KATHMANDU UNIVERSITY

SCHOOL OF ENGINEERING
DEPARTMENT OF MECHANICAL ENGINEERING

REPORT ON

ANALYSIS OF A BAR ELEMENT BY USING FINITE ELEMENT METHODS

ANSYS WORKBENCH AND MATLAB

SUBMITTED TO: -
SUBMITTED BY: - Dr. Krishna Prasad Shrestha
Abishkar Shrestha (31110) Assistant Professor
Kushal Sedhai (31109) Department of Mechanical Engineering
Bandhan Dangi (31)
Group: - Design and Manufacturing

August 2021
TABLE OF CONTENTS
CHAPTER 1 INTRODUCTION...............................................................................................1

1.1 Background.....................................................................................................................1

1.2 Objectives.......................................................................................................................2

CHAPTER 2 DISCUSSION.....................................................................................................3

2.1 Problem definition..........................................................................................................3

2.2 Methodology...................................................................................................................3

2.2.1 Direct method...........................................................................................................3

2.2.2 Variation Method or Rayleigh method or Potential energy method........................6

2.2.3 Galerkin’s Method....................................................................................................7

2.2.4 H-Method...............................................................................................................11

2.2.5 P-Method................................................................................................................13

2.2.6 R-Method...............................................................................................................18

2.3 ANSYS WORKBENCH Solution................................................................................18

2.4 MATLAB Solution.......................................................................................................26

CHAPTER 3 RESULTS AND ANALYSIS...........................................................................28

3.1 Nodal Deflection distribution.......................................................................................28

CHAPTER 4 CONCLUSION.................................................................................................28

REFERENCES........................................................................................................................30
CHAPTER 1 INTRODUCTION

1.1 Background
Bar element is a structural member used to support load which may be fixed at any point.
Load may be in different directions i.e. axial or transverse. With the application of load
deflection is observed in the bar element. Due to application of transverse load bending occurs
while in case of application of axial load elongation or contraction occurs depending upon the
direction of application of load.
Finite Element Methods (FEM) can be used to solve deflection in a bar element. It is able to
calculate the deflections observed in different directions. With approximation in determines
different parameters of a bar element problem such as deflection, reaction and stresses. [1]
This paper studies the Finite Element Method for a bar stepped bar element with uniformly
varying load. Various methods used to solve the bar element problem. The various methods
used to solve the problem are listed below:
i. Direct method
ii. Rayleigh method (Potential energy method)
iii. Galerkin method
iv. H-method
v. P-method
vi. R-method
vii. ANSYS WORKBENCH
viii. MATLAB

The problem is also solved using ANSY and MATLAB. The results obtained from various
methods used for calculation are compared and the deviation is observed.
Finite Element Method is an approximation method so results obtained using different
methods will surely vary but the deviation from different methods must be within tolerable
range. Also, all the finite element methods might not be applicable to nay one problem as
their potential energy might not be available which restricts the use of Rayleigh method on
that problem. So, different finite element methods are used for different problems.

The advantages of the finite element method are the ability to

➢ Model irregularly shaped bodies quite easily
➢ Handle general load conditions without difficulty
➢ Model bodies composed of several different materials because element equations are
evaluated individually
➢ Handle unlimited numbers and kinds of boundary conditions

1
 ➢ Vary the size of the elements to make it possible to use small elements where
necessary.

1.2 Objectives
The objectives of this Finite Element Method based project are:
A. To develop mathematical models for structural problem (bar element)
B. To determine the deflection at different locations using FEM
C. To learn to solve problems using ANSYS and MATLAB
D.

2
 CHAPTER 2 DISCUSSION

2.1 Problem definition

A 2 m long stepped bar with different cross-sectional areas is fixed between two ends. It is
made up of steel with following dimensions:
Young’s Modulas = 100 GPa
Poisson’s ratio = 0.3
Respective area of element (1), (2) and (3) = 10−3 m2 ,3 ×10−3 m2∧10−3 m2
Point load in axial direction acting on element (2)
X = 10 KN
Y=8 KN

Figure 1 Unit Stepped Shaft

2.2 Methodology
We solved this stepped bar problem using Finite Element Method and calculations are done in
ANSYS WORKBENCH and MATLAB software’s.

2.2.1 Direct method

..As one of the method of structural analysis, the direct stiffness matrix method, also known as
the matrix stiffness method. It is a matrix method that makes use of the member’s stiffness
relation for computing member forces and displacements in structures. In applying the
method, the system must be modeled as a set of simpler, idealized elements interconnected at
the nodes. The material stiffness properties of these elements are than, through matrix
mathematics, compiled into a single matrix equation which governs the behavior of the entire
idealized structure. The structure’s unknown displacements and forces can then be determined
by solving this equation.
Solving the above equation from the given figure

3
Element stiffness matrix of

For element (1),

EA 1 −1 100× 109 ×10−3 1 −1 6 1

[ k 1 ]= [
l 1 −1 1 ]
=
0.5 −1 1 [
=200 ×10 ]
−1
−1 1 [ ]
For element (2),
EA 1 −1 100× 109 ×3 ×10−3 1 −1 6 1
[ k 1 ]= [
l 1 −1 1 ]
=
1 −1 1 [
=300 ×10
−1 1]
−1
[ ]
For element (3),
EA 1 −1 100× 109 ×10−3 1 −1 6 1
[ k 1 ]= l −1
1
[
1
= ]
0.5 −1 1 [
=200 ×10
−1
−1 1 ] [ ]
Then Global Stiffness Matrix is given by

200 −200 0 0
[ K ] =10 200

0
6

[
500 −300 0
0 −500 500 −200
0 −200 200
]
Now, we know, In matrix form with boundary conditions,
200 −200 0 0 u1 0
[K] {u}= {F}

[
106 200 500 −300
0 −500 500 −200 u3
0 0
0

−200 200 u4
u2
]{ } { }
= −10000
8000
0

Reducing we get
500 −300 u2 −10000
106 [
−300 500 u 3
= ]{ } {
8000 }
Solving above matrix we get;

4
u2=−16.25× 10−6 m
u3=6.25× 10−6 m
Again, solving we get;
f1=?
f2=?
200 −200 0 0 0 f1
106
[
200 500 −300 0 −16.25 ×10
0 −300 500 −200 6.25 ×10−6
0 0 −200 200 0
−6

]{
= −10000
8000
f4
}{ }
f 1=−200 ×−16.25 ×10−6 ×10 6=3250 N
f 4 =−200 ×6.25 ×10−6 ×106=−1250 N

Reaction forces
Element 1
f 1(1 ) 0
{ }
f2
(1 )
=10
6 200
[−200 −200
]{
200 −16.25 ×10−6 }
f 1(1)=3250 N
f 2(2)=−3250 N
Element 2
f 1(2) −300 −16.25 ×10
−6

{ } [
f 2(2)
=10
6 300
−300 300 6.25 ×10−6 ]{ }
f 1(1)=6750 N
f 2(2)=−6750 N
Element 3
f 1(3 )
{ } [
f 2(3)
=10
6 200 −200 6.25 ×10−6
−200 200 0 ]{ }
f 1(3)=1250 N
f 2(3)=−1250 N

5
2.2.2 Variation Method or Rayleigh method or Potential energy method
STEP 1
Function potential Elemental for each
element are:
E × A 1 100 ×10 9 ×10−3
k 1= = =200 ×106 N /m
L1 0.5
E × A 2 100 ×10 9 × 3× 10−3 6
k 2= = =300 ×10 N /m
L2 1
E × A 3 100 ×10 9 × 10−3
k 3= = =200 ×106 N /m
L3 1

Now, total potential energy for the bar element is given as

1 1 1
π p= k 1( u2−u1 )2+ k 2 (u3−u 4)2 − k 3 ( u5−u6 ) 2−p 1 u1 −p 2 u2 −p 3 u3 −p 4 u4
2 2 2

On differentiating we get
∂πp
=−k 1 ( u2−u1 ) −p 1--------------------------------(1)
∂ u1
Similarly;
∂πp
=−k 1 ( u2−u1 ) −k 2 ( u3−u 4 )− p 2
∂ u2
−k 1 u1 + ( k 1+ k 2 ) u 2−k 2 u3− p2 =0-----------------------(2)
∂π/∂𝑢3=k2(𝑢3−𝑢2 )−k2(𝑢4−𝑢3)− 𝑝 3
−k2𝑢2+
(k3+k2)𝑢3−𝐸A1𝐿1 𝑢4−𝑝3=0---------------------
----(3)
∂π/∂𝑢4=k3(𝑢4−𝑢3 )−𝑝4
−k3𝑢3+k3𝑢4−𝑝4= 0-------------------------------
--------------- (4)
We have p2 =−10000 𝑁 , 𝑝 3 =8000 𝑁 and 𝑢1=𝑢2=0, and p1 , p2are unknown Now, writing the
above equations in matrix form, we get

[𝐾]{𝑢} = {𝑃}

200 −200 0 0 0 0

[
106 200 500 −300 0 u
0 −500 500 −200 u3
0 0 −200 200 0
2
= −10000
8000
0
]{ } { }
6
Applying boundary conditions and Reducing the matrix, we get,

500 −300 u2 −10000

106 [−300 ]{ } {
500 u 3
=
8000 }
Solving the above matrix, we get
u2=−1.625× 10−5 m
u3=6.25× 10−6 m
Again, solving for the global nodal forces,
200 −200 0 0 0 p1
106
[
200 500 −300 0
0 −500 500 −200 6.25 ×10
0 0 −200 200 0
]{
−16.25 ×10
−6
−6
= −10000
8000
p4
}{ }
p1=−200 ×−16.25 ×10−6 × 106=3250 N
p 4=−200× 6.25× 10−6 ×106 =−1250 N

7
2.2.3 Galerkin’s Method
In mathematics, in the are of numerical analysis, Galerkin method is a class of method for
converting a continuous operator problem (such as a differential equation) to a discrete
problem. In principle, it is the equivalent of applying the method of variation of governing
equation for the bar element is given by,
d du
( AE )+b=0
dx dx
b

∫ ∂∂x ( AE ∂∂ux ) ×w dx=0

a
b b b
∂ ∂u ∂w ∂ du
w ×∫
a
∂x
(AE
∂x
) dx−∫
a
∫
∂x a ∂x
b
AE
dx
dx ) dx=0
(
du ∂ w du
w × AE
dx
− AE ∫
a ∂ x dx
dx=0 ( )
Now,
b
k ij= AE ∫
a
( ∂∂wx ) dudx
a

To find local matrices,

b
∂ Ni ∂ N j
k ij=EA ∫ dx
a ∂x ∂ x

For element (1),

0.5
9 ∂ N1 ∂ N1
−3
K 11 =100× 10 ×10 ∫ dx
0 ∂x ∂x
x x
K 11 =10 8
o .5

∫
( )( )
∂ 1− ∂ 1−
l l
dx
0 ∂x ∂x
8
10
k 11= × [ 0.5−0 ]
0.52
k 11=200 ×106

K 12=108 ∫
o .5 ( xl ) ∂ ( xl ) dx
∂ 1−

0 ∂x ∂x

108
k 12= × [ 0.5−0 ]
0.52
k 12=−200× 106

Similarly,
k 21=200× 106

8
k 22=200× 106

For element (2),

1
∂ N1 ∂ N 1
K 11 =100× 109 ×3 ×10−3∫ dx
0.5 ∂ x ∂ x

x x
K 11 =3 ×108 ∫
1 ∂ 1−
l ( )( )
∂ 1−
l
0.5 ∂x ∂x
3 ×108
k 11= × [ 1−0.5 ]
12
k 11=300 ×106
k 12=−300× 106
k 21=−300× 106
k 22=300× 106

For element (3),

2
9 −3 ∂ N1 ∂ N1
K 11 =100× 10 ×10 ∫ dx
1.5 ∂ x ∂x
x x
K 11 =108
o .5

∫
( )( )
∂ 1− ∂ 1−
l l
0 ∂x ∂x
8
10
k 11= × [ 0.5−0 ]
12
k 11=200 ×106

Similarly,

k 12=−200× 106

k 21=−200× 106
k 22=200× 106

So ,global stiffness matrix becomes:

200 −200 0 0
6
[ K ] =10
[
200 500 −300 0
0 −500 500 −200
0 0 −200 200
]
[𝐾]{𝑢} = {f}

9
 200 −200 0 0 0 0
10
[
6 200 500 −300 0 u
0 −500 500 −200 u3
0 0 −200 200 0
2
=
]{ } { }
−10000
8000
0

Applying boundary conditions and Reducing the matrix, we get,

500 −300 u2 −10000

106 [ ]{ } {
−300 500 u 3
=
8000 }
Solving the above matrix, we get
u2=−1.625× 10−5 m
u3=6.25× 10−6 m
Again, solving for the global nodal forces,
200 −200 0 0 0 f1
106
[
200 500 −300 0 −16.25
0 −500 500 −200 6.25 ×10−6
0
Solving,
0 −200 200 0
]{
×10 −6
= −10000
8000
f4
}{ }
f 1=−200 ×−16.25 ×10−6 ×10 6=3250 N
f 4 =−200 ×6.25 ×10−6 ×106=−1250 N

10
2.2.4 H-Method
The H-method improves results by using a finer mesh of the same type of element in
different areas. This method refers to decreasing the characteristic length (h) of elements,
dividing each existing element into two or more elements without changing the type of
elements used .The number of elements must be increased in areas where the stress changes
quickly over a small distance.

b
∂ Ni ∂ N j
k ij=EA ∫ dx
a ∂x ∂ x

For element (1),

0.5
9 ∂ N1 ∂ N1
−3
K 11 =100× 10 ×10 ∫ dx
0 ∂x ∂x
x x
K 11 =108 ∫
o .5 ( )( )
∂ 1− ∂ 1−
l l
0 ∂x ∂x
8
10
k 11= × [ 0.5−0 ]
12
k 11=200 ×106
Similarly,
k 12=200× 106
k 21=−200× 106
k 22=200× 106

For element (2),

1
∂ N1 ∂ N 1
K 11 =100× 109 ×3 ×10−3∫ dx
0.5 ∂ x ∂ x
x x
K 11 =3 ×108 ∫
1 ∂ 1−
l
∂ 1− ( )( )
l
0.5 ∂x ∂x
3 ×108
k 11= × [ 1−0.5 ]
l2
k 11=600 ×10 6k 12=−600× 106

11
k 21=−600× 106
k 22=600× 106

For element (3),

1.5
9 ∂ N1 ∂ N 1
−3
K 11 =100× 10 ×3 ×10 ∫ dx
1 ∂x ∂x
x x
K 11 =3 ×108 ∫
( )( )
1.5 ∂ 1−
l
∂ 1−
l
1 ∂x ∂x
3 ×108
k 11= × [ 1.5−1 ]
22
k 11=600 ×10 6
k 12=−600× 106
k 21=−600× 106
k 22=600× 106

For element (4),

2
9 ∂ N1 ∂ N1
−3
K 11 =100× 10 ×10 ∫ dx
1.5 ∂ x ∂x
x x
K 11 =108 ∫
( )( )
2 ∂ 1−
l
∂ 1−
l
1.5 ∂x ∂x
108
k 11= × [ 2−1.5 ]
12
k 11=200 ×106
Similarly,
k 12=200× 106
k 21=−200× 106
k 22=200× 106

Then, Global stiffness matrix becomes

200 −200 0 0 0

[−200 800 −600

[ K ] =106 0
0
0
0
0 0
0
−600 1200 −600
0
0
−200 200 −200
−200 200
]
Global system of equation is

12
 [K] {u}= {F}

200 −200 0 0 0 u1 f1

10
[6
−200 800 −600
0
0
0
0
0 0
0
−600 1200 −600
0
0
u2 f2
u3 = f 3
−600 200 −200 u4
−200 200 u5
f4
f5
]{ } { }
Using boundary conditions; u1 = 0, u5 = 0, and deleting the respective rows and columns,
we get
800 −600 0 u2

]{ } { }
−1000
6

[
10 −600 1200 −600 u3 =
0 −600 200 u 4
0
8000
Solving we get,

u2=−16.25× 10−6 m
u3=5× 10−6 m
u4 =6.25 ×10−6 m

Again, solving for global nodal forces,

200 −200 0 0 0 0

10
[6
−200 800 −600
0
0
0
0
0 0
0
−600 1200 −600
0
0

−200 200
]{
LINK Word . Document .12 C:\\Users\\abish\\Downloads\\fem rep
5× 10−6 m
−600 200 −200 LINK Word . Document .12 C:\\Users\\abish\\Downloads\\fem re
0

On solving, we get,
f 1=3250 N
f 3=O
f 5=−1250 N

2.2.5 P-Method
The p-method improves results by using the same mesh but increases the polynomial order
of the shape function to improve the accuracy. This method refers to increasing the degree

13
of the highest complete polynomial (p) within an element without changing the number of
elements used.
Here,
Assuming the element to be quadratic i.e. only three shape functions N1, N2, N3
So,
For element (1),
E A 7 −8 1
[
k 1= 1 1 −8 16 −8
3 L1
1 −8 7 ]
100 ×10 9 × 10−3 7 −8 1
k 1=
3 ×0.5 [
−8 16 −8
1 −8 7 ]
7 −8 1
k 1=66.66 ×106 −8 16 −8
1 −8 7 [ ]
466.2 −533.28 66.66
6

[
𝑘1¿ 10 −533.28 1066.56 −533.28
66.66 −533.28 466.2 ]
For element (2),
E A 7 −8 1
[
k 2= 2 2 −8 16 −8
3 L2
1 −8 7 ]
100 ×10 9 × 3× 10−3 7 −8 1
k 2=
3 ×1 [
−8 16 −8
1 −8 7 ]
7 −8 1

[
k 2=100× 106 −8 16 −8
1 −8 7 ]
700 −800 100

[
k 2=106 −800 1600 −800
100 −800 700
For element (3),
]
E 1 A1 7 −8 1
k 3=
3 L1 [−8 16 −8
1 −8 7 ]
100 ×10 9 × 10−3 7 −8 1
k 3=
3 ×0.5 [
−8 16 −8
1 −8 7 ]
7 −8 1
3
k =66.66 ×10 −8 16 −8
1 −8 7
6

[ ]
14
 466.2 −533.28 66.66

[
k 3=106 −533.28 1066.56 −533.28
66.66 −533.28 466.2 ]
Then,
Global stiffness matrix is given by:

k =k 1 +k 2+ k 3

466.62 −533.28 0 0 0 0 0

[ ]
−533.28 1066.56 −533.28 0 0 0 0
66.66 −532.28 1166.62 −800 100 0 0
[ K ] =106 0 0 −800 1600 −800 0 0
0 0 100 −800 1166.62 −533.28 66.66
0 0 0 0 −533.28 1066.56 −533.28
0 0 0 0 66.66 −533.28 466.2

Again,
Forces are given by:
F=FI +Fb
Where , FI= Traction force & Fb=Body force
Now,
Here 𝛾 = 𝑝𝑜𝑖𝑠𝑖𝑜𝑛 𝑟𝑎𝑡𝑖𝑜 = 0.3
For element (1),

2 2

[] [] []
1
6 6
2
4 4
F l1 =γ × A 1 × l 1 =0.3 ×10−3 × 0.5 =10−4 1
6 6
1
2 2
2
6 6
For element (2),

2 2

F l2 =γ × A 2 × l 2

[] [ ] []
6
4
6
2
6
=0.3 ×3 × 10−3 × 1
6
4
6
2
6
3
=10−4 6
3

15
For element (3),

2 2

[] [] []
1
6 6
2
4 4
F l3 =γ × A 3 × l 3 =0.3× 10−3 ×0.5 =10−4 1
6 6
1
2 2
2
6 6
[
Then,

0.5 ×10−4

{ }
1× 10−4
3.5 ×10− 4
F a= 6 ×10−4
3.5 ×10− 4
1× 10−4
0.5 ×10−4

And Fb is given by:

{}
0
−10000
F a= 0
8000
0
0

0.5 ×10−4

{ }
1 ×10−4
−10000+3.5 ×10−4
⸫F= 6 ×10−4
8000+3.5 ×10−4
1 ×10−4
0.5 ×10−4

Then, the global stiffness equation with global nodal forces and displacement becomes:

16
 466.62 −533.28 0 0 0 0 0 0 0.5 ×10−4

[ ]{ } { }
−532.28 1066.56 −533.28 0 0 0 0 u2 1 ×10−4
66.66 −532.28 1166.62 −800 1000 0 0 u3 −10000+3.5 ×10− 4
6
10 0 0 −800 1600 −800 0 0 u4 = 6 ×10−4
0 0 100 −800 1166.62 −533.28 66.66 u5 8000+3.5 ×10− 4
0 0 0 0 −533.28 1066.56 −533.28 u 6 1 ×10−4
0 0 0 0 66.66 −533.28 466.2 0 0.5 ×10−4

}
Applying boundary condition. We get the global stiffness matrix with global nodal forces and
displacement equation is

1066.56 −533.28 0 0 0 u2 1 ×10−4

106
[−533.28 1166.62 −800
0
0
0
−800
100
0
1600
100
−800
0
0
u3
u4 =
−800 1166.62 −533.28 u 5
0 −533.28 1066.56 u 6
]{ } {
−10000+ 3.5× 10−4
6 ×10−4
8000+ 3.5× 10−4
1 ×10−4
}
And solving the above matrix from engineering equation solver (EES),
We get

u2=−8.13 ×10−6
u3 =−16.26 ×10−6
u4 =−5.01 ×10−6
u5=6.24 ×10−6
−6
u6 =3.122× 10

17
2.2.6 R-Method

Stiffness for each element are:

E × A1
K 1=
L1
100 × 109 × 10−3 6
∴ K 1= =400× 10 N /m
0.25

E × A2
K 2=
L2
∴ K 2=400 ×10 6 N /m

Now,
E× A 3 100× 106 × 3 ×10−3
∴ K 3= = =300 ×10 6 N /m
L3 1

E × A 4 100× 109 × 10−3 6

∴ K 4= = =400× 10 N /m
L4 0.25

Similarly, ∴ K 5=400 × 106 N /m

Global stiffness matrix is given by;

400 −400 0 0 0 0

K=10 6 0
0
0
0
[
−400 800 −400
−400
0
0
0
700
0
−300
0
0
−300 700 −400
0
0
0
0
0
−400 300 −400
0 −400 400
]
We have, [ K ] {u }={F }
Applying boundary conditions in stiffness equations, we get,

18
 800 −400 0 0 u2 0
10
[
6 −400
0
0
700 −300 0 u3
−300 700 −400 u 4
0 −400 800 u5
=
]{ } { }
−10000
8000
0

Solving we get,
u2=−16.25× 10−6 m
u3=−5× 10−6 m
u 4=6.25 ×10−6
u5=3.125× 10−6 m

Force is calculated as;

u1=0 F1

]{ } { }
400 −400 0 0 0 0

[
−6
−400 800 −400 0 0 0 u 2 =−16.25 ×10 F2
−6
u ×10
106 0 −400 700 −300 0 0 =−5
3
= −10000
0 0 −300 700 −400 0 u 4=6.25× 10 −6
8000
0 0 0 −400 300 −400 u =3.125 ×10−6 F5
5
0 0 0 0 −400 400 F6
u6=0

By solving the above matrix equation, reaction forces are;

F 1=6500 N
F 2=−11000 N
F 5=0 N
F 6=1250 N

19
 2.3 ANSYS WORKBENCH Solution
From given data :
Area of element 1=10-3 m2
Area of element 2=3x10-3 m2
Area of element 3=10-3 m2

Length of element 1=0.5 m

Length of element 2=1 m
Length of element 3=0.5 m
Modulus of elasticity=100 GPa
Poisson’s ratio=0.3

Step1: Static Structural was selected from Analysis systems and design the stepped bar
and defining the material, nodes, elements and nodal forces.

20
21
22
23
Step 2: Displacement solution

Step 3: Reactions solution

24
Step 4: Stress solution at individual node

25
 2.4 MATLAB Solution
Solution of stepped bar element using MATLAB

E=100e9;
A1=10^-3

A2=3*10^-3

A3=10^-3

L1=0.5

L2=1

L3=0.5
K1=barstiffness(E,A1,L1);
K2=barstiffness(E,A2,L2);
K3=barstiffness(E,A3,L3);
K=zeros(4,4);
K=GlobalStiffness(K,K1,1,2);
K=GlobalStiffness(K,K2,2,3);
K=GlobalStiffness(K,K3,3,4)
%Applying Boundry Condition
k=[K(2:3,2:3)]; f=[-
10000;8000]; u=k^-1*f
U=[0;u(1);u(2);0]; F=K*U
u1=[0;u(1)]; u2=[u(2);0];
f1=K1*u1 f2=K2*u2
function [b] =
GlobalStiffness(K,k,i,j)
K(i,i) = K(i,i) + k(1,1);
K(i,j) = K(i,j) + k(1,2);
K(j,i) = K(j,i) + k(2,1);
K(j,j) = K(j,j) + k(2,2);
b = K; end

function [a] = barstiffness(E,A,L)

a=E*A/L *[1 -1;-1 1]; end

26
OUTPUT

27
28
 CHAPTER 3 RESULTS AND ANALYSIS

Nodal deflection was obtained using various finite element techniques and the results are
compared. The maximum deviation of 0.56% is found in the solution using different
techniques.

3.1 Nodal Deflection distribution

The deflection at different nodes was found to increase while moving in both sides axial
direction from the fixed support.
In both end the displacement is 0 m because both end are fixed.
But at node 2 the displacement is 1.3×10-5m opposite direction and at node 3 , the
displacement is 0.5×10-5m

Figure: Normal Stress results obtained from ANSYS

CHAPTER 4 CONCLUSION

Deflection at nodes along the length of a bar element was calculated using different finite
element techniques as well as ANSYS and MATLAB. A maximum variation of <1%.was
obtained from the solutions obtained from different solutions which is acceptable range.
Knowledge regarding different finite elements methods for solving problems were gained.
This project helped to learn to solve structural problems using ANSYS and MATLAB
software’s.

29
The maximum displacement in the problem was found to be 1. 3 × 10−5 m using P method
which has the highest level of accuracy and the accuracy of the displacement using all other
methods was found to be within an error percentage of <1%.

Knowledge regarding different finite elements methods for solving problems were gained.
This project helped to learn to solve structural problems using ANSYS and MATLAB
software’s.

30
 REFERENCES
[1] D. L. Logan, in Finite Element Method, Cengage Learning, 2015, pp. 25-486.

[2] C. David E Penney, in Differential Equations and Boundary Value Problems, Pearson
Education Canada, 1995, pp. 56-91.

31