Você está na página 1de 29

NONLINEAR FINITE ELEMENT

ANALYSIS
SCRIPT OF LECTURES

Doc. Ing. Vladimír Ivančo, PhD.

Faculty of Mechanical Engineering,


Technical University of Košice, Slovakia

HS Wismar, June 2011


4 NONLINEAR FINITE ELEMENT ANALYSIS

1. STRUCTURAL NONLINEARITIES
1.1 Introduction
Solution of many engineering problems is based on linear approximations. In
structural analyses, these approximations are represented by consideration that
• displacements are small and can be neglected in equilibrium
equations,
• the strain is proportional to the stress (linear Hookean material
model),
• loads are conservative, independent on displacements,
• supports of the structure remain unchanged during loading.
Consequently, in the Finite Element Analysis (FEA) the set of equations,
describing the structural behavior is then linear
K d = F, (1.1)
where K is the stiffness matrix of the structure, d is the nodal displacements vector
and F is the external nodal force vector. Characteristics of solution of this linear
problem is that
• the displacements are proportional to the loads,
• the stiffness of the structure is independent on the value of the
load level.
In reality, behavior of structures is nonlinear, but divergences from linear
response are usually small and may be neglected in most practical problems.
On other hand, solution of many engineering problems needs abandonment of
linear approximations. For example, displacements of slender structures (like crane
towers, masts etc.) may be so large that changes of the structure shape (or
configuration changes) cannot be neglected. Many materials behave nonlinearly or
linear material model cannot be used if stress exceeds some value. Moreover,
loads may change their orientations according to displacements and supports may
change during loading. Consequently, structure behaves nonlinearly. If these
phenomena are included in a FEA, the set of equilibrium equations becomes
nonlinear and instead of set of linear equations (1.1) we obtain a set of nonlinear
algebraic equations
R (d ) = F . (1.2)
NONLINEAR FINITE ELEMENT ANALYSIS 5

1.2 Types of structural nonlinearities


Structural nonlinearities can be specified as
1. Geometrical nonlinearities: The effect of large displacements on
the overall geometric configuration of the structure.
2. Material nonlinearities: Material behavior is nonlinear. Possible
material models are:
a) nonlinear elastic,
b) elastoplastic,
c) viscoelastic,
d) viscoplastic.
3. Boundary nonlinearities, i.e. displacement dependent boundary
conditions. The most frequent boundary nonlinearities are
encountered in contact problems.

Consequences of nonlinear structural behavior that have to be recognized are:


a) The principle of superposition cannot be applied. Thus, for
example, the results of several load cases cannot be combined.
Results of the nonlinear analysis cannot be scaled.
b) Only one load case can be handled at a time.
c) The sequence of application of loads (loading history) may be
important. Especially, plastic deformations depend on a manner
of loading. This is a reason for dividing loads into small
increments in nonlinear FE analysis.
d) The structural behavior can be markedly non-proportional to the
applied load.
e) The initial state of stress (e.g. residual stresses from heat
treatment, welding, cold forming etc.) may be important.

1.3 Concept of time curves


For nonlinear static analysis, the loads are applied in incremental steps using time
curves. The “time” variable represents a pseudo time, which denotes the intensity
of the applied loads at certain step.
For nonlinear dynamic analysis and nonlinear static analysis with time-
dependent material properties,1 “time” represents the real time associated with the
loads’ application.

1
i.e. analysis of creep and relaxation problems by use of viscoelastic or viscoplastic material
models.
6 NONLINEAR FINITE ELEMENT ANALYSIS

As an example, time curves of forces F1 and F2 loading simple beam are


displayed in Figure 1.1. Values of forces at any time are defined as

F1 = λ1 (t ) f1 and F2 = λ2 (t ) f 2

where f1 and f2 are input values of forces and λ1 and λ1 are load parameters that are
functions of time t.

Figure 1.1: Example of time curves

The choice of time step size depends on several factors such as the level of
nonlinearities2 of the problem and the solution procedure. Generally, sufficiently
small steps are necessary to simulate nonlinear response of a structure with
satisfactory accuracy. On the other hand, large number of too small time steps
uselessly increases consumption of CPU time. Computer programs are usually
equipped with an adaptive automatic stepping algorithm to facilitate the analysis
and to reduce the solution time demands.

2
Highly nonlinear problems need smaller load increments.
NONLINEAR FINITE ELEMENT ANALYSIS 7

2. GEOMETRICALLY NONLINEAR
FINITE ELEMENT ANALYSIS
2.1 Large displacement and small strain behavior
To examine geometrically nonlinear behavior we will start with an example. We
assume large displacement, but small rotation and, what is the most important,
small strain. The structure is very simple – only one bar truss as is shown in Figure
2.1. At the beginning, when the force P is zero, the axial force N in the bar is zero
too and bar has its initial length L0.

Figure 2.1: Example of nonlinear structure – single bar truss

Using the free body diagram shown in Figure 2.1 the equilibrium equation is
N sin α − P = 0
and after substituting sin α = ( h + u ) L
h+u
N − P = 0. (2.1)
L
Assume that material is linearly elastic with Young’s modulus E. The
assumption of small strains means here that changes of the bar cross sectional area
A can be neglected. Then axial force in the bar is
8 NONLINEAR FINITE ELEMENT ANALYSIS

N = E A0 ε (2.2)

where A0 is the initial cross sectional area and ε is the engineering strain defined
as
L − L0
ε= . (2.3)
L0
As lengths are given as

L0 = a 2 + h 2 and L = a 2 + ( h + u )2 (2.4)

the expression for strain is getting rather complicated. We can overcame this
problem by introducing Green’s strain defined as

L2 − L20
εG = (2.5)
2 L20
which for our problem becomes
2
h u 1 u 
εG = +   . (2.6)
L0 L0 2  L0 
Use of this new measure of strain is possible because we can define strain
arbitrarily. The only condition is that the strain measure must be objective, which
means that is have to be independent on choice of coordinate system and
insensitive to a rigid body movement. From equations (2.3) and (2.5), it follows
that

L2 − L20 L − L0 L + L0 1 L + L0 1  L + L0 
εG = = = ε = ε + 1 − 1 =
2 L20 2 L0 L0 2 L0 2  L 
 0 
1 L−L 
= ε  0 + 2 L0 
2  L L0 
 0
or
1
εG = ε + ε 2 . (2.7)
2
Noting that the constitution equation was measured as
N
σ= =Eε (2.8)
A0
the same constitutive equation when using Green’s strain should be
NONLINEAR FINITE ELEMENT ANALYSIS 9

ε ε E
σ =E εG = E εG = εG . (2.9)
εG 1 2
ε+ ε
1
1+ ε
2 2
This means that we should use value
E
E∗ =
1+ (1 / 2)ε
instead of E in the constitution equation. Fortunately, we can ignore this
complication now because for small engineering strain is the difference between
engineering and Green’s strain negligible.
For example, consider that ε = 0,002 (e.g. mild steel yields at about this value),
then εG = 0,002 + 0,5⋅0,0022 = 0,002002. This means that difference is only 0,1%
i.e. a value that can be usually neglected. Assuming that strain is small, we can
write σ ≈ E ε G and according to equation (2.6)
E A0  1 
N= 2 
h u + u2  . (2.10)
L0 2 

Substituting (2.10) to equilibrium equation (2.1) and assuming that for small strain
is L ≈ L0 gives the equilibrium equation
E A0
2 L30
(u3 + 3 h u2 + 2 h2u) = P. 2.11)

Obviously, the equation is nonlinear with respect to displacement u. That means


that relation between load P and displacement u is represented not by a straight
line as it is when changes of configuration are neglected but by a curve. This
nonlinear characteristic for E = 2,1⋅105 MPa, A0 = 100 mm2, a = 200 mm and
h = 20 mm is shown in Figure 2.2.

2500 P [N]

2000

1500

1000

500

0
-10 -5 0 5 10

u [m m]
-500

-1000

Figure 2.2: Geometrically nonlinear behavior of a single bar truss


10 NONLINEAR FINITE ELEMENT ANALYSIS

There is another possibility to obtain equation of equilibrium (2.1) or (2.11).


From principle of virtual displacements, it follows that when the structure is in
equilibrium, virtual works of internal and external forces are equal for every
kinematic admissible set of virtual displacements. For our structure with one
degree of freedom, only one virtual displacement δ u is possible and principle of
virtual displacements has a form

∫ σ δε G dV = P δ u (2.12)
V

where δε G is virtual strain corresponding to virtual displacement δ u . The virtual


strain can be expressed from equation (2.6) as

d  h u 1 u  
2
dε G h+u
δε G = δu =  +    δ u = 2 δ u. (2.13)
du du  L0 L0 2  L0   L0
 
It is assumed in principle of virtual displacements that virtual displacement is
infinitesimal and hence the stress σ = ( N / A) remains unchanged. Noting that σ
and δε G are constant over the whole volume V in this case and assuming that
changes of the volume can be neglected due to small strain, i.e. V ≈ V0 = A0 L0 ,
equation (2.12) becomes
N h+u
δ u A L0 = P δ u
A0 L20

and from this equation it follows that


h+u
N =P.
L0

This is the same equation as the equation of equilibrium (2.1). After substituting
for N from (2.10) the equation (2.11) will be received again.
Utilization of principle of virtual displacements (PVD) is a convenient way to
obtain conditions of equilibrium for complex structures. For general three-
dimensional case we have three components of displacement u, v, w and six
components of Green’s strain

∂u 1  ∂u   ∂v   ∂w  
2 2 2
εx = +   +   +    ,
∂x 2  ∂x   ∂x   ∂x  
 

∂v 1  ∂u   ∂v   ∂w  
2 2 2
ε y = +   +   +   
 ,
∂y 2  ∂y   ∂y   ∂y  
 

∂w 1  ∂u   ∂v   ∂w  
2 2 2
εx = +   +   +    , (2.14)
∂z 2  ∂z   ∂z   ∂z  
 
NONLINEAR FINITE ELEMENT ANALYSIS 11

∂u ∂v ∂u ∂u ∂v ∂v ∂w ∂w
γ xy = + + + + ,
∂y ∂x ∂x ∂y ∂x ∂y ∂x ∂y

∂u ∂w ∂u ∂u ∂v ∂v ∂w ∂w
γ xz = + + + + ,
∂z ∂x ∂x ∂z ∂x ∂z ∂x ∂z
∂v ∂w ∂u ∂u ∂v ∂v ∂w ∂w
γ yz = + + + + .
∂z ∂y ∂y ∂z ∂y ∂z ∂y ∂z
In finite element method are displacement interpolated within the finite
elements as
u = ∑ N i ui , v = ∑ N i vi , w = ∑ N i wi , (2.15)
i i i
where ui, vi, wi are nodal displacements and Ni are shape functions. Substituting
these equations into expressions of Green’s strain components, we obtain
1
ε = ( BL + B N ) d . (2.16)
2
In matrix equation (2.16)
 εx 
ε 
 y
 ε z 
ε =  , (2.17)
γ yz 
γ xz 
 
γ xy 

and d is matrix of nodal displacements. Matrix BL is the usual small displacement


matrix and matrix BN reflects the fact that Green’s strain is a nonlinear function of
displacements. Elements of this matrix are linear functions of nodal displacements
d. It might be shown that virtual strain corresponding to the virtual nodal
displacements δ d is
δ ε = ( BL + B N ) δ d = B δ d . (2.18)
According to the principle of virtual displacements, virtual work of internal
forces must be equal to virtual work of external forces if the structure is in
equilibrium. This is represented by the equation

∫δ ε
T
σ dV = δ d T F (2.19)
V
where F is matrix of nodal forces.
We suppose linear relation between stress and strain components, hence
σ = Dε
where D is matrix of material elastic constants.
12 NONLINEAR FINITE ELEMENT ANALYSIS

Substituting (2.18) into (2.19) gives


δ d T BT σ d V = δ d T F
V
(2.20)

for any kinematic admissible set of virtual displacements δ d . Then

∫ B σ dV = F .
T
(2.21)
V

The last equation is a matrix representation of a set of nonlinear algebraic


equations for unknown nodal displacements d.
R (d ) = F . (2.22)

2.2 Incremental-iterative solutions


We have seen that assumption of large displacements leads to nonlinear equation
of equilibrium (2.1) or (2.11) for a simple bar truss example. Generally, in finite
element analysis we have a set of nonlinear equations (2.22).
Let us start with the bar-truss example. The equation of equilibrium (2.1) or (2.11)
can be written in a form
R (u ) = P (2.23)
where

R (u ) = N
h+u
L0
=
EA
2 L30
(u3 + 3h u2 + 2 h2u). (2.24)

represents a component of internal force.


The basic step to solve the nonlinear equation (2.24) is a linear approximation
for small increment of force and corresponding increment of displacement.
Assume that for a prescribed value of force P we managed to find (e.g. by error
and trial method) a displacement u satisfying the equation (2.23). Internal force
R(u + d u ) for new external force P1 ( P + d P ) can be approximated by the linear
function
dR
R ( u + d u ) = R (u ) +   du
 d u u
and approximate condition of equilibrium is
dR
R (u ) +   du = P + d P .
 d u u
Assuming equation (2.23) gives

dR
  du = d P (2.25)
 d u u
NONLINEAR FINITE ELEMENT ANALYSIS 13

or
KT (u ) d u = d P (2.26)
where
dR
KT =   (2.27)
 d u u
is called the tangent stiffness. For the particular case of the bar-truss, tangent
stiffness can be easily found as
d u+h u+h dN
KT =   N +   .
d u  L0   0  du
L
Using the equation (2.10) gives
d N EA u + h
=
d u L0 L0
from which
K T = K 0 + K u + Kσ (2.28)
where
2
EA h 
K0 =   is the linear stiffness
L0  L0 

2
E A  u u  h 
2
Ku = 2 +      is the initial displacement stiffness
L0  h  h    L0 
 

N
Kσ = is the initial stress stiffness.
L0

The linear stiffness, which is independent on displacement, is familiar from


small displacement structural analysis. The initial displacement stiffness reflects
the effect of displacement on stiffness3. The initial stress stiffness reflects the fact
that there is an axial force in the bar prior to load increment.
In like manner, we can precede in a general case described by the equation
(2.21) or (2.22) and derive
R (d + d d ) = R (d ) + KT d d
and
K T dd = d F (2.29)
where
3
e.g. it can be seen from the diagram in Figure 2.2 that for compressive load P stiffness decreases
and for tensional force P increases.
14 NONLINEAR FINITE ELEMENT ANALYSIS

∂R
KT =
∂d
is the tangent stiffness matrix. We can also find out that
K T = K 0 + K u + Kσ . (2.30)
where KT is linear stiffness matrix, Ku is initial displacement stiffness matrix and
Kσ is initial stress stiffness matrix.
Introduction of tangent stiffness matrix is crucial for solution of nonlinear
equations (2.22). The most widely used methods are briefly introduced in the
following text:

2.2.1 Incremental method


The load is divided into a set of small increments ∆Fi . Increments of displacements
∆di are calculated from the set of linear simultaneous equations

KT (i −1) ∆d i = ∆Fi .31)

and an updated solution is obtained as


d i = d i −1 + ∆d i . 32)

Figure 2.3: Incremental method


NONLINEAR FINITE ELEMENT ANALYSIS 15

The procedure is shown in Figure 2.3. It is obvious that solution error, i.e.
difference from exact solution gradually cumulates. To reduce error, large number
of small incremental steps has to be done that is inefficient. On the other hand,
division of loading process into sufficiently small increments is necessary to model
load path dependent behavior of a structure. Dependence of response on a manner
of loading, not only of final values of loads is typical for problems with plastic
deformation and with friction. In these problems, incremental method is usually
combined with one of following methods.

2.2.2 Newton-Raphson method


Suppose that initial displacements d 0 are known. The first guess of nodal
displacements for load F is calculated by solving set of linear algebraic equations

KT ( 0) d1 = F (2.33)

where

KT ( 0) = KT (d 0 )

is tangent stiffness matrix calculated for initial displacements.


As the displacements d1 are most probably not accurate, the equilibrium
equation (2.22) is not satisfied and
R (d1 ) ≠ F
that means there are unbalanced (or residual) nodal forces
r1 = R (d1 ) − F . (2.34)
By computing new tangential stiffness matrix
KT (1) = KT (d1 )

and solving new set of algebraic linear equations


KT (1) ∆d1 = r1 (2.35)

we will obtain an improved solution


d 2 = d1 + ∆d1 . (2.36)
If r2 = R (d 2 ) − F ≠ 0 the procedure is repeated until the sufficiently accurate
solution is obtained. The iterations are schematically shown in Figure 2.4.
This method, known as Newton-Raphson4 method (NR) is often combined with
incremental method as displayed in Figure 2.5.

4
Joseph Raphson (1648-1715) was an English mathematician, a Fellow of the Royal Society of
London and friend of Newton.
16 NONLINEAR FINITE ELEMENT ANALYSIS

Figure 2.4: Standard Newton-Raphson (NR) method.

Figure 2.5: Combination of Newton-Raphson and incremental methods.

2.2.3 Modified Newton-Raphson method


The standard Newton-Raphson method, although effective in many cases, needs
the solution of the set of linear equations (2.35) which is time demanding for large
systems. Modified Newton-Raphson method (MNR) differs from standard NR
algorithm in that the stiffness matrix is only updated occasionally. In the example
shown in Figure 2.6, the tangential stiffness matrix is formed and decomposed at
the beginning and used throughout the iterations. Advantage of the method is in
saving computer time, because factorization of the tangent stiffness matrix is
NONLINEAR FINITE ELEMENT ANALYSIS 17

performed only once for the load increment. On the other hand, number of
iterations needed is usually larger.

Figure 2.6: Modified Newton-Raphson(MNR) method.

2.2.4 Quasi-Newton methods


There exist many other methods for solution of the set of nonlinear algebraic
equations, so called quasi-Newton methods. The most popular among them is
Broyden – Fletcher – Goldfarb – Shanno (BFGS) method.

2.3 Linear stability analysis


Theoretically, below a certain critical load a structure is in position of stable
equilibrium, whilst above that load the equilibrium may be unstable. Unstable
equilibrium means that though the structure is in equilibrium, any arbitrary small
disturbance will cause loss of this equilibrium. In many practical problems, the
displacements are small for load less than critical and behavior of the structure can
be considered as a linear function of applied load. The typical example is Euler
strut buckling, Figure 2.7.
For axial force N that is less than critical, the strut is in stabile equilibrium. This
equilibrium is possible if a lateral load P then deflects the strut as well. If the
lateral load is removed, the strut will return to its straight shape.
18 NONLINEAR FINITE ELEMENT ANALYSIS

If the force N is greater than critical, the strut can remain (theoretically) straight
but its equilibrium is unstable, any small lateral load will cause deflection
increasing until the collapse5.

Figure 2.7: Buckling of a strut

For load less than critical small longitudinal (in plane) and lateral displacements
allow the initial displacement stiffness matrix Ku to be ignored. The equilibrium
equation can be written as
[K 0 + λK σ (N )] d = P (2.37)
The elastic critical (buckling) load is given by the lowest value of load parameter λ
for which d ≠ 0 when the lateral load P = 0. Physically this means that
equilibrium is possible with very small lateral displacements in the absence of any
lateral load. In mathematical sense, we have to solve the eigenvalue problem
[K 0 + λKσ (N )] d = 0 (2.38)
where λ is the eigenvalue and d is the corresponding eigenvector.
It should be noted that due to assumptions accepted the solution represents itself
only an estimation of the upper bound of the structure load capacity.

2.4 Large displacement and large strain behavior


When strain is large, it is inadmissible to neglect shape and volume changes of a
structure. For example, in the simple bar example we have to introduce current
cross sectional A instead of initial A0 and current length L instead of initial length
L0 in the equations (2.10) and (2.11).
Accordingly, integration in the equation (2.19) expressing the principle of
virtual displacements has to be taken over the current volume. This brings
problems, as the current volume is unknown, because it depends on displacements
that are unknown too and must be calculated first. To solve this problem, it is
necessary to introduce a transformation so that integrals are taken over known
volume. Two possible ways are briefly described bellow:

5
In reality, unstable equilibrium is due to initial imperfections (e.g. eccentricity of force N, initial
curvature of the strut etc.) impossible, but estimation of critical load may be useful in many cases.
NONLINEAR FINITE ELEMENT ANALYSIS 19

2.4.1 Total Lagrangian formulation


In a Total Lagrangian (TL) formulation all integrals are calculated with respect to
the initial undeformed configuration of the structure

∫ δ εG σ P d V = δ d
T T
F (2.39)
V0

where V0 is the initial volume. Due to transformation, new measure for stress so
called second Piola-Kirchhoff stress tensor σP has to be introduced with Green’s
strain tensor εG.

2.4.2 Updated Lagrangian formulation


In an Updated Lagrangian (UL) formulation, a known deformed configuration i is
taken as an initial state for subsequent configuration (i+1) and this is continually
updated as the calculation proceeds

∫δ ( )
(i +1) T (i +1)
εA σC d V = δ d T F (2.40)
V0

In the left side of the equation (2.40), σC is Cauchy stress tensor and εA is Almansi
strain tensor respectively. Notation (i+1)εA and (i+1)σC means that the strain and
stress are in configuration (i+1). Integration is done over volume Vi that is in
current configuration i.
Use of different measures for stress and strain in TL and UL formulation
follows condition that virtual work of internal forces must be the same irrespective
of the volume over which is integration taken6.

6
That means that stress and strain measures must be work conjugate.
20 NONLINEAR FINITE ELEMENT ANALYSIS

3. MATERIAL NONLINEARITIES
3.1 Introduction
Linear elastic FE analysis is based on linear constitutive stress-strain equations
σ = Dε (3.1)
in which the terms of material matrix D are expressed as functions of constant
values of modulus of elasticity and Poisson’s ratio. The constant D matrix leads to
a constant stiffness matrix K, which is for strain-displacement relationship
ε = Bd (3.2)
given by
T
K = ∫ B D B dV (3.3)
V
Departure from linear elasticity implies that the linear elastic constitutive
equations are no longer valid, as the material matrix is no longer constant. The
non-constant material matrix D represents nonlinear constitutive equations
corresponding to the adopted nonlinear material model. Consequently, the
conditions of equilibrium derived in FEM from principle of virtual displacements
are nonlinear like equations (2.21) and (2.22). Solution of these equations is based
on the same methods as in geometrically nonlinear case. Usually it is necessary to
divide load into increments and perform equilibrium iterations (e.g. by MNR or
NR method) for each increment. Moreover, for each load increment there must be
performed stress iterations, as the material matrix is function of strain. The strain is
unknown a priori and will be computed only.
Material nonlinearities are often combined with geometrical and/or boundary
nonlinearities.

3.2 Nonlinear elasticity models


Nonlinear elastic behavior of materials can be formulated in several ways. The
simplest is total formulation, where the stress and strains are defined in terms of
the secant modulus of elasticity Es, see Figure 3.1,
σ = E s (ε ) ε . (3.4)
In hypo-elastic formulation, the relationship between the increments of stress
and strain are defined by the tangential modulus of elasticity Et
d ε = Et (ε ) d σ (3.5)
NONLINEAR FINITE ELEMENT ANALYSIS 21

The nonlinear elastic material law can also be formulated in terms of


hyperelastic formulation, which assumes the existence of strain energy density
function U and the corresponding complementary energy density function U∗ such
that

dU dU∗
σ = and ε= (3.6)
dε dσ
The hyperelastic material model is usually used for rubber-like materials7.

Figure 3.1: Nonlinear elasticity model.

Material models for multiaxial states of stress are usually based on


generalization of one-dimensional concepts. For example, in a hyperelastic
formulation components of stress tensor are computed as
∂U
σ= (3.7)
∂ε
that means
∂U ∂U ∂U
σx = , σy = , τ xy = etc.
∂ε x ∂ε y ∂γ xy

Figure 3.2: Strain energy density functions U and U∗

7
An example is the Mooney-Rivlin material model used for modelling rubber-like materials.
22 NONLINEAR FINITE ELEMENT ANALYSIS

For any nonlinear elastic material model, it is possible to define relation


between stress and strain increments as
d σ = DT d ε (3.8)
where matrix DT is function of strains ε. Consequently, a set of equilibrium
equations we receive in FEM is nonlinear and must be solved by use of any
method (e.g. NR) described above.

3.3 Elastoplastic material model


3.3.1 Yielding criterion
Experiments indicate that linear elastic model is acceptable only within a limited
range of stress. As an example, the stress-strain curve from tension test of steel
specimen is shown in Figure 3.3. Until the yield stress represented by point A (in
the given case σy = 280 MPa) the deformations are elastic and stress-strain relation
may be described as σ = E ε. When the stress level exceeds the yield stress, an
elastoplastic constitutive law governs the relationship between increments of stress
and strain.
Due to lack of information,8 approximate stress-strain curves are usually used in
analysis. Bilinear approximation defined by yield stress, modulus of elasticity E
and tangential modulus ET is shown in Figure 3.4. If ET = 0, material model is
elastic-perfectly plastic. If ET ≠ 0 material model assumes strain hardening.

450

400

350

300
A
σ [MPa]

250

200

150

100

50

0
0.00 0.02 0.04 0.06 0.08 0.10 0.12
ε
Figure 3.3: Typical stress-strain curve for mild steel

It should be noted that curves in Figures 3.3 and 3.4 are for tensile behavior. It
is usually assumed that similar curves for compressive behavior are applicable if
there has been no history of plastic deformation.

8
In a design process, the real material curve is usually unknown, only basic values like yield stress
etc. are available. Moreover, the material properties slightly differ by different supplies.
NONLINEAR FINITE ELEMENT ANALYSIS 23

Figure 3.4: Elastoplastic model with linear strain hardening

The indication of yielding under multiaxial conditions in metals is obtained


from experiments usually conducted on cylindrical samples subjected to combined
axial load and torque. Experiments suggest that there is no significant difference in
behavior of metals in tension or compression and no volume change associated
with yielding and no effect of mean stress level on yielding can be assumed.
In a mathematical description, onset of yielding may be represented by a scalar
function termed the yield function F. The yield function is written in a form, which
leads to the conditions

F < 0 for elastic and F = 0 for plastic deformation. (3.9)

In engineering practice, two following conditions for yielding are most


frequently used:

● Von Mises yield criterion


F = (σ 1 − σ 1 ) 2 + (σ 2 − σ 3 ) 2 + (σ 3 − σ 1 ) 2 − 2σ y = 0 (3.10)

where σ1, σ2 and σ3 are principal stresses. Thus, yield occurs when the
effective stress σeff reaches the yield stress value σy
1
σ eff = (σ1 − σ1 )2 + (σ 2 − σ 3 )2 + (σ 3 − σ1 )2 = σ y . (3.11)
2

● Tresca yield criterion


[ ][ ][ ]
F = (σ 1 − σ 2 ) 2 − σ y2 (σ 2 − σ 3 ) 2 − σ y2 (σ 3 − σ 1 ) 2 − σ y2 = 0 . (3.12)

The largest difference between these two classical yield criteria is about 15%
for the pure shear stress state. For other stress states is the difference less. Hence,
both criteria are frequently considered as equivalent in engineering practice.
Any yield condition that is function of stress tensor components σ and material
parameters κ
24 NONLINEAR FINITE ELEMENT ANALYSIS

F (σ , κ ) = 0 (3.13)
defines a yield surface in principal stress space, see Figure 3.5. Stress points that
lie inside the yield surface are associated with elastic stress states whereas those
that lie on the surface represent plastic stress states. No stress point can be outside
the yield surface.

Figure 3.5: Yield surface

3.3.2 Post yielding behavior


The fundamental assumption in describing post-yielding behavior is the
decomposition of the total strain increment into an elastic (recoverable) part and a
plastic (irreversible) part. For uniaxial stress state is, according to Figure 3.6

= ET ,

(3.14)

=E
dε e
and plastic strain increment is then
E − ET
dε p = dε − dε e = dσ . (3.15)
E ET

Figure 3.2: Decomposition of the total strain increment


NONLINEAR FINITE ELEMENT ANALYSIS 25

By analogy, in multiaxial stress state the total strains we decompose into elastic
and plastic parts too

ε = εe + ε p (3.16)
In multiaxial cases, subsequent loading after first yield produces further plastic
deformation that can result in a modification of the shape and/or position of the
yield surface.
For a perfectly plastic material, the yield surface remains unchanged during
plastic deformation. For a strain hardening material, plastic deformation produces
a change in shape and position of the yield surface. This means that initial yield
surface is gradually replaced by the subsequent yield surfaces. A modified yield
function is adopted which has a form such as
F (σ , ε p , K ) = 0 . (3.17)
This yield function depends on the stresses but also the plastic strains and a
hardening parameter K. The way in which the plastic strains modify the yield
function is defined by hardening rules:

Figure 3.7: Isotropic hardening.

1. An isotropic hardening law implies that the yield surface increases in


size but maintains its original shape under loading conditions.
Schematic representation of isotropic hardening for uniaxial and biaxial
stress state is shown in Figure 3.7.

2. In kinematic hardening, the original yield surface is translated to a new


position in stress space with no change of its shape and size as shown in
Figure 3.8. Kinematic hardening has paramount importance in
modelling cyclic behavior.
26 NONLINEAR FINITE ELEMENT ANALYSIS

3. The combination of the two principal hardening laws leads to a mixed


hardening law, where the initial yield surface both expands and
translates as a consequence of plastic flow.

Figure 3.8: Kinematic hardening.

3.3.3 Constitutional equations of elastoplastic material


The yield criterion says whether plastic deformation will occur but says nothing
about the plastic behavior of a material after onset of plastic deformations. This is
defined by so-called flow rule in which is the rate and the direction of plastic
strains is related to the stress state and the stress rate. This relation can be
expressed as
∂Q
d ε ijp = dλ (3.17)
∂σ ij

or in matrix form as
∂Q
d ε p = dλ (3.18)
∂σ
where dλ is a scalar value (to be determined) and Q is a scalar valued function of
stress components called plastic potential.
For metals, the so-called associated flow rule, in which the plastic potential
surface coincides with the yield surface, i.e.
Q=F

can be adapted to model plastic flow. For some other materials, non-associated
flow rule in which Q ≠ F has to be used to model plastic flow adequately. In the
following text we will deal with associated flow rule
∂F
d ε p = dλ (3.19)
∂σ
NONLINEAR FINITE ELEMENT ANALYSIS 27

Consider a uniaxial stress state first. The plastic behavior of material is


described as
dσ = ET d ε (3.20)
where ET is constant for a bilinear material as obvious from the equations (3.14)
and (3.15).
In a multiaxial stress state, we can formulate a similar constitutive equation
d σ = DT d ε (3.21)
where tangential material matrix DT can be derived from known stress tensor σ,
strain tensor ε and constitutive matrix D from equation (3.1) in following way:
The first step is strain decomposition into elastic dε e and plastic part dε p
dε = dεe + dε p . (3.22)
From constitutive law it follows that
dσ = D dε e
hence
(
dσ = D dε − dε p . ) (3.24)
From associated plastic flow rule, it follows that
∂F
d ε p = dλ = adλ (3.25)
∂σ
where
 ∂F 
a= . (3.26)
 ∂σ 

Using equations (3.24) and (3.25) we obtain


d σ = D (d ε − a d λ ) . (3.27)
The stress point must lie in yield surface9 and hence the following consistency
conditions must be fulfilled
T T
 ∂F   ∂F  p
dF =  dσ +  p  dε = 0 (3.29)
 ∂σ   ∂ε 
or with respect to equations (3.25) and (3.25)
T
T  ∂F 
d F = a dσ +   adλ = 0.
 ∂ε p 
After substituting from equation (3.27) we obtain

9
note that yield surface may change according to hardening rule
28 NONLINEAR FINITE ELEMENT ANALYSIS

T
 ∂F 
d F = a D (d ε − a d λ ) + 
T
 adλ = 0
 ∂ε p 
or
d F = a T D (d ε − a d λ ) − A d λ = 0 (3.30)
where scalar quantity A is defined as
T
 ∂F 
A = −  a. (3.31)
 ∂ε p 
Now, we can derive parameter dλ from equation (3.29)

aT D d ε
dλ = (3.32)
A + aT D a
and substituting this expression for dλ into equation (3.28) we finally obtain
 D aaT D 

dσ = D − dε . (3.33)
 A + aT D a 
 
When compare the last equation with equation (3.21) we can see that

D a aT D
DT = D − . (3.34)
A + aT D a

Note that material matrix D is symmetric, i.e. D T = D , hence matrix DT is


symmetric as well.

3.3.4 Integration of constitutive equations


We have derived that for infinitesimal increments of stress and strain it holds
d σ = DT d ε .
In FE analysis we need to work with finite increments ∆σ and ∆ε for which is
the relation above approximate only, so if we use relation
∆σ = DT ∆ε
for large increments of stress and strain, an error occurs as stress σ + ∆σ in
subsequent step will not satisfy constitutive law and consistency condition. Hence,
we need to integrate over the increment of pseudo-time

∆σ = ∫ d σ
∆t
where, according to equation (3.24)
(
dσ = D dε − dε p . )
NONLINEAR FINITE ELEMENT ANALYSIS 29

It is important to note that if plastic flow is present, the DT changes during


increment ∆t and as a result, the ratio between total and plastic strain changes too.
To obtain correct results, various stress increment integration schemes that
differs in the degree of approximation have been developed. Frequently used are
the following schemes:

3.3.4.1 Generalized trapezoidal rule


Consider that we know stress σ n , total strain ε n and plastic strain ε np at time step
n. Then at step n+1

(
σ n +1 = σ n + ∆σ n +1 = σ n + D ∆ε n +1 − ∆ε np+1 ) (3.35)

∆ε np+1 = ∆λ [(1 − α )an + α an +1 ] (3.36)

Fn +1 = 0 (3.37)

3.3.4.2 Generalized mid point rule

(
σ n +1 = σ n + ∆σ n +1 = σ n + D ∆ε n +1 − ∆ε np+1 ) (3.35)

∆ε np+1 = ∆λ an +α (3.36)

Fn +α = 0 (3.37)

In both rules, α is a parameter ranging from 0 to 1.


For α = 0 we obtain explicit forward Euler integration scheme. Advantage of
this algorithm is in its simplicity; disadvantage is that it is conditionally stable
only. That means that step increment has to be smaller than some critical value to
avoid instability of the solution.
For α = 1 we obtain implicit backward Euler integration scheme

(
∆σ n +1 = D ∆ε n +1 − ∆ε np+1 )
∆ε np+1 = ∆λ an +1

Fn +1 = F (σ n +1 , ε n +1 ) = 0
30 NONLINEAR FINITE ELEMENT ANALYSIS

It is obvious that in difference with forward scheme, we deal with values


defined at the end of the increment, which are unknown at start of it. Hence, the
procedure is of an iterative nature. This means that at beginning of the increment,
the trial stress is estimated by assuming elastic deformation and computed values
are then checked whether consistency condition and constitutional equation are
satisfied. If not, the process is repeated with improved values until the conditions
are satisfied.

3.3.5 Numerical procedures


The tangential material matrix DT is used to form a tangential stiffness matrix KT.
When the tangential stiffness matrix is defined, the displacement increment is
obtained for a known load increment
K T ∆d = ∆F (3.22)
As load and displacement increments are final, not infinitesimal, displacements
obtained by solution of this set of linear algebraic equation will be approximate
only. That means, conditions of equilibrium of internal and external nodal forces
will not be satisfied and iterative process is necessary. Any of methods mentioned
above may be used.
The problem that arises now is the fundamental problem in computational
elastoplasticity - not only equilibrium equations but also constitutive equations of
material must be satisfied. That means that within the each equilibrium iteration
step check of stress state and iterations to find elastic and plastic part of strains at
every integration point must be included. The iteration process continues until
both, equilibrium conditions and constitutive equations are satisfied
simultaneously. The converged solution at the end of load increment is then used
at the start of new load increment.
NONLINEAR FINITE ELEMENT ANALYSIS 31

REFERENCES
[1] Hinton, E: Introduction to Nonlinear Finite Element Analysis. NAFEMS,
Glasgow 1992

[2] Crisfield, M. A.: Non-linear Finite Element Analysis of Solids and Structures.
John Wiley & Sons 1991

[3] Bittnar, Z., Šejnoha, J.: Numerické metody mechaniky (Numerical methods of
mechanics). ČVUT, Praha 1992

[4] Okrouhlík, M., Höshl, C., Plešek, J., Pták, S., Nadrchal, J.: Mechanika
poddajných těles, numerická matematika a superpočítače (Mechanic of solids,
numerical mathematics and supercomputers). Czech Academy of Science,
Prague 1997.

[5] Okrouhlík, M.: Implementation of Nonlinear Continuum Mechanics in Finite


Element Codes. Institute of Thermodynamics, Prague 1995.

[6] Hinton, E., Ezatt, M., H.: Fundamental Tests for Two and Three Dimensional,
Small Strain, Elastoplastic Finite Element Analysis. NAFEMS, Glasgow 1987.

[7] Electronic documentation of program COSMOS/M, version 2.95. SRAC, Los


Angeles 2005, www,cosmosm.com.

[8] Falzon, B., G., Hitchings, D.: An Introduction to Modeling Buckling and
Collapse. NAFEMS, Glasgow 2006
[9] Eurocode 3: Design of steel structures - Part 1-1: General rules and rules for
buildings. CEN, Brussels 2005.

Você também pode gostar