Professional Documents
Culture Documents
Article1379754909 - Vanam Et Al
Article1379754909 - Vanam Et Al
Department of Mechanical Engineering, Sir CRR College of Engineering, Eluru, Andhra Pradesh-534007, India.
2
Department of Mechanical Engineering, National Institute of Technology -Rourkela, Odissa-769008, India.
Accepted 16 January, 2012
This research work aims to analyze the static analysis of an isotropic rectangular plate with various
boundary conditions and various types of load applications. In this paper, finite element analysis has
been carried out for an isotropic rectangular plate by considering the master element as a four noded
quadrilateral element. Numerical analysis (finite element analysis, FEA) has been carried out by
developing programming in mathematical software MATLAB and the results obtained from MATLAB are
giving good agreement with the results obtained by classical method - exact solutions. Later, for the
same structure, analysis has been carried out using finite element analysis software ANSYS. This job is
helpful for obtaining the results not only at node points but also the entire surface of the rectangular
plate. Finally, comparison has been done between the results obtained from FEA numerical analysis,
and ANSYS results with classical method - exact solutions. Numerical results showed that, the results
obtained by finite element analysis and ANSYS simulation results are in close agreement with the
results obtained from exact solutions from classical method. During this analysis, the optimal thickness
of the plate has been obtained when the plate is subjected to different loading and boundary
conditions.
Key words: Finite element analysis (FEA), isotropic rectangular plate, ANSYS, static analysis.
INTRODUCTION
Finite element analysis (FEA) is a powerful computational
technique used for solving engineering problems having
complex geometries that are subjected to general
boundary conditions. While the analysis is being carried
out, the field variables are varied from point to point, thus,
possessing an infinite number of solutions in the domain.
So, the problem is quite complex. To overcome this
difficulty FEA is used; the system is discretized into a
finite number of parts known as elements by expressing
the unknown field variable in terms of the assumed
approximating functions within each element. For each
element, systematic approximate solution is constructed
by applying the variational or weighted residual methods.
These functions (also called interpolation functions) are,
Vanam et al.
149
and
. However
150
zero. Right angles in the plane of the plate may not remain right
angles after loading. The plate may twist in the plane.
ii. Change in plate thickness can be neglected and normal undergo
no extension. It means:
.
iii. Normal stress
and
in
(1)
(2)
Vanam et al.
151
(3)
(4)
and
(14)
(5)
Using the definitions of in-plane strains, the in-plane strain/
displacement equations are:
Where
x=
(6)
The governing differential equations are:
y=
(7)
(15)
and
xy=
(8)
The two points of Equations 1 and 2 are used in beam theory. The
remaining equations (Equations 3 to 8) are new to plate theory.
According to Kirchhoff theory, the plane stress equations for an
isotropic material are:
and
are
(9)
(16)
(10)
From Equation 16, the solution of thin-plate bending is a function of
the transverse displacement w.
(11)
Analytical solution for rectangular thin plate
Where
The in-plane normal stresses and shear stress are acting on the
edges of the plate as shown in Figure 2. The stresses are varying
linearly in the Z-direction from the mid surface of the plate. Although
transverse shear deformation is neglected, transverse shear
stresses
yz
and
(12)
w=0 and
(17)
w=0 and
(18)
An infinite Fourier series for the deflection and distributed load can
be defined as follows:
(19)
(20)
152
Where
and
Fourier
coefficients
one
can
multiply
by
(21)
It can be shown by direct integration that:
(22)
Vanam et al.
(31)
(23)
Using the property of Equations 22 and 23, one can get a general
expression for the coefficient of distributed load as shown thus:
(24)
Using the assumed Equation 24 and inserting them in the
governing equation, one can obtain the following expression:
153
(32)
The value of depends on the boundary condition of rectangular
plate as in Equation 32, where p is the value of distributed load. As
a result, will be 0.00516 for simple supported and 0.00148 for
fixed or clamped edges.
(25)
FEA formulation for 4-noded quadrilateral element
After simplifying and rearranging coefficient of the deflection is as
follows:
, ) coordinates:
(26)
,
(33)
Inserting the expression obtained for coefficients of deflection and
distributed load in Equation 16 to get a solution for the governing
equation for a general distributed load, we get:
(27)
(34)
For a uniform distributed load, the expression in Equation 27 can be
reduced further by specifying the general coefficient of distributed
load:
and
(36)
(30)
A detailed calculation to reduce Equation 30 is a special case
where the concentrated load is applied at the canter of a simple
supported square plate. Accordingly, the maximum deflection for a
square plate is given thus:
(38)
(39)
154
(44)
(45)
Figure 5. Rectangular element geometry.
(46)
Where
(40)
(41)
(42)
(43)
In deriving this result, it is simpler to use the expression in Equation
42 for w and substitute for {a} after performing the integration. A
typical integral is then the element stiffness matrix, and thus;
Vanam et al.
155
Where
(47)
Figure 6. Geometry of shell63 element.
As in the case of the inertia matrix, it is simpler to use the
expression in Equation 42 for w and substitute for {a} after
performing the integration using Equation 47. This procedure has
been generalised for a number of plate elements with isotropic
material properties in reference is the element equivalent nodal
force matrix. Assuming pz to be constant substituting for [N] from
Equation 44 and integrating gives:
(48)
[k]=
[P]=
(51)
(49)
156
Problem specification
Clamped edge and simple support boundary conditions
are applied for the considered rectangular plate. The detailed description of the problem is described in Figure 7.
Uniform distributed load on thin plate
Vanam et al.
Table 1. All sides fixed or clamped plate subjected to uniform distributed load (pressure).
S/No.
1
2
3
4
5
6
7
0.01
0.03
0.06
0.09
0.12
0.15
0.18
9.86*10
3.65*10-4
-5
4.56*10
1.35*10-5
-6
5.70*10
-6
2.92*10
-6
1.69*10
-3
From ANSYS
deflection (m)
-3
9.69*10
-4
3.5*10
-5
4.47*10
1.33*10-5
-6
5.559*10
-6
2.86*10
-6
1.66*10
9.53*10
3.56*10-4
-5
4.43*10
1.29*10-5
-6
5.62*10
-6
2.74*10
-6
1.54*10
157
158
Table 2. Deflection of plate: All sides simply supported to uniform distributed load (pressure).
S/No.
1
2
3
4
5
6
7
0.01
0.03
0.06
0.09
0.12
0.15
0.18
3.15*10-5
-5
1.15*10
1.43*10-6
4.25*10-7
-7
1.79*10
9.18*10-8
-8
5.3*10
From ANSYS
deflection (m)
3.1*10-5
-5
1.15*10
1.44*10-6
4.26*10-7
-7
1.8*10
9.19*10-8
-8
5.32*10
Figure 9. Deflection of plate: All sides clamped or fixed supported subjected to UDL pressure.
Figure 10. Deflection of plate: All sides clamped or fixed supported under UDL pressure
Vanam et al.
Figure 11. Deflection vs. thickness graph of fixed plate subjected to uniform distributed load.
Figure 12. Deflection of plate: All sides simply supported subjected to uniform distributed load (pressure).
159
160
Figure 13. Deflection vs. thickness graph for all sides simply supported subjected to UDL.
Figure 14. All sides simply supported plate and concentrated load at centre of the plate.
Conclusion
This paper mainly focused on the finite element model for
finding field variables of an isotropic rectangular plate.
Vanam et al.
161
Figure 15. Deflection vs. thickness graph for a simply supported plate concentrated load.
Table 3. Deflection of plate: All sides simply supported subjected to concentrated load.
S/No.
1
2
3
4
5
6
7
0.01
0.03
0.06
0.09
0.12
0.15
0.18
7.84*10-3
2.90*10-4
3.63*10-5
1.03*10-5
4.54*10-6
-6
2.32*10
1.34*10-6
From ANSYS
deflection (m)
8.8*10-3
3.3*10-4
4.1*10-5
1.22*10-5
5.1*10-6
-6
2.6*10
1.5*10-6
162