tI = sI + tm ,

tII = sII + tm ,

tIII = sIII + tm . (1.86)

Noting the above relations, the octahedral shear stress (1.77) can be rewritten in terms of

the principal deviatoric stress as

1 2 
02 = sI + s2II + s2III . (1.87)
3
38
Definition: Three invariants of a given stress tensor t are

It = tr(t),
1h i
II t = (tr(t))2 tr(t2 ) ,
2
III t = det(t). (1.88)

For the deviatoric stress tensor, we can write

Is = 0,
1
II s = tr(t2 ),
2
III s = det(s). (1.89)

Notation: We define J2 = II s = 21 tr(s2 ) = 12 sij sij .

With respect to the principal directions

s = sij (ei ej );



sI 0 0


sij

=
0 sII 0 .
(1.90)



0 0 sIII

J2 can be simplified as

1 2 
J2 = sI + s2II + s2III . (1.91)
2

This leads to

2 2
02 = J2 = II s . (1.92)
3 3
39
1.7.1 von Mises Yield Criterion

Figure 1.28: A unit vector defining octahedral planes in the principal space.

The von Mises yield criterion states that yielding occurs when the octahedral shear stress

reaches a critical value, denoted by k0 for example. The yield surface in this case can be

expressed as

= 02 k02 = 0, (1.93)

where 0 denotes the octahedral sher stress and k0 a constant to be determined experimentally.

Noting previous relations for the octahedral shear stress we can rewrite the von Mises yield

criterion as

1 2  1
(tI , tII , tIII , k0 ) = tI + t2II + t2III (tI + tII + tIII )2 k02 = 0; (1.94)
3 9
40
Figure 1.29: von Mises yield surface represented in principal directions of stress tensor.

1 2 
(sI , sII , sIII , k0 ) = sI + s2II + s2III k02 = 0; (1.95)
3
2
(J2 , k0 ) = J2 k02 = 0; (1.96)
3
2 2
(It , II t , k0 ) = It2 II t k02 = 0. (1.97)
9 3

At this point we introduce a special plane for graphics purpose.

Definition: The -plane is defined by the unit normal vector v = 1 (e + e2 + e3 ) and the
3 1

mean stress tm = 31 trt = 0. Thus, it is the octahedral plane located at the origin.

From the above definition, we see that any stress vector t = tI e1 + tII e2 + tIII e3 can be

decomposed into a mean stress vector and a stress vector in the -plane.

As demonstrated in Figure 1.29

1) Clearly the vector component on the -plane is independent of the mean stress;
41
2) The -plane vector is precisely the octahedral shear stress.

Looking down v on the -plane we see Figure 1.30.

Figure 1.30: Projection of von Mises yield surface on the -plane.

Recall that the von Mises yield criterion is given by equation (1.93) or 02 = k02 . Conse-

quently, the von Mises yield criterion appears as a circle on the -plane as shown. It can be

further shown that, the radius of this circle is R = 3k0 . The proof is left as an exercise.

(Hint: consider any stress vectors belonging to the -plane, such as t = { 12 , 12 , 0}R or

t = { 16 , 16 , 26 }R and substitute back to the yield criterion (1.94).)

1.7.2 Tresca Yield Criterion

The Tresca yield criterion states that plastic yielding initiates when the maximum shear stress

(max ) reaches a critical value, denoted by k1 for example. The yield surface in this case can be
42
expressed as

2
= max k12 = 0. (1.98)

In analogy to k0 for the case of von Mises, k1 is a constant to be determined experimentally. We

can further generalize the above relation in terms of principal stress states as

= (12 k12 )(22 k12 )(32 k12 ) = 0, (1.99)

where

1
1 = (tI tII ),
2
1
2 = (tII tIII ),
2
1
3 = (tIII tI ). (1.100)
2

We can show that of 1 , 2 and 3 , one must be max by examining a 3-D Mohrs circle.

The Tresca yield criterion plots as a hexagon on the -plane.

1.7.3 Properties of Initial Yield Surfaces

Now let us examine some properties of the yield surface based on isotropic arguments.

1) The yield surface implies yielding is the same in tension and compression;

2) From isotropy the yield surface must be the same in each 600 sector;

3) Our stability postulates tell us the yield surface must be convex, this implies
43
a. The Tresca yield surface represents the inner bound of convexity.

b. The outer bound on convexity is represented by the dashed line in Figure 1.31.

c. The maximum difference between the inner bound and the outer bound is

2 R 3 R
3 2
= 25%. (1.101)
2 R
3

d. The maximum difference between the Tresca and von Mises occurs in pure shear

and is

R 23 R
13.4%. (1.102)
R

4) The yield surface takes on different appearances depending on what stress space coordi-

nates are used. For example, let us consider biaxial testing where tI , tII 6= 0 and tIII = 0.

Under such a simplified condition, the von Mises yield criterion (1.94) becomes

9
t2I + t2II tI tII k02 = 0, (1.103)
2

while three candidates of the maximum shear stress in Tresca criterion (1.99) now take

the form

1
1 = (tI tII ),
2
1
2 = tII ,
2
1
3 = tI . (1.104)
2

Figure 1.32 plots the yield surfaces of both von Mises and Tresca criterion on the same

biaxial stress space.


44
5) To determine the values for k0 and k1 , let us further suppose that the Tresca and von Mises

yield criterion are chosen to agree in a uniaxial tension test, i.e. tI 6= 0 and tII , tIII = 0.

This is conventional but absolutely not mandatory. For convenience, we will let 0 denote

the initial yield stress. In this case, the von Mises yield criterion (1.103) can be further

simplified to

9 2
= 02 k02 = 0 = k0 = 0 . (1.105)
2 3

For uniaxial testing, the maximum shear stress max = 0 /2. Hence, the yield criterion of

Tresca (1.99) is

2
0 1

= k12 = 0 = k1 = 0 . (1.106)
2 2

These results are of importance in the comparison of von Mises and Tresca yield criterion

for especially biaxial stress states. As shown in Figure 1.32, the von Mises criterion is

more economic while the Tresca criterion is more conservative based on a simple com-

parison on the area inside their yield contours.

1.8 Subsequent Yield Surfaces

After initial yielding, the subsequent loading produces changes in the the yield surface. To

further study this, let us consider a plane stress formulation consisting of a normal stress and a

shear stress. A stress state of this nature is readily achieved using combined tension and torsion

on a thin-walled tube. For this, let us use polar coordinates (r, , z). For this case, we have
45
Figure 1.31: Simultaneous projection of von Mises and Tresca yield surface on the -plane.

Figure 1.32: Yield surfaces of both von Mises and Tresca criterion plotted on a single biaxial

stress coordinate.

P P
tzz = = , (1.107)
A 2R
TR T
tz = = , (1.108)
J 2R2
46
where P , T , R, A, J, denote the axial load, twisting moment, median radius, cross-sectional

area, polar moment of inertia, and tube wall thickness, respectively. The complete stress state

can be written



0 0 0



tij =
0 0 .
tz (1.109)



0 tz tzz

Clearly, we can use the following equivalent stress state for such a case

t11 t12 0



tij = t12

0 .
0 (1.110)



0 0 0

Let us find the corresponding principal stress components.


s
2
t11 t11

tI,II = tavg R = + t212 , tIII = 0. (1.111)
2 2

Note the order of the three principal stresses tI > tIII > tII . Now consider the two yield

criteria.

1) von Mises:

= 02 k02
1 2  1
= tI + t2II + t2III (tI + tII + tIII )2 k02
3 9
2 2
!
1 t11 t11 1 2
= + + t212 2 (t11 )2 02
3 4 4 9 9
2 2 2 2
= t11 + t212 02
9 3 9
= 0. (1.112)
47
We normalize as

= t211 + 3t212 02 = 0, (1.113)

or

t211 t212
+ 2 = 1, (1.114)
02

0

3

as graphically represented by an ellipse in Figure 1.33.

Figure 1.33: The yield surface of a thin-walled tube under combined axial and torsional loading

predicted by von Mises yield criterion.

2) Tresca:

2
= max k12
1 1
= (tI tII )2 02 for tI tIII tII
4 4
1 2
  1
= t11 + 4t212 02 (1.115)
4 4
= 0 (1.116)
48
Normalization gives

= t211 + 4t212 02 = 0, (1.117)

or

t211 t212
= + = 1. (1.118)
02 (0 /2)2

A comparison on the graphical representations of von Mises and Tresca yield criteria are

shown in Figure 1.34.

Figure 1.34: The yield surface of a thin-walled tube under combined axial and torsional loading

predicted by von Mises and Tresca yield criteria.

In what follows we will write the initial yield surface as

= t211 + a t212 02 = 0, (1.119)

where a = 3 for von Mises and a = 4 for Tresca criterion, respectively. For subsequent yield

surface, two popular models have evolved


49
1) isotropic hardening

2) kinematic hardening

1.8.1 Isotropic Hardening

This model, while simple, appears to work well for most metals subject to monotonic loading

and generally small plastic strains. This model assumes that

1) the shape of the yield surface does not change,

2) the yield surface remains centered at the origin, i.e. it does not translate,

3) the yield surface merely expands, i.e. from

= t211 + a t212 02 = 0, (1.120)

to

= t211 + a t212 2 = 0, (1.121)

where is the largest value reached by (t211 + at212 ).

1.8.2 Kinematic Hardening

To accommodate a yield surface which translates we can write

= (t11 )2 + a (t12 )2 02 = 0, (1.122)


50
where and are state variables reflecting the history of the inelastic loading, i.e. = {, }.

From a geometric point of view, and represent the coordinates of the center of the ellipse.

The yield function may be expressed as

= (t11 , t12 , , ), (1.123)

Recall that during loading, = 0 and d = 0, hence


d = dt11 + dt12 + d + d = 0. (1.124)
t11 t12

Substituting for from equation (1.122) results in

(t11 )dt11 + a (t12 )dt12 = (t11 )d + a (t12 )d . (1.125)

This gives us one equation to determine increments in and , i.e. (d, d ). To find another

relation, recall from Section 1.1.3 that we assumed

d = b T(t, ),
(1.126)

i.e. we assumed t and determined the direction of d.


The simplest form that has been

observed to work well is


(t, )
di = . (1.127)
i

The above implies the vector d is parallel to the gradient of yield function with respect to .

This is an experimental result. For our 2-D theory, it can be further simplified to


d = ,


d = , (1.128)

51
or, diving these two state vectors gives

d /
= . (1.129)
d /

Rearrangement of the above yields

/
d = d . (1.130)
/

With the aid of the translated yield function (1.122), the above equation reads

(t11 )
d = d . (1.131)
a (t12 )

The above equation represents another relation necessary for solving d and d . Joining of

equations (1.131) and (1.125) produces

" #
(t11 )dt11 + a (t12 )dt12
d = (t11 )
(t11 )2 + a2 (t12 )2
" #
(t11 )dt11 + a (t12 )dt12
d = a (t12 ) . (1.132)
(t11 )2 + a2 (t12 )2

Note that the above is only valid under the condition of loading , i.e.

= 0,

dt11 + dt12 > 0. (1.133)
t11 t12

For both unloading and neutral loading d = d = 0.


52
1.8.3 Isotropic Hardening vs. Kinematic Hardening

Let us now explore the difference between isotropic and kinematic hardening under the general

framework of uniaxial testing for stress component t11 , while maintaining the shear component

as a constant in our example of combined tension and torsion on thin-walled tubes.

1) Isotropic Hardening:

As shown in Figure 1.35, the yield surface is a monotonically increasing function for

isotropic hardening model. To account for this mathematically, we can relate hardening

to the increments in plastic work, i.e.

dW p = tij dpij . (1.134)

Hence, we can write = (t, W p ). Since W p is also a monotonically increasing function

we often refer to isotropic hardening as work hardening.

Figure 1.35: Initial and subsequent yield surfaces of isotropic modeling from a uniaxial test.

53
2) Kinematic hardening:

As shown in Figure 1.36, after one complete cycle p11 = 0 and moreover the yield surface

is the same as the original yield surface. This suggests that the yield surface for kinematic

hardening can be expressed as a function of the plastic strains = (t, ) = 0. Later

we will prove there is an one-to-one relationship between (, ) and (p11 , p12 ) in our 2-D

theory. For this reason, we refer to kinematic hardening as strain hardening.

Figure 1.36: Initial and subsequent yield surfaces of kinematic modeling from a uniaxial test.

1.9 Incremental Formulation for Isotropic Hardening

Engineering shear strain:

Recall the constitutive law for plastic deformation

!

dpij =G : dt (1.135)
t tij
54
Implicit in the above equation is symmetry with respect to shear components of stress and strain.

For instance, for our 2-D theory weve written:

= t211 + at212 2 = 0. (1.136)

In order to use equation (1.135) we must write the yield function as

a
= t211 + (t212 + t221 ) 2 = 0. (1.137)
2

For instance:


= t211 + at12 t21 2 = 0








 

dp11 = G(t, )

dt11 + dt12 + dt21


t11 t12 t21 t11

 
(1.138)



dp12
= G(t, )
dt11 +
dt12 +
dt21


t11 t12 t21 t12


 

dp21 = G(t, )


dt11 + dt12 + dt21



t11 t12 t21 t21

However, we can use equation (1.136) by recognizing the shear strain as the engineering shear

strain, 12 . Hence, we can write:




i) = t211 + at212 2 = 0







 
ii) dp11 = G
t11
dt11 +
t12
dt12
t11
(1.139)



 
p


d12 =G dt11 + dt12


iii) t11 t12 t12

Exercise: Expand the derivatives in equations (1.138) and (1.139) and verify

p
d12 = 2dp12 (1.140)

Lets now develop a constitutive law for our 2-D plasticity theory, where t11 , t12 6= 0 and all

other tij = 0.
55
Elastic strains:

1+
deij = dtij dtkk ij (1.141)
E E
dtkk = dt11 + dt22 + dt33 = dt11 (1.142)
1+
de = dt (tr dt)1 (1.143)
E E
dt11 1+
de11 = , de12 = dt12 (1.144)
E E

e 2(1+) dt12
or d12 = E
dt12 =
. (1.145)

Plastic strains:


dp11 = C G( dt11 + dt12 ) (1.146)
t11 t12 t11
p
d12 = C G( dt11 + dt12 ) (1.147)
t11 t12 t12


1, loading



where C=



0, unloading

Remarks:

1) Remember, during loading we have both elastic and plastic strain increments.

2) Notice the coupling between normal and shear in stress and strain for the plastic constitutive

law.

For our 2D theory the yield function is:

= t211 + at212 2 (1.148)


56


3 for Von Mise



a=



4 for Treaca
To completely specify the constitutive law, a functional for the scalar hardening coefficient must

be determined. We have from earlier work:

G = G(t11 , t12 , 2 )

From isotropy arguments one can prove:

G = G( 2 ) only! (1.149)

We can determine a form for G using experimental data. Lets consider a uniaxial tension

test, dt11 6= 0, dt12 = 0. Then

= t211 2

The normal plastic strain increment is:

 
dp11 = G
t11
dt11
t11


We have t11
= 2t11

Therefore: dp11 = 4Gt211 dt11 .

Further during loading t211 = 2 .

Hence, d11 p = 4G 2 dt11 .


57
The total strain increment is given by:

d11 = de11 + dp11 (1.150)

1
d11 = dt11 + 4G 2 dt11 (1.151)
E

d11 1
= + 4G 2 (1.152)
dt11 E

Definition: The tangent modulus, ET , is the slope of the stress strain curve in the nonlinear

region.

dt11
ET = = ET () (1.153)
d11

We have
" #
1 1 1 1 1
= + 4G 2 G= 2 0 (1.154)
ET () E 4 ET () E

Note: during loading 2 = t211 + at212

Remarks:

1) We assume the above form of G is valid for all 2-D stress states. This assumption is

essential to the success (or failure) of plasticity theory.

2) We could just as well have used a torsion test to define G().


58
Figure 1.37: The schematic of a thin-walled tube simultaneously subjected to axial and torsional

loads.

1
G= [1
4a 2 T
1 ]

Example: Path dependence of plastic strains.

Consider a tension-torsion test of a thin-walled tube as shown in Figure 1.37. We will use

the Von Mises yield criterion (a = 3). From equilibrium

P T
t11 = , t12 = (1.155)
2R 2R2

We will consider two different paths in stress space:

1) The first path is


59
t11 = t12 , 0 20

P T
2R
= 2R2
T = PR

Figure 1.38: Two possible loading paths of the tension-torsion test of a thin-walled tube.

2) We have two parts to the second path as shown in Figure 1.38.

a. t11 = 0 20 ; t12 = 0

b. t11 = 20 ; t12 = 0 20

1
For simplicity, we will assume G() = =constant, e.g. ( 1
4 2 ET
E1 ) =

E
ET = 4 2 E+1

This type of behavior is characterized by Figure 1.39.


60
Figure 1.39: Characterization of loading path 1 of the tension-torsion test of a thin-walled tube.

Path (a): = t211 + 3t212 2 . Since t11 = t12 we can write:

1) dt11 = dt12

2) The plastic strains are:


dp11 = G( dt11 + dt12 ) = 16Gt211 dt11 (1.156)
t11 t12 t11

p
d12 = G( dt11 + dt12 ) = 48Gt211 dt12 (1.157)
t11 t12 t12

We require the initial yield point be the lower limit of integration.

= t211 + 3t212 02 = 0 (1.158)

1
t11 = t12 = 0 (1.159)
2

We can now integrate equation (1.156) and (1.157)

61
p11 =
R 20
0 16Gt211 dt11 = 42G03
2

p R 20
12 = 0 48Gt212 dt12 = 126G03
2

p
Path (b): 1) t11 : 0 20 , t12 = 0 12 =0

dp11 = G( t11 dt11 ) t11 = 4Gt211 dt11

p11 =
R 20 28
0 4Gt211 dt11 = 3
G03

2) t11 = 20 , t12 : 0 20

dp11 = G( t11 dt11 +


t12
dt12 ) t11 = G(2t11 dt11 + 6t12 dt12 )2t11 = 24G0 t12 dt12
p
d12 = G(6t12 dt12 )6t12 = 36Gt212 dt12

p11 = 28 R 20
3
G03 + 0 24G0 t12 dt12 = (57 13 )G03
p R 20
12 = 0 36Gt212 dt12 = 96G03

Path dependence!

Example: Tension-torsion of a thick-walled cylinder with inner radius a and outer radius b, as

shown in Figure 1.40. An elastic-linear hardening behavior is assumed, as illustrated in

Figure 1.41.

Lets look at a differential element in cylindrical coordinate as illustrated in Figure 1.42. From

St. Venants torsion theory:

tzz 6= 0, tz 6= 0, all other tij = 0.

we can use our 2-D theory.

62
Figure 1.40: The schematic of a thick-walled tube simultaneously subjected to axial and tor-

sional loads.

Elastic strains:

1
dezz = E
dtzz
(1.160)
e 2(1+)
dz = E
dtz

63
Figure 1.41: Elastic-linear hardening behavior of a thick-walled tube simultaneously subjected

to axial and torsional loads.

Figure 1.42: A differential element of a thick-walled tube simultaneously subjected to axial and

torsional loads.

64
Plastic strains:

dpzz = C G( tzz dtzz +


tz
dtz ) tzz
(1.161)
p
dz =C G( tzz dtzz +
tz
dtz ) tz

Assuming a Von Mises yield criterion, we can write = t2zz + 3t2z 2 and = 0 for

initial yield.

The scalar hardening coefficient is

1 1 1 1 1
G= ( ) = ( 1) (1.162)
4 2 ET E 4 2 E

Carrying out the differentiation of

C
dpzz = (1
E 2
1)(t2zz dtzz + 3tzz tz dtz )
(1.163)
p C
dz = (1
E 2
1)(3tz tzz dtzz + 9t2z dtz )
The total strain increments:

dzz = dezz + dpzz


(1.164)
e p
dz = dz + dz
Hence, collecting terms:

dzz = [ E1 + C
(1
E 2
1)t2zz ]dtzz + [ EC2 ( 1 1)3tzz tz ]dtz = 1 dtzz + 2 dtz
(1.165)
dz = [ EC2 ( 1 1)3tz tzz ]dtzz + [ 2(1+)
E
+ C
(1
E 2
1)9t2z ]dtz = 1 dtzz + 2 dtz
In matrix form:


dzz 1 2 dtzz




= (1.166)






dz
1 2 dtz

65
Note: We can not solve these equations in their present form. The reason is, we do not know

the shear stress distribution. Therefore lets invert the equations.



dtzz 2 2 dzz

1



=
(1.167)
1 2 2 1


1


dtz
1
dz

Now letting and denote the axial elongation and angle of twist per unit length, respectively.

Then

zz = L , z = Lr , (Ref. St. Venant theory of torsion)

Rewriting the above in an incremental form:


zz = L
, z = Lr



tzz 2 2 zz

1



=
(1.168)
1 2 2 1


1


tz
1
z

The loads required to produce this deformation are


R R b R 2
p = tzz A and T = A tz r dA = a 0 tz r rddr

*********************************

1.10 Incremental Formulation for Kinematic Hardening

Recall the yield function for kinematic hardening is

= (t11 )2 + a(t12 )2 02 (1.169)


66
In the above equation, (, ) denotes the coordinates of the center of the yield function (an

ellipse). Assume initially that (, ) = (0, 0). The evolution equations for (, ) are

d = (t11 ) (t11(t)dt 11 +a(t12 )dt12


2 2
11 ) +a (t12 )
2
(1.170)
d = a(t12 ) (t11(t)dt 11 +a(t12 )dt12
2 2
11 ) +a (t12 )
2

The plastic strains are

dp11 = G( t11 dt11 +


t12
dt12 ) t11
(1.171)
p
d12 = G( t11 dt11 +
t12
dt12 ) t12

Figure 1.43: Movement of the yield surface due to kinematic hardening.


Differentiating the yield function gives: t11
= 2(t11 ), t12
= 2a(t12 ). The plastic

strain, hence, can be rewritten as

dp11 = 4G(t11 )[(t11 )dt11 + a(t12 )dt12 ]


(1.172)
p
d12 = 4Ga(t12 )[(t11 )dt11 + a(t12 )dt12 ]

If we compare equations (1.170) and (1.172) we find

dp11
d = (1.173)
4G[(t11 )2 + a2 (t12 )2 ]
67
p
d12
d = (1.174)
4G[(t11 )2 + a2 (t12 )2 ]

These equations can be integrated to provide one to one relation for

p
p11 , 12 . (1.175)

Hence, we can express the yield function in either of the following two forms

p
= (t11 , t12 , , ) = (t11 , t12 , p11 , 12 ). (1.176)

In the above equations the scalar hardening coefficient, G, is still arbitrary. Therefore, we can

write

G0
G = G(t11 , t12 , , ) = , (1.177)
(t11 )2 + a2 (t12 )2

where G0 is the principle

G0 = G0 (t11 , t12 , , ). (1.178)

The simplest possible form for G0 is to assume it is a constant. This implies, however, that the

stress-strain curve is bilinear as will be shown. Returning to equations (1.170) and (1.172) we

find

dp11 p
d12
d = , d = . (1.179)
4G0 4G0

Now let us determine a value for G0 using a tension test. We will approximate the curve as

being bilinear, as shown in Figure 1.44. For this case,

1
d11 = dt11 + 4G(t11 )2 dt11 (1.180)
E
68
Figure 1.44: An elastic and linear-hardening material.

Solving for G gives

1 1 1
G= 2
( ) (1.181)
4(t11 ) ET E

Furthermore, for this stress state the form assumed for G is given by

G0
G= . (1.182)
(t11 )2

This implies that

1 1 1
G0 = ( ) constant. (1.183)
4 ET E

Example: (path dependence) consider an elastic and linear-hardening material as illus-

trated in Figure 1.44. E = 30MPa, = 0.3, = 0.2, 0 = 20MPa. Well examine two load

paths

1) t11 = t12 , 0 20

2) t12 = 0, t11 = 0 20

t11 = 20 , t12 = 0 20
69
Figure 1.45: Two possible loading paths of an incremental kinematic hardening model.

Path (a): t11 = t12 , initial yield surface: = t211 + 3t212 02 = 0

(t11 )dt11 + 3(t12 )dt12


d = (t11 )
(t11 )2 + 9(t12 )2
(t11 )dt11 + 3(t12 )dt12
d = 3(t12 ) (1.184)
(t11 )2 + 9(t12 )2

The above equations must be integrated numerically.

Path (b): 1) t12 = 0, t11 = 0 20

d = 0, d = dt11
R 20
= 0 dt11 = 0

= 0 , = 0

2) t11 = 20 , t12 = 0 20
70
d =
(1.185)
d =

Numerical integration is subsequently performed...

Results:

Path (a): = 23 KPa, = 33.8 KPa; p11 = 3.03(103 ), 12


p
= 4.53(103 )

p
Path (b): = 35.2 KPa, = 29.6 KPa; p11 = 4.39(103 ), 12 = 3.93(103 )

1.11 Three Dimensional Plasticity

In 3D, the elastic strain increments are

1+
de = dt (trt)1 (1.186)
E E

For the inelastic strains we have


dp = G( : dt) (1.187)
t t

is the yield function, and G is the scalar hardening coefficient (G > 0).
where = (t, )

Recall that during loading


: dt = dtkl > 0 (1.188)
t tkl

Note: the yield function is a scalar function of a second order tensor t. This implies that

or the principal
may be expressed as a function of the invariants of t, i.e. = (It , IIt , IIIt , ),
71
Now note, we have assumed that plastic deformation is
values of t, i.e. = (tI , tII , tIII , ).

independent of hydrostatic effects. This implies

V
trp = pkk = 0 = (1.189)
V

Recall that the constitutive law is given by equation (1.187). Taking the trace of dp gives


tr(dp ) = G( : dt)tr( ) = 0 (1.190)
t t

During loading G > 0,


t
: dt > 0 tr(
t
)=0
tkk
=0


The above results may be used to show is not a function of IT , i.e. = (IIt , IIIt ).

The proof is left as an exercise. Since the volumetric effect does not play a role in , it is

more appropriate to express the yield function in terms of deviatoric stress. Recall that s =

and Is = 0 = J1 , IIs = 1 sij sij = J2 , IIIs = det(s) = J3 . Therefore =


t 13 tr(1)I) 2

(J2 , J3 , ).

1.11.1 Initial Yielding

1) Von Mises yield criterion is given by

= 02 k02 , (1.191)

where 0 is the octahedral shear stress

1 2
02 = sij sij = J2 . (1.192)
3 3

Hence, we have

2
= J2 k02 . (1.193)
3
72
2) Tresca yield criterion is given by

= 4J23 27J32 + 96k14 J2 64k16 . (1.194)

This function is generally not used due to its complexity. However, the Tresca criterion

assumes a much simpler form when expressed in terms of Principal stresses. Hence, it

may be used when the principal stress are known (e.g. Thick walled pressure vessels).

1.11.2 Isotropic Hardening

For initial yielding: = 23 J2 k02 = 0, ( 32 J2 = k02 = 92 02 J2 = 31 02 ). For subsequent

yielding: = 23 (J2 J2 ), where J2 is the highest recorded value of J2 for all time.

Note: J2 13 02 as this is the value of J2 at initial yield in a uniaxial tension test. Hence, our

internal state vector is

= J2 . (1.195)

Let us now look at the constitutive law


dpij = G( : dt) (1.196)
t tij

Definition: let e denote the deviatoric strain given by

1
e =  (tr)1. (1.197)
3

Since pkk = 0, we have

1
ep = p pkk 1 ep = p . (1.198)
3
73
= (s, ),
Also, since = (J2 , ) we can write equation (1.196) as

!

depij =G : ds . (1.199)
s sij


Its proof is left as an exercise. (Hint: t
: dt = s
: ds and t
= s
)

To complete the problem, we need a form for the scalar hardening function G


G = G(s, ). (1.200)

Recall that = J2 . The form we adopt subsequently is


From isotropy: G = G(J2 , J3 , ).

G = G(J2 ). This is consistent with our 2-D theory where G = G(). Now let us expand the

derivatives in equation (1.199)

2 2
= J2 J2 , (1.201)
3 3
1
J2 = sij sij , (1.202)
2
J2 2
= = sij , (1.203)
sij J2 sij 3
2 2
 
depij = G skl dskl sij , (1.204)
3 3
4
depij = G (skl dskl ) sij . (1.205)
9

Now consider a tension test



t11 2

0 0



t
3 11
0 0



tij = 0

0 0
sij =
0 13 t11 0
(1.206)



0 0 0 0 0 13 t11

74
4
dep11 = G(s11 ds11 + s22 ds22 + S33 ds33 )S11
9
 " 2  2  2 #
4 2 2 1 1

= G t11 t11 dt11 + t11 dt11 + t11 dt11
9 3 3 3 3
 4
2
= Gt211 dt11 (1.207)
3

The total strain increment is

d11 = de11 + dp11


 4
1 2
= dt11 + Gt211 dt11 (1.208)
E 3

Now let ET = dt11 /d11 and hence

1 1 2 4 2
 
=+ Gt11
ET E 3
1 1 1
 
G = 2 4 2
( 3 ) t11 ET E
1 1 1
 
= 2 3
( 3 ) 2J2 ET E
27 1 1
 
= = G(J2 ), (1.209)
16J2 ET E

where ET = ET (J2 ).

Note: During loading

1
J2 = J2 = sij sij
2
1 4 2 1 2 1 2
 
= t + t + t
2 9 11 9 11 9 11
1 2
= t
3 11
t211 = 3J2 . (1.210)

Recall that

4
depij = G (skl dskl ) sij , (1.211)
9
75
Hence,

3 1 1
 
depij = (skl dskl )sij . (1.212)
4J2 ET E

1.11.3 Kinematic Hardening (3D)

Assuming a Von Mises yield criterion, the initial yield surface is

1
 
= 02 k02 02 = sij sij , (1.213)
3

where s is the deviatonic stress tensor. Therefore, = (s) for initial yielding. For subsequent

yielding we assume the yield function

= (s, ep ). (1.214)

This is similar to our 2D theory, where we assumed

= (t11 , t12 , , ), (1.215)

or

= (t11 , t12 , p11 , 12


p
). (1.216)

since there is an one-to-one correspondence between (, ) and (p11 , 12


p
). Recall the plastic

strain increments are given by


depij = G( dskl ) (1.217)
skl sij

We also have a requirement during loading that


d = dskl + dk = 0 (1.218)
skl k
76
In this case:

= (1 , 2 , 3 , ..., 9 ) = (ep11 , ep12 , ..., ep33 ). (1.219)

Noting equation (1.218) we can write


dskl = p depkl (1.220)
skl ekl

Substitute equation (1.220) into (1.217) gives

!
p
depij = G de , (1.221)
epkl kl sij

or

!

depij = G depkl . (1.222)
epkl sij

The above implies

!

G = ik jl . (1.223)
epkl sij

From the above we can readily show

1
G= . (1.224)
epkl skl

Note that the scalar hardening function G is completely determined by the yield function .

Pragers Hardening rule

A commonly assumed form of the yield function is

(s, ep ) = (s hep ) : (s hep ) 3k02 , (1.225)


77
where h is a scalar. Note that when ep = 0, we can set

= sij sij 3k02


2
= sij sij 02 . (1.226)
3

The scalar hardening coefficient becomes

1
G= . (1.227)
epij sij

Noting equation (1.225),


= 2h(sij hepij ), (1.228)
epij

and


= 2(sij hepij ), (1.229)
sij

we obtain

1 1
G= = . (1.230)
4h(sij hepij )(sij hepij ) 12hk02

The scalar h may be determined from a tension test and this is left as an exercise.

1.11.4 Combined Isotropic and Kinematic Hardening

We can develop a combined isotropic/kinematic theory by assuming

= (sij , epij , ), (1.231)

where is the plastic work given by


Z
= sij depij , (1.232)
78
integrated over the entire stress history. Hence our internal state vector becomes

= (ep11 , ep12 , ..., ep33 , ). (1.233)

Our constitutive law is still given by


depij = G( dskl ) (1.234)
skl sij

Now consider the case of loading


d = dskl + p depkl + d = 0 (1.235)
skl ekl
" #
p
dskl = p dekl + d (1.236)
skl ekl

Furthermore, d = skl depkl . Noting the above one can readily show

1
G= (1.237)
epkl skl
+ skl skl

A combined isotropic/kinematic hardening model can be constructed by assuming

= (sij hepij )(sij hepij ) 3k0 (k0 g). (1.238)

where g is a constant.

79
Chapter 2

Implementation in the Finite Element

Method

Here we present an introduction to the finite element method as it applies to plasticity in struc-

tural mechanics. To do this we must first establish a principle of virtual work.

2.1 Principle of Virtual Work

This principle is the fundamental variational principle of solid mechanics. Consider a body

B in static equilibrium under the action of specified body forces and surface tractions. Let S

denote the bounding surface

S = St + Su , (2.1)
80
where St and Su denote surface portions with tractions t(n) or displacements u(x) specified,

respectively, as illustrated in Fig. 2.1.

Figure 2.1: Tractions and displacements specified boundaries of a solid body.

Recall that the equilibrium equations are

t + f = 0,
(2.2)

where f = b and b is the body force density (usually g if generated due to gravity).

A statically admissible stress field is one which satisfies the equilibrium equations over the

interior of B and all stress boundary conditions on St ;


81
A kinematically admissible displacement field is one satisfying all displacement boundary

conditions on Su and possessing continuous first-order derivative in the interior of B.

Now consider a body with statically admissible stress distributions and subjected to kine-

matically admissible virtual displacements. The virtual work due to external loads is

Z Z Z Z
WE = f udV + t(n) udS = f udV + (n t(n) ) udS, (2.3)
V S V S

or in indicia form
Z Z
(n)
WE = fi ui dV + nj tji ui dS. (2.4)
V S

Applying the divergence theorem, i.e.


R
n rdS =
R
rdV , to the above equation we have

S V

Z Z

WE = fi ui dV + (tji ui )dV
V V xj
!
Z
ui tji
= tji + + fi ui dV. (2.5)
xj xj

Noting that

tji
+ fi = 0;
xj
" #
ui 1 ui uj 1 ui uj
ui = = ( + )+ ( ) = ij + wij ;
xj xj 2 xj xi 2 xj xi
tij wij = 0 (t = tT , w = wT ), (2.6)

we can conclude
Z Z Z
(n)
fi ui dV + ti ui dS = tij ij dV. (2.7)
V S V

The above relation is an alternative statement of equilibrium. Sometimes we write this as

wE = wI . (2.8)
82
Namely, the virtual work done by statically admissible external loads against kinematically

admissible virtual displacements is equal to the virtual work done by internal stresses against

virtual strain field.

Remarks:

1) In equation (2.7) the body forces, stresses, and the surface tractions are constants;

2) The stresses are independent of the virtual deformation, i.e. there is no relationship be-

tween the statically admissible stress field and the kinematically admissible displacement

field;

3) This is not an energy principle! It is valid when energy is not conserved (plasticity);

4) This principle is independent of any constitutive law.

Example: Consider an isotropic, linear elastic cantilever beam. Deriving the governing equa-

tions for the deflection of the beam along with the boundary conditions.

Figure 2.2: A cantilever beam subjected to a few generalized loads.

83
Solution: we begin by writing the statement of the virtual work for the system

Z a !
Z
dv
tij ij dV = qvdx p0 v(b) + M0 . (2.9)
V 0 dx x=l

For small deflections we have

d2 v
 = y
dx2
Z Z lZ
d2 v
tij ij dV = y dAdx
V 0 A dx2
Z l 2 Z l 2
d v Z d v
= 2
ydAdx = M dx
0 dx A 0 dx2
Z l" 2
# !
d v dv
M 2
+ qH(a x)v + p0 (x b)v dx M0 = 0. (2.10)
0 dx dx x=l

where H(a x) is a unit step (heavy side) function and (x b) is the Dirac delta function.

Integrating the first term by parts twice yields:

Z l" 2 # !
dM dv
+ qH(a x) + p 0 (x b) vdx M0
0 dx2 dx x=l
" ! #l
dv dM
+ M v =0 (2.11)
dx dx 0

or

Z l" 2 # !
dM dv
2
+ qH(a x) + p0 (x b) vdx + (M M0 )
0 dx dx x=l
! ! !
dM dv dM
v M + v =0 (2.12)
dx x=l
dx x=0
dx x=0

Using the standard variational arguments there follows

d2 M
1) dx2
= qH(a x) p0 (x b) (The Differential Equation);
84
dv
2) at x = 0; v = 0, dx =0

at x = l; M = M0 , dM
dx
=0 (Boundary Conditions)

Now assuming linearly elastic behavior we have

d2 v
M = EI
dx2
2
d2 v
!
d
EI 2 = qH(a x) p0 (x b)
dx2 dx
dv
v(0) = 0, (0) = 0
dx
d2 v d2 v
!
d
at x = l; EI 2 = EIM0 , EI 2 = 0. (2.13)
dx dx dx

2.2 Principle of Minimum Total Potential Energy

In what follows we develop a special case of the principle of virtual work which is restricted to

elastic materials. Recall, for elastic materials there exists a strain energy density, U0 (ij ), such

that tij = U0 /ij . For instance, for a linearly elastic material

1
U0 = tij ij . (2.14)
2

Noting the above, let us invoke the principle of virtual work

Z Z
U0 Z Z
tij ij dV = ij dV = U0 dV = U0 dV = U
V V ij V V
Z Z
(n)
= fi ui dV + ti ui dS (2.15)
V S

Definition: we define the potential energy of the applied loads as

Z Z
(n)
V = fi ui dV ti ui dS. (2.16)
V S

85
(n)
For fi and ti prescribed there follows

Z Z
(n)
V = fi ui dV ti ui dS. (2.17)
V S

Hence, in view of equation (2.15) the principle of virtual work becomes

(U + V ) = = 0. (2.18)

The above relation is often referred to as the principle of minimum total potential energy.

Remarks:

1) is the total potential energy (PE) of the system;

2) The total potential energy is actually a minimum at equilibrium;

3) The principle of minimum total potential energy (2.18) is restricted to elastic materials.

2.3 Introduction to Finite Element Method

Here we present a brief review of the finite element method as it applies to linearly elastic

structural mechanics. The field equations to be solved are

t + b = 0;
Equilibrium:

1
h i
Strain-displacement relationship:  = 2
u + u ;

(1+)
Constitutive law:  = E
t E
tr(t)1 + 0 , where 0 is the non-elastic strain, i.e. thermal

strain. The inverse relation is t = [tr() tr(0 )] 1 + 2( 0 ).


86
The finite element method consists of dividing the body into a number of sub-domains

called elements, each of which contains a few grid points (nodes). Assuming each element has

n nodes and furthermore, each node is allowed to have m generalized degrees of freedom (dof),

e.g. typically displacements and/or slopes.

. . T .
Notation: column vectors: {..}; row vectors: b c = {..} ; matrices: [ . . ].

Examples of 2D elements:

Constant stress triangle: as shown in Fig. 2.3. The generalized nodal displacements for this

case can be chosen as







ui ui























vi vi










{ui }















uj uj








ue = {uj } = = . (2.19)


















vj








vj



{uk }















uk uk































vk





vk

Rectangular plate element: as shown in Fig. 2.4. The generalized nodal displacements for
87
Figure 2.3: A constant-stress triangular element.

this case can be chosen as








wi wi






























wi




wi





x




x












w wi



i




y y




















w wj



j






















wj wj

{ui }

















x




x












wj wj

{uj }









y y

ue = = = . (2.20)

{uk }

w wk



k

































wk wk
{ul }











x

x






















wk






wk



y y




















w wl



l


























wl




wl





x





x












w wl



l



y y

88
Figure 2.4: A rectangular plate element.

We can also have elements containing mid-side and/or interior nodes. Curved surfaces can

be readily handled with the aid of more complex elements, e.g. an 8-node iso-parametric 2D

element.

Consider an element with n-nodal points. We select one interpolation function for each node

point (N1 , N2 , , Nn ) which relate the field variables of an arbitrary point, i.e. displacements

u(x), to the generalized nodal displacement ue


89


u1






















v1



















w1














u2










u(x)













v2



h i


{u(x)} = = [N ]33n {ue }3n1 = N1 1, N2 1, , Nn 1


v(x)









w2








w(x)






31

..

.















un

























vn














wn






Pn
i=1 Ni ui












=
Pn
i=1 Ni vi
, (2.21)






P
n
i=1 Ni wi



where Ni represents the interpolation function for the ith node. Naturally, we require u(x =

xi ) = ui .

Once the interpolation functions are successfully established, the strains are derived using

1h i
 = u + u . (2.22)
2

Let bc = b11 , 22 , 33 , 12 , 23 , 13 c. We can write the strains in terms of the displacements in

matrix form as

{} = [B]{ue }, (2.23)


90
where [B] typically contains partial differential operators with respect to spatial coordinates

{x1 , x2 , x3 }. Next we write the constitutive law in matrix form, e.g. for a linearly elastic solid

we have

{t} = [D] ({} {0 }) , (2.24)

= [D][B]{ue } [D]{0 }, (2.25)

where {0 } may represent thermal strains and {} the total strain. Also, [D] is a matrix of elastic

constants. For simplicity we assume that the generalized displacements field, u(x), contains

displacements only (no derivatives)

{u(x)} = [N ]{ue }. (2.26)

Now let us invoke the principle of virtual work, i.e.

Z Z Z
(n)
fi ui dV + ti ui dS = tij ij dV. (2.27)
V S V

(n)
Let ti = qi denote the surface fractions and note the following

bu(x)c = bue c[N ]T , (2.28)

bc = bue c[B]T , (2.29)

(2.30)

we have

Z Z Z
fi ui dV = bu(x)c{f}dV = bue c [N ]T {f}dV ; (2.31)
V V V
| {z }
{RB }

91
Z Z Z
qi ui dS = bu(x)c{q}dS = bue c [N ]T {q}dS ; (2.32)
S S S
| {z }
{RS }
Z Z Z
tij ij dV = bc{t}dV = bue c [B]T [D][B]dV {ue }
V V V
| {z }
[Ke ]
Z
bue c [B]T [D]{0 }dV . (2.33)
|V {z }
{RI }

Plugging the above relations into the principle of virtual work (2.27) and rearranging terms give

us the element equations

[Ke ]{ue } = {Re }, (2.34)

where {Re } = {RB } + {RS } + {RI }. The remaining steps are then

1) Overlay the elements equations into a global set (assembly);

2) Impose the boundary conditions;

3) Solve the global equations [K]{u} = {R};

4) Post processing to get the displacements, strains and stresses.

2.4 An Example of Constant Stress Triangular Elements

For this example we choose the constant stress triangular element, as shown in Fig. 2.3. We

have two degrees of freedom per node and hence a total of six for the element. Assuming


u = 1 + 2 x + 3 y



, (2.35)



v = 4 + 5 x + 6 y

92
we will let the displacements to vary linearly with coordinates. Hence







1










2














u(x, y) 1 x y 0 0 0 3




= . (2.36)







v(x, y)
0 0 0 1 x y


4










5














6


The six constants are evaluated by imposing displacement conditions at the three nodal points.

For u(x, y) we have




1











u(x, y) = b1, x, yc 2 . (2.37)





3


We impose the displacement component u = ui at the three nodal points, hence




ui 1 xi yi 1















uj =
1 xj yj

2 . (2.38)









uk 1 xk y 3



k

Inverting the above equation yields




1 ai aj ak ui











1




2 = 2
yjk yki yij

uj , (2.39)









3 xkj xik xji uk



93
where xij = xi xj , ai = xj yk xk yj , aj = xk yi xi yk , ak = xi yj xj yi , and



1 xi yi



1


= yj , (2.40)
2 1 xj






1 xk yk

denotes the area of the triangular element. Substituting equation (2.39) and a similar one for

4 , 5 and 6 into (2.36) and rearranging the order of the generalized coordinates, we have





ui















vi















u(x, y) uj



h i




= Ni 1, Nj 1, Nk 1 , (2.41)






v(x, y)





vj











uk

















vk




where

1
Ni = (ai + yjk x + xkj y),
2
1
Nj = (aj + yki x + xik y),
2
1
Nk = (ak + yij x + xji y). (2.42)
2
94
This element satisfies C 0 continuity. The strain components at any point of the element is given

by x = u/x, y = v/y, and xy = v/x + u/y. Hence










ui






ui














vi vi








x yjk 0 yki 0 yij 0













uj uj

1


h i


= B = . (2.43)
y 0 xkj 0 xik 0 xji



2











vj








vj



xy xkj yjk xik yki xji yij












uk uk























vk vk



Note: Since the matrix [B] is a constant one the strains remain constant throughout the element.

The determination of stress field depends on the way how the constitutive law is defined.

For simple 2-D elements, two types of assumptions exist: plane strain elements and plane stress

elements.

2.4.1 Plane Strain Elements

The plane strain assumption is represented by zx = zy = zz = 0. Furthermore, we assume

isotropic thermal expansion, i.e. 0x = 0y = 0z = T and 0xy = 0yz = 0xz = 0. Collectively,

the constitutive law for this case may be written as

{t} = [D]{ 0 }, (2.44)


95
where the elastic matrix



1 0

E


[D] =
1 0 (2.45)
(1 + )(1 2)





0 0 (1 2)/2
and the thermal strain components


T












0 = (1 + ) T . (2.46)







0



2.4.2 Plane Stress Elements

The plane stress assumption on the other hand is represented by tzx = tzy = tzz = 0. Isotropic

thermal expansion per stated above is still assumed. The corresponding constitutive law for this

case may then be written as

{t} = [D]{ 0 }, (2.47)

where the elastic matrix





1 0

E


[D] =
1 0 (2.48)
(1 2 )





0 0 (1 )/2
and the thermal strain components


T












0 = T . (2.49)







0



96
Note: This form of elastic matrix is only valid for isotropic materials. If the material is not

isotropic, the elastic matrix [D] generally assumes less symmetry. For example, for an

orthotropic solid, [D] has the general form





D11 D12 D13



[D] =
D22 .
D23 (2.50)



Symm. D33

Having identified the fundamental matrices, i.e. [B], [D] and [N ], we can now proceed to define

the corresponding equilibrium equations.

T
R
1) Stiffness matrix: [Ke ] = V [B] [D][B]dV . Note that matrices [B] and [D] have no

spatial dependence which implies that the above integral can be simplified as [Ke ] =

t [B]T [D][B], where t denotes the nominal thickness of the triangular element;

T
2) Nodal forces from initial strains: {RI } = [D]{0 }dV . For a constant strain ele-
R
V [B]

ment, {0 } = constant. Hence {RI } = t [B]T [D]{0 };

3) Nodal forces from body forces: {RB } = T


{f}dV ;
R
V [N ]

T
4) Nodal forces from surface tractions: {RS } = {q}dS.
R
S [N ]

The total element assembly then becomes

[Ke ]{ue } = {Re }, (2.51)

where {Re } = {RB } + {RS } + {RI }.


97
2.5 Incremental Plasticity Finite Element

In what follows, we restrict the discussion to a purely mechanical development, i.e. no thermal

strains. Recall from the principle of virtual work we can write

Z Z Z
(n)
fi ui dV + ti ui dS = tij ij dV. (2.52)
V S V

This principle holds for all time instants and is valid for both linear and nonlinear material be-

haviors within the context of small stains. The fundamental difference versus an elastic problem

lies in the constitutive law. For elasticity we write

{t} = [D]{}, (2.53)

where bc = b11 , 22 , 33 , 12 , 23 , 13 c. For plasticity we write

{t} = [Dep ]{}, (2.54)

where Dep represents the instantaneous elastic-plastic material matrix.

Let us now consider the isotropic hardening formulation where the yield function is given

by

= (t, K). (2.55)

Here, K represents the plastic work, i.e.

Z
K=W P
= tij dpij . (2.56)
98
Note that, the above formulation may be shown identical to that developed earlier in notes where

we assume = (t, J2 ). For convexity of the yield surface and normality of the plastic strains

we can write


dpij = . (2.57)
tij

Also, during loading we have

= 0, (2.58)

and


0 = d = dtij + dK. (2.59)
tij K

Let

1
A= dK, (2.60)
K

then the consistency condition of equation (2.58) becomes


dtij A = 0, (2.61)
tij

or in the matrix form

$ %

{dt} A = 0, (2.62)
t

where

$ % $ %

= , , ,2 ,2 ,2 , (2.63)
t t11 t22 t33 t12 t23 t13
99
and

{dt} = {dt11 , dt22 , dt33 , dt12 , dt23 , dt13 } . (2.64)

To evaluate the increment in stress, we begin by writing

h i h i
{dt} = D {de } = D ({d} {dp }) . (2.65)

Noting the constitutive law for plastic strains (2.57) we can write
( )!
h i
{dt} = D {d} . (2.66)
t
j k

Multiplying the above equation by t
yields
$ % $ % $ % ( )
h i h i
{dt} = D {d} D . (2.67)
t t t t

Substituting equation (2.62) into the above we have


$ % $ % ( ) !
h i h i
D {d} D + A = 0. (2.68)
t t t

Solving for produces


j kh i

t
D
= j k h i n o {d} . (2.69)
t
D t + A

Combining equations (2.69) and (2.66) reads


j kh in o


h i
t
D t
{dt} = D 1 j k h i n o {d
} . (2.70)
t
D t
+ A

Comparing the above equation with (2.54) we may write the total material matrix as
j kh in o


h i
t
D t
[Dep ] = D 1 j k h i n o . (2.71)
t
D t
+ A

100
Significance of parameter A:

Case (a) - elastic perfectly plastic materials: As shown in Fig. 1.10, the yield function =

(t) only for this case. Hence, we have K
= 0 and A = 0. As a result, the elastic-plastic
h i h i
matrix is well defined for perfectly plastic materials: Dep = D .

Case (b) - work hardening materials: The uniaxial stress-strain curve for this type of materi-

als is given in Fig. 1.8. Recall that


dpij = , (2.72)
tij

and

dK = tij dpij . (2.73)

Hence,


dK = tij , (2.74)
tij

and

1
A= dK = tij . (2.75)
K K tij

Take von Mises yield function with isotropic hardening as an example. The yield function

in terms of a general 3-D stress state is given by


101
s
1 1 1
= (t11 t22 )2 + (t22 t33 )2 + (t33 t11 )2 + 3t212 + 3t223 + 3t213 e
2 2 2
s
3 q
= sij sij e = 3J2 e , (2.76)
2

where e is defined as the effective stress and is the largest recorded value of the square roots. It

can further be shown that prior to initial yielding e = 0 , where 0 represents the initial yield

stress of a material subjected to uniaxial tensile tests (left as an exercise!).

To proceed further, we assume that there exists an effective stress-strain curve such that:

dK = e dp , where dp represents the effective plastic strain.

1) The above assumption is equivalent to the assumption in our theory where the scalar

hardening coefficient is in the form G = G(J2 );

2) For isotropic materials the effective stress-strain curve is identical to the uniaxial stress-

strain curve, i.e. t11 e and p11 p .

2.6 Numerical Implementation of the Elastoplastic Constitu-

tive Relations

Typical FE formulations evaluate the stress at integration (Gauss) points.

Here we look at implementing the relation

{dt} = [Deq ] {d} (2.77)


102
With in the context of FE. A nonlinear FE problem must be solved incrementally.

i.e. Consider tensile loading of a notched bar.

For a nonlinear analysis, we divide the loading P into a number of increments (P ).

In a typical load step, say m+1, assume the stress, strain, and displacement are known at the

previous load step.

i.e. {t}m , {}m , {u}m known

Now apply a load increment.

The global equations are of the form:

[k]m {u}m = {R}m (2.78)

Hence, we can estimate the change in displacement field by solving the above.

Strain increment:

{} = [B] {u} (2.79)

where we have dropped the superscripts as they are unnecessary here. We need to determine

{t} associated with {}

The first step is to determine whether loading into the plastic strain zone has occured. 2 cases:

case A: Stress at load step m is elastic, we begin by assuming a trial stress increment based on

elastic behavior. i.e.

{te } = [D] {t}


103
Next check for yielding



<0












e , e ) = 0
(t + t (2.80)









>0



If > 0, the the point in question has undergone plastic deformation.

Assume > 0, we scale back the stress by a factor r, such that

e , e ) = 0
(t + rt

We can evaluate r by expanding the above into a Taylor series about (t, e ) and neglecting the
(t,e )
higher order terms = (t, e ) + t
: (rte ) (H.O.T)

m
(t + rte , e ) = (tm , e ) + t
{te } r + H.O.T

(tm ,e ) m
r= {te }
t

Remark: A more accurate estimate of r may be strained by retaining quadratic in Taylor series.

2.7 Iteration on the Global Equations

104
Chapter 3

Limit Analysis

3.1 Limit Functions and Surfaces

3.2 Generalized Stresses and Strains

3.3 Limit Theorems

3.4 Limit Analysis for Beams and Frames

105

Vous aimerez peut-être aussi