Structural Analysis

1. Consider the triangular element shown below. The coordinates are in the units of
millimeters. Assume plane stress conditions. Let 𝑬 = 𝟐𝟎𝟎 𝑮𝑷𝒂, 𝒗 = 𝟎. 𝟑, and thickness
𝒕 = 𝟒 𝒎𝒎. Assume the element nodal displacements (after deformation) have been
determined to be as follows:

𝒖𝟏 = 𝟎 ; 𝒗𝟏 = 𝟎. 𝟎𝟎𝟑𝟖𝟎𝟒 𝒎𝒎

𝒖𝟐 = 𝟎. 𝟎𝟑𝟏𝟓𝟒𝟖 𝒎𝒎 ; 𝒗𝟐 = −𝟎. 𝟎𝟎𝟔𝟖𝟎𝟒 𝒎𝒎

𝒖 𝟑 = 𝟎 ; 𝒗𝟑 = 𝟎

Determine the element stresses (𝝈𝒙 , 𝝈𝒚 , and 𝝉𝒙𝒚 ).

Initially, for plane stress:

1 𝑣 0 1 0.3 0
𝐸 𝑣 1 0 200 0.3 1 0
[𝐸] = 2
[ 1 − 𝑣] = 1 − 0.32 [ 1 − 0.3]
1−𝑣
0 0 0 0
2 2
1 0.3 0
20000𝑥103 0.3 1 0
[𝐸] = [ 1 − 0.3] 𝑀𝑃𝑎
91
0 0
2

Consequently, the following is determined:

 𝛽 0 𝛽𝑗 0 𝛽𝑚 0
1 𝑖
[𝐵] = [0 𝛾𝑖 0 𝛾𝑗 0 𝛾𝑚 ]
2𝐴
𝛾𝑖 𝛽𝑖 𝛾𝑗 𝛽𝑗 𝛾𝑚 𝛽𝑚

Where:

𝛽𝑖 = 𝑦𝑗 − 𝑦𝑚 = 0 − 30 = −30

𝛽𝑗 = 𝑦𝑚 − 𝑦𝑖 = 30 − 0 = 30

𝛽𝑚 = 𝑦𝑖 − 𝑦𝑗 = 0 − 0 = 0

𝛾𝑖 = 𝑥𝑚 − 𝑥𝑗 = 0 − 60 = −60

𝛾𝑗 = 𝑥𝑖 − 𝑥𝑚 = 0 − 0 = 0

𝛾𝑚 = 𝑥𝑗 − 𝑥𝑖 = 60 − 0 = 60

1
𝐴= ∙ |𝑥𝑖 ∙ (𝑦𝑗 − 𝑦𝑚 ) + 𝑥𝑗 ∙ (𝑦𝑚 − 𝑦𝑖 ) + 𝑥𝑚 ∙ (𝑦𝑖 − 𝑦𝑗 )|
2
1
𝐴= ∙ |0 ∙ (0 − 3) + 60 ∙ (3 − 0) + 0 ∙ (0 − 0)|
2
1
𝐴= ∙ |60 ∙ (30)| = 900
2

Hence:

1 −30 0 30 0 0 0
[𝐵] = [ 0 −60 0 0 0 60]
2(900)
−60 −30 0 30 60 0

1 −30 0 30 0 0 0 1
[𝐵] = [ 0 −60 0 0 0 60]
1800 𝑚𝑚
−60 −30 0 30 60 0

Consequently, the stresses are obtained as:

𝜎𝑥
{𝜎} = { 𝜎𝑦 } = [𝐸][𝐵][𝑑]
𝜏𝑥𝑦

Where
 𝑢1 0
𝑣1 0.003804
𝑢
[𝑑] = 𝑣2 = 0.031548 𝑚𝑚
2 −0.006804
𝑢3 0
{𝑣3 } { 0 }

Therefore:

0
𝜎𝑥 1 0.3 0 0.003804
20000𝑥109 0.3 1 0 1 −30 0 30 0 0 0
0.031548
{𝜎} = { 𝜎𝑦 } = [ 1 − 0.3 1800 [ 0
] −60 0 0 0 60]
𝜏𝑥𝑦 91 −0.006804
0 0 −60 −30 0 30 60 0
2 0
{ 0 }

Finally:
𝜎𝑥 107.20
{𝜎} = { 𝜎𝑦 } = { 6.80 } 𝑀𝑃𝑎
𝜏𝑥𝑦 −13.6

𝝈𝒙 = 𝟏𝟎𝟕. 𝟐𝟎 𝑴𝑷𝒂 ; 𝝈𝒚 = 𝟔. 𝟖𝟎 𝑴𝑷𝒂 ; 𝝉𝒙𝒚 = −𝟏𝟑. 𝟔 𝑴𝑷𝒂

2. Determine the nodal forces for the varying pressure 𝑷𝒙 by evaluating the surface
integral:
𝑻
{𝒇𝒔 } = ∬[𝑵] {𝑻𝒔 }𝒅𝑺

Assume the element thickness is equal to t.

For this case:

𝑦
−𝑁𝑖 𝑃0
𝑁𝑖 0 𝐿
0 𝑁𝑖 0
𝑁𝑗 0 −𝑃0 𝑦 −𝑁𝑗 𝑃0
𝑦
{𝑓𝑠 } = ∬ { 𝐿 } 𝑑𝑧𝑑𝑦 = 𝑡 ∫ 𝐿 𝑑𝑦
0 𝑁𝑗
0 0
𝑁𝑚 0 𝑦
[ 0 𝑁𝑚 ] −𝑁𝑚 𝑃0
𝐿
[ 0 ]

Where:

1
𝑁𝑖 = (𝛼 + 𝛽𝑖 𝑥 + 𝛾𝑖 𝑦)
2𝐴 𝑖
1
𝑁𝑗 = (𝛼 + 𝛽𝑗 𝑥 + 𝛾𝑗 𝑦)
2𝐴 𝑗
1
𝑁𝑚 = (𝛼 + 𝛽𝑚 𝑥 + 𝛾𝑚 𝑦)
2𝐴 𝑚
For this case:

𝑥𝑖 = 0 ; 𝑦𝑖 = 𝐿

𝑥𝑗 = 0 ; 𝑦𝑗 = 0

𝑥𝑚 = 𝑎 ; 𝑦𝑚 = 𝐿

Therefore:

𝛼𝑖 = 𝑥𝑗 𝑦𝑚 − 𝑦𝑗 𝑥𝑚 = (0)(𝐿) − (0)(𝑎) = 0

𝛼𝑗 = 𝑥𝑚 𝑦𝑖 − 𝑦𝑚 𝑥𝑖 = (𝑎)(𝐿) − (𝐿)(0) = (𝑎)(𝐿)

𝛼𝑚 = 𝑥𝑖 𝑦𝑗 − 𝑦𝑖 𝑥𝑗 = (0)(0) − (𝐿)(0) = 0

𝛽𝑖 = 𝑦𝑗 − 𝑦𝑚 = 0 − 𝐿 = −𝐿

𝛽𝑗 = 𝑦𝑚 − 𝑦𝑖 = 𝐿 − 𝐿 = 0

𝛽𝑚 = 𝑦𝑖 − 𝑦𝑗 = 𝐿 − 0 = 𝐿

𝛾𝑖 = 𝑥𝑚 − 𝑥𝑗 = 𝑎 − 0 = 𝑎

𝛾𝑗 = 𝑥𝑖 − 𝑥𝑚 = 0 − 𝑎 = −𝑎

𝛾𝑚 = 𝑥𝑗 − 𝑥𝑖 = 0 − 0 = 0

1
𝐴= ∙ |𝑥𝑖 ∙ (𝑦𝑗 − 𝑦𝑚 ) + 𝑥𝑗 ∙ (𝑦𝑚 − 𝑦𝑖 ) + 𝑥𝑚 ∙ (𝑦𝑖 − 𝑦𝑗 )|
2
1
𝐴= ∙ |0 ∙ (0 − 𝐿) + 0 ∙ (𝐿 − 𝐿) + 𝑎 ∙ (𝐿 − 0)|
2
1 (𝑎)(𝐿)
𝐴= ∙ |𝑎 ∙ (𝐿)| =
2 2

Therefore:

(−𝐿𝑥 + 𝑎𝑦)
𝑁𝑖 =
(𝑎)(𝐿)

(𝐿 − 𝑦)
𝑁𝑗 =
𝐿
𝑥
𝑁𝑚 =
𝑎
Consequently:

(−𝐿𝑥 + 𝑎𝑦) 𝑦 𝑦 𝑦
− 𝑃0 − 𝑃0
(𝑎)(𝐿) 𝐿 𝐿 𝐿
0 0
(𝐿 − 𝑦) 𝑦 (𝐿 − 𝑦) 𝑦
{𝑓𝑠 } = 𝑡 ∫ − 𝑃0 𝑑𝑦 = 𝑡 ∫ − 𝑃0 𝑑𝑦
𝐿 𝐿 𝐿 𝐿
0 0
𝑥 𝑦
− 𝑃0 0
𝑎 𝐿 [ 0 ]
[ 0 ]𝑥=0,𝑦=𝑦

𝑦=𝐿
𝑃0 𝑃0 𝑦 3
− 2 𝑦2 − 2
𝐿 𝐿 3
0 0
𝑃0 𝑃 𝑦2 𝑦3
{𝑓𝑠 } = 𝑡 ∫ − (𝐿𝑦 − 𝑦 2 ) 𝑑𝑦 = 𝑡 − 0 (𝐿 − )
𝐿2 𝐿2 2 3
0 0
0 0
[ 0 ] [ ]𝑦=0
0

𝑃0 𝐿3 𝑃0 𝐿
− −
𝐿2 3 3
0 0
𝑃 𝐿2 𝐿3 𝐿 𝐿
{𝑓𝑠 } = 𝑡 − 0 (𝐿 − ) = 𝑡 −𝑃0 ( − )
𝐿2 2 3 2 3
0 0
0 0
[ 0 ] [ 0 ]

Finally:

𝑷𝟎 𝑳𝒕
𝒇𝒔𝟏𝒙 −
𝟑
𝒇𝒔𝟏𝒚 𝟎
𝒇 𝑷 𝑳𝒕
{𝒇𝒔 } = 𝒔𝟐𝒙 = − 𝟎
𝒇𝒔𝟐𝒚 𝟔
𝒇𝒔𝟑𝒙 𝟎
[𝒇𝒔𝟑𝒚 ] 𝟎
[ 𝟎 ]
3. Consider the thin plate shown below, consisting of two triangular elements, with 𝑬 =
𝟐𝟎𝟎 𝑮𝑷𝒂, 𝒗 = 𝟎. 𝟑, and thickness 𝒕 = 𝟒 𝒎𝒎. Assume plane stress conditions apply.

Use ANSYS Workbench to:

a) Determine the displacements of nodes 1, 2, and 3 (in x- and y-directions). Compare

the results with the displacements given in Problem 1 and comment on this comparison.

b) Determine the stresses (𝝈𝒙 , 𝝈𝒚 , and 𝝉𝒙𝒚 ) in element 1 (only element 1). Compare these
stresses with the ones obtained in Problem 1 and comment on this comparison.

Material Properties

𝑌𝑜𝑢𝑛𝑔′ 𝑠 𝑀𝑜𝑑𝑢𝑙𝑢𝑠 (𝐸) = 200𝑥109 𝑃𝑎

𝑃𝑜𝑖𝑠𝑜𝑛′ 𝑠 𝑅𝑎𝑡𝑖𝑜 (𝑣) = 0.3

Geometry / Dimensions

Mesh / Element Type:

Loads / Boundary Conditions

Results

(a) Determine the displacements of nodes 1, 2, and 3.

Deformation in x-direction
Deformation in y-direction

Displacement - Displacement -
Relative error
Node Direction Problem 1 ANSYS
(%)
(mm) (mm)

1 x 0 0 0

1 y 0.003804 0.003804 0

2 x 0.031548 0.031548 0

2 y -0.006804 -0.006804 0

3 x 0 0 0

3 y 0 0 0

As it can be seen in Table above, the results obtained from ANSYS have a relative error
of 0% when compared to those provided as input data in Problem 1 of this project, indicating
that ANSYS is completely reliable when evaluating the structural behavior of this type of
system under specific supporting and loading conditions, being able to provide exactly the
same results as those obtained analytically. Also, this type of analysis can be effectively use
to determine the magnitude of a load applied to a body based on its resulting deformations,
based on the results obtained when using the commercial software mentioned.

(b) Determine the stresses in element 1.

Normal stress in x-direction

Normal stress in y-direction

Shear stress in xy-plane

Stress Problem 1 (MPa) ANSYS (MPa) Relative error (%)

𝜎𝑥 107.2 107.2 0

𝜎𝑦 6.8 6.8 0

𝜏𝑥𝑦 -13.6 -13.6 0

Similar to what was observed with the displacement results, the normal and shear
stresses determined on element 1 when using ANSYS Workbench provided exactly the same
results as those obtained analytically when applying the Finite Element equations, confirming
once again the accuracy and reliability of the commercial software ANSYS when evaluating
the structural behavior of a system. Additionally, the ANSYS method requires significant
less effort, as all the calculations are automatically performed by the software, allowing
saving high amounts of time that increase with the complexity of the system being analyzed,
being this one of the major advantages of employing finite element analysis software for this
purpose.
4. Consider the following two models of thin plates having a central hole, fixed from the
top with load F applied on the lower edge:

Use ANSYS Workbench with the following instructions to perform the analysis of the
models:

 Use symmetry in your FE models.

 Apply the load on the edge (not node):
 Use 𝑬 = 𝟏𝟗𝟑 𝑮𝑷𝒂, 𝒗 = 𝟎. 𝟑, and 𝒕 = 𝟓 𝒎𝒎.
 Assume plane stress conditions and perform 2D analysis using triangular
quadratic elements (PLANE183).
 Use a global mesh size of 4 mm for both models.

Based on the finite element analysis of the two models:

a) Find the location and value of maximum equivalent (von-Mises) stress in both
models and compare the results.
b) Find the maximum deflections at Point A in both models and compare the
results.
c) Perform mesh sensitivity study for Model 1 as follows: Do the analysis six times
with six global element sizes of 15 mm, 10 mm, 6 mm, 4 mm, 2 mm and 1 mm
and compare the values of equivalent stress at Point B for the six cases. For the
comparison, plot the values of equivalent stress at Point B versus the number of
elements in six cases. What could be the optimum element size for this study?
Why?
Material Properties

𝑌𝑜𝑢𝑛𝑔′ 𝑠 𝑀𝑜𝑑𝑢𝑙𝑢𝑠 (𝐸) = 193𝑥109 𝑃𝑎

𝑃𝑜𝑖𝑠𝑜𝑛′ 𝑠 𝑅𝑎𝑡𝑖𝑜 (𝑣) = 0.3

Geometry / Dimensions

Model 1: W300H200
Model 2: W200H300

Mesh / Element Type:

Loads / Boundary Conditions

Results

(a) Find the location and value of maximum equivalent (von-Mises) stress in both
models and compare the results.

Model 1: W300H200

𝝈𝒎𝒂𝒙 = 𝟏𝟗𝟏. 𝟐𝟖 𝑴𝑷𝒂 − 𝑳𝒐𝒄𝒂𝒕𝒆𝒅 𝒂𝒕 𝑷𝒐𝒊𝒏𝒕 𝑩

Model 2: W200H300

𝝈𝒎𝒂𝒙 = 𝟑𝟎𝟗. 𝟔𝟗 𝑴𝑷𝒂 − 𝑳𝒐𝒄𝒂𝒕𝒆𝒅 𝒂𝒕 𝑷𝒐𝒊𝒏𝒕 𝑩

As it can be seen from the results obtained, the maximum stress is located at point B in
both models. This point corresponds to the theoretically expected location of the maximum
stress in a plate subjected to a tension load while having a centered hole, which acts as a
stress amplifier known as stress concentrator. This physical discontinuity causes a significant
increment in the stress applied, resulting in this region being the first to reach the yielding
point of the material, and hence, presenting failures.

The stress encountered in the second design is higher than that of the first design, and
this occurs due to the fact that the cross-section area transversal to the load applied is smaller
in the second design, thus leading to a smaller resistance to the load applied and, as a
consequence, higher level of equivalent stresses.
b) Find the maximum deflections at Point A in both models and compare the results.

Model 1: W300H200

𝜹𝑨 = 𝟎. 𝟎𝟔𝟓𝟓𝟐𝟒 𝒎𝒎

Model 2: W200H300

𝜹𝑨 = 𝟎. 𝟏𝟐𝟕𝟓𝟒 𝒎𝒎

Similar to what was observed in the stress comparison, it can be appreciated that the
deflection at point A is higher in model 2 than in model 1, and this occur due to the fact that
the second design presents a smaller resistance to the load applied by having a smaller cross-
section area, thus being easier to apply higher deformations to model 2 than to model 1 when
the same load is applied. It can also be appreciated that the deformation at point A is not the
maximum of the system, but it indeed is the higher at the adjacencies of the hole, since it is
the closest point to the load application region.
c) Perform mesh sensitivity study for Model 1.

After varying the global element size as 15 mm, 10 mm, 6 mm, 4 mm, 2 mm and 1 mm and
recording the equivalent stress at point B for Model 1, the following results were obtained.

As it can be seen in this Figure, as the number of elements increases, the equivalent stress
at point B converges to a value of around 210 MPa. An increment in the number of elements
produces a rise in the quality of the mesh and accuracy of the results; however, this also
increases significantly the amount of computational resources required for the calculations,
as well as the time needed for it. Therefore, it can be stated that the optimum element size
for the system analyzed 2 mm, because the following element size of 1 mm does not produce
any significant variation in the equivalent stress obtained, but requires a lot more of
computational resources, since it increases the number of elements in the mesh from 15,000
to 60,000, indicating that the calculations will take four times the time needed to be
performed when using an element size of 2 mm.

As a final conclusion, the element size of 2 mm is the most adequate for the system,
being it capable of providing accurate results when minimizing the amount of resources
needed for successfully completing the numerical analysis of the structural behavior of the
system.