FINETE ELEMENT METHODS IN

ENGINEERING DESIGN
ΜΜΥ416
6. 4-node quadrilateral isoparametric
elements
-
Τετράπλευρα ισοπαραμετρικά
στοιχεία 4ων κόμβων
1
 Orthogonal elements - Ορθογώνια στοιχεία - Introduction

Orthogonal element: 4 nodes, 8 DoF. Two elements share a common side and 2
common nodes, apart from the elements at the boundaries of the solid. 1st node is
assigned randomly, but the order of nodes follows the anticlockwise direction. The
degrees of freedom of the element are the displacement in the u and v in x directions
Chapter Summary
4 3
Quadrilateral orthogonal elements. Formulation:
y
• Interpolation (shape) function
• Strain for stress calculation
• Stiffness matrix (through the TMTPE)
• Force vector (through TMTPE) 1 2
x
Quadrilateral isoparametric elements. Formulation:
• Interpolation function and geometry
• Strain for stress calculation
• Jacobian matrix and determinant (Ιακωβιανού και ορίζουσας)
• Stiffness matrix (through the TMTPE)
• Force vector (through TMTPE)
• Numerical integration (Gauss Quadrature)

2
 Orthogonal elements – Interpolation function formulation

• All sides are parallel to the coordinate directions: not practical for non-uniform
geometries, however the rectangular element forms the basis for the development
of the quadrilateral isoparametric elements.

4 3
• Interpolation function y
- Polynomials in terms of x and y
- 4 nodes, therefore 4 boundary
conditions for each directions, therefore 1 2
ponomials with 4 coeficients will be use x
in the x and y:

u  1   2 x  3 y   4 xy
v  1  2 x  3 y  4 xy (6.1)

3
 Orthogonal elements – Interpolation function formulation
1ος method for the formulation of the function
• Διατύπωση μετατοπίσεων u σε κάθε κόμβο:
 u1  1   2 x1   3 y1   4 x1y1 4 3
u     x   y   x y y
 2 1 2 2 3 2 4 2 2
 (6.2)
u3  1   2 x 3   3 y 3   4 x 3 y 3
u4  1   2 x 4   3 y 4   4 x 4 y 4
1 2
x
• In matrix form:
u1 1 𝑥1 𝑦1 𝑥1𝑦1 α1
u2 1 𝑥2 𝑦2 𝑥2𝑦2 α2
= (6.3)
u3 1 𝑥3 𝑦3 𝑥3𝑦3 α3
u4 1 𝑥4 𝑦4 𝑥4𝑦4 α4

𝐮 = 𝐗𝐘 𝐚

• For the solution of the coefficients α1 to α4 the inverse of 𝐗𝐘 is needed:

𝐗𝐘 −𝟏 𝐮=𝐚 (6.4)
4
 Orthogonal elements – Interpolation function formulation
2ος method for the interpolation folmulation:
• Lagrange interpolations. 4 3
y
– The objective is the formulation of the
displacement in terms of the shape
functions: u I
1 2
x
 u1 
u 
 
u(x,y)  [N1 N2 N3 N4 ]  2  (6.5)
u3 
u4 
Shape
functions

– For the formulation of the functions initially the displacement in one

side (1-2) of the element is calculated, where y = y1 , therefore the
displacement will be function of x only:
 u1  (6.6)
uI (x,y1 )  [n1(x) n2 (x)]  
u2  5
 Orthogonal elements – Interpolation function formulation

– For the sides with constant y (y1=constant) the shape functions are
defined as the unidirectional element:

x1 x x2
u ( x)  N1 ( x)u1  N 2 ( x)u2
u1 u(x) u2

x  x2 x  x1 lx x
N1 ( x)  , N 2 ( x)  u ( x)  u1  u2 (6.7)
x1  x2 x2  x1 l l
– The same result give the Lagrange interpolations for one dimension
(y1=constant): 𝑛

𝑝𝑛 𝑥 = 𝑦1 𝐿𝑖 𝑥 (6.8)
𝑖=0

Li are the Lagrange polynomials for n+1 interpolation points:

𝑥 − 𝑥0 … 𝑥 − 𝑥𝑖−1 𝑥 − 𝑥𝑖+1 … (𝑥 − 𝑥𝑛)
𝐿𝑖 𝑥 = , 𝑖 = 0, … , 𝑛 (6.9)
𝑥𝑖 − 𝑥0 … 𝑥𝑖 − 𝑥𝑖−1 𝑥𝑖 − 𝑥𝑖+1 … (𝑥𝑖 − 𝑥𝑛)
6
 Orthogonal elements – Interpolation function formulation

– Therefore for the side 1-2 (y = y1):

uII
 u1  4 3
uI (x,y1 )  [n1(x) n2 (x)]   (6.10) y
u2 
x  x2 x  x1
n1(x)  , n2 (x)  (6.11) uI
x1  x 2 x 2  x1
1 2
x
– For the side 4-3, where y = y3 (constant)
u 
uII (x,y 3 )  [n4 (x) n3 (x)]  4  (6.12)
u3 
x  x3 x  x4
n4 (x)  , n3 (x)  (6.13)
x 4  x3 x3  x 4

Equations 6.10 and 6.12 are the interpolation equations for the displacement
within sides 1-2 και 3-4 .

7
 Orthogonal elements – Interpolation function formulation

- Similarly the interpolations for the sides 1-4 και uII

4 3
2-3 in Y-direction (x = constant) are formulated: y

 uI (x,y1 ) 
u(x,y)  [n1(y) n4 (y)]   (6.14)
 II
u (x,y 3 
)
1 uI x 2
y  y4 y  y1
n1(y)  , n4 (y)  (6.15)
y1  y 4 y 4  y1
- Combining the equations 6.10, 6.12 and 6.14 the interpolation equation is
formed u(x,y) in the inside of the element:
Nodal displacements
  u1  
 [n1(x) n2 (x)]   
 u2  
u(x,y)  [n1(y) n4 (y)]   (6.16)
[n (x) n (x)]  4 
u
 4 3  
 u3 

8
 Orthogonal elements – Interpolation function formulation

- Eq. 6.16 is formulated in terms of the Lagrange interpolation functions and

the displacement vector:  u1 
u 
 
u(x,y)  [n1(x)n1(y) n2 (x)n1(y) n3 (x)n4 (y) n 4 (x)n 4 (y)]  2  (6.16)
u3 
u4 
- The products of the Lagrange interpolations will give the shape functions of
the rectangular (orthogonal) element.

The shape functions have  1

been formed for the N
 1  n1 (x)n1 (y)  (x 3  x)(y 3  y)
A
coordinates of two nodes as 
x1=x4, x2=x3, y1=y2 and y3=y4. N  n (x)n (y)   1 (x  x)(y  y)
 2 2 1
A
1 3
(6.17)
Α=(x3-x1)(y3-y1) is the area of 
the rectangular element. N  n (x)n (y)  1 (x  x)(y  y)
 3 3 4
A
1 1

 1
N4  n4 (x)n4 (y)   (x 3  x)(y1  y)
 A
9
 Orthogonal elements – Interpolation function formulation

• Properties of the shape functions N1(x, y) :

– 1 at node 1 and 0 at the rest of the nodes.
– Linear function of x along the side 1-2 and linear function of y along the
length of the 1-4 .
– Zero along the side 2-3 and 3-4.

N1

4 3

1 2

• The rest of the functions behave similarly. Due to the above properties the
shape function is used for the both the u and v displacements in the same
node.
10
 Orthogonal elements – Interpolation function formulation

• Displacement equation u1 

v 
 1
u2 
  {u}  [N]28 {q}81
u  N1 0 N2 0 N3 0 N4 0  v 2 
{u}       
v  0 N1 0 N2 0 N3 0 N4  u3 
v 3 
  (6.18)
u4 
Displacement Shape v 4 
within the function
element (x,y) matrix
Node
displacemen
t vector

11
 Orthogonal elements – Interpolation function formulation

Example 6.1: Calculation of shape functions for orthogonal element.

y
 1
N
 1  n1 (x)n1 (y)  (x 3  x)(y 3  y)
A

N  n (x)n (y)   1 (x  x)(y  y) 4 (0,2) 3 (3,2)
 2 2 1
A
1 3


N  n (x)n (y)  1 (x  x)(y  y)
 3 3 4
A
1 1

 1
N4  n4 (x)n4 (y)   (x 3  x)(y1  y) x
 A 1 (0,0) 2 (3,0)

( x  3)( y  2)  x( y  2)
N1  N2 
6 6
xy  y ( x  3)
N3  N4 
6 6
12
 Orthogonal elements – Interpolation function formulation

Example 6.1: Calculation of shape functions for orthogonal element.

y

( x  3)( y  2)  x( y  2)
N1  N2  4 (0,2) 3 (3,2)
6 6
xy  y ( x  3)
N3  N4 
6 6
x

1 (0,0) 2 (3,0)

N1 N2

13
 Orthogonal elements – Strain equation formulation
• Strain displacement relations
– Similarly to the triangular element, i.e. by differentiating the shape
functions in the x and y directions.

u   4  4 Ni
 xx    
x x  i1
Ni (x,y)ui   
 i1 x
ui

u1 
• Similarly the strains εyy and γxy are defined and the strain vector v 
is formulated:  1
u 
 y  y3 0 y3  y 0 y  y1 0 y1  y 0  2
v 
{ }   0 x3  x   2 
1
x  x3 0 x1  x 0 x  x1 0
A  u
 x  x 3 y  y3 x1  x y 3  y x  x1 y  y1 x 3  x y1  y   3 
v 3 
 
u4 
v 4 
 [B]{q} (6.19)
Matrix [B] is a linear function in terms of x or y, therefore the strain changes linearly
within the element in the x or y (not linear in both x and y).
14
 Orthogonal elements – Stress equation formulation
• Stress calculation
– Plane stress or strain conditions are used (εξ. 5.5):

- Plane stress:

 xx  1  0   xx 
  E  1   

 yy     {}  [C ]{}
  yy 
0 (5.5)
1   2 
  0 0  
 xy  2 (1   )   xy 
1

Plane strain
 xx  1    0   xx 
  E     

 yy   1  0  yy   {}  [C ]{} (5.7)
 
  (1   )(1  2 )  0  
 xy  2      xy 
1
 0

15
 Orthogonal elements – Stress equation formulation
• Stiffness matrix from strain energy, εξ. 5.25, και εξ. 6.19:

h h (e) T
U(e)
  { } [C]{ } dA  {q }  [B]8T3 [C]33 [B]38 dA{q(e) }
T (e)

2A 2 A

1 (e) T (e)
 {q } [k ]88 {q(e) } (6.20)
2
The calculation of [k(e)] for a general quadrilateral element is performed
with numerical integration, however, for the orthogonal element in plane
stress condition (where C=Cσ) the analytical integration results in:

 36 1
8  312 1 3 
8
3 
12  1
8

6
13 
8 
 1 3  13  
 1 3  1 3 
 312 
 8 6 8 6 8 12 8

  312 13 
8
3 
6  1
8

6
1 3 
8
3 
12
1
8

 3  
Eh  183  
 1 3  13 
 312 1
12 
[k ] 
(e) 6 8 6 8 8

1   2  312  1  13  3  1

 312 1 3  

 1  
8 6 8 6 8 8
(6.21)
 8  312
3  1 3  1 3  13 
12 8 8 6 8 6 
 6 1 3  3  1
 312 13  3 
 1 
 13  
8 12 8 8 6 8

 8  312 1
8
3 
12
13 
8

6  1
8
3 
6   16
 Orthogonal elements – Force vector formulation
• Distributed forces are defined as in the triangular element.
• Nodal forces (concentrated) (εξ. 5.27):
ND= συνολικός αριθμός κόμβων στο στερεό
ND
V   (Fixui  Fiy v i )  {Q s } T {FN } {FN }  [F1x F1y FNDx FNDy ]T
i1

• Uniformly distributed forces in side of the element (εξ. 5.30, 5.31):

(l  s) / l 0  Tx 
 0 T 
(l  s) / l  y
 
l l s/l 0  Tx  hl Tx 
{fT }  h [N] {T } ds  h  
(e) T
 T  ds   
0 0
 0 s /l  y 2 Ty 
 0 0  0
   
 0 0   0 

NS
V   {q(e) } T {fT(e) }  {Q s } T {FT } NS= συνολικός αριθμός πλευρών όπου
e 1 ασκούνται κατανεμημένα φορτία στο στερεό 17
 Orthogonal elements – Force vector formulation
• Body forces b = {bx, by}T:

b x  b x 
V (e)
 h [u v]   dA  {q } h [N] dA  
(e) T T

A b y  A b y  (6.22)
 {q(e) } T {fb(e) }

1 0 b x 
0  b 

1
  y
1 0 b x 
  b  
hA 0 1   x  hA b y  i.e. the force magnitude
{fb } 
(e)
   
4 1 
0 b y  4 b x  is equally distributed in (6.23)
  b y  the το the 4 nodes
0 1
 
1 0  b x 
  b 
0 1
 y

18
 Orthogonal elements – Potential energy formulation
• The potential energy of the body forces has to be formulated as {fb(e)} from
all elements (εξ.5.32):
NE
V   {q(e) } T {fb(e) }  {Q s } T {FB }
e 1

• The theorem of the minimum of the total potential energy is used for the
system equations as in the triangular element case:

- Total potential energy

1
  U  V  {Q s } T [Ks ]{Q s }  {Q s } T {FN  FT  FB } (5.33)
2

- TMTPE

 0  [K s ]{Q s }  {FN  FT  FB } (5.34)
 {Q s }

19
 Orthogonal elements – Example 6.2
Quadrilateral orthogonal solid is y
studied as one element. The solid is
subjected to a distributed force f in
side 3-4 and it is firmly supported in
the side 1-2. Rolling exists in the x- 4 f 3
direction for the 3-4 side, and
therefore only shear deformation is
1m
expected. 1m

Calculate the stresses in the solid. 1 2

Comment the on the result in view of x
the expected shear deformation.

• E = 10 GPa, v = 0.25, h=0.1m

• F = 100 kN/m2
20
 Orthogonal elements – Example 6.2
Solution method y

• Stress {σ} needs to be calculated.

From εξ. 5.5 for plane stress: 4 3
f
 xx  1  0   xx 
   1  
0   yy   {}  [C ]{}
E
 yy    
  1 
2
0 0      xy  1m
 xy 
1
(1 )
2  
1m
• Therefore εxx, εyy και γxy, i.e. {ε} (εξ.
6.19), also: 1 2
x
u1 
v 
 1
u 
 y  y3 0 y3  y 0 y  y1 0 y1  y 0  2
1 v 
{ }  0 x  x3 0 x1  x 0 x  x1 0 x3  x   2 
A  u
 x  x 3 y  y3 x1  x y 3  y x  x1 y  y1 x 3  x y1  y   3 
v 3 
 
u4 
v 4 
 [B]{q}

• For {ε} we need the coordinates and the displacement vector of the
system {Qs} = {u1, v1, u2, v2, u3, v3, u4, v4}T. 21
 Orthogonal elements – Example 6.2
• For {Qs} we use TMTPE: y

4 f 3

 0  [K s ]{Q s }  {FN  FT  FB } 1m
 {Q s } 1m

0 0 1 2
x
• Stifness matrix (εξ. 6.21):
u1 v1 u2 v2 u3 v3 u4 v4
 36 1
8  312 1 3 
8
3 
12  1
8

6
13 
8  u1
 1 3  13  
 1 3  1 3 
 312  v1
 8 6 8 6 8 12 8

  312 13 
8
3 
6  1
8

6
1 3 
8
3 
12
1
8
 u2
 3  
Eh  183  
 1 3  13 
 312 1
12  v2
[k ] 
(e) 6 8 6 8 8

1   2  312  1  13  3  1

 312 1 3  
u3
 1  
8 6 8 6 8 8

 8  312
3  1 3  1 3  13 
12 8 8 6 8 6  v3
 6 1 3  3  1
 312 13  3 
 1  u4
 13  
8 12 8 8 6 8

 8  312 1
8
3 
12
13 
8

6  1
8
3 
6   v4
Eh  36  312  u3
Zero DoF: u3 και u4. [K]    3 3  
1  2  12 6  u4 22
 Orthogonal elements – Example 6.2
y
• Calculation of nodal equivalent forces [FT] due to
f
4 f 3
(l  s) / l 0  Tx 
 0 T  1m
(l  s) / l  y
1m
 
l l s/l 0  Tx  hl Tx  1 2
{fT }  h [N] {T } ds  h  
(e) T
 T  ds    x
0 0
 0 s / l  y 2 Ty 
 0 0  0
   
 0 0   0 

0 0 0 fx1
0 0 0 fy1
0 0 0 fx2
hl 0 0 0 fy2
{fT(e)}= = 0.05 = Δείκτες
2 Tx3 100000 5000 fx3
0 0 0 fy3
Tx4 100000 5000 fx4
0 0 0 fy4
23
 Orthogonal elements – Example 6.2
y
• System equation:

 4.88 2.88  u3  5,000  4 f 3

 
8
10   
 2.88 4.88  4 
u 5,000  1m
1m
• Displacement calculation: u3  u4  0.025mm 1 2
x

{Qs} = {u1, v1, u2, v2, u3, v3, u4, v4}T= {0, 0, 0, 0, 0.025, 0, 0.025, 0}T

• Strain calculations εxx, εyy και γxy ({ε}) from eq. 6.19:
0 
0 
 
0 
y  1 0 1 y 0 y 0 y 0    0 
   
{ }   0 1 x 
0
x 1 0 x 0 x 0    0 
  2.5  10 5
 x  1 y  1  x 1  y x y 1  x  y    2.5  10 5 
0   
 5 
 2.5  10 
0 
24
 Orthogonal elements – Example 6.2
• Stress calculation from εξ. 5.5:

 xx   1 0.25 0  0   0 
  1010 0.25    
 yy   1 0  0    0  Pa
 
  1  0.25
2

 0 0.375  2.5  105  105 

 xy 
0

• Only shear strain is observed γxy therefore the stresses are shear only.
The orthogonal element is accurate for simple shear deformation
calculations.

g xy Deformed
Initial geometry geometry

25
 Orthogonal elements – Example 6.3
y
The solid in the diagram is subjected to
f
equal and opposite forces f at the
nodes 3-4 and is firmly supported in 4 3
the side 4-1.
1m
1m

Note.: The application of the forces is 1 2 f

x
analogous to the bending moment
applied in a beam, therefore the E = 10 GPa, v = 0.25, h=0.1m
deformation should be expected to be f = 100 kN
bending.

Calculate the stresses in the solid using one element. Comment on the
result in view of the expected bending.

26
 Orthogonal elements – Example 6.3
y
Problem solution
f
4 3
• Stresses {σ} from eq. 5.5 for plane
stress:
1m
 xx  1m
1  0   xx 
   1  
0   yy   {}  [C ]{}
E
 yy    
  1 
2
0 0      xy 
1 2 f
 xy 
1
(1 ) x
2  

• Strains are needed εxx, εyy and γxy,

i.e. {ε} (εξ. 6.19):
u1 
v 
 1
u 
 y  y3 0 y3  y 0 y  y1 0 y1  y 0  2
1 v 
{ }  0 x  x3 0 x1  x 0 x  x1 0 x3  x   2 
A  u
 x  x 3 y  y3 x1  x y 3  y x  x1 y  y1 x 3  x y1  y   3 
v 3 
 
u4 
v 4 
 [B]{q}

• For the calculation of {ε} the coordinates and the displacements {Qs} =
{u1, v1, u2, v2, u3, v3, u4, v4}T are needed. 27
 Orthogonal elements – Example 6.3
y
• Calculate {Qs} with using TMTPE:
f
0 0 4 3

 0  [K s ]{Q s }  {FN  FT  FB }
 {Q s } 1m
1m

• Calculate the stiffness matrix (εξ. 1 2 f

x
6.21):
u1 v1 u2 v2 u3 v3 u4 v4
 36 1
8  312 1 3 
8
3 
12  1
8

6
13 
8  u1
 1 3  13  
 1 3  1 3 
 312  v1
 8 6 8 6 8 12 8

  312 13 
8
3 
6  1
8

6
1 3 
8
3 
12
1
8
 u2
 3  
Eh  183  
 1 3  13 
 312 1
12  v2
[k ] 
(e) 6 8 6 8 8

1   2  312  1  13  3  1

 312 1 3  
u3
 1  
8 6 8 6 8 8

 8  312
3  1 3  1 3  13 
12 8 8 6 8 6  v3
 6 1 3  3  1
 312 13  3 
 1  u4
 13  
8 12 8 8 6 8

 8  312 1
8
3 
12
13 
8

6  1
8
3 
6   v4

Non-zero DoF: u2, v2, u3 και v3.

28
 Orthogonal elements – Example 6.3
y

• Nodal forces [FN] from eq. 5.27: f

4 3
ND
V   (Fixui  Fiy v i )  {Q s } T {FN } 1m
i1 1m

{FN }  [F1x F1y FNDx FNDy ]T 1 2 f

x

{ FN} = {F1x, F1y, F2x, F2y, F3x, F3y, F4x, F4y}T

= {0, 0, 100000,0,-100000, 0, 0, 0}T

0 fx1
0 fy1
100000 fx2

{fN(e)}= 0 fy2
−100000 fx3 Δείκτες
0 fy3
0 fx4
0 fy4
29
 Orthogonal elements – Example 6.3

• Formulation of the system equations:

 4.89 1.67 0.44 0.33  u2   100,000 

 1.67 4.89  
0.33 2.89  v 2  
 0 

108      
 0.44 0.33 4.89 1.67  u3  100,000 
 
 0.33 2.89 1.67 4.89  v 3  
  0 

• Displacement calculations:
y
u2  0.4091mm, v 2  0.4091mm
f
u3  0.4091mm, v 3  0.4091mm
4 3

1m
1m

1 2 f
x
30
 Orthogonal elements – Example 6.3
• Strains εxx, εyy and γxy ({ε} from eq. 6.19 and srains from 5.5:
0 
0 
 
0.41 
y  1 0 1 y 0 y 0 y 0   0.41 10 3 (1  2y) 
0.41   
{ }   0 x 1 0 x 0 x 0 1 x   3

 0.41  
10 0

 x  1 y  1  x 1  y x y 1  x  y    0.41 10 (1  2x) 

3

0.41 
 
 0 
0 
 xx   1 0.25 0  0.41 10 3 (1  2y) 4.4(1  2y)
  1010
0.25    
 yy   2 
1 0
  0    1.1(1  2y)  MPa
  1  0.25
  0 0 0.375  0.41 10 3 (1  2x) 1.6(1  2x) 
 xy

Αρχική
γεωμετρία

Παραμορφωμέν
η γεωμετρία

31
 Orthogonal elements – Example 6.3

The quadrilateral element cannot reproduce bending because of the

linear sides. This is the results of the linear deformation in the element.
The element is deformed as a trapezoidal geometry and as a result
non-zero shear stresses and strains in the y direction develop.
Comparing to the solution with the analytical one is observed a
discrepancy of:
(xx)max /(xx)exact = 4.364/6.0 (73%)
Therefore the element shows a greater stiffness from the expected one.

Αρχική
γεωμετρία

Παραμορφωμέν
η γεωμετρία

32
 Quadrilateral elements

Chapter Summary
Quadrilateral orthogonal elements. Formulation:
• Interpolation (shape) function
• Strain for stress calculation


• Stiffness matrix (through the TMTPE)
• Force vector (through TMTPE)
Quadrilateral isoparametric elements. Formulation:
• Interpolation function and geometry
• Strain for stress calculation
• Jacobian matrix and determinant (Ιακωβιανού και ορίζουσας)
• Stiffness matrix (through the TMTPE)
• Force vector (through TMTPE)
• Numerical integration (Gauss Quadrature)

33
 Quadrilateral isoparametric elements

• Quadrilateral elements
– The most common element in FEM for non-uniform geometries.

- For the 2D problem of the plate under tension with a hole in the centre
ADINA automatically uses mostly 4-sided elements.

34
 Quadrilateral isoparametric elements
• General 4-sided geometry
– Generalisation of the element is needed because of the difference in
the geometry between the elements. For this reason a mapping
method is developed for the transformation of the general 4sided to an
orthogonal quadrilateral element (στοιχείο αναφοράς - reference
element).
– With mapping, all elements with different geometries are transformed
to rectangular elements.
– Relationships between (x, y) and (s, t) must be formulated.
– All formulations are performed
v
in the reference element.
3
y v4 t
3 u3
4 4 (-1,1) 3 (1,1)
u4

v1
s

1 u1 v2

u2 1 (-1,-1) 2 (1,-1)
2 x
Physical element (x, y) Reference element (s, t) 35
 Quadrilateral isoparametric elements
• All angles should be smaller than 180ο
• Nodal order as in the rectangular element.
• Every node has 2 DoF, therefore, the element has 8 ΒΕ.
• The centre of the s-t coordinate system is in the center of the element.
• The sides range from -1 to1.

v3
y v4 t
3 u3
4 4 (-1,1) 3 (1,1)
u4

v1
s

1 u1 v2

u2 1 (-1,-1) 2 (1,-1)
2 x

Physical element (x, y) Reference element (s, t) 36

 Quadrilateral isoparametric elements
• Mapping should ensure the mapping of all physical elements at the
reference element.
• The shape functions are defined in the reference element, and apply
to all physical elements.
• The shape functions in the isoparametric element point to the
deformation and location of the physical element.
• The first node is located at the bottom left corner.

v3
y v4 t
3 u3
4 4 (-1,1) 3 (1,1)
u4

v1
s

1 u1 v2

u2 1 (-1,-1) 2 (1,-1)
2 x
Physical element (x, y) Reference element (s, t) 37
 Quadrilateral isoparametric elements
• Shape function formulation
– Shape functions N1, N2, N3, and N4 are formed at the issoparametric
element (reference)for s and t.
– N1, N2, N3, and N4 are formulated using Larange interpolations similarly
to the rectangular element (εξ. 6.17).
t
 1
 N1  n1(x)n1(y)  A (x 3  x)(y 3  y) 4 (-1,1) 3 (1,1)

N  n (x)n (y)   1 (x  x)(y  y)
 2 2 1
A
1 3


N  n (x)n (y)  1 (x  x)(y  y) s 1 t 1 s
 3 3 4
A
1 1
N1(s,t)  
 1
N4  n4 (x)n4 (y)   (x 3  x)(y1  y)
1  1 1  1
 A
 1
 N1(s,t)  4 (1  s)(1  t) 1 (-1,-1) 2 (1,-1)

N (s,t)  1 (1  s)(1  t)
 2 4
 (6.24)
N (s,t)  1 (1  s)(1  t)
 3 4
 1
N4 (s,t)  (1  s)(1  t)
 4 38
 Quadrilateral isoparametric elements
• Isoparametric mapping
– 4 nodes from the physical element are mapped in the 4 nodes of the
reference element.
– Every point within the physical element is mapped within the reference
element.

y t
3
4 4 (-1,1) 3 (1,1)

1
1 (-1,-1) 2 (1,-1)
2 x

• The mapping is proportional

- For known values of (x,y), the (s,t) values are found.
- For known (s,t), the (x,y) values are found.
39
 Quadrilateral isoparametric elements
• The shape functions in the isoparametric element map the
displacement of the element also and its physical location:

Displacement Interpolation
 u1  Geometry interpolation  x1 
u  x 
   
u(s,t)  [N1 N2 N3 N4 ]  2  x(s,t)  [N1 N2 N3 N4 ]  2 
u3   x3 
u4   x 4 
(6.25) (6.26)

 v1   y1 
v  y 
   
v(s,t)  [N1 N2 N3 N4 ]  2  y(s,t)  [N1 N2 N3 N4 ]  2 
v 3   y3 
v 4   y 4 

• At node 1 (s,t) = (-1, -1) , N1 =1 and N2= N3= N4 = 0, therefore x(-1,-1)=x1,

και y(-1,-1)=y1 .
• For known (s,t), x(s,t), y(s,t), u(s,t) and v(s,t) can be calculated.
• The calculation of (s,t), for known (x,y) it is not straight forward. 40
 Quadrilateral isoparametric elements – Geometry interpolation
- Example 6.4: calculation of (x,y) and (s,t)
• Calculation of the point at the physical element
– At Α, (s, t) = (0.5, 0.5). Shape functions:
1 3 9 3
N1( 21 , 21 ) 
, N2 ( 21 , 21 )  , N3 ( 21 , 21 )  , N4 ( 21 , 21 ) 
16 16 16 16
– Physical coordinated (x, y):
4
1 3 9 3
x( 21 , 21 )   NI ( 21 , 21 )xI  6  4 2  0  2.25 (6.27)
I1 16 16 16 16
4
1 3 9 3
y( , )   NI ( 21 , 21 )yI 
1 1
2 2 0  4 4 0  3 (6.28)
I1 16 16 16 16

y t
3 (2,4) 4 (−1,1) 3 (1,1)
2 (4,4)
A (.5,.5)
B (1,2) s

x
4 (0,0) 1 (6,0) 1 (−1,−1) 2 (1,−1)
41
 Quadrilateral isoparametric elements – Geometry interpolation
- Example 6.4: calculation of (x,y) and (s,t)
• Mapping of B at the reference element
– At B, (x, y) = (1, 2)
4
x  1   NI (s,t)xI  41 (1  s)(1  t)  6  41 (1  s)(1  t)  4
I1

 41 (1  s)(1  t)  2  41 (1  s)(1  t)  0
 st  2t  3
4
y  2   NI (s,t)yI  41 (1  s)(1  t)  0  41 (1  s)(1  t)  4
I1

 41 (1  s)(1  t)  4  41 (1  s)(1  t)  0
 2  2s

1) st-2t+3=1 t=1
2) 2+2s=2 s=0

– Άρα, B, (x, y) = (1, 2) (s, t) = (0, 1)

42
 Quadrilateral isoparametric elements – Geometry interpolation
- Example 6.4: calculation of (x,y) and (s,t)
• Mapping of B at the reference element
– Therefore, B, (x, y) = (1, 2) (s, t) = (0, 1)

y t
3 (2,4) 4 (−1,1) 3 (1,1)
2 (4,4)
A (.5,.5)
B (1,2) s

x
4 (0,0) 1 (6,0) 1 (−1,−1) 2 (1,−1)

43
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and determinat

Shape functions are defined in terms of (s,t). For the strain and stress
calculation the differentiation of u(s,t) and v(s,t) in terms of (x,y) is needed,
i.e. differentiation of Ν (εξ. 6.24) in terms of (x,y). Since Ν=Ν(s,t), s=s(x,y)
και t(x,y), i.e. Ν(s(x,y), t(x,y)), the chain rule is used:

NI NI x NI y NI NI x NI y

    (6.29)
s x s y s t x t y t
In matrix form:

 NI   x y   NI   NI 


 s   s s   x 
  x 
 
     [ J]   (6.30)
 NI   x 
y   I 
N N
 I

 t   t t  
 y 
  y 
 

Matrix [J] is the Jacobian (Ιακωβιανό μητρώο) of mapping.

44
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and determinat

Calculation of Jacobian [J]. From 6.26:

4
1 1 1 1
x(s,t)= NI s,t xI = 1 − 𝑠 1 − 𝑡 x1 + 1 + 𝑠 1 − 𝑡 x2 + 1 + 𝑠 1 + 𝑡 x3 + 1 − 𝑠 1 + 𝑡 x4
4 4 4 4
I=1
(6.31)
4
1 1 1 1
y(s,t)= NI s,t yI = 1 − 𝑠 1 − 𝑡 y1 + 1 + 𝑠 1 − 𝑡 y2 + 1 + 𝑠 1 + 𝑡 y3 + 1 − 𝑠 1 + 𝑡 y4
4 4 4 4
I=1

Differentiate x(s,t) and y(s,t) in terms of s and t:

x 4
N 1 t
  I xI  (  x1  x 2  x 3  x 4 )  (x1  x 2  x 3  x 4 )
s I1 s 4 4
x 4
N 1 s
  I xI  (  x1  x 2  x 3  x 4 )  (x1  x 2  x 3  x 4 )
t I1 t 4 4
4
(6.32)
𝜕y 𝜕NI 1 t
= yI = (−y1 + y2 + y3 − y4) + (y1 − y2 + y3 − y4)
𝜕s 𝜕s 4 4
I=1
4
𝜕y 𝜕NI 1 s
= y = (−y1 − y2 + y3 + y4) + (y1 − y2 + y3 − y4)
𝜕t 𝜕s I 4 4
I=1
45
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and determinat
• The shape differentiations in terms of (x,y) are needed, therefore eq. 6.30
is reformulated:

 NI   NI   y y   NI 

 
 x  
1  s 
 1  t s   s 

   [ J]      (6.33)
N
 I  NI  J   x x   NI 
 y 
  
 t   t s  
 t 


• Determinant |J| του Ιακωβιανού μητρώου [J] (Ιακωβιανό ή Jacobian):

x y x y (6.34)
J 
s t t s
• What if |J| = 0 ή |J| < 0?
– No differentiation is possible when |J| = 0 at any point in the element
– The mapping relations between (x, y) and (s, t) are not valid if |J| = 0 or
|J| < 0 at any point in the element where (–1 ≤ s, t ≤ 1).
46
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and
determinant

• The Jacobian is an important criterion for the assessment of the validity of

the mapping and quality of the element.
• Every point in the reference element should correspond to one element in
inside of the physical element. Κάθε σημείο του στοιχείου αναφοράς
πρέπει να αντιστοιχεί σε σημείο στο εσωτερικό του φυσικού στοιχείου.
• When internal point at the reference element (s, t) corresponds to a point
outside the physical element (x, y) the Jacobian becomes negative.
• When multiple points at the reference element (s, t) correspond to one
point in the physical element (x, y) the Jacobian becomes 0 at this point.
• Therefore it is important the physical elements to acquire geometries
which will have positive Jacobian everywhere in the element.

47
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and
determinant – Example 6.5: Assessment of mapping validity and
element quality
Jacobian should be non-zero in all points in the reference element (-1 ≤ s, t ≤
1)
y
• Coordinates at (x,y): 3(2, 2)
x1  0, x 2  1, x 3  2, x 4  0
y1  0, y 2  0, y 3  2, y 4  1
4(0, 1)
• Isoparametric mapping (εξ. 6.31):
4
1 x
x   NI xI  N2  2N3  (3  3s  t  st) 1(0, 0) 2(1, 0)
I1 4
4
1
y   NI yI  2N3  N4  (3  s  3t  st)
I1 4
• Jacobian:  x y 
 s s  1 3  t 1  t 
[ J]   
 x y  4 1  s 3  s 
 t t  48
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and
determinant – Example 6.5: Assessment of mapping validity and
element quality
• Determinant (Ιακωβιανό) from εξ. 6.34:
1 1 1 1
J [(3  t)(3  s)  (1  t)(1  s)]   s  t
4 2 8 8

– Note |J| > 0 για –1 ≤ s ≤ 1 and –1 ≤ t ≤ 1.

Σταθερό s
t
4 (-1,1) 3 (1,1)

Σταθερό t
1 (-1,-1) 2 (1,-1)

Acceptable quality
49
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and
determinant – Example 6.6: Assessment of mapping validity and
element quality y
• Coordinates at (x,y): 4(0, 5) 3(5, 5)
x1  0, x 2  1, x 3  5, x 4  0
y1  0, y 2  4, y 3  5, y 4  5 2(1, 4)

• Mapping (εξ. 6.31):

4
1
x   NI xI  (1  s)(3  2t)
I1 2
4
1
y   NI yI  (7  2s  3t  2st) 1(0, 0)
I1 2 x

• Determinant:

1
J (5  10s  10t) |J| = 0 at 5 – 10s + 10t = 0, for s – t = 1/2
4

50
 Quadrilateral isoparametric elements – Differentiation of the shape
functions for the strains and stresses. - Jacobian matrix and
determinant – Example 6.6: Assessment of mapping validity and
element quality
Constant s
• In general the geometry of the physical
element is not valid when the Jacobian
(determinant) is zero or negative
enywhere in the reference element. Constant t
• Problems also appear when the
Jacobian approaches zero due to
rounding of values (bad geometry. Invalid mapping
• To avoid problems due to bad geometry
is suggested to use elements with o
internal angles > 15˚and < 165˚ > 15

< 165o

51
 elements – Displacement interpolation
INTERPOLATION
Quadrilateral isoparametric

• Displacement interpolation for the 8-node  u1 

isoparametric(8-DOF) is similar to the v 
rectangulaer element:  1
 u2 
 
u  N1 0 N2 0 N3 0 N4 0  v 2 
      [N]{q} (6.35)
v   0 N1 0 N2 0 N3 0 N4  u3 
v 3 
 
u 4 
Shape functions: v 4 

• Interpolation in the (s, t) system

• The behaviour of the isoparametric element is similar to the rectangular
(slide 10 – Ιδιότητες συναρτήσεων μορφής) since both elements are
defined based on the Lagrange interpolations. 52
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations
• Strain {ε} is defined by the differentiation of displacements in terms of
(x,y):

 xx   u x 
u x  1 0 0 0  
      u y 
{}   yy    v y 
   v 
0 0 0 1 (6.36)
  u y  v x  0 1 1 0   x 
  xy      v
 y 
Displacements u and v are defined in terms of (s,t), and cannot be calculated
directly in terms of (x,y), therefore the inverse Jacobian is used:

∂u ∂y ∂y ∂u ∂v ∂y ∂y ∂v
∂x 1 ∂t − ∂s ∂s ∂x 1 ∂t − ∂s ∂s (6.37)
∂u = J ∂x ∂x ∂u ∂v = J ∂x ∂x ∂v
− −
∂y ∂t ∂s ∂t ∂y ∂t ∂s ∂t

Note.: The shape functions are the same as in εξ. 6.25 και 6.26, and
therefore eq. 6.33 is used.
53
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations
• Eq. 6.37 combined to one:

u x   y t  y s 0 0  u s 
u y  1   x t x s 0  u t 
 0
   (6.38)
 
v x  J  0 0 y t  y s  v s 
v  
y   0 0  x t x s  v t 

• Strain {ε} is defined in terms of (s,t)

 xx   y t  y s 0 0  u s  u s 
1 0 0 0 
  1   x t x s 0 0  u t  u
 t 
 yy   0 0 0 1      [ A]   (6.39)
  J 0 1 1 0   0 0 y t  y s  v s  v s 
  xy    0
 0  x t x s

 v t  v t 

Εξ.6.34 Υπολογίζεται από την παραγώγιση

Εξ.6.32 της εξ. 6.35 ως προς (s,t), δηλ.,
ομοίως του αποτελέσματος στην εξ.
6.32 με τα x και y αντικατεστημένα
από u και v. 54
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations
 u1 
• Matrix form v 
 1
u s   1  t 0 1 t 0 1 t 0 1  t 0   u2 
u t    
  1  1  s 0 1  s 0 1 s 0 1 s 0  v 2 
      [G]{q}
v s  4  0 1  t 0 1 t 0 1 t 0 1  t  u3 
v t   
 0 1  s 0 1  s 0 1 s 0 1  s  v 3 
  (6.40)
• Strain of element: u 4 
v 4 

 xx  u s 
  u t 
(6.41)


 yy   [ A ]    [ A][G]{q}  [B]{q} Μητρώο τροπής-μετατόπισης
  v s 
  xy  v t 

• Matrix [B] needs to be calculated because the [A] matrix is needed which
requires the inverse of the Jacobian matrix.
• Matrix [B] is not constant and therefore the strains and stresses will vary
in the element.
55
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
Displacements at the nodes of the 4-sided given as:

{q} = {u1, v1, u2, v2, u3, v3, u4, v4} = {0, 0, 1, 0, 2, 1, 0, 2}

Calculate the strains and the stresses for the point (s,t)=(1/3, 0) in the
reference element.
y t
4 (0,2) 3 (3,2) 4 (-1,1) 3 (1,1)

s
x
1 (0,0) 2 (3,0) 1 (-1,-1) 2 (1,-1)

56
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
1
Strain (εξ.6.41) : {ε}= [A][G]{q}
J
 xx   y t  y s 0 0  u s  u s 
1 0 0 0  
  1   x t x s 0 0  u t  u
 t  (6.39)
 yy   0 0 0 1      [ A]  
  J 0 1 1 0   0 0 y t  y s  v s  v s 
  xy    0
 0  x t x s

 v t  v t 

Συντεταγμένες (s,t)
Εξ.6.34 Εξ.6.32 γνωστές: (s,t)=(1/3, 0)
x y x y x N
J 
4
1 t
  I xI  (  x1  x 2  x 3  x 4 )  (x1  x 2  x 3  x 4 )
s t t s s I1 s 4 4
x 4
N 1 s
  I xI  (  x1  x 2  x 3  x 4 )  (x1  x 2  x 3  x 4 )
t I1 t 4 4
4
Συντεταγμένες 𝜕y 𝜕NI 1 t
= yI = (−y1 + y2 + y3 − y4) + (y1 − y2 + y3 − y4)
(x,y) από τη 𝜕s 𝜕s 4 4
I=1
γεωμετρία του 4
𝜕y 𝜕NI 1 s
φυσικού 𝜕t
=
𝜕s
yI = (−y1 − y2 + y3 + y4) + (y1 − y2 + y3 − y4)
4 4
στοιχείου I=1 57
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
1
Strain (εξ.6.41): {ε}= [A][G]{q}
J
 xx   y t  y s 0 0  u s  u s 
1 0 0 0  
  1   x t x s 0 0  u t  u
 t  (6.39)
 yy   0 0 0 1      [ A]  
  J 0 1 1 0   0 0 y t  y s  v s  v s 
  xy    0
 0  x t x s

 v t  v t 

Εξ.6.34 Από εξ.6.32

Εξ.6.40
 u1 
v 
 1
u s   1  t 0 1 t 0 1 t 0 1  t 0   u2 
u t    
  1  1  s 0 1  s 0 1 s 0 1 s 0  v 2 
      [G]{q}
v s  4  0 1  t 0 1 t 0 1 t 0 1  t  u3 
v t   
 0 1  s 0 1  s 0 1 s 0 1  s  v 3 
 
u 4 
Συντεταγμένες (s,t) Μετατοπίσεις γνωστές: v 4 
γνωστές: (s,t)=(1/3, 0) {q}={0, 0, 1, 0, 2, 1, 0, 2} 58
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
Calculation of εξ. 6.32 x 4
N 1 t
 I
xI  (  x1  x 2  x 3  x 4 )  (x1  x 2  x 3  x 4 )
s I1 s 4 4
x 4
N 1 s
x1=0, x2=3, x3=3, x4=0   I xI  (  x1  x 2  x 3  x 4 )  (x1  x 2  x 3  x 4 )
t I1 t 4 4
y1=0, y2=0, y3=2, y4=2 4
𝜕y 𝜕NI 1 t
s=1/3 = yI = (−y1 + y2 + y3 − y4) + (y1 − y2 + y3 − y4)
𝜕s 𝜕s 4 4
t=0 I=1
4
𝜕y 𝜕NI 1 s
= y = (−y1 − y2 + y3 + y4) + (y1 − y2 + y3 − y4)
𝜕t 𝜕s I 4 4
I=1

𝜕x 1 0 3
= −0 + 3 + 3 − 0 + 0 − 3 + 3 − 0 =
𝜕s 4 4 2
𝜕x 1 1
= −0 − 3 + 3 + 0 + 0−3+3−0 =0
𝜕t 4 12
𝜕y 1 0
= −0 + 0 + 2 − 2 + 0 − 0 + 2 − 2 = 0
𝜕s 4 4
𝜕y 1 1
= −0 − 0 + 2 + 2 + 0−0+2−2 =1
𝜕t 4 12 59
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
• Calculation of Jacobian matrix, determinant, and its inverse:

x y x y
J  =3/2
s t t s

𝜕𝑦 𝜕𝑦
1 𝜕𝑡 − 1 0 2
J −1 = 𝜕𝑠 = 2 3 = 3 0
J 𝜕𝑥 𝜕𝑥 3 0
− 2 0 1
𝜕𝑡 𝜕𝑠

– The Jacobian is positive and the mapping is valid

– The Jacobian matrix is constant within the element and has values in
the diagonal therefore the element is rectangular.

60
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
Calculation of [G]{q}:
 u1 
v 
 1
u s   1  t 0 1 t 0 1 t 0 1  t 0   u2 
u t    
  1  1  s 0 1  s 0 1 s 0 1 s 0  v 2 
      [G]{q}
v s  4  0 1  t 0 1 t 0 1 t 0 1  t  u3 
v t   
 0 1  s 0 1  s 0 1 s 0 1  s  v 3 
 
u 4 
v 4 
(s,t)=(1/3, 0) {q}={0, 0, 1, 0, 2, 1, 0, 2}
𝜕𝑢 0
𝜕𝑠 0
𝜕𝑢 −1 0 1 0 1 0 −1 0 1 0.75
𝜕𝑡 = 1 −0.67 0 −1.33 0 1.33 0 0.67 0 0
=
0.33
𝜕𝑣 4 0 −1 0 1 0 1 0 −1 2 −0.25
𝜕𝑠 0 −0.67 0 −1.33 0 1.33 0 0.67 1 0.67
𝜕𝑣 0
𝜕𝑡 2
61
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
1
Strain (εξ.6.41): {ε}= [A][G]{q}
J

 xx   y t  y s 0 0  u s  u s 
1 0 0 0  
  1   x t x s 0 0  u t  u
 t  (6.39)
 yy   0 0 0 1      [ A ]  
  J 0 1 1 0   0 0 y t  y s  v s  v s 
  xy    0
 0

 x t x s  v t  v t 

𝜀𝑥𝑥 1 0 0 0 0.75
2 1 0 0 0 0 3\2 0 0 0.50
𝜀𝑦𝑦 = 0 0 0 1 0.33
0 = 0.67
0 1
𝛾𝑥𝑦 3 0 1 1 0 0 −0.25
0.17
0 0 0 3\2 0.67

62
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
• For the displacements use eq. 6.25. Calculate the shape functions in the
(s,t)=(1/3, 0):

1 1
N1(s,t)  (1  s)(1  t) N2 (s,t)  (1  s)(1  t)
4 4
1 1
N3 (s,t)  (1  s)(1  t) N4 (s,t)  (1  s)(1  t)
4 4
 u1 
1 1 1 1 u 
N1  , N2  , N3  , N4   
6 3 3 6 u(s,t)  [N1 N2 N3 N4 ]  2 
u3 
From u(s,t) and v(s,t) the u4 
displacement are calculated:
 4
1 1 1 1  v1 


u   Nu
I I 
6
 0 
3
 1 
3
 2 
6
0 1
v 

I1
 
4 v(s,t)  [N1 N2 N3 N4 ]  2 
v  N v  1  0  1  0  1  1  1  2  2
  I I v 3 
I1 6 3 3 6 3 v 4 
63
 Quadrilateral isoparametric elements - Displacement interpolation
- Strain equations – Example 6.7.
• Points (x,y) calculated from eξ. 6.26 (6.31):

 4
1 1 1 1


x  
I1
N x
I I 
6
 0 
3
 3 
3
 3 
6
0  2
 4
y  N y  1  0  1  0  1  2  1  2  1
 
I1
I I
6 3 3 6

y t
4 (0,2) 3 (3,2) 4 (-1,1) 3 (1,1)

s
x
1 (0,0) 2 (3,0) 1 (-1,-1) 2 (1,-1)

64
 Quadrilateral isoparametric elements

Περίγραμμα κεφαλαίου
Τετράπλευρα ορθογώνια στοιχεία
• Διατύπωση συνάρτησης παρεμβολής μετατόπισης
• Διατύπωση τροπής για υπολογισμό τάσεων


• Διατύπωση μητρώου ακαμψίας (μέσω ΕΟΔΕ)
• Διατύπωση διανύσματος δυνάμεων (μέσω ΕΟΔΕ)
Τετράπλευρα ισοπαραμετρικά στοιχεία
• Διατύπωση συνάρτησης παρεμβολής μετατόπισης και γεωμετρίας
• Διατύπωση τροπής για υπολογισμό τάσεων
• Διατύπωση Ιακωβιανού μητρώου και ορίζουσας
• Διατύπωση μητρώου ακαμψίας (μέσω ΕΟΔΕ)
• Διατύπωση διανύσματος δυνάμεων (μέσω ΕΟΔΕ)
• Διατύπωση αριθμητικής ολοκλήρωσης (Gauss Quadrature)

65
 Quadrilateral isoparametric elements – Stiffness matrix formulation

• The strain and stress equations require the known values for the {q(e)}
vector. TMTPE will be used
• The stiffness matrix is defined from the strain energy as in the triangular
element.

h
2 
U(e)  {  } T
[ C]{  } dA (e)

h (e) T
 {q }  [B]8T3 [C]33 [B]38 dA{q(e) } (6.42
2 A

1 (e) T (e)
 {q } [k ]88 {q(e) }
2
• [k(e)] is the stiffness matrix of the element
For the integration of the eq 6.42 there are two basic difficulties:
1. The geometry is, therefore cannot be done analytically.
2. The displacement/strain matrix is defined in the (s,t) coordinates
whereas the area Α of the 4-sieded in the (x,y) coordinates.

66
 Quadrilateral isoparametric elements – Stiffness matrix formulation

• One way to overcome the mentioned problems is to integrate in the

(s,t) system.
• Therefore a change in the variable A is needed ως προς τις (s,t)
coordinates:
x και y are functions of s and t. If vectors 𝑑r 1 and 𝑑r 2 are defined in terms
of (s,t) then the analogy dA= 𝑑r 1 x 𝑑r 2 ισχύει.

r =xi+yj, dr =dxi+dyj
y t=ti
Since x(s,t) και y(s,t):
𝜕𝑥 𝜕𝑥
s=si dx= 𝜕𝑠 𝑑𝑠 + 𝜕𝑡 𝑑𝑡
t=t2
𝜕𝑦 𝜕𝑦
t=t1 dy= 𝜕𝑠 𝑑𝑠 + 𝜕𝑡 𝑑𝑡
𝑑r 1 dΑ ∂x ∂x ∂y ∂y
dr 1= ds+ dt i+ ds+ dt j
∂s ∂t ∂s ∂t
𝑑r 2 s=s2 In r 1 direction t is constant therefore dt=0:
∂x ∂y
s=s1 dr 1= ds i+ ds j
∂s ∂s
x
67
Quadrilateral isoparametric elements – Stiffness matrix formulation

Similarly 𝑑r 2:
∂x ∂y
dr 2= dt i+ dt j
∂t ∂t
The area is defined with the cross product of the vectors
∂x ∂y ∂y ∂x
dA =𝑑r 1 x 𝑑r 2 = dsdt − dsdt k
∂s ∂t ∂s ∂t

y t=ti
∂x ∂y ∂y ∂x
dA =dA= − dsdt
s=si ∂s ∂t ∂s ∂t
t=t2
∂x ∂y ∂y ∂x
t=t1 dA= − dsdt
∂s ∂t ∂s ∂t
𝑑r 1 dΑ
(6.43)
dA=dxdy= J dsdt
𝑑r 2 s=s2
s=s1
x
68
 Quadrilateral isoparametric elements – Stiffness matrix formulation

• Change of variable:

1 1
[k(e) ]  h [B]T [C][B]dA  h   [B]T [C][B] J dsdt (6.44)
A 1 1

dA = J dsdt

The change of variable dA to J dsdt improved the integration however, the

integration is not simple because the variables cannot be expressed as
separate variables of s και t. In addition matrix [Β] involves the inverse of the
Jacobian matrix which is difficult to be integrated.

One option is to used numerical integration (αριθμητική ολοκλήρωση

βασισμένη στη μέθοδο Gauss Quadrature).

69
 Quadrilateral isoparametric elements – Formulation of the force vectors
using the potential energy of element
Nodal forces and distributed forces as in the triangular element
• Nodal forces - Δυνάμεις κόμβων (εξ. 5.27):
ND= συνολικός αριθμός κόμβων στο στερεό
ND
V   (Fixui  Fiy v i )  {Q s } T {FN } {FN }  [F1x F1y FNDx FNDy ]T
i1

• Nodal forces from uniformly distributed loads in the side of element (εξ.
5.30, 5.31): (l  s) / l 0  Tx 
 0 T 
(l  s) / l  y
 
l l s/l 0  Tx  hl Tx 
{fT }  h [N] {T } ds  h  
(e) T
 T  ds   
0 0
 0 s /l  y 2 Ty 
 0 0  0
   
 0 0   0 

NS
V   {q(e) } T {fT(e) }  {Q s } T {FT } NS= συνολικός αριθμός πλευρών όπου
e 1 ασκούνται κατανεμημένα φορτία στο στερεό 70
Quadrilateral isoparametric elements – Formulation of the force vectors
using the potential energy of element
• Body forces, b = {bx, by}T, nodal forces:

b x  b x 
V (e)
 h [u v]   dA  {q } h [N] dA  
(e) T T
(6.22)
A b y  A b y 
 {q(e) } T {fb(e) }

• In the present case the integration should be performed in the (s,t) since
the [Ν] is a function of (s,t). The change of variable is performed using εξ.
6.43:

dA = J dsdt
1
bx
V (e) =− q e Th N T J dsdt b = q e T fb e (6.45)
y
−1
• Note.: For uniformly distributed forces the load is not equally distributed at
the nodes because the matrix J is not constant within the element.
71
Quadrilateral isoparametric elements – Formulation of the force vectors
using the potential energy of element - TMTPE
• Potential energy due to body forces {fb(e)} from all elements should be
assembled as in eq. 5.32:
NE
V   {q(e) } T {fb(e) }  {Q s } T {FB }
e 1

• The TMTPE is used for the formulation of the system equations as in the
triangular element:

- Total potential energy

1
  U  V  {Q s } T [Ks ]{Q s }  {Q s } T {FN  FT  FB } (5.33)
2

- TMTPE for the calculation of the displacements


 0  [K s ]{Q s }  {FN  FT  FB } (5.34)
 {Q s }

72
 Quadrilateral isoparametric elements – Strain and stress calculation

• Strain from eq. 5.32:

 xx   y t  y s 0 0  u s  u s 
1 0 0 0  
  1   x t x s 0 0  u t  u
 t  (6.39)
 yy   0 0 0 1      [ A ]  
  J 0 1 1 0   0 0 y t  y s  v s  v s 
  xy    0
 0

 x t x s  v t  v t 

• Stress from:

 xx  1  0   xx 
  E  1    (5.5)

 yy     {}  [C ]{}
  yy 
0
1   2 
  0 0  
 xy  2 (1   )   xy 
1

 xx  1    0   xx 
  E     

 yy   1  0  yy   {}  [C ]{}
  (5.7)
  (1   )(1  2 )  0     
 xy     xy 
1
 0 2

73
 Quadrilateral isoparametric elements

Chapter Summary
Quadrilateral orthogonal elements. Formulation:
• Interpolation (shape) function
• Strain for stress calculation


• Stiffness matrix (through the TMTPE)
• Force vector (through TMTPE)
Quadrilateral isoparametric elements. Formulation:
• Interpolation function and geometry
• Strain for stress calculation
• Jacobian matrix and determinant (Ιακωβιανού και ορίζουσας)
• Stiffness matrix (through the TMTPE)
• Force vector (through TMTPE)

• Numerical integration (Gauss Quadrature)

74
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature
• The stiffness matrix and the vector of forces due to body forces involve εξ.
6.44 and 6.45 involve integrations in the (s,t) field from -1 to1.
• The integral is extremely difficult to be performed analytically
• There are various methods for numerical integration: in the present case
Gaussian Quadrature will be used, which is widely accepted in FEM.
• Definition of the Gauss Quadrature for one dimension:
1 n
I   f(s)ds   w i f(si )
1 i1

(6.46)
– The f(s) function is integrated for s=-1 to 1 using specific values and
weighting coefficients.
– si: integration points (σημεία ολοκληρώματος Gauss)
– wi: integration coefficients (συντελεστές ολοκληρώματος Gauss)
– f(si): values of the function at the integration points (τιμή συνάρτησης στα
σημεία ολοκληρώματος Gauss)
– n: total number of points (συνολικός αριθμός σημείων ολοκληρώματος).
75
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature
• The points of integration and the coefficients are chosen from the table
bellow.
Exact for degree -
Coefficients -
Points - Σημεία Ολοκλήρωμα ακριβές
n Συντελεστές
ολοκληρώματος (si) για πολυώνυμο
στάθμισης (wi)
βαθμού:
1 0.0 2.0 1
2 .5773502692 1.0 3
.7745966692 .5555555556 5
3
0.0 .8888888889
.8611363116 .3478546451 7
4
.3399810436 .6521451549
.9061798459 .2369268851 9
5 .5384693101 .4786286705
0.0 .5688888889
• The points and the coefficients have been calculated so that if f(s) is of
2n-1 degree the integration is exact.
76
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature - – Example 6.8
• Constant f(s) = 4.
– Use one point s1 = 0 with coefficient w1 = 2
1
I   4ds  w1f(s1 )  2  4  8
1

– The integration is exact.

• Linear function (1st degree): f(s) = 2s + 1
– 2n-1, therefore n=1, point s1 = 0 and coefficient w1 = 2

1
I   (2s  1)ds  w1f(s1 )  2  1  2
1

– Integral is exact.
• Gaussian Quadrature of one point integrates constant and 1st degree
functions exactly.

77
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature – Example 6.8
• Parabolic function (2ου βαθμού): f(s) = 3s2 + 2s + 1
– For Gauss Quadrature of one point:

1
I   (3s2  2s  1)ds  4
1

w1f(s1 )  2  1  2

– Exact solution.
– Two point: w1 = w2 = 1 and -s1 = s2 = 5773502692

w1f(s1 )  w 2 f(s2 )  1 f(  1
3
)  1 f( 13 )
 3  31  2
3
 1  3  31  2
3
1 4

• The points and the coefficients have been calculated so that if f(s) is of
2n-1 degree the integration is exact.
78
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature – Example 6.8
• 7th degree:
1
I   (8x 7  7x 6 )dx  2
1

– With 1 point Gauss:

s1  0, f(s1 )  0, w1  2
I  w1f(s1 )  2  0  0

– With 2 point Gauss:

s1  .577, f(s1 )  8( .577)7  7( .577)6  .0882, w 1  1

s2  .577, f(s2 )  8(.577)7  7(.577)6  .4303, w2  1
I  w1f(x1 )  w 2 f(x 2 )  .0882  .4303  .5185

79
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature – Example 6.8
– 3 point Gauss:

s1  .577, f(s1 )  8( .577)7  7( .577)6  .0882, w1  1

s2  .577, f(s2 )  8(.577)7  7(.577)6  .4303, w2  1
I  w1f(x1 )  w 2 f(x 2 )  .0882  .4303  .5185

– 4 point Gauss:

s1  .8611, f(s1 )  .0452, w 1  .3479

s2  .3400, f(s2 )  .0066, w 2  .6521
s3  .3400, f(s3 )  .0150, w 3  .6521
s3  .8611, f(s3 )  5.6638, w 3  .3479
I  w1f(s1 )  w 2 f(s2 )  w 3 f(s3 )  w 4 f(s 4 )  2.0

– 4 point gives the exact solution.

80
Quadrilateral isoparametric elements – Numerical integration Gaussian
Quadrature – 2D
• Integral in 2D is performed with the combination of two 1D integrals.

1 1 1 m n m
I   f(s,t)dsdt    w f  s ,t  dt   w w f(s ,t )
1 1 1 i1
i i
j1 i1
i j i j (6.47)

– Total number of integration points = m×n.

– n = point number in the t direction
– m = point number in the s direction
– wi and wj = weighting coefficients
t t t

s s s

(a) 11 (b) 22 (c) 33 81

 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix
• Application of the integration on the system stiffness equation. 2x2 points:

1 1
[k ]  h   [B]T [C][B] J dsdt
(e)

1 1
2 2
 h w i w j [B(si ,t j )]T [C][B(si,t j )] J(si,t j ) (6.48)
i1 j1

• Example 6.9
Calculate the stiffness matrix of the rectangular element with thickness 0.1m
using 1x1 Gauss integrals, and 2x2 points. E=10GPa, ν=0.25.

y
2m
4 3
2m
1 2
x 82
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9

The dimensions are the same with those of the reference element.

y
2m t
4 (-1,1) 3 (1,1)
4 3

2m
s
1 2
1 (-1,-1) 2 (1,-1)
x

1 1
[k ]  h   [B]T [C][B] J dsdt
(e)

1 1
2 2
 h w i w j [B(si ,t j )]T [C][B(si,t j )] J(si,t j )
i1 j1

83
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
𝟏 𝟏

Integral Gauss1x1: [𝐤 (𝐞) ] ≈ 𝐡 𝐰𝟏 𝐰𝟏 𝐁 𝐬𝟏 , 𝐭 𝟏 𝐓 [𝐂] 𝐁 𝐬𝟏 , 𝐭 𝟏 J(𝐬𝟏 , 𝐭 𝟏 )

𝐢=𝟏 𝐣=𝟏

Ολοκλήρωμα ακριβές
Σημεία ολοκληρώματος Συντελεστές
n για πολυώνυμο
(si , ti) στάθμισης (wi)
βαθμού:
1 0.0 2.0 1

[𝐤 (𝐞) ] ≈ 𝐡𝒘𝟏 𝒘𝟏 𝐁 𝟎, 𝟎 𝐓[𝐂] 𝐁 𝟎, 𝟎 J(𝟎, 𝟎)

𝐁 = [𝚨] 𝑮

Υπολογίζεται
[𝚨] από την
εξ.6.32

84
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
y
Integral Gauss1x1: 2m
𝜕x 1 0 4 3
= −0 + 2 + 2 − 0 + 0 − 2 + 2 − 0 = 1
𝜕s 4 4 2m
Υπολογίζεται
𝜕x 1 0 από την 1 2
= −0 − 2 + 2 + 0 + 0−2+2−0 =0 x
𝜕t 4 12 εξ.6.32
𝜕y 1 0
= −0 + 0 + 2 − 2 + 0 − 0 + 2 − 2 = 0
𝜕s 4 4
𝜕y 1 0
= −0 − 0 + 2 + 2 + 0−0+2−2 =1
𝜕t 4 12

1 0 0 0
1 1 0 0 0 0 1 0 0 1 1 0 0 0 1 0 0 0
𝐴 = 0 0 0 1 = 0 0 0 1 = 0 0 0 1
J 0 0 0 1 0 J 0 1
1 1 0 0 1 0 0 1 1 0
0 0 1
x y x y
J  =1
s t t s

85
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
Integral Gauss1x1:

𝑮 Υπολογίζεται από την εξ.6.40

−1 0 1 0 1 0 −1 0
1 −1 0 −1 0 1 0 1 0
[G] =
4 0 −1 0 1 0 1 0 −1
0 −1 0 −1 0 1 0 1

𝐁 = [𝚨] 𝑮

−1 0 1 0 1 0 −1 0
1 1 0 0 0 −1 0 −1 0 1 0 1 0
= 0 0 0 1
4 0 1 1 0 −1 0 1 0 1 0 −1
0
0 −1 0 −1 0 1 0 1

1 −1 0 1 0 1 0 −1 0
𝐁 = 0 −1 0 −1 0 1 0 1
4
−1 −1 −1 1 1 1 1 −1

86
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
Integral Gauss1x1:  xx  1  0   xx 
   1  
0   yy   {}  [C ]{}
E

 yy  
𝐂 from εξ.5.5:  
  1 
2
0 0  
2 (1   )   xy 
 xy 
1

1010 1 0.25 0 1.067 0.27 0

C𝝈 = 0.25 1 0 = 1010
0.27 1.067 0
(1 − 0.252) 0 0 0.5(1 − 0.25) 0 0 0.4

[𝐤 (𝐞) ] ≈ 𝐡𝒘𝟏 𝒘𝟏 𝐁 𝟎, 𝟎 𝐓[𝐂] 𝐁 𝟎, 𝟎 J(𝟎, 𝟎)

≈ 𝟎. 𝟒 𝐁 𝟎, 𝟎 𝐓 𝐂 𝐁 𝟎, 𝟎 J 𝟎, 𝟎

−1 −1 −1
0 0 −1
1 −1 −1
1 0 1.067 0.27 0 1 −1 0 1 0 1 0 −1 0
0 1 10
≈ 0.4 10
0.27 1.067 0 0 −1 0 −1 0 1 0 1
0.4 4 −1 −1 −1 1 1 1 1 −1
4 1 1 1
0 0
0 0 1
−1 1 1
0 0 −1

87
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
Integral Gauss1x1:

[𝐤 (𝐞) ] ≈
8x3

−1 0 −1 3x3
0 −1 −1 3x8
1 0 −1
0 −1 1
1.067 0.27 0 −1 0 1 0 1 0 −1 0
≈ 𝟐. 𝟓 𝟏𝟎𝟖
1 0 1
0.27 1.067 0 0 −1 0 −1 0 1 0 1
0 1 1
0 0 0.4 −1 −1 −1 1 1 1 1 −1
−1 0 1
0 −1 −1
−1.067 −0,267 −0.4
−0,267 −1.067 −0.4
0,267 0,267 −0.4
−0,267 −1.067 0.4 −1 0 1 0 1 0 −1 0
≈ 𝟐. 𝟓 𝟏𝟎𝟖
1.067 0,267 0.4 0 −1 0 −1 0 1 0 1
0,267 1.06 0.4 −1 −1 −1 1 1 1 1 −1
−1.067 −0,267 0.4
0,267 1.067 −0.4

88
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
Integral Gauss1x1:
8x3
−1.067 −0,267 −0.4
−0,267 −1.067 −0.4 3x8
0,267 0,267 −0.4
−0,267 −1.067 0.4 −1 0 1 0 1 0 −1 0
≈ 𝟐. 𝟓 𝟏𝟎𝟖
1.067 0,267 0.4 0 −1 0 −1 0 1 0 1
0,267 1.06 0.4 −1 −1 −1 1 1 1 1 −1
−1.067 −0,267 0.4
0,267 1.067 −0.4

0,367 0,167 -0,167 -0,033 -0,367 -0,167 0,167 0,033

0,167 0,367 0,033 0,167 -0,167 -0,367 -0,033 -0,167
0,033 0,033 0,167 -0,167 -0,033 -0,033 -0,167 0,167
[𝐤 (𝐞) ] ≈109 -0,033 0,167 -0,167 0,367 0,033 -0,167 0,167 -0,367
-0,367 -0,167 0,167 0,033 0,367 0,167 -0,167 -0,033
-0,167 -0,367 -0,033 -0,167 0,167 0,367 0,033 0,167
0,167 -0,033 -0,367 0,167 -0,167 0,033 0,367 -0,167
0,033 -0,167 0,167 -0,367 -0,033 0,167 -0,167 0,367

Συμμετρικό

89
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9
𝟐 𝟐

Integral Gauss 2x2: [𝐤 (𝐞) ] ≈ 𝐡 𝐰𝐢 𝐰𝐣 𝐁 𝐬𝐢 , 𝐭 𝐣 𝐓 [𝐂] 𝐁 𝐬𝐢 , 𝐭 𝐣 J(𝐬𝐢 , 𝐭 𝐣 )

𝐢=𝟏 𝐣=𝟏

Ολοκλήρωμα ακριβές
Σημεία ολοκληρώματος Συντελεστές
n για πολυώνυμο
(si, ti) στάθμισης (wi)
βαθμού:
2 .5773502692 1.0 3

𝟐 𝟐

𝐤 𝐞 ≈𝐡 𝐰𝐢 𝐰𝐣 𝐁 𝐬𝐢 , 𝐭 𝐣 𝐓 𝐂 𝐁 𝐬𝐢 , 𝐭 𝐣 J 𝐬𝐢 , 𝐭 𝐣
𝐢=𝟏 𝐣=𝟏
≈ 𝐡 𝐰𝟏 𝐰𝟏 𝐁 𝐬𝟏 , 𝐭 𝟏 𝐓 𝐂 𝐁 𝐬𝟏 , 𝐭 𝟏 J 𝐬𝟏 , 𝐭 𝟏
+𝐡 𝐰𝟏 𝐰𝟐 𝐁 𝐬𝟏 , 𝐭 𝟐 𝐓 𝐂 𝐁 𝐬𝟏 , 𝐭 𝟐 J 𝐬𝟏 , 𝐭 𝟐
+𝐡 𝐰𝟐 𝐰𝟏 𝐁 𝐬𝟐 , 𝐭 𝟏 𝐓 𝐂 𝐁 𝐬𝟐 , 𝐭 𝟏 J 𝐬𝟐 , 𝐭 𝟏
+𝐡 𝐰𝟐 𝐰𝟐 𝐁 𝐬𝟐 , 𝐭 𝟐 𝐓 𝐂 𝐁 𝐬𝟐 , 𝐭 𝟐 J 𝐬𝟐 , 𝐭 𝟐

Note: for a two point integration the

combinations of wi wj =1(i,j=1,2).
90
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9

Integral Gauss 2x2:

0,489 0,167 -0,289 -0,033 -0,033 -0,244 0,167 0,044

0,489 0,033 0,044 -0,167 -0,244 -0,033 -0,289
0,489 -0,167 -0,044 -0,033 -0,244 0,167
0,489 0,033 -0,289 0,167 -0,244
[𝐤 (𝐞) ] ≈109 0,489 0,167 -0,289 -0,033
0,489 0,033 0,044
0,489 -0,167
0,489

Συμμετρικό

Note: In general for the 4sided elements the integral is not accurate
(so it is not included in the table). In the above case is accurate
because of the geometry of the element.

91
 Quadrilateral isoparametric elements – Numerical integration of the
stiffness matrix – Example 6.9

Integral Gauss 3x3:

Ολοκλήρωμα ακριβές
Σημεία ολοκληρώματος Συντελεστές
n για πολυώνυμο
(si) στάθμισης (wi)
βαθμού:
.7745966692 .5555555556 5
3
0.0 .8888888889
𝟑 𝟑

𝐤 𝐞 ≈𝐡 𝐰𝐢 𝐰𝐣 𝐁 𝐬𝐢 , 𝐭 𝐣 𝐓 𝐂 𝐁 𝐬𝐢 , 𝐭 𝐣 J 𝐬𝐢 , 𝐭 𝐣
𝐢=𝟏 𝐣=𝟏
≈ 𝐡 𝐰𝟏 𝐰𝟏 𝐁 𝐬𝟏 , 𝐭 𝟏 𝐓 𝐂 𝐁 𝐬𝟏 , 𝐭 𝟏 J 𝐬𝟏 , 𝐭 𝟏
+𝐡 𝐰𝟏 𝐰𝟐 𝐁 𝐬𝟏 , 𝐭 𝟐 𝐓 𝐂 𝐁 𝐬𝟏 , 𝐭 𝟐 J 𝐬𝟏 , 𝐭 𝟐
+𝐡 𝐰𝟏 𝐰𝟑 𝐁 𝐬𝟏 , 𝐭 𝟑 𝐓 𝐂 𝐁 𝐬𝟏 , 𝐭 𝟑 J 𝐬𝟏 , 𝐭 𝟑
+𝐡 𝐰𝟐 𝐰𝟏 𝐁 𝐬𝟐 , 𝐭 𝟏 𝐓 𝐂 𝐁 𝐬𝟐 , 𝐭 𝟏 J 𝐬𝟐 , 𝐭 𝟏
+𝐡 𝐰𝟐 𝐰𝟐 𝐁 𝐬𝟐 , 𝐭 𝟐 𝐓 𝐂 𝐁 𝐬𝟐 , 𝐭 𝟐 J 𝐬𝟐 , 𝐭 𝟐
+𝐡 𝐰𝟐 𝐰𝟑 𝐁 𝐬𝟐 , 𝐭 𝟑 𝐓 𝐂 𝐁 𝐬𝟐 , 𝐭 𝟑 J 𝐬𝟐 , 𝐭 𝟑
+𝐡 𝐰𝟑 𝐰𝟏 𝐁 𝐬𝟑 , 𝐭 𝟏 𝐓 𝐂 𝐁 𝐬𝟑 , 𝐭 𝟏 J 𝐬𝟑 , 𝐭 𝟏
+𝐡 𝐰𝟑 𝐰𝟐 𝐁 𝐬𝟑 , 𝐭 𝟐 𝐓 𝐂 𝐁 𝐬𝟑 , 𝐭 𝟐 J 𝐬𝟑 , 𝐭 𝟐
+𝐡 𝐰𝟑 𝐰𝟑 𝐁 𝐬𝟑 , 𝐭 𝟑 𝐓 𝐂 𝐁 𝐬𝟑 , 𝐭 𝟑 J 𝐬𝟑 , 𝐭 𝟑
92
 Quadrilateral isoparametric elements – Problem 6.1

For the rectangular element of the

diagram the displacements are {u1
v1 u2 v2 u3 u3 u4 v4 }Τ
={0, 0, 1, 0, 2, 1, 0, 2}. Calculate the
displacements (u,v) and the stain εxx
at the point (x,y)=(2,1).

93
 Quadrilateral isoparametric elements – Problem 6.2

The rectangular element of the diagram (right) is assumed that it has been
deformed by bending moments (left diagram). The displacements are: {u1 v1
u2 v2 u3 u3 u4 v4 }T = {-1, 0, 1, 0, -1, 0, 1, 0}.
1. Calculate the strains εxx, εyy, γxy as functions of (x,y)
2. Can the element represent a pure bending case?

94
 Quadrilateral isoparametric elements – Problem 6.3

The displacements at the nodes of the element are: {u1 v1 u2 v2 u3 u3 u4 v4 }T

= {-1, 0, -1, 0, 0, 1, 0, 1}.
1. Calculate the (s,t) coordinates of point Α (0.5, 0) using isoparametric
mapping.
2. Calculate the displacement of the point Β which has coordinates (s,t)=(0,
0.5)
3. Calculate the Jacobian matrix [J] at point Β.

95
 Quadrilateral isoparametric elements – Problem 6.4

The node order of the element is 51 52 63 64 and the displacements are:

{0, 0, 0.1, 0, 0.1, 0.1, 0, 0}.
1. Calculate displacement at point (x,y)=(0.75, 0.75).
2. Calculate the Jacobian matrix [J] at point (x,y)=(0.75, 0.75).
3. Calculate the strain εyy at the centre of the element.

96