# INTRODUCTION TO NONLINEAR FINITE

ELEMENT ANALYSIS
Faculty of Mechanical Engineering,
Technical University of Košice, Slovakia
HS Wismar, June 2009
NONLINEAR FINITE ELEMENT ANALYSIS 2
NONLINEAR FINITE ELEMENT ANALYSIS 3
CONTENTS
1. Structural nonlinearities ................................................................................................... 4
1.1 Introduction ............................................................................................................ 4
1.2 Types of structural nonlinearities ........................................................................... 5
1.3 Concept of time curves ........................................................................................... 5
2. Geometrically nonlinear finite element analysis ............................................................... 7
2.1 Large displacement and small strain behavior ........................................................ 7
2.2 Incremental - iterative solutions ........................................................................... 12
2.3 Linear stability analysis ........................................................................................ 17
2.4 Large displacement and large strain behaviour .................................................... 19
3. Material nonlinearities .................................................................................................... 20
3.1 Introduction .......................................................................................................... 20
3.2 Nonlinear elasticity models .................................................................................. 20
3.3 Elastoplastic material model ................................................................................ 22
Appendix...........................................................................................................................30
Examples for COSMOS/M program.........................................................................31
NONLINEAR FINITE ELEMENT ANALYSIS 4
1. STRUCTURAL NONLINEARITIES
1.1 Introduction
Many engineering problems can be solved using a linear approximation. In the Finite
Element Analysis (FEA) the set of equations, describing the structural behaviour is then
linear
F d K ·
(1.1)
In this matrix equation, K is the stiffness matrix of the structure, d is the nodal
displacements vector and F is the external nodal force vector.
Characteristics of linear problems is that
• the displacements are proportional to the loads,
• the stiffness of the structure is independent on the value of the load
level.
Though behaviour of real structures is nonlinear, e.g. displacements are not proportional to
the loads; nonlinearities are usually unimportant and may be neglected in most practical
problems.
The set of linear algebraic equations (1.1) is received if assuming 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, and supports of
the structure remain unchanged.
Solution of many problems needs abandonment of these approximations. The
displacements may be so large 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
structure behaves nonlinearly. If these phenomena are included in a FEA, the set of
equilibrium equations becomes nonlinear.
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 behaviour 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 behaviour 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.
important. Especially, plastic deformations depend on a manner of
nonlinear FE analysis.
d) The structural behaviour can be markedly non-proportional to the
e) The initial state of stress (e.g. residual stresses from heat treatment,
welding 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
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.
NONLINEAR FINITE ELEMENT ANALYSIS 6
As an example, time curves of forces F
1
and F
2
Figure 1.1. Values of forces at any time are defined as
( )
1 1 1
f t F λ · and
( )
2 2 2
f t F λ ·
where f
1
and f
2
are nominal (input) values of forces and λ
1
and λ
1
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 nonlinearities
2
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 behaviour 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 L
0
.
Figure 2.2: Example of nonlinear structure – single bar truss
Using the free body diagram shown in Figure 2.1 the equilibrium equation is
0 sin · − P N α
or
. 0 · −
+
P
L
u h
N (2.1)
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
NONLINEAR FINITE ELEMENT ANALYSIS 8
ε
0
A E N · (2.2)
where A
0
is the initial cross sectional area and
ε
is the engineering strain defined as
0
0
L
L L −
· ε
. (2.3)
As lengths are given as
2 2
0
h a L + ·
and
2 2
) ( u h a L + + · (2.4)
the expression for strain is getting rather complicated. We can overcame this problem by
introducing Green’s strain defined as
2
0
2
0
2
2L
L L
G

· ε (2.5)
which for our problem becomes
2
0 0 0
2
1

,
_

¸
¸
+ ·
L
u
L
u
L
h
G
ε . (2.6)
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

,
_

¸
¸
+

·

,
_

¸
¸
− +
+
·
+
·
+ −
·

·
0
0
0
0
0
0
0
0
0
0
0
0
2
0
2
0
2
2
2
1
1 1
2
1
2
1
2 2 L
L
L
L L
L
L L
L
L L
L
L L
L
L L
L
L L
G
ε ε ε ε
or
2
2
1
ε ε ε + ·
G
. (2.7)
Noting that the constitution equation was measured as
ε σ E
A
N
· ·
0
(2.8)
the same constitutive equation when using Green’s strain should be
G G G
G
E
E E ε
ε
ε
ε ε
ε
ε
ε
ε
σ
2
1
1
2
1
2
+
·
+
· ·
. (2.9)
NONLINEAR FINITE ELEMENT ANALYSIS 9
This means that we should use value ] ) 2 / 1 ( 1 /[ ε + ·

E E 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.
ε
G
= 0,002 + 0,5⋅ 0,002
2
= 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
G
E ε σ ≈ and
according to equation (2.6)

,
_

¸
¸
+ ·
2
2
0
0
2
1
u u h
L
A E
N
. (2.10)
Substituting (2.10) to equilibrium equation (2.1) and assuming that for small strain is
0
L L ≈ gives the equilibrium equation
( ) . 2 3
2
2 2 3
3
0
0
P u h u h u
L
A E
· + +
(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⋅ 10
5
MPa, A
0
= 100 mm
2
, a = 200 mm and h = 20 mm is shown in Figure 2.2.
-1000
-500
0
500
1000
1500
2000
2500
-10 -8 -6 -4 -2 0 2 4 6 8 10
u[mm]
P [N]
Figure 2.3: Geometrically nonlinear behaviour of a single bar truss
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 kinematically admissible
NONLINEAR FINITE ELEMENT ANALYSIS 10
set of virtual displacements. For our structure with one degree of freedom, only one virtual
displacement
u δ
is possible and principle of virtual displacements has a form
u P V
V
G
δ δε σ ·

d
(2.12)
where
G
δε is virtual strain corresponding to virtual displacement
u δ
. The virtual strain
can be expressed from equation (2.6) as
.
2
1
d
d
d
d
2
0
2
0 0 0
u
L
u h
u
L
u
L
u
L
h
u
u
u
G
G
δ δ δ
ε
δε
+
·
1
1
]
1

¸

,
_

¸
¸
+ · ·
(2.13)
It is assumed in principle of virtual displacements that virtual displacement is infinitesimal
and hence the stress
) / ( A N · σ
remains unchanged. Noting that
σ
and
G
δε 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.
0 0 0
L A V V · ≈ , equation (2.12) becomes
u P L A u
L
u h
A
N
δ δ ·
+
0
2
0 0
and from this equation it follows that
P
L
u h
N ·
+
0
.
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
1
1
]
1

¸

,
_

¸
¸

+
,
_

¸
¸

+
,
_

¸
¸

+

·
2 2 2
2
1
x
w
x
v
x
u
x
u
x
ε
,
1
1
]
1

¸

,
_

¸
¸

+

,
_

¸
¸

+

,
_

¸
¸

+

·
2 2 2
2
1
y
w
y
v
y
u
y
v
y
ε
,
1
1
]
1

¸

,
_

¸
¸

+
,
_

¸
¸

+
,
_

¸
¸

+

·
2 2 2
2
1
z
w
z
v
z
u
z
w
x
ε
, (2.14)
NONLINEAR FINITE ELEMENT ANALYSIS 11
y
w
x
w
y
v
x
v
y
u
x
u
x
v
y
u
xy

+

+

+

+

· γ
,
z
w
x
w
z
v
x
v
z
u
x
u
x
w
z
u
xz

+

+

+

+

· γ ,
z
w
y
w
z
v
y
v
z
u
y
u
y
w
z
v
yz

+

+

+

+

· γ
.
In finite element method are displacement interpolated within the finite elements as
i
i
i
u N u

·
,
i
i
i
v N v

·
,
i
i
i
w N w

·
, (2.15)
where u
i
, v
i
, w
i
are nodal displacements and N
i
are shape functions. Substituting these
equations into expressions of Green’s strain components, we obtain
d B B ε )
2
1
(
N L
+ · . (2.16)
In matrix equation (2.16)
¹
¹
¹
¹
¹
¹
¹
¹
¹
;
¹
¹
¹
¹
¹
¹
¹
¹
¹
¹
'
¹
·
yz
xz
xy
z
y
x
γ
γ
γ
ε
ε
ε
ε , (2.17)
and d is matrix of nodal displacements. Matrix
L
B is the usual small displacement matrix
and matrix
N
B 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
d δ
is
. ) ( d B d B B ε δ δ δ · + ·
N L
(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
F d σ ε
T
V
T
V δ δ

· d
(2.19)
where F is matrix of nodal forces.
NONLINEAR FINITE ELEMENT ANALYSIS 12
We suppose linear relation between stress and strain components, hence
Dε σ ·
where D is matrix of material elastic constants.
Substituting (2.18) into (2.19) gives
F d σ B d
T T
V
T
V δ δ

· d
(2.20)
for any kinematically admissible set of virtual displacements
d δ
. Then
F σ B
T

·
V
V d
. (2.21)
The last equation is a matrix representation of a set of nonlinear algebraic equations for
unknown nodal displacements d.
F d R · ) (
. (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
P u R · ) (
(2.23)
where
( ). 2 3
2
) (
2 2 3
3
0 0
u h u h u
L
A E
L
u h
N u R + + ·
+
·
(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 + du) for new
external force P
1
= P + dP and related displacement u
1
= u + du can be approximated by
the linear function
NONLINEAR FINITE ELEMENT ANALYSIS 13
u
u
R
u R u u R
u
d
d
d
) ( ) d (

,
_

¸
¸
+ · +
and approximate condition of equilibrium is
P P u
u
R
u R
u
d d
d
d
) ( + ·

,
_

¸
¸
+
.
Assuming equation (2.23) gives
P u
u
R
u
d d
d
d
·

,
_

¸
¸
(2.25)
or
P u u K
T
d d ) ( · (2.26)
where
u
T
u
R
K

,
_

¸
¸
·
d
d
(2.27)
is called the tangent stiffness. For the particular case of the bar-truss, tangent stiffness can
be easily found as
u
N
L
h u
N
L
h u
u
K
T
d
d
d
d
0 0

,
_

¸
¸ +
+

,
_

¸
¸ +
·
.
Using the equation (2.10) gives
0 0
d
d
L
h u
L
EA
u
N +
·
from which
σ
K K K K
u T
+ + ·
0
(2.28)
where
2
0 0
0

,
_

¸
¸
·
L
h
L
A E
K is the linear stiffness
2
0
2
0
2

,
_

¸
¸
1
1
]
1

¸

,
_

¸
¸
+ ·
L
h
h
u
h
u
L
A E
K
u
is the initial displacement stiffness
0
L
N
K ·
σ is the initial stress stiffness.
The linear stiffness, which is independent on displacement, is familiar from small
displacement structural analysis. The initial displacement stiffness reflects the effect of
NONLINEAR FINITE ELEMENT ANALYSIS 14
displacement on stiffness
3
. 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
d K d R d d R d ) ( ) d (
T
+ · +
and
F d K d d ·
T
(2.29)
where
d
R
K

·
T
is the tangent stiffness matrix. We can also find out that
σ
K K K K + + ·
u T 0
. (2.30)
where
T
K is linear stiffness matrix,
u
K is initial displacement stiffness matrix and
σ
K 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
i
F ∆ . Increments of displacements
i
d ∆
are calculated from the set of linear simultaneous equations
i i i T
F d K ∆ · ∆
− ) 1 (
.31)
and an updated solution is obtained as
.
1 i i i
d d d ∆ + ·

32)
The procedure is shown in Figure 2.3. It is obvious that solution error – 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 behaviour of a
is typical for problems with plastic deformation and with friction. In these problems,
incremental method is usually combined with one of following methods.
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.
NONLINEAR FINITE ELEMENT ANALYSIS 15
Figure 2.4: Incremental method
2.2.2 Newton-Raphson method
Suppose that initial displacements 0
d
are known. The first guess of nodal displacements
for load F is calculated by solving set of linear algebraic equations
F d K ·
1 ) 0 ( T
(2.33)
where ) (
0 ) 0 (
d K K
T T
· is tangent stiffness matrix calculated for initial displacements.
As the displacements
1
d are (most probably) not accurate, the equilibrium equation (2.22)
is not satisfied and
F d R ≠ ) (
1
that means there are unbalanced (or residual) nodal forces
F d R r − · ) (
1 1
. (2.34)
By computing new tangential stiffness matrix ) (
1 ) 1 (
d K K
T T
· and solving new set of
algebraic linear equations
1 1 ) 1 (
r d K · ∆
T (2.35)
we will obtain an improved solution
NONLINEAR FINITE ELEMENT ANALYSIS 16
1 1 2
d d d ∆ + · . (2.36)
If 0 F d R r ≠ − · ) (
2 2
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-Raphson
4
method (NR) is often combined with
incremental method as displayed in Figure 2.5.
Figure 2.4: Standard Newton-Raphson (NR) method.
Figure 2.5: Combination of Newton-Raphson and incremental methods.
4
Joseph Raphson (1648-1715) was an English mathematician, a Fellow of the Royal Society of
London and friend of Newton.
NONLINEAR FINITE ELEMENT ANALYSIS 17
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 factorisation
of the tangent stiffness matrix is 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
NONLINEAR FINITE ELEMENT ANALYSIS 18
critical and behaviour 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
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
collapse
5
.
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 K
u
to be ignored. The equilibrium equation can be
written as
P d N K K · + )] ( [
0 σ
λ (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
0 d N K K · + )] ( [
0 σ
λ (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.
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 Large displacement and large strain behaviour
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
0
and current length L instead of initial length L
0
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:
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
F d σ ε
T
V
P
T
G
V δ δ

·
0
d
(2.39)
where V
0
is the initial volume. Due to transformation, new measure for stress so called
second Piola-Kirchhoff stress tensor σ
P
has to be introduced with Green’s strain tensor ε
G
.
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
( ) F d σ ε
T
V
C
i T
A
i
V δ δ

·
+ +
0
d
) 1 ( ) 1 (
(2.40)
In the left side of the equation (2.40), σ
C
is Cauchy stress tensor and ε
A
is Almansi strain
tensor respectively. Notation
(i+1)
ε
A
and
(i+1)
σ
C
means that the strain and stress are in
configuration (i+1). Integration is done over volume V
i
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 taken
6
.
6
That means that stress and strain measures must be work conjugate.
NONLINEAR FINITE ELEMENT ANALYSIS 20
3. MATERIAL NONLINEARITIES
3.1 Introduction
Linear elastic FE analysis is based on linear constitutive stress-strain equations
ε D σ ·
(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
d B ε ·
(3.2)
given by
V
T
V
d B D B K

·
(3.3)
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 format, where the stress and strains are defined in terms of the secant modulus of
elasticity E
s
, see Figure 3.1,
ε ε σ ) (
s
E · (3.4)
NONLINEAR FINITE ELEMENT ANALYSIS 21
In hypo-elastic format, the relationship between the increments of stress and strain are
defined by the tangential modulus of elasticity E
t
σ ε ε d ) ( d
t
E · (3.5)
The nonlinear elastic material law can also be formulated in terms of hyperelastic format,
which assumes the existence of strain energy density function U and the corresponding
complementary energy density function U

such that
ε
σ
d
dU
·
and
σ
ε
d
d

·
U
(3.6)
The hyperelastic material model is usually used for rubber-like materials
7
.
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
ε
σ

·
U
(3.7)
that means
x
x
U
ε
σ

·
,
y
y
U
ε
σ

·
,
xy
xy
U
γ
τ

·
etc.
7
An example is the Mooney-Rivlin material model used for modelling rubber-like materials.
NONLINEAR FINITE ELEMENT ANALYSIS 22
Figure 3.2: Strain energy density functions U and U

For any nonlinear elastic material model, it is possible to define relation between stress and
strain increments as
ε D σ d d
T
· (3.8)
Matrix D
T
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
σ
y
= 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 E
T
is shown in Figure 3.4. If E
T
= 0, material model is elastic-perfectly
plastic. If E
T
≠ 0 material model assumes strain hardening.
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
0
50
100
150
200
250
300
350
400
450
0.00 0.02 0.04 0.06 0.08 0.10 0.12
ε
σ

[
M
P
a
] A
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 behaviour. It is usually
assumed that similar curves for compressive behaviour are applicable if there has been no
history of plastic deformation.
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 behaviour 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
NONLINEAR FINITE ELEMENT ANALYSIS 24
0 < F for elastic behavior and 0 · F for plastic deformation. (3.9)
In engineering practice, two following conditions for yielding are most frequently used:
Von Mises yield criterion
0 2 ) ( ) ( ) (
2
1 3
2
3 2
2
1 1
· − − + − + − ·
y
F σ σ σ σ σ σ σ (3.10)
where σ
1
, σ
2
and

σ
3
are principal stresses. Thus, yield occurs when the effective stress σ
eff
reaches the yield stress value σ
y
y eff
σ σ σ σ σ σ σ σ · − + − + − ·
2
1 3
2
3 2
2
1 1
) ( ) ( ) (
2
1
. (3.11)
Tresca yield criterion
[ ] [ ] [ ] 0 ) ( ) ( ) (
2 2
1 3
2 2
3 2
2 2
2 1
· − − − − − − ·
y y y
F σ σ σ σ σ σ σ σ σ . (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
0 ) , , (
3 2 1
· σ σ σ F (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
NONLINEAR FINITE ELEMENT ANALYSIS 25
3.3.2 Post yielding behaviour
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
T
E ·
ε
σ
d
d
, E
e
·
ε
σ
d
d
(3.14)
and plastic strain increment is then
σ ε ε ε d d d d
T
T e p
E E
E E −
· − ·
. (3.15)
Figure 3.2: Decomposition of the total strain increment
By analogy, in multiaxial stress state the total strains are decomposed into elastic and
plastic parts too
p e
ε ε ε + ·
(3.16)
In multiaxial cases, subsequent loading after first yield produces plastic deformation that
can be accompanied by a modification of the shape and/or position of the yield surface.
For a perfectly plastic material, there is no change in the yield surface during plastic
deformation. For a strain hardening material, plastic deformation produces a change in the
yield surface. The initial yield surface is replaced by the subsequent yield surface. A
modified yield function is adopted which has a form such as
0 ) , , ( · K F
p
ε σ . (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:
NONLINEAR FINITE ELEMENT ANALYSIS 26
Figure 3.7: Isotropic hardening.
An isotropic hardening law implies that the yield surface increases in size but maintains its
for uniaxial and biaxial stress state is shown in Figure 3.7.
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.
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.
NONLINEAR FINITE ELEMENT ANALYSIS 27
3.3.3 Constitutional equations of elastoplastic material
The yield criterion says whether plastic deformation will occur but says nothing about the
plastic behaviour 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
ij
p
ij
Q
σ
λ ε

· d d
(3.17)
or in matrix form as
σ
ε

·
Q
p
λ d d (3.18)
where dλ is a scalar value (to be determined) and Q is a scalar valued function of stress
components called plastic potential.
3.3.4 Numerical procedures
In a uniaxial stress state, the plastic behaviour of material can be described as
ε σ d d
T
E · (3.18)
For a bilinear material is E
T
constant as obvious from the equations (3.14) and (3.15).
In a multiaxial stress state a similar constitutive equation can be formulated as
ε D σ d d
T
· (3.19)
The tangential material matrix D
T
is used to form a tangential stiffness matrix K
T
. When
the tangential stiffness matrix is defined, the displacement increment is obtained for a
F d K ∆ · ∆
T
(3.20)
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 28
NONLINEAR FINITE ELEMENT ANALYSIS 29
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.
NONLINEAR FINITE ELEMENT ANALYSIS 30
APPENDIX
NONLINEAR FINITE ELEMENT ANALYSIS 31
EXAMPLES FOR COSMOS/M PROGRAM
1.1 Compressed slender beam
Problem description: Slender elastic beam is compressed by force F and laterally loaded by
a small force P. The forces increase according to time curves
9
TC 1 and TC 2. Force
P = 200 N is associated with the curve TC 1 and force F = 100000 N with the curve
TC 2
10
. Dimensions of the beam are L = 1600 mm, a = 80 mm and t = 3 mm. Modulus of
elasticity is E = 2,1⋅ 10
5
MPa.
Fig. 1: Compressed beam

0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
TC 1
TC 2
Fig. 2: Time curves
9
The time is now only a parameter controlling the loading, because the problem is static.
10
Association with a time curve means that at every time the value of force is multiplied by
instantaneous value of time curve ordinate.
NONLINEAR FINITE ELEMENT ANALYSIS 32
Modelling hints:
1. The BEAM2D element group is used to model beam. As the problem is
geometrically non-linear, finer finite element mesh must be used than for a linear
problem. Geometric nonlinearities are taken into account if option 6 for the
element group BEAM2D is set to 1 (large displacement formulation).
2. It is useful to perform linear and buckling analyses before the nonlinear analysis.
For comparison, the Euler critical force is given by
2
2
) 2 ( L
I E
F
z
cr
π
· . For the
dimensions and modulus of elasticity given above is F
cr
= 185088,7 N.
3. Load should be divided into small increments and NR, or MNR iterations of
equilibrium should be used for each incremental step.
4. Finer load increments are necessary when F →F
cr
.
5. Restart from the last successfully accomplished load increment can be performed.
Commands:
C* 1. Geometry creation
VIEW,0,0,1,0
PT,1,0,0,0
CREXTR,1,1,1,Y,1600
CREXTR,2,2,1,Y,1600
C* 2. Defining material properties
MPROP,1,EX,2.1E5
C* 3. Defining elemnt group and beam section
EGROUP,1,BEAM2D,0,0,0,0,0,1,0,0
BMSECDEF,1,1,4,1,9,80,80,3,3,0,0,0,0,0
C* 4. Creating FE mesh
M_CR,1,2,1,2,5,1
NMERGE,1,12,1,1,0,0,0
NCOMPRESS,1,12
C* 5. Defining boundary conditions
C* 5.1 prescribed displacements
DPT,1,UX,0,3,2,
DPT,1,UY,0,1,1,
C* 5.2 defining force P
CURDEF,TIME,1,1,0,0,.1,1,10,1
FPT,2,FX,200,2,1
C* 5.3 force F
CURDEF,TIME,2,1,0,0,10,10,10
FPT,3,FY,-1E5,3,1
A_NONLINEAR,S,1,1,20,0.001,0,N,0,0,1E+010,0.001,0.01,0,1,0,0
NL_PLOT,1,100,1,0
C* 7. specification of load increments
TIMES,0,1.7,.1
C* 7. nonlinear analysis
R_NONLIN
C* restart from the last step
NONLINEAR FINITE ELEMENT ANALYSIS 33
RESTART,1
TIMES,1.7,1.8,0.02
R_NONLIN
C*
RESTART,1
TIMES,1.8,1.82,0.005
R_NONLIN
Results: Typical nonlinear response is obvious from the graph (Figure 3) showing beam
deflection at the middle of span versus load parameter (time).
0
5
10
15
20
25
0 0,5 1 1,5 2
TIME
U
X

[
m
m
]
Fig. 3: Deflection versus time
1.2 Thick-walled pipe subjected to internal pressure
Problem description: The long cylindrical pressure vessel is subjected to internal pressure.
Both ends are assumed fixed. Compute stress distribution during pressure test and during
normal operation. Assume small plastic deformations according to bilinear material model
if yield stress is σ
Y
= 260 MPa, modulus of elasticity is E = 2,1.10
5
MPa and tangent
modulus is E
T
= 100 MPa.
Fig. 4: Thick walled pipe
NONLINEAR FINITE ELEMENT ANALYSIS 34
0
50
100
150
200
250
300
0 0.005 0.01 0.015 0.02 0.025 0.03
strain
s
t
r
e
s
s

Fig. 5: Bilinear material mode
Time curve reflects load history. During test is internal pressure increased to
1,2 × nominal value, which is 120 MPa.
Fig. 6: Load history – time curve
Modelling hints:
1. Due to axial symmetry, only a part of the cylinder longitudinal section may be
modelled.
2. 8 node PLANE2D elements are used with option 5 set to “Von Mises kinematic
hardening”.
3. Load incrementation with NR iterations should be used.
4. Restart is used to change load (time) increment.
NONLINEAR FINITE ELEMENT ANALYSIS 35
Fig. 7: FE mesh and boundary conditions
Commands:
C* GEOMETRY CREATION
VIEW,0,0,1,0
PT,1,45,0,0
CREXTR,1,1,1,X,35
SFEXTR,1,1,1,Y,10
C* DEFINING ELEMENT GROUP
EGROUP,1,PLANE2D,0,2,1,0,2,0,0,0
C* DEFINING MATERIAL PROPERTIES
MPROP,1,EX,2.1E5
MPROP,1,NUXY,.3
MPROP,1,ETAN,100
MPROP,1,SIGYLD,260
C* MESHING
M_SF,1,1,1,8,8,2,1,1
C* DEFINING BOUNDARY CONDITIONS
DCR,2,UY,0,1,1,
CURDEF,TIME,1,1,0,0,1,1.2,1.1,1.2,2,0,3,1
PCR,3,120,3,1,120,4
NL_CONTROL,0,1
NL_PRINT,0,0,0,0,0,0,0
NL_PLOT,1,50,1,0
C* TEST SIMULATON
TIMES,0,1,.2
R_NONLIN
C*
RESTART,1
TIMES,1,1.1,.1
R_NONLIN
TIMES,1.1,2,0.09
R_NONLIN
C*
TIMES,2,3,0.2
R_NONLIN
C*
NONLINEAR FINITE ELEMENT ANALYSIS 36
Results: Note that
1. Linear analysis gives non-realistic values of stress (maximal von Mises stress is about
304 MPa).
3. After test, residual stress is present in the cylinder. This redistribution of stress is
useful and desired.
4. Due to stress redistribution, only stress within elastic range is present during nominal
1.3 Deformation of a thin-walled tank
Problem description: Thin-walled pressure vessel – tank is subjected to internal pressure.
Compute stress distribution according to linear and nonlinear static analysis. In the
nonlinear analysis both geometrical and material nonlinearities should be taken into
account. Material is elastoplastic, the same as in previous example (Figure 5). Nominal
value of internal pressure is 0,6 MPa. Symmetric half of the tank is shown in Figure 8.
Diameter of the manhole cover (thickness of 25 mm) is 600 mm.
Fig. 8: Thin-walled tank
NONLINEAR FINITE ELEMENT ANALYSIS 37
Modelling hints:
1. SHELL4T elements with options for large displacement, small plastic strain,
von Mises yield criterion and kinematic hardening are used.
2. Due to symmetry, only one eighth of the tank may be modelled.
3. Stresses are calculated in element co-ordinate systems. To receive compatible co-
ordinate systems curves are reoriented.
4. Element co-ordinate system of SHELL4T element is defined according to Figure 9.
The element x-axis goes from the first node to the second. The y-axis lies in the plane
defined by the first three nodes perpendicular to the x-axis toward the fourth node. The
z-axis completes a right-hand Cartesian system. When meshing surfaces, element x-
axes are parallel to the first surface curve marked by asterisks in Figure 10. FE mesh is
shown in Figure 11.
5. The automatic time stepping and restart options are used for nonlinear FE analysis.
Figure 9: Element co-ordinate system (ECS).
1
2
4
3
x
y
z
TOP FACE
BOTTOM FACE
Fig. 9: Element co-ordinate system (ECS)

x

x
-
a
x
i
s
z

x
-
a
x
i
s
y

x
-
a
x
i
s
Figure 10: Element co-ordinate system.
Fig. 10: Orientation of elements
NONLINEAR FINITE ELEMENT ANALYSIS 38
Fig. 11:
Finite
element
mesh
Commands:
C*
C* 1. DEFINING GEOMETRY
C*
C* 1.1. creating points and curves
C*
VIEW,0,0,1,0
PT,1,0,0,0
PT,2,2900,0,0
PLANE,Z,0,1
CRPCIRCLE,1,2,1,2900,-80,1
PT,4,0,1700,0
C*
CREXTR,4,4,1,X,2450
C* 1.1.1 intersecting curves
CRINTCC,1,2,2,1,2,5E-005
C* 1.1.2 deleting useless curves
CRDEL,4,2,2
C* 1.1.3 creating fillet etc.
CRFILLET,4,1,3,400,1,0,1E-006
PT,10,-20,300,0
CREXTR,10,10,1,X,100
CRINTCC,1,5,5,1,0,5E-005
CRDEL,1,1,1
CRDEL,5,5,1
CREXTR,12,12,1,Y,-300
C* 1.1.4 changing orientation of curves 3 and 7 in order to receive desired
C* ECS orientation
CRREPAR,7,3,4
VIEW,-1,2,3,0
CRCOMPRESS,1,7
C*
C* 1.2 creating surfaces
C*
NONLINEAR FINITE ELEMENT ANALYSIS 39
SFSWEEP,1,4,1,X,90,2
C*
C* 2. FE MESH DEFINITION
C* 2.1 defining element type and real constant sets
EGROUP,1,SHELL4T,1,0,0,1,2,1,0,0
RCONST,1,1,1,6,6,0,0,0,0,0
RCONST,1,2,1,6,25,0,0,0,0,0
C* 2.2 defining material properties
MPROP,1,EX,2.1E5,NUXY,.3,SIGYLD,260
MPROP,1,ETAN,100
C* 2.3 meshing
ACTSET,RC,1
M_SF,5,6,1,4,10,6,1.8,1
VIEW,-1.5,2,2,0
ACTSET,RC,2
MA_NUSF,7,8,1,4,4,6,0
MASFCH,7,8,1,Q,4,1,0.4,1
ACTSET,RC,1
M_SF,3,4,1,4,3,6,1,1
M_SF,1,2,1,4,10,6,2,1
NMERGE,1,400,1,1,0,0,0
NCOMPRESS,1,400
NCOMPRESS,1,330
ECOMPRESS,1,315
C* 2.4 checking real constants assignment
ACTECLR,1,RC,1
EPLOT,1,ELMAX,1
ACTECLR,0
C*
C* 3. DEFINING BOUNDARY CONDITIONS
C*
C* 3.1 displacements
SELPIC,CR,20,16,12,6,0
DCR,1,SY,0,CRMAX,1,
INITSEL,CR,1,1
SELPIC,CR,4,3,2,1,0
DCR,1,SZ,0,CRMAX,1,
INITSEL,ALL,1,1
DCR,8,SX,0,10,2,
CLS,1
C* 3.2 pressure
CURDEF,TIME,1,1,0,0,10,10
PSF,1,0.6,8,1,0.6,0.6,4
C*
C* 4. ANALYSIS
C* 4.1 preliminary linear static analysis
C* R_STAT
C*
NL_AUTOSTEP,1,.001,.1,5
NL_PLOT,1,20,1,0
NL_PRINT,0,0,0,0,0,0,0
A_NONLINEAR,S,1,1,20,0.001,0,N,0,1,1E+010,0.001,0.01,0,1,0,0
C* 4.3 running nonlinear analysis
NONLINEAR FINITE ELEMENT ANALYSIS 40
TIMES,0,1,.1
R_NONLIN
C*
C* 4.4 continuing in NL analysis
RESTART,1
TIMES,1,1.5,.1
R_NONLIN
C*
C* 5. POSTPROCESSING
C* plotting stress, displacements and xy graphs.
C*
Results:
1. There are large differences between results of linear and nonlinear analysis. In linear
analysis the von Mises stress is above the yield stress 260 MPa. Difference between
linear and nonlinear analysis results is a consequence of different solution methods.
While in the linear analysis is stress considered as elastic and geometry of the tank is
considered unchanged, in the nonlinear analysis changes of the tank shape are taken
into account together with material nonlinearities.
2. Onset of plastic deformation is obvious from the graph in Fig. 12.
0
40
80
120
160
200
240
280
0 0.3 0.6 0.9 1.2 1.5
time
σ

[
M
P
a
]
Fig. 12: Course of equivalent von Mises stress in node 177