Chap-3 FEA for Nonlinear Elastic Problems.pptx

1
Prof. Samirsinh Parmar
Asst. Professor, Dept. of Civil Engg.
Dharmasinh Desai University, Nadiad, Gujarat, INDIA
Mail: samirddu@gmail.com
CHAPTER- 3

2
Table of Contents
• 3.2. Stress and Strain Measures in Large Deformation
• 3.3. Nonlinear Elastic Analysis
• 3.4. Critical Load Analysis
• 3.5. Hyperelastic Materials
• 3.6. Finite Element Formulation for Nonlinear Elasticity
• 3.7. MATLAB Code for Hyperelastic Material Model
• 3.8. Nonlinear Elastic Analysis Using Commercial Finite
Element Programs
• 3.9. Fitting Hyperelastic Material Parameters from Test
Data
• 3.9. Summary
• 3.10.Exercises

3
Stress and Strain Measures
Section 3.2

4
Goals – Stress & Strain Measures
• Definition of a nonlinear elastic problem
• Understand the deformation gradient?
• What are Lagrangian and Eulerian strains?
• What is polar decomposition and how to do it?
• How to express the deformation of an area and volume
• What are Piola-Kirchhoff and Cauchy stresses?

5
Mild vs. Rough Nonlinearity
• Mild Nonlinear Problems (Chap 3)
– Continuous, history-independent nonlinear relations between
stress and strain
– Nonlinear elasticity, Geometric nonlinearity, and deformation-
dependent loads
• Rough Nonlinear Problems (Chap 4 & 5)
– Equality and/or inequality constraints in constitutive relations
– History-dependent nonlinear relations between stress and strain
– Elastoplasticity and contact problems

6
What Is a Nonlinear Elastic Problem?
• Elastic (same for linear and nonlinear problems)
– Stress-strain relation is elastic
– Deformation disappears when the applied load is removed
– Deformation is history-independent
– Potential energy exists (function of deformation)
• Nonlinear
– Stress-strain relation is nonlinear
(D is not constant or do not exist)
– Deformation is large
• Examples
– Rubber material
– Bending of a long slender member
(small strain, large displacement)

7
Reference Frame of Stress and Strain
• Force and displacement (vector) are independent of the
configuration frame in which they are defined (Reference
Frame Indifference)
• Stress and strain (tensor) depend on the configuration
• Lagrangian or Material Stress/Strain: when the
reference frame is undeformed configuration
• Eulerian or Spatial Stress/Strain: when the reference
frame is deformed configuration
• Question: What is the reference frame in linear
problems?

8
Deformation and Mapping
• Initial domain W0 is deformed to Wx
– We can think of this as a mapping from W0 to Wx
• X: material point in W0 x: material point in Wx
• Material point P in W0 is deformed to Q in Wx
 
x X u
displacement
   
( ,t) ( ,t)
x X X u X
W0
Wx
X x
u
P
Q

1
, :

  One-to-one mapping
Continuously differentiable

9
Deformation Gradient
• Infinitesimal length dX in W0 deforms to dx in Wx
• Remember that the mapping is continuously differentiable
• Deformation gradient:
– gradient of mapping 
– Second-order tensor, Depend on both W0 and Wx
– Due to one-to-one mapping:
– F includes both deformation and rigid-body rotation
W0
Wx
u dx
dX
P
Q
P'
Q'
d d d d

  

x
x X x F X
X
i
ij
j
x
F
X


 0

    

u
F 1 1 u
X
det J 0.
 
F
ij
0 x
[ ],
,
 
 
   
 
1
X x
1
d d


X F x

10
Example – Uniform Extension
• Uniform extension of a cube in all three directions
• Continuity requirement: Why?
• : uniform expansion (dilatation) or contraction
• Volume change
– Initial volume:
– Deformed volume:
1 1 1 2 2 2 3 3 3
x X , x X , x X
     
1
2
3
0 0
0 0
0 0

 
 
 
 
 

 
F
i 0
 
1 2 3
    
0 1 2 3
dV dX dX dX

x 1 2 3 1 2 3 1 2 3 1 2 3 0
dV dx dx dx dX dX dX dV
        

11
Green-Lagrange Strain
• Why different strains?
• Length change:
• Right Cauchy-Green Deformation Tensor
• Green-Lagrange Strain Tensor
2 2 T T
T T T
T T
d d d d d d
d d d d
d ( )d
  
 
 
x X x x X X
X F F X X X
X F F 1 X
Ratio of length change
T

C F F
1
( )
2
 
E C 1
dX
dx
The effect of rotation is eliminated
To match with infinitesimal strain

12
Green-Lagrange Strain cont.
• Properties:
– E is symmetric: ET = E
– No deformation: F = 1, E = 0
– When ,
– E = 0 for a rigid-body motion, but
 
T T
T T
1
0 0 0 0
2
1
2
 
   
  
 
   
 
      
u u u u
E
X X X X
u u u u
j
i
ij
j i
u
u
1
2 X X

 

  
 
 
 
 
Displacement gradient
Higher-order term
0 1
 
u  
T
0 0
1
ε
2
    
E u u

ε 0

13
Example – Rigid-Body Rotation
• Rigid-body rotation
• Approach 1: using deformation gradient
   
   

1 1 2
2 1 2
3 3
x X cos X sin
x X sin X cos
x X
  
 
 
  
 
 
 
cos sin 0
sin cos 0
0 0 1
F
 
 
  
 
 
T
1 0 0
0 1 0
0 0 1
F F
  
T
1
2
( )
E F F 1 0
Green-Lagrange strain removes rigid-body rotation from deformation


14
Example – Rigid-Body Rotation cont.
• Approach 2: using displacement gradient
      
      
  
1 1 1 1 2
2 2 2 1 2
3 3 3
u x X X (cos 1) X sin
u x X X sin X (cos 1)
u x X 0
   
 
 
    
 
 
 
0
cos 1 sin 0
sin cos 1 0
0 0 0
u
 
 
 
    
 
 
 
T
0 0
2(1 cos ) 0 0
0 2(1 cos ) 0
0 0 0
u u
       
T T
1
0 0 0 0
2
( )
E u u u u 0

15
Example – Rigid-Body Rotation cont.
• What happens to engineering strain?
      
      
  
1 1 1 1 2
2 2 2 1 2
3 3 3
u x X X (cos 1) X sin
u x X X sin X (cos 1)
u x X 0
 
 
 
  
 
 
 

cos 1 0 0
0 cos 1 0
0 0 0
Engineering strain is unable to take care of rigid-body rotation

16
Eulerian (Almansi) Strain Tensor
• Length change:
• Left Cauchy-Green Deformation Tensor
• Eulerian (Almansi) Strain Tensor
 
 

  
 
 
 
2 2 T T
T T T 1
T T 1
T 1
d d d d d d
d d d d
d ( )d
d ( )d
x X x x X X
x x x F F x
x 1 F F x
x 1 b x
 T
b FF

  1
1
( )
2
e 1 b
Reference is deformed (current) configuration
b–1: Finger tensor

17
Eulerian Strain Tensor cont.
• Properties
– Symmetric
– Approach engineering strain when
– In terms of displacement gradient
• Relation between E and e
 
 
   
  
 
   
 
      
T T
T T
x x x x
1
2
1
2
u u u u
e
x x x x
u u u u

 

x
x
Spatial gradient
 T
E F eF



1
u
x

18
Example – Lagrangian Strain
• Calculate F and E for deformation in the figure
• Mapping relation in W0
• Mapping relation in Wx
1.5
1.0
X
Y
Undeformed
element
Deformed element
2.0
0.7



  




  




4
I I
I 1
4
I I
I 1
3
X N (s,t)X (s 1)
4
1
Y N (s,t)Y (t 1)
2



  




  




4
I I
I 1
4
I I
I 1
x(s,t) N (s,t)x 0.35(1 t)
y(s,t) N (s,t)y s 1

19
Example – Lagrangian Strain cont.
• Deformation gradient
• Green-Lagrange Strain
W0
Wx
u dx
dX
P
Q
P'
Q'
( , )
s t
X
( , )
s t
x
Reference
domain (s, t)
  
 
  

   
    
   

 
  
 
0 .35 4 / 3 0
1 0 0 2
0 0.7
4 / 3 0
x x s
F
X s X
 
    

 
T 0.389 0
1
( )
2 0 0.255
E F F 1

20
Example – Lagrangian Strain cont.
• Almansi Strain
• Engineering Strain
 
    
 
T 0.49 0
0 1.78
b F F
 
 
 
    
 
1
1
2
0.52 0
0 0.22
e 1 b
 
 
     

 
0
1 0.7
1.33 1
u F 1
 

 
      

 
 T
1
0 0
2
1 0.32
0.32 1
u u
Which strain is consistent with actual deformation?

21
Example – Uniaxial Tension
• Uniaxial tension of incompressible material (1 =   1)
• From incompressibility
• Deformation gradient and deformation tensor
• G-L Strain

          1/2
1 2 3 2 3
1



 
 
 
 
 

 
1/2
1/2
0 0
0 0
0 0
F


 

 
 
 
 

 
2
1
1
0 0
0 0
0 0
C


 
 
 
  
 
 
 
 
2
1
1
1 0 0
1
0 1 0
2
0 0 1
E
 
 
 
1 1 1
2 2 2
3 3 3
x X
x X
x X

22
• Almansi Strain (b = C)
• Engineering Strain
• Difference

 
 
 
  
 
 
 
 
2
1 0 0
1
0 1 0
2
0 0 1
e


 

 
 
 
 

 
2
1
0 0
0 0
0 0
b


 
 
 
  
 
 
 
 
 1/2
1/2
1 0 0
0 1 0
0 0 1

         
2 2
11 11 11
1 1
E ( 1) e (1 ) 1
2 2

23
Polar Decomposition
• Want to separate deformation from rigid-body rotation
• Similar to principal directions of strain
• Unique decomposition of deformation gradient
– Q: orthogonal tensor (rigid-body rotation)
– U, V: right- and left-stretch tensor (symmetric)
• U and V have the same eigenvalues (principal stretches),
but different eigenvectors
 
F QU VQ

24
Polar Decomposition cont.
• Eigenvectors of U: E1, E2, E3
• Eigenvectors of V: e1, e2, e3
• Eigenvalues of U and V:1, 2, 3
Q
Q
V
U
E1 E2
E3
λ1E1
λ2E2
λ3E3
e1
e2
e3
λ1e1
λ2e2
λ3e3

F QU

F VQ
  
  
d d
d
x Q U X
V Q X

25
• Relation between U and C
– U and C have the same eigenvectors.
– Eigenvalue of U is the square root of that of C
• How to calculate U from C?
• Let eigenvectors of C be
• Then, where
 
2
U C U C
1 2 3
[ ]
 E E E

T
 C
  
 

 
  
 
 

 
2
1
2
2
2
3
0 0
0 0
0 0
Deformation tensor in
principal directions

26
• And
• General Deformation
1. Stretch in principal directions
2. Rigid-body rotation
3. Rigid-body translation
   
d d d
x F X b QU X b
   T
U

 
 
  
 
 

 
1
2
3
0 0
0 0
0 0

  

3
i i i
i 1
U E E

  

3
i i i
i 1
V e e

 

3
i i
i 1
Q e E

  

3
i i i
i 1
F e E

  

3
2
i i i
i 1
C E E

  

3
2
i i i
i 1
b e e
Useful formulas

27
Generalized Lagrangian Strain
• G-L strain is a special case of general form of Lagrangian
strain tensors (Hill, 1968)
 
 
2m
m
1
2m
E U 1

28
Example – Polar Decomposition
• Simple shear problem
• Deformation tensor
• Find eigenvalues and eigenvectors of C
 




 

1 1 2
2 2
3 3
x X kX
x X
x X

2
k
3
X1, x1
X2, x2
 
  
 
1 k
0 1
F
 
 
 
  
 
  
   
2
3
T
2 7
2
3
3
1
1 k
k k 1
C F F
   
   
  
1 2
3 3
1 1
1 2
2 2 2 2
3, 1 3
,
E E
X1
X2
E2
E1
60o

29
Example – Polar Decomposition cont.
• In E1 – E2 coordinates
• Principal Direction Matrix
• Deformation tensor in principal directions
• Stretch tensor
 
    
 

3 0
0 1 3
C
1 2
1 2 3 2
[ ]
3 2 1 2
 

    
 
 
E E
  
  
T
C
3 0
0 1 3
 
  
 
 

 
     
 
 
  T 3 2 1 2
1 2 5 2 3
U

30
• How U deforms a square?
• Rotational Tensor
– 30o clockwise rotation
1 3 2 1 2
1 2 3 2

 
    

 
 
Q F U
1 2
1 0
3 2
,
0 1
1 2 5 2 3
   
   
   
       
   
   
U U
X1, x1
X2, x2
30o
X1, x1
X2, x2
30o
1 2
1 1.15
3 2
,
0 1
1 2 5 2 3
   
   
   
       
   
   
Q Q
5 3 6 1 2
1 2 3 2
T
 
    
 
 
V F Q

31
• A straight line will deform to
• Consider a diagonal line: q = 45o
• Consider a circle
 q
2 1
X X tan
 
q
  
   q
  
1 1 2 2 2
2 1 2
1
1 2
tan
X x kx , X x
x (x kx )tan
x k x
     

2
1
x 1
tan 24.9
x 1 k
 
  
   
2 2 2
1 2
2 2 2
1 2 2
2 2 2 2
1 1 2 2
X X r
(x kx ) x r
x 2kx x (1 k )x r Equation of ellipse
X1, x1
X2, x2
25o
X1, x1
X2, x2

32
Deformation of a Volume
• Infinitesimal volume by three vectors
– Undeformed:
– Deformed:
   
1 2 3 1 2 3
0 rst r s t
dV d (d d ) e dX dX dX
X X X
   
1 2 3 1 2 3
x ijk i j k
dV d (d d ) e dx dx dx
x x x


  
  
 
   
  
  
   
 
 


  


1 2 3
x ijk i j k
j
1 2 3
k
i
ijk r s t
r s t
j 1 2 3
k
i
ijk r s t
r s t
1 2 3
rst r s t
0
dV e dx dx dx
x x
x
e dX dX dX
X X X
x x
x
e dX dX dX
X X X
e J dX dX dX
JdV
From Continuum Mechanics

ijk ir js kt rst
e a a a e deta
    
1 2 3
J detF
dX1
dX3
dX2
dx1
dx3
dx2

33
Deformation of a Volume cont.
• Volume change
• Volumetric Strain
• Incompressible condition: J = 1
• Transformation of integral domain

 
x 0
0
dV dV
J 1
dV

x 0
dV JdV
W W
W  W
  0
d d
x
f fJ

34
Example - Uniaxial Deformation of a Beam
• Initial dimension of L0×h0×h0 deforms to L×h×h
• Constant volume
   
   
   
1 1 1 1 0
2 2 2 2 0
3 3 3 3 0
x X L / L
x X h / h
x X h / h

 
 
 
 
 

 
1
2
3
0 0
0 0
0 0
F
    
 
 
 
 
1 2 3
2
0 0 0 0
J det
L h LA
L h L A
F
   
0 0
0 0
L L
J 1 h h A A
L L
L0
h0
h0
L
h
h

35
Deformation of an Area
• Relationship between dS0 and dSx
dSx
dx1
n
Sx
x
dS0
dX1
N
S0
X
F(X)
dX2 dx2
Undeformed Deformed
  
  
1 2 1 2
0 i 0 ijk j k
1 2 1 2
x r x rst s t
dS d d NdS e dX dX
dS d d n dS e dx dx
N X X
n x x
 

 
j 1 2
k
i 0 ijk s t
s t
X X
NdS e dx dx
x x
 
 

   
j 1 2
k
i i
i 0 ijk s t
r r s t
X X
X X
NdS e dx dx
x x x x



i
r
X
x

36
Deformation of an Area cont..
• Results from Continuum Mechanics
• Use the second relation:

 
 
 
   
1
j 1 2 1 2
k
i i
i 0 ijk s t rst s t
r r s t
X X
X X
NdS e dx dx e dx dx
x x x x
F

  

  
 


  
r s t
ijk rst
i j k
j
1 k
i
rst ijk
r s t
x x x
e e
X X X
X X
X
e e .
x x x
F
F
r x
n dS

 
T
x 0
dS J dS
n F N




  

T
T
T
/ /
F N
n F N n
F N

 T
x 0
dS J ( ) ( ) dS
F x N X

37
Stress Measures
• Stress and strain (tensor) depend on the configuration
• Cauchy (True) Stress: Force acts on the deformed config.
– Stress vector at Wx:
– Cauchy stress refers to the current deformed configuration as a
reference for both area and force (true stress)
P
N
∆S0
P n
∆Sx
∆f
Undeformed configuration Deformed configuration
 

 

x
S 0 x
lim
S
f
t n

Cauchy Stress, sym

38
Stress Measures cont.
• The same force, but different area (undeformed area)
– P refers to the force in the deformed configuration and the area
in the undeformed configuration
• Make both force and area to refer to undeformed config.
 

 

0
T
S 0 0
lim
S
f
T P N
First Piola-Kirchhoff Stress
Not symmetric
  T
x 0
d dS dS
f n P N
 
 
T
x 0
dS J dS
n F N

 1
J
P F 

 
T T
0 0
d (J dS ) dS
f F N P N

: Relation between  and P

39
• Unsymmetric property of P makes it difficult to use
– Remember we used the symmetric property of stress & strain
several times in linear problems
• Make P symmetric by multiplying with F-T
– Just convenient mathematical quantities
• Further simplification is possible by handling J differently
  
    
T 1 T
J
S P F F F

Second Piola-Kirchhoff Stress, symmetric
   T
1
J
F S F

    T
J F S F
 
Kirchhoff Stress, symmetric

40
• Example
• Observation
– For linear problems (small deformation):
– For linear problems (small deformation):
– S and E are conjugate in energy
– S and E are invariant in rigid-body motion
W W W
W  W  W
  
x 0 0
x 0 0
: d : Jd : d
     
Integration can be done in W0
  
P S
 
 
E e


41
• Cauchy (true) stress: , 22 = 33 = 12 = 23 = 13 = 0
• First P-K stress
• Second P-K stress
L0
h0
h0
L
h
h
F
 
11
F
A

 

 

 
  
 
 

 
1
1
1 1
2
1
3
0 0
0 0 , J 1
0 0
F
 
      


2
1 T
11 11 2 2 2
0 1
1 0 0
F 1 F A FA F
S (J )
A A A
A A
F F


  

1
11 11
1 0 0
F 1 F A F
P (J
A A A A
F  

42
Summary
• Nonlinear elastic problems use different measures of
stress and strain due to changes in the reference frame
• Lagrangian strain is independent of rigid-body rotation,
but engineering strain is not
• Any deformation can be uniquely decomposed into rigid-
body rotation and stretch
• The determinant of deformation gradient is related to the
volume change, while the deformation gradient and
surface normal are related to the area change
• Four different stress measures are defined based on the
reference frame.
• All stress and strain measures are identical when the
deformation is infinitesimal

43
Nonlinear Elastic Analysis
Section 3.3

44
Goals
• Understanding the principle of minimum potential energy
– Understand the concept of variation
• Understanding St. Venant-Kirchhoff material
• How to obtain the governing equation for nonlinear elastic
problem
• What is the total Lagrangian formulation?
• What is the updated Lagrangian formulation?
• Understanding the linearization process

45
Numerical Methods for Nonlinear Elastic Problem
• We will obtain the variational equation using the principle
of minimum potential energy
– Only possible for elastic materials (potential exists)
• The N-R method will be used (need Jacobian matrix)
• Total Lagrangian (material) formulation uses the
undeformed configuration as a reference, while the
updated Lagrangian (spatial) uses the current
configuration as a reference
• The total and updated Lagrangian formulations are
mathematically equivalent but have different aspects in
computation

46
Total Lagrangian Formulation
• Using incremental force method and N-R method
– Total No. of load steps (N), current load step (n)
• Assume that the solution has converged up to tn
• Want to find the equilibrium state at tn+1
0W
nW
X x
nu
∆u
Undeformed configuration
(known)
Last converged configuration
(known)
Current configuration
(unknown)
0P
nP

  
n 1 n n
f f f

47
Total Lagrangian Formulation cont.
• In TL, the undeformed configuration is the reference
• 2nd P-K stress (S) and G-L strain (E) are the natural choice
• In elastic material, strain energy density W exists, such
that
• We need to express W in terms of E



W
stress
strain

48
Strain Energy Density and Stress Measures
• By differentiating strain energy density with respect to
proper strains, we can obtain stresses
• When W(E) is given
• When W(F) is given
• It is difficult to have W() because  depends on rigid-
body rotation. Instead, we will use invariants in Section
3.5



W( )
E
S
E
   
     
   
T
W W W
:
E
F F S P
F E F E
Second P-K stress
First P-K stress

49
St. Venant-Kirchhoff Material
• Strain energy density for St. Venant-Kirchhoff material
• Fourth-order constitutive tensor (isotropic material)
– Lame’s constants:
– Identity tensor (2nd order):
– Identity tensor (4th order):
– Tensor product:
 1
2
W( ) : :
E E D E Contraction operator:  ij ij
: a b
a b
    
2
D 1 1 I

   
     
E E
(1 )(1 2 ) 2(1 )
 ij
[ ]
1
     
1
ijkl ik jl il jk
2
I ( )
 
    
ii 11 22 33
: , 2nd-order sym.
: tr( ) a a a a
I a a a
1 a a
  ij kl
a a (4th-order)
a a

50
St. Venant-Kirchhoff Material cont.
• Stress calculation
– differentiate strain energy density
– Limited to small strain but large rotation
– Rigid-body rotation is removed and only the stretch tensor
contributes to the strain
– Can show

     

W( )
: tr( ) 2
E
S D E E 1 E
E
     
T T T 2
1 1 1
2 2 2
( ) ( ) ( )
E F F 1 U Q Qu 1 U 1
 
 
 
W W
2
S
E C
Deformation tensor

51
Example
• E = 30,000 and  = 0.3
• G-L strain:
• Lame’s constants:
• 2nd P-K Stress:
1.5
1.0
X
Y
Undeformed
element
Deformed element
2.0
0.7
 
  

 
0.389 0
0 0.255
E

     
     
E E
17,308 11,538
(1 )(1 2 ) 2(1 )
   
        
   

   
 
  

 
1 0 .389 0
tr( ) 2 (.389 .255) 2
0 1 0 .255
11,296 0
0 3,565
S E 1 E

 
   
 
T 1,872 0
1
J 0 21,516
FSF


52
Example – Simple Shear Problem
• Deformation map
• Material properties
• 2nd P-K stress
   
1 1 2 2 2 3 3
x X kX , x X , x X
X1, x1
X2, x2
 
    
 
T
2
0 k
1 1
( )
2 2 k k
E F F 1
 
  
 
1 k
0 1
F

     
     
E E
40MPa 40MPa
(1 )(1 2 ) 2(1 )
 
      
 
 
2
2
k 2k
tr( ) 2 20 MPa
2k 3k
S E 1 E
 
 
   

 
 
2 4 3
T
3 2
5k 3k 2k 3k
1
20 MPa
J 2k 3k 3k
FSF

-0.4 -0.2 0.0 0.2 0.4
20
10
0
-10
-20
Cauchy
2nd P-K
Shear parameter k
Shear
stress

53
Boundary Conditions
• Boundary Conditions
• Solution space (set)
• Kinematically admissible space
 
 
h
T s
, on
, on
u g
t P N
You can’t use S
Essential (displacement) boundary
Natural (traction) boundary
 

  W 
V h
1 3
[H ( )] ,
u u u g
 

  W 
Z h
1 3
[H ( )] , 0
u u u

54
Variational Formulation
• We want to minimize the potential energy (equilibrium)
Pint: stored internal energy
Pext: potential energy of applied loads
• Want to find u that minimizes the potential energy
– Perturb u in the direction of ū proportional to 
– If u minimizes the potential, P(u) must be smaller than P(u) for all
possible ū
W W 
P  P  P
 W  W  
   s
0 0 o
int ext
T b T
( ) ( ) ( )
W( )d d d
u u u
E u f u t
   
u u u

55
Variational Formulation cont.
• Variation of Potential Energy (Directional Derivative)
– P depends on u only, but P depends on both u and ū
– Minimum potential energy happens when its variation becomes
zero for every possible ū
– One-dimensional example

P  P  
 0
d
( , ) ( )
d
u u u u We will use “over-bar” for variation
P(u)
ū
u
ū
At minimum, all directional
derivatives are zero

56
Example – Linear Spring
• Potential energy:
• Perturbation:
• Differentiation:
• Evaluate at original state:
P    
2
1
2
(u) k u f u
P          
2
1
2
(u u) k (u u) f (u u)
P         
 
 

d
(u u) k (u u) u f u
d

P        
 
 
 0
d
(u u) k u u f u 0
d
k
f
u
Variation is similar to differentiation !!!

57
• Variational Equation
– From the definition of stress
– Note: load term is similar to linear problems
– Nonlinearity in the strain energy term
• Need to write LHS in terms of u and ū
W W 

P  W  W   

   s
0 0 o
T b T
W( )
( , ) : d d d 0
E
u u E u f u t
E
for all ū
W W 
W  W  
   s
0 0 o
T b T
: d d d
S E u f u t
Variational equation in TL formulation

58
• How to express strain variation
 
        
T T
1 1
0 0 0 0
2 2
( ) ( )
E u C 1 u u u u
 
 
 

  

         
       
   
0
T T T
1
0 0 0 0 0 0
2
T T
1
0 0 0 0
2
T T
1
0 0
2
d
( , ) ( )
d
( ) ( )
E u u E u u
u u u u u u
1 u u u 1 u
F u u F
  T
0
( , ) sym( )
E u u u F
Note: E(u) is nonlinear, but is linear
( , )
E u u

59
• Linear in terms of strain if St. Venant-Kirchhoff material
is used
• Also linear in terms of ū
• Nonlinear in terms of u because displacement-strain
relation is nonlinear
W W 
W  W  
   s
0 0 o
T b T
: d d d
S E u f u t for all ū
a( , )
u u ( )
u
Energy form Load form
  
a( , ) ( ),
u u u u Z

60
Linearization
• We are still in continuum domain (not discretized yet)
• Residual
• We want to linearize R(u) in the direction of u
– First, assume that u is perturbed in the direction of u using a
variable . Then linearization becomes
– R(u) is nonlinear w.r.t. u, but L[R(u)] is linear w.r.t. u
– Iteration k did not converged, and we want to make the residual at
iteration k+1 zero
 
R( ) a( , ) ( )
u u u u

   
 
  
 
 
 
T
0
R( ) R
L[R( )]
u u
u u
u
  

   
 

 
T
k
k 1 k k
R( )
R( ) R( ) 0
u
u u u
u

61
Linearization cont.
• This is N-R method (see Chapter 2)
• Update state
• We know how to calculate R(uk), but how about ?
– Only linearization of energy form will be required
– We will address displacement-dependent load later
 

  
 

 
T
k
k k
R( )
R( )
u
u u
u

 
  
 
k 1 k k
k 1 k 1
u u u
x X u
 

 

 
k
R( )
u
u
 
 
 
[R( )] [a( , ) ( )]
u u u u
u u

62
Linearization cont.
• Linearization of energy form
– Note that the domain is fixed (undeformed reference)
– Need to express in terms of displacement increment u
• Stress increment (St. Venant-Kirchhoff material)
• Strain increment (Green-Lagrange strain)
W W
 
 W     W
 
 
0 0
L[a( , )] L : d [ : : ]d
u u S E S E S E

    

: :
S
S E D E
E
    
T T
1
2
( )
E F F F F
   
   
        
   
  
   
0
( )
x X u u
F u
X X X

63
Linearization cont.
• Strain increment
• Inc. strain variation
• Linearized energy form
– Implicitly depends on u, but bilinear w.r.t. u and ū
– First term: tangent stiffness
– Second term: initial stiffness
    
     
  
T T
1
2
T T
1
0 0
2
T
0
( )
( )
sym( )
E F F F F
u F F u
u F
!!! Linear w.r.t. u
   
  
   
T
0
T
0
T
0 0
[sym( )]
sym( )
sym( )
E u F
u F
u u !!! Linear w.r.t. u
W
    W  
0
*
L[a( , )] [ : : : ]d a ( ; , )
u u E D E S E u u u

64
• N-R Iteration with Incremental Force
– Let tn be the current load step and (k+1) be the current iteration
– Then, the N-R iteration can be done by
– Update the total displacement
• In discrete form
• What are and ?
Linearization cont.
    
* n k k n k
a ( ; , ) ( ) a( , ),
u u u u u u u Z

  
n k 1 n k k
u u u
 
T n k k T n k
T
{ } [ ]{ } { } { }
d K d d R
n k
T
[ ]
K n k
{ }
R

65
Example – Uniaxial Bar
• Kinematics
• Strain variation
• Strain energy density and stress
• Energy and load forms
• Variational equation
L0=1m
1 2
F = 100N
x
 
2 2
du du
u , u
dX dX
 
   
 
 
2
2
11 2 2
du 1 du 1
E u (u )
dX 2 dX 2
  2
1
11 11
2
W(E ) E (E )
  
    
 
  
2
11 11 2 2
11
W 1
S E E E u (u )
E 2
   
11 2 2
du du du
E u (1 u )
dX dX dX
  

0
L
11 11 11 0 2 2
0
a(u,u) S E AdX S AL (1 u )u  2
(u) u F
 
    
2 11 0 2 2
R u S AL (1 u ) F 0, u

66
• Linearization
• N-R iteration
     
11 11 2 2
S E E E(1 u ) u   
11 2 2
E u u
 
       
    

0
L
*
11 11 11 11
0
2
0 2 2 2 11 0 2 2
a (u; u,u) E E E S E AdX
EAL (1 u ) u u S AL u u
     
k 2 k k k k
2 11 0 2 11 2 0
[E(1 u ) S ]AL u F S (1 u )AL

  
k 1 k k
2 s 2
u u u

67
(a) with initial stiffness
Iteration u Strain Stress conv
0 0.0000 0.0000 0.0000 9.999E01
1 0.5000 0.6250 125.00 7.655E01
2 0.3478 0.4083 81.664 1.014E02
3 0.3252 0.3781 75.616 4.236E06
(b) without initial stiffness
0 0.0000 0.0000 0.0000 9.999E01
1 0.5000 0.6250 125.00 7.655E01
2 0.3056 0.3252 70.448 6.442E03
3 0.3291 0.3833 76.651 3.524E04
4 0.3238 0.3762 75.242 1.568E05
5 0.3250 0.3770 75.541 7.314E07

68
Updated Lagrangian Formulation
• The current configuration is the reference frame
– Remember it is unknown until we solve the problem
– How are we going to integrate if we don’t know integral domain?
• What stress and strain should be used?
– For stress, we can use Cauchy stress ()
– For strain, engineering strain is a pair of Cauchy stress
– But, it must be defined in the current configuration
 
 
   
 
 
 
 
T
x
1
sym( )
2
u u
u
x x


69
Variational Equation in UL
• Instead of deriving a new variational equation, we will
convert from TL equation
 
  
   
T
1 T
1
J
J
F S F
S F F


 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
   
 
 
 
   
 
 
 
 
 
 
 
 
  
T
T
T
T T 1
T T
T
T
T
T
1
2
1
2
1
2
1
2
u u
E F F
X X
u u
F F F F
X X
X u u X
F F
x X X x
u u
F F
x x
F F

    
 
 
  
 
 
 
 
T
T
1
2
E F F
u u
x x


Similarly

70
Variational Equation in UL cont.
• Energy Form
– We just showed that material and spatial forms are
mathematically equivalent
• Although they are equivalent, we use different notation:
 
W W
 W  W
   
0 0
1 T T
a( , ) : d (J ) : ( )d
u u S E F F F F
 
         
1 1
ik kl jl mi mn nj mk nl kl mn mn mn
F F F F
W W W
W  W  W
  
0 0 x
: d : Jd : d
S E    
W
 W
  
x
a( , ) : d
u u
  
a( , ) ( ),
u u u u Z What happens to load form?
Is this linear or nonlinear?

71
 
kl( , )
u u
Linearization of UL
• Linearization of will be challenging because we
don’t know the current configuration (it is function of u)
• Similar to the energy form, we can convert the linearized
energy form of TL
• Remember
• Initial stiffness term
x
a ( , )
u u
W
     W
0
* 0
a ( ; , ) [ : : : ]d
u u u E D E S E
 
 
 
   
  
 
 
   
 
 
   
  
 
 
   
 
   
 
  
 
   
 
T T
1 T
m m m m
1 1
ik kl jl
i j i j
m m m m
kl
k l k l
1
: J( ) :
2
u u u u
1
JF F
2 X X X X
u u u u
1
J
2 x x x x
u u u u
S E F F
X X X X


72
4th-order spatial
constitutive tensor
Linearization of UL cont.
• Initial stiffness term
• Tangent stiffness term
  
: J : ( , )
S E u u
      
T
x x
( , ) sym( )
u u u u

      
  
 
  
 
 
 
T T
ki kl lj ijmn pm pq qn
kl ki lj ijmn pm qn pq
( : : ) ( ) : : ( )
F F D F F
1
J F F D F F
J
E D E F F D F F
  
 
: : J : :
E D E c
where 
ijkl ir js km ln rsmn
1
c F F F F D
J

73
Spatial Constitutive Tensor
• For St. Venant-Kirchhoff material
• It is possible to show
• Observation
– D (material) is constant, but c (spatial) is not
–
              
rsmn rs mn rm sn rn sm
( ) 2 ( )
D 1 1 I D
 
    
 
ijkl ij kl ik jl il jk
1
c G G (G G G G ) .
J
 
 
: , :
S D E c

74
Linearization of UL cont.
• From equivalence, the energy form is linearized in TL and
converted to UL
• N-R Iteration
• Observations
– Two formulations are theoretically identical with different
expression
– Numerical implementation will be different
– Different constitutive relation
W
   W
    
0
L[a( , )] [ : : : ]Jd
u u c
W
    W
    
x
*
a ( ; , ) [ : : : ]d
u u u c
    
* n k k n k
a ( ; , ) ( ) a( , ),
u u u u u u u Z

75
• Kinematics
• Cauchy stress:
• Strain variation:
• Energy & load forms:
• Residual:
L0=1m
1 2
F = 100N
x
 
 
2 2
2 2
u u
du du
,
dx 1 u dx 1 u
    
11 2 2
dx
F 1 u , J 1 u
dX
    
2
11 11 11 11 2 2 2
1 1
F S F E(u u )(1 u )
J 2
 
  

T 1 2
11 11 11 11
2
u
(u) F E F
1 u
    

L
11 11 11 2
0
a(u,u) (u)Adx Au  2
(u) u F
 
    
2 11 2
R u A F 0, u

76
• Spatial constitutive relation:
• Linearization:
   3
1111 11 11 11 11 2
1
c F F F F E (1 u ) E
J
     

L 2
11 1111 11 2 2 2
0
(u)c ( u)Adx EA(1 u ) u u

    


L
11
11 11 2 2
0
2
A
( u,u)Adx u u
1 u
 
        

    


L
*
11 1111 11 11
0
2 11
2 2 2 2 2
2
a (u; u,u) (u)c ( u) ( u,u) Adx
EA(1 u ) u u Au u
1 u
0 0.0000 0.0000 0.000 9.999E01
1 0.5000 0.3333 187.500 7.655E01
2 0.3478 0.2581 110.068 1.014E02
3 0.3252 0.2454 100.206 4.236E06

77
Hyperelastic Material Model
Section 3.5

78
Goals
• Understand the definition of hyperelastic material
• Understand strain energy density function and how to use
it to obtain stress
• Understand the role of invariants in hyperelasticity
• Understand how to impose incompressibility
• Understand mixed formulation and perturbed Lagrangian
formulation
• Understand linearization process when strain energy
density is written in terms of invariants

79
What Is Hyperelasticity?
• Hyperelastic material - stress-strain relationship derives
from a strain energy density function
– Stress is a function of total strain (independent of history)
– Depending on strain energy density, different names are used,
such as Mooney-Rivlin, Ogden, Yeoh, or polynomial model
• Generally comes with incompressibility (J = 1)
– The volume preserves during large deformation
– Mixed formulation – completely incompressible hyperelasticity
– Penalty formulation - nearly incompressible hyperelasticity
• Example: rubber, biological tissues
– nonlinear elastic, isotropic, incompressible and generally
independent of strain rate
• Hypoelastic material: relation is given in terms of stress
and strain rates

80
Strain Energy Density
• We are interested in isotropic materials
– Material frame indifference: no matter what coordinate system is
chosen, the response of the material is identical
– The components of a deformation tensor depends on coord. system
– Three invariants of C is independent of coord. system
• Invariants of C
– In order to be material frame indifferent, material properties
must be expressed using invariants
– For incompressibility, I3 = 1
         
2 2 2
1 11 22 33 1 2 3
I tr( ) C C C
C
 
          
 
2 2 2 2 2 2 2 2
1
2 1 2 2 3 3 1
2
I (tr ) tr( )
C C
    
2 2 2
3 1 2 3
I detC
No deformation
I1 = 3
I2 = 3
I3 = 1

81
Strain Energy Density cont.
• Strain Energy Density Function
– Must be zero when C = 1, i.e., 1 = 2 = 3 = 1
– For incompressible material
– Ex: Neo-Hookean model
– Mooney-Rivlin model

  
   
 m n k
1 2 3 mnk 1 2 3
m n k 1
W(I ,I ,I ) A (I 3) (I 3) (I 1)

 
  
 m n
1 2 mn 1 2
m n 1
W(I ,I ) A (I 3) (I 3)
 
1 10 1
W(I ) A (I 3)
   
1 2 10 1 01 2
W(I ,I ) A (I 3) A (I 3)


10
A
2

82
Strain Energy Density cont.
• Strain Energy Density Function
– Yeoh model
– Ogden model
– When N = 1 and a1 = 1, Neo-Hookean material
– When N = 2, 1 = 2, and 2 = 2, Mooney-Rivlin material
     
2 3
1 1 10 1 20 1 30 1
W(I ) A (I 3) A (I 3) A (I 3)
 
  


         

 i i i
N
i
1 1 2 3 1 2 3
i 1 i
W ( , , ) 3

   

N
i i
i 1
1
2

83
Example – Neo-Hookean Model
• Uniaxial tension with incompressibility
• Energy density
• Nominal stress
       
1 2 3 1 /
            

2 2 2 2
10 1 10 1 2 3 10
2
W A (I 3) A ( 3) A ( 3)
 
  
        
 
 
   
   
10 2 2
W 1 1
P 2A 1
(1 )
-0.8 -0.4 0 0.4 0.8
-250
-200
-150
-100
-50
0
50
Nominal strain
Nominal
stress
Neo-Hookean
Linear elastic

84
Example – St. Venant Kirchhoff Material
• Show that St. Venant-Kirchhoff material has the following
strain energy density
• First term
• Second term

  
 
 
2 2
W( ) tr( ) tr( )
2
E E E

 

tr( )
tr( ) :
E
E 1 E 1
E
  
    
  
2
W( ) tr( ) tr( )
tr( )
E E E
S E
E E E

     

tr( )
tr( ) ( : ) ( ) :
E
E 1 1 E 1 1 E
E

        

ij ji
ik jl ji ij jk il lk lk lk
kl
E E
E E E E 2E
E

85
Example – St. Venant Kirchhoff Material cont.
• Therefore
 
   
 
    
    
 
 
2
tr( ) tr( )
tr( )
( ) : 2
( ) 2 :
E E
S E
E E
1 1 E E
1 1 I E
D

86
Nearly Incompressible Hyperelasticity
• Incompressible material
– Cannot calculate stress from strain. Why?
• Nearly incompressible material
– Many material show nearly incompressible behavior
– We can use the bulk modulus to model it
• Using I1 and I2 enough for incompressibility?
– No, I1 and I2 actually vary under hydrostatic deformation
– We will use reduced invariants: J1, J2, and J3
• Will J1 and J2 be constant under dilatation?
 
   
1/3 2/3 1/2
1 1 3 2 2 3 3 3
J I I J I I J J I

87
Locking
• What is locking
– Elements do not want to deform even if forces are applied
– Locking is one of the most common modes of failure in NL analysis
– It is very difficult to find and solutions show strange behaviors
• Types of locking
– Shear locking: shell or beam elements under transverse loading
– Volumetric locking: large elastic and plastic deformation
• Why does locking occur?
– Incompressible sphere under hydrostatic pressure
sphere
p
Volumetric strain
Pressure No unique pressure
for given displ.

88
How to solve locking problems?
• Mixed formulation (incompressibility)
– Can’t interpolate pressure from displacements
– Pressure should be considered as an independent variable
– Becomes the Lagrange multiplier method
– The stiffness matrix becomes positive semi-definite
4x1 formulation
Displacement
Pressure

89
Penalty Method
• Instead of incompressibility, the material is assumed to be nearly
incompressible
• This is closer to actual observation
• Use a large bulk modulus (penalty parameter) so that a small volume
change causes a large pressure change
• Large penalty term makes the stiffness matrix ill-conditioned
• Ill-conditioned matrix often yields excessive deformation
• Temporarily reduce the penalty term in the stiffness calculation
• Stress calculation use the penalty term as it is
Volumetric strain
Pressure
Unique pressure
for given displ.
7
1
10
[K]
1
1
 
 
 

 
 
 

90
Example – Hydrostatic Tension
• Invariants
• Reduced invariants
 


 

  

1 1
2 2
3 3
x X
x X
x X

 
 
 
 
 

 
0 0
0 0
0 0
F
 

 
 
 
 

 
2
2
2
0 0
0 0
0 0
C
     
2 4 6
1 2 3
I 3 I 3 I


 
 
  
1/3
1 1 3
2/3
2 2 3
1/2 3
3 3
J I I 3
J I I 3
J I
I1 and I2 are not constant
J1 and J2 are constant

91
Strain Energy Density
• Using reduced invariants
– WD(J1, J2): Distortional strain energy density
– WH(J3): Dilatational strain energy density
• The second terms is related to nearly incompressible
behavior
– K: bulk modulus for linear elastic material
 
1 2 3 D 1 2 H 3
W(J ,J ,J ) W (J ,J ) W (J )
  2
H 3 3
K
W (J ) (J 1)
2
   
2
3
2
H 3 3
1
W (J ) (J 1)
2D
 
Abaqus:

92
Mooney-Rivlin Material
• Most popular model
– Initial shear modulus ~ 2(A10 + A01)
– Initial Young’s modulus ~ 6(A10 + A01) (3D) or 8(A10 + A01) (2D)
– Bulk modulus = K
• Hydrostatic pressure
– Numerical instability for large K (volumetric locking)
– Penalty method with K as a penalty parameter
 
     
1 2 3 D 1 2 H 3
2
10 1 01 2 3
W(J ,J ,J ) W (J ,J ) W (J )
K
A (J 3) A (J 3) (J 1)
2


   
 
H
3
3 3
W
W
p K(J 1)
J J

93
Mooney-Rivlin Material cont.
• Second P-K stress
– Use chain rule of differentiation

 
   
   
      
   
3
1 2
1 2 3
10 1, 01 2, 3 3,
J
J J
W W W W
J J J
A J A J K(J 1)J
E E E
S
E E E E 


,
a
aE
E
 
 

 
 

1/3 4/3
1
1, 3 1, 1 3 3,
3
2/3 5/3
2
2, 3 2, 2 3 3,
3
1/2
1
3, 3 3,
2
J (I )I I (I )I
J (I )I I (I )I
J (I )I
E E E
E E E
E E

  
   
1,
2,
9
3, imn jrs mr ns
4
I 2
I 4(1 tr ) 4
I (2 4tr ) 4 [ e e E E ]
E
E
E
1
E 1 E
E 1 E





1/3
1 1 3
2/3
2 2 3
1/2
3 3
J I I
J I I
J I


 

1,
2, 1
1
3, 3
I 2
I 2(I )
I 2I
E
E
E
1
1 C
C

94
Example
• Show
• Let
• Then
• Derivatives
and

    1
1, 2, 1 3, 3
I 2 , I 2(I ), I 2I
E E E
1 1 C C
  
1 1
1 2 3
2 3
I tr( ), I tr( ), I tr( )
C CC CCC
     
2 3
1 1
1 1 2 1 2 3 3 1 1 2
2 6
I I , I I I , I I I I I

 
   
  
3
1 2
ij ji jk ki
ij ij ij
I
I I
, C , C C
C C C


 
     
  
1
3
1 2
ij 1 ij ji 3 ji
ij ij ij
I
I I
, I C , I C
C C C
 

 
2
C E

95
Mixed Formulation
• Using bulk modulus often causes instability
– Selectively reduced integration (Full integration for deviatoric
part, reduced integration for dilatation part)
• Mixed formulation: Independent treatment of pressure
– Pressure p is additional unknown (pure incompressible material)
– Advantage: No numerical instability
– Disadvantage: system matrix is not positive definite
• Perturbed Lagrangian formulation
– Second term make the material nearly incompressible and the
system matrix positive definite
 
H 3 3
W (J ,p) p(J 1)
   2
H 3 3
1
W (J ,p) p(J 1) p
2K

96
Variational Equation (Perturbed Lagrangian)
• Stress calculation
• Variation of strain energy density
• Introduce a vector of unknowns:
  
10 1, 01 2, 3,
A J A J pJ
E E E
S
 
   
, ,p
3
W W W p
p
: (J 1 )p
K
EE
S E
 ( ,p)
r u
W
 
  W
 
 0
0
a ( , ) : pH d
r r S E
  
3
p
H J 1
K
Volumetric strain
       2
1 2 3 10 1 01 2 3
1
W(J ,J ,J ) A (J 3) A (J 3) p(J 1) p
2K

97
Example – Simple Shear
• Calculate 2nd P-K stress for the simple shear deformation
– material properties (A10, A01, K)
   
   
  
   
   
   
T
1 1 0 1 1 0
0 1 0 1 2 0
0 0 1 0 0 1
F C F F



 
 
   
 
 
 

 
 
  
 
 
 
1,
2, 1
1
3, 3
I 2
6 2 0
I 2(I ) 2 4 0
0 0 6
4 2 0
I 2I 2 2 0
0 0 2
E
E
E
1
1 C
C
X1, x1
X2, x2
45o
  
1 2 3
I 4, I 4, I 1

98
Example – Simple Shear cont.

 
 
   
 
 

 

 
 
   
 
 
 
1, 1, 3,
8
2, 2, 3,
3
5 4 0
4 2
J I I 4 1 0
3 3
0 0 1
7 5 0
2
J I I 5 2 0
3
0 0 1
E E E
E E E
   
  
 
 
   
 
 
 
 
10 1, 01 2, 3 3,
10 01 10 01
10 01 10 01
10 01
A J A J K(J 1)J
5A 7A 4A 5A 0
2
4A 5A A 2A 0
3
0 0 A A
E E E
S
Note: S11, S22 and S33 are not zero



 
 
 
1/3
1 1 3
2/3
2 2 3
1/2
3 3
J I I 4
J I I 4
J I 1

99
Stress Calculation Algorithm
• Given: {E} = {E11, E22, E33, E12, E23, E13}T, {p}, (A10, A01)
  
T
{ } {1 1 1 0 0 0} { } 2{ } { }
1 C E 1
  
     
     
1 1 2 3
2 1 2 1 3 2 3 4 4 5 5 6 6
3 1 2 4 4 3 4 6 1 5 5 4 5 2 6 6
I C C C
I C C C C C C C C C C C C
I (C C C C )C (C C C C )C (C C C C )C

      
   
  
1,
2, 2 3 3 1 1 2 4 5 6
2 2 2
3, 2 3 5 3 1 6 1 2 4
5 6 3 4 6 4 1 5 4 5 2 6
{I } 2{1 1 1 0}
{I } 2{C C C C C C C C C }
{I } 2{C C C C C C C C C
C C C C C C C C C C C C }
E
E
E
 
 

 
 

1/3 4/3
1
1, 3 1, 1 3 3,
3
2/3 5/3
2
2, 3 2, 2 3 3,
3
1/2
1
3, 3 3,
2
{J } I {I } I I {I }
{J } I {I } I I {I }
{J } I {I },
E E E
E E E
E E
  
10 1, 01 2, 3,
{ } A {J } A {J } p{J }
E E E
S
For penalty method, use
K(J3 – 1) instead of p

100
Linearization (Penalty Method)
• Stress increment
• Material stiffness
• Linearized energy form
W
 
     W
 
 0
*
a ( ; , ) : : : d
u u u E D E S E
    
, ,
W : :
E E
S E D E

      
 10 1, 01 2, 3 3, 3, 3,
A J A J K(J 1)J KJ J
EE EE EE E E
S
D
E

101
Linearization cont.
• Second-order derivatives of reduced invariants
   
   
 
       
       
   
4 7 4
1
3 3 3 3
5 8 5
2
3 3 3 3
3 1
2 2
1, 1, 1, 3, 3, 1, 1 3, 3, 1 3,
3 3 3 3
2, 2, 2, 3, 3, 2, 2 3, 3, 2 3,
3 3 3 3
3, 3, 3, 3,
3 3
1 4 1
J I I I (I I I I ) I I I I I I I
3 9 3
2 10 2
J I I I (I I I I ) I I I I I I I
3 9 3
1 1
J I I I I I
4 2
EE EE E E E E E E EE
EE EE E E E E E E EE
EE E E EE
   

  
  
1,
2,
1 1 1 1
3, 3 3
I
I 4
I 4I I
EE
EE
EE
0
1 1 I
C C C IC

102
MATLAB Function Mooney
• Calculates S and D for a given deformation gradient
%
% 2nd PK stress and material stiffness for Mooney-Rivlin material
%
function [Stress D] = Mooney(F, A10, A01, K, ltan)
% Inputs:
% F = Deformation gradient [3x3]
% A10, A01, K = Material constants
% ltan = 0 Calculate stress alone;
% 1 Calculate stress and material stiffness
% Outputs:
% Stress = 2nd PK stress [S11, S22, S33, S12, S23, S13];
% D = Material stiffness [6x6]
%

103
Summary
• Hyperelastic material: strain energy density exists with
incompressible constraint
• In order to be material frame indifferent, material
properties must be expressed using invariants
• Numerical instability (volumetric locking) can occur when
large bulk modulus is used for incompressibility
• Mixed formulation is used for purely incompressibility
(additional pressure variable, non-PD tangent stiffness)
• Perturbed Lagrangian formulation for nearly
incompressibility (reduced integration for pressure term)

104
Finite Element Formulation for
Nonlinear Elasticity
Section 3.6

105
Voigt Notation
• We will use the Voigt notation because the tensor
notation is not convenient for implementation
– 2nd-order tensor vector
– 4th-order tensor matrix
• Stress and strain vectors (Voigt notation)
– Since stress and strain are symmetric, we don’t need 21 component
 T
11 22 12
{ } {E E 2E }
E
 T
11 22 12
{ } {S S S }
S

106
4-Node Quadrilateral Element in TL
• We will use plane-strain, 4-node quadrilateral element to
discuss implementation of nonlinear elastic FEA
• We will use TL formulation
• UL formulation will be discussed in Chapter 4
Finite Element at
undeformed domain
Reference Element
X1
X2
1 2
3
4
s
t
(–1,–1) (1,–1)
(1,1)
(–1,1)

107
Interpolation and Isoparametric Mapping
• Displacement interpolation
• Isoparametric mapping
– The same interpolation function is used for geometry mapping

 
e
N
I I
I 1
N ( )
u s u

 
e
N
I I
I 1
N ( )
X s X
Nodal displacement vector (uI, vI)
Interpolation function
Nodal coordinate (XI, YI)
  
  
  
  
1
1 4
1
2 4
1
3 4
1
4 4
N (1 s)(1 t)
N (1 s)(1 t)
N (1 s)(1 t)
N (1 s)(1 t)
Interpolation (shape) function
• Same for all elements
• Mapping depends of geometry

108
Displacement and Deformation Gradients
• Displacement gradient
– How to calculate
– Both displacement and deformation gradients are not symmetric




 

e
N
I
I
I 1
N ( )
s
u
u
X X 
 
e
N
i,j I,j Ii
I 1
u N ( )u
s
  T
0 1,1 1,2 2,1 2,2
{u u u u }
u
   
T T
11 12 21 22 1,1 1,2 2,1 2,2
{ } {F F F F } {1 u u u 1 u }
F


I
N ( )
?
s
X

109
Green-Lagrange Strain
• Green-Lagrange strain
– Due to nonlinearity,
– For St. Venant-Kirchhoff material,
 
 
 
 
 
   
   
   
  
   
1
1,1 1,1 1,1 2,1 2,1
2
11
1
22 2,2 1,2 2,1 2,2 2,2
2
12 1,2 2,1 1,2 1,1 2,1 2,2
u (u u u u )
E
{ } E u (u u u u )
2E u u u u u u
E

{ } [ ]{ }
E B d

{ } [ ]{ }
S D E
   
 
 
    
 
 

 
2 0
[ ] 2 0
0 0
D

110
Variation of G-R Strain
• Although E(u) is nonlinear, is linear
 N
{ } [ ]{ }
E B d
  T
0
( , ) sym( )
E u u u F
( , )
E u u
Function of u
Different from linear strain-displacement matrix
 
 
 
 
 

 
 

     
 
11 1,1 21 1,1 11 2,1 21 2,1 11 4,1 21 4,1
N 12 1,2 22 1,2 12 2,2 22 2,2 12 4,2 22 4,2
11 1,2 21 1,2 11 2,2 21 2,2 11 4,2 21 4,2
12 1,1 22 1,1 12 2,1 22 2,1 12 4,1 22 4,1
F N F N F N F N F N F N
[ ] F N F N F N F N F N F N
B



111
Variational Equation
• Energy form
• Load form
• Residual
W
W
 W
 W



0
0
T T
N
T int
a( , ) : d
{ } [ ] { }d
{ } { }
u u S E
d B S
d F
 
W 
W 

 W  
 W  

 
  
S
0 0
e
S
0 0
T b T
N
T b
I I I
I 1
T ext
( ) d d
N ( ) d N ( ) d
{ } { }
u u f u t
u s f s t
d F
   Z
T int T ext
h
{ } { ( )} { } { }, { }
d F d d F d

112
Linearization – Tangent Stiffness
• Incremental strain
• Linearization
  
N
{ } [ ]{ }
E B d
W W
 
 W  W 
 
 
 
0 0
T T
N N
: : d { } [ ] [ ][ ]d { }
E D E d B D B d
W W
 
 W  W 
 
 
 
0 0
T T
G G
: d { } [ ] [ ][ ]d { }
S E d B B d

 
 
 

 
 
 
11 12
12 22
11 12
12 22
S S 0 0
S S 0 0
[ ]
0 0 S S
0 0 S S

 
 
 

 
 
 
 
1,1 2,1 3,1 4,1
1,2 2,2 3,2 4,2
G
1,1 2,1 3,1 4,1
2,1 2,2 3,2 4,2
N 0 N 0 N 0 N 0
N 0 N 0 N 0 N 0
[ ]
0 N 0 N 0 N 0 N
0 N 0 N 0 N 0 N
B

113
Linearization – Tangent Stiffness
• Tangent stiffness
• Discrete incremental equation (N-R iteration)
– [KT] changes according to stress and strain
– Solved iteratively until the residual term vanishes
W
 
  W
 
0
T T 0
T N N G G
[ ] [ ] [ ][ ] [ ] [ ][ ] d
K B D B B B

     Z
T T ext int
T h
{ } [ ]{ } { } { }, { }
d K d d F F d

114
Summary
• For elastic material, the variational equation can be
obtained from the principle of minimum potential energy
• St. Venant-Kirchhoff material has linear relationship
between 2nd P-K stress and G-L strain
• In TL, nonlinearity comes from nonlinear strain-
displacement relation
• In UL, nonlinearity comes from constitutive relation and
unknown current domain (Jacobian of deformation
gradient)
• TL and UL are mathematically equivalent, but have
different reference frames
• TL and UL have different interpretation of constitutive
relation.

115
MATLAB Code for
Hyperelastic Material Model
Section 3.7

116
HYPER3D.m
• Building the tangent stiffness matrix, [K], and the residual
force vector, {R}, for hyperelastic material
• Input variables for HYPER3D.m
Variable Array size Meaning
MID Integer Material Identification No. (3) (Not used)
PROP (3,1) Material properties (A10, A01, K)
UPDATE Logical variable If true, save stress values
LTAN Logical variable If true, calculate the global stiffness matrix
NE Integer Total number of elements
NDOF Integer Dimension of problem (3)
XYZ (3,NNODE) Coordinates of all nodes
LE (8,NE) Element connectivity

117
function HYPER3D(MID, PROP, UPDATE, LTAN, NE, NDOF, XYZ, LE)
%***********************************************************************
% MAIN PROGRAM COMPUTING GLOBAL STIFFNESS MATRIX AND RESIDUAL FORCE FOR
% HYPERELASTIC MATERIAL MODELS
%***********************************************************************
%%
global DISPTD FORCE GKF SIGMA
%
% Integration points and weights
XG=[-0.57735026918963D0, 0.57735026918963D0];
WGT=[1.00000000000000D0, 1.00000000000000D0];
%
% Index for history variables (each integration pt)
INTN=0;
%
%LOOP OVER ELEMENTS, THIS IS MAIN LOOP TO COMPUTE K AND F
for IE=1:NE
% Nodal coordinates and incremental displacements
ELXY=XYZ(LE(IE,:),:);
% Local to global mapping
IDOF=zeros(1,24);
for I=1:8
II=(I-1)*NDOF+1;
IDOF(II:II+2)=(LE(IE,I)-1)*NDOF+1:(LE(IE,I)-1)*NDOF+3;
end
DSP=DISPTD(IDOF);
DSP=reshape(DSP,NDOF,8);
%
%LOOP OVER INTEGRATION POINTS
for LX=1:2, for LY=1:2, for LZ=1:2
E1=XG(LX); E2=XG(LY); E3=XG(LZ);
INTN = INTN + 1;
%
% Determinant and shape function derivatives
[~, SHPD, DET] = SHAPEL([E1 E2 E3], ELXY);
FAC=WGT(LX)*WGT(LY)*WGT(LZ)*DET;

118
% Deformation gradient
F=DSP*SHPD' + eye(3);
%
% Computer stress and tangent stiffness
[STRESS DTAN] = Mooney(F, PROP(1), PROP(2), PROP(3), LTAN);
%
% Store stress into the global array
if UPDATE
SIGMA(:,INTN)=STRESS;
continue;
end
%
% Add residual force and tangent stiffness matrix
BM=zeros(6,24); BG=zeros(9,24);
for I=1:8
COL=(I-1)*3+1:(I-1)*3+3;
BM(:,COL)=[SHPD(1,I)*F(1,1) SHPD(1,I)*F(2,1) SHPD(1,I)*F(3,1);
SHPD(2,I)*F(1,2) SHPD(2,I)*F(2,2) SHPD(2,I)*F(3,2);
SHPD(3,I)*F(1,3) SHPD(3,I)*F(2,3) SHPD(3,I)*F(3,3);
SHPD(1,I)*F(1,2)+SHPD(2,I)*F(1,1)
SHPD(1,I)*F(2,2)+SHPD(2,I)*F(2,1) SHPD(1,I)*F(3,2)+SHPD(2,I)*F(3,1);
SHPD(2,I)*F(1,3)+SHPD(3,I)*F(1,2)
SHPD(2,I)*F(2,3)+SHPD(3,I)*F(2,2) SHPD(2,I)*F(3,3)+SHPD(3,I)*F(3,2);
SHPD(1,I)*F(1,3)+SHPD(3,I)*F(1,1)
SHPD(1,I)*F(2,3)+SHPD(3,I)*F(2,1) SHPD(1,I)*F(3,3)+SHPD(3,I)*F(3,1)];
%
BG(:,COL)=[SHPD(1,I) 0 0;
SHPD(2,I) 0 0;
SHPD(3,I) 0 0;
0 SHPD(1,I) 0;
0 SHPD(2,I) 0;
0 SHPD(3,I) 0;
0 0 SHPD(1,I);
0 0 SHPD(2,I);
0 0 SHPD(3,I)];
end

119
%
% Residual forces
FORCE(IDOF) = FORCE(IDOF) - FAC*BM'*STRESS;
%
% Tangent stiffness
if LTAN
SIG=[STRESS(1) STRESS(4) STRESS(6);
STRESS(4) STRESS(2) STRESS(5);
STRESS(6) STRESS(5) STRESS(3)];
SHEAD=zeros(9);
SHEAD(1:3,1:3)=SIG;
SHEAD(4:6,4:6)=SIG;
SHEAD(7:9,7:9)=SIG;
%
EKF = BM'*DTAN*BM + BG'*SHEAD*BG;
GKF(IDOF,IDOF)=GKF(IDOF,IDOF)+FAC*EKF;
end
end; end; end;
end
end

120
Hyperelastic Material Analysis Using ABAQUS
• *ELEMENT,TYPE=C3D8RH,ELSET=ONE
– 8-node linear brick, reduced integration with hourglass control,
hybrid with constant pressure
• *MATERIAL,NAME=MOONEY
*HYPERELASTIC, MOONEY-RIVLIN
80., 20.,
– Mooney-Rivlin material with A10 = 80 and A01 = 20
• *STATIC,DIRECT
– Fixed time step (no automatic time step control)
x
y
z

121
*HEADING
- Incompressible hyperelasticity (Mooney-
Rivlin) Uniaxial tension
*NODE,NSET=ALL
1,
2,1.
3,1.,1.,
4,0.,1.,
5,0.,0.,1.
6,1.,0.,1.
7,1.,1.,1.
8,0.,1.,1.
*NSET,NSET=FACE1
1,2,3,4
*NSET,NSET=FACE3
1,2,5,6
*NSET,NSET=FACE4
2,3,6,7
*NSET,NSET=FACE6
4,1,8,5
*ELEMENT,TYPE=C3D8RH,ELSET=ONE
1,1,2,3,4,5,6,7,8
*SOLID SECTION, ELSET=ONE,
MATERIAL= MOONEY
*MATERIAL,NAME=MOONEY
*HYPERELASTIC, MOONEY-RIVLIN
80., 20.,
*STEP,NLGEOM,INC=20
UNIAXIAL TENSION
*STATIC,DIRECT
1.,20.
*BOUNDARY,OP=NEW
FACE1,3
FACE3,2
FACE6,1
FACE4,1,1,5.
*EL PRINT,F=1
S,
E,
*NODE PRINT,F=1
U,RF
*OUTPUT,FIELD,FREQ=1
*ELEMENT OUTPUT
S,E
*OUTPUT,FIELD,FREQ=1
*NODE OUTPUT
U,RF
*END STEP

122
• Analytical solution procedure
– Gradually increase the principal stretch  from 1 to 6
– Deformation gradient
– Calculate J1,E and J2,E
– Calculate 2nd P-K stress
– Calculate Cauchy stress
– Remove the hydrostatic component of stress

 
 
 
 
 

 
0 0
0 1 / 0
0 0 1 /
F
 
10 1, 01 2,
A J A J
E E
S
   T
1
J
F S F

    
11 11 22

123
• Comparison with analytical stress vs. numerical stress

124
Fitting Hyperelastic Material
Parameters from Test Data
Section 3.9

125
Elastomer Test Procedures
• Elastomer tests
– simple tension, simple compression, equi-biaxial tension, simple
shear, pure shear, and volumetric compression
0 1 2 3 4 5 6 7
0
10
20
30
40
50
60
70
Nominal strain
Nominal
stress
uni-axial
bi-axial
pure shear

126
F
F L
Simple tension test
F
F
L
Pure shear test
L
F
Equal biaxial test
F
L
Volumetric compression test
Elastomer Tests
• Data type: Nominal stress vs. principal stretch

127
Data Preparation
• Need enough number of independent experimental data
– No rank deficiency for curve fitting algorithm
• All tests measure principal stress and principle stretch
Experiment Type Stretch Stress
Uniaxial tension Stretch ratio  = L/L0 Nominal stress TE = F/A0
Equi-biaxial
tension
Stretch ratio  = L/L0 in y-
direction
Nominal stress TE = F/A0
in y-direction
Pure shear test Stretch ratio  = L/L0 Nominal stress TE = F/A0
Volumetric test Compression ratio  = L/L0 Pressure TE = F/A0

128
Data Preparation cont.
• Uni-axial test
• Equi-biaxial test
• Pure shear test
       
1 2 3
, 1 /


     

3
10 01
U
T 2(1 )(A A )
   
 
         
 
 
10
T 2 3
10 01
01
A
T(A ,A , ) { } { } 2( ) 2(1 )
A
x b
       2
1 2 3
, 1 /


      

5 2
10 01
1 U
T 2( )(A A )
2
       
1 2 3
, 1, 1 /


     

3
10 01
U
T 2( )(A A )

129
Data Preparation cont.
• Data Preparation
• For Mooney-Rivlin material model, nominal stress is a
linear function of material parameters (A10, A01)


      
1 2 3 i i 1 NDT
E E E E E E E
1 2 3 i i 1 NDT
Type 1 1 1 4 4 4
T T T T T T T

130
Curve Fitting for Mooney-Rivlin Material
• Need to determine A10 and A01 by minimizing error
between test data and model
• For Mooney-Rivlin, T(A10, A01, lk) is linear function
– Least-squares can be used
 

 

10 01
NDT
2
E
k 10 01 k
A ,A k 1
minimize T T(A ,A , )
 

 
 
 

   
  
   
   
  
 
   
T
1
1
T
2 1
T
NDT NDT
( )
T
T ( )
{ } { } [ ]{ }
T ( )
x
x
T b X b
x
 
  
 
10
01
A
{ }
A
b
 
 
 
  
 
 
 
E
1
E
E 2
E
NDT
T
T
{ }
T
T

131
Curve Fitting cont.
• Minimize error(square)
• Minimization  Linear regression equation
  
  
  
T E T E
E T E
E T E T T E T T
{ } { } { } { }
{ } { }
{ } { } 2{ } [ ] { } { } [ ] [ ]{ }
e e T T T T
T Xb T Xb
T T b X T b X X b

T T E
[ ] [ ]{ } [ ] { }
X X b X T

132
Stability of Constitutive Model
• Stable material: the slope in the stress-strain curve is
always positive (Drucker stability)
• Stability requirement (Mooney-Rivlin material)
• Stability check is normally performed at several specified
deformations (principal directions)
• In order to be P.D.

 
d : : d 0
D
     
1 1 2 2
d d d d 0
 

   
  
 
  
   
11 12 1
1 2
21 22 2
D D d
d d 0
D D d
 
 
11 22
11 22 12 21
D D 0
D D D D 0

Chap-3 FEA for Nonlinear Elastic Problems.pptx

More Related Content

What's hot (20)

Similar to Chap-3 FEA for Nonlinear Elastic Problems.pptx (20)

More from Dharmsinh Desai of University (20)

Recently uploaded (20)

Chap-3 FEA for Nonlinear Elastic Problems.pptx