3 Questions on Finite Element Analysis - Midterm Exam 1 | AE 420, Exams of Aerospace Engineering

Material Type: Exam; Professor: Geubelle; Class: Finite Element Analysis; Subject: Aerospace Engineering; University: University of Illinois - Urbana-Champaign; Term: Spring 2007;

Typology: Exams

Pre 2010

Uploaded on 03/16/2009

koofers-user-gxi
koofers-user-gxi 🇺🇸

5

(1)

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
AE420/ME471 – Spring 2007 – First Midterm
Friday, February 23, 2007
50 minutes – Closed notes/closed books/no calculator
Question 1: Rayleigh-Ritz Method (7 points)
Consider the simple beam buckling problem described in Figure 1.
Figure 1
The beam is of length L, stiffness E and moment of inertia I. It is simply supported at both ends, x=0 and
x=L. It is subjected to a compressive axial force P applied at both ends and to a distributed transverse load
q(x) (as shown in Figure 1). Note: we only solve for the transverse deflection w, not for the axial
displacement. The boundary conditions for this problem are simply
w0
( )
=!!
w0
( )
=w L
( )
=!!
w L
( )
=0
.
The potential energy for this problem can be shown to be
!=1
2EI d2w
dx2
"
#
$%
&
'
2
(1
2Pdw
dx
"
#
$%
&
'
2
(qw
)
*
+
+
,
-
.
.dx
0
L
/
.
a) Denoting the approximate deflection solution as
!
w x
( )
=aifix
( )
i=1
N
!
,
where the
ai
are the unknown coefficients and
fi
are the basis functions, derive the general expression of the
linear system
K
[ ]
A
{ }
=R
{ }
used to solve the N-component vector
A=a1a2... aN
. Write the matrix [K] and vector {R} both in
“matrix/vector” (i.e., [K]=…, {R}=…) and component (i.e.,
…,
Ri=
…) forms. What are the
conditions on the basis functions to make them admissible?
b) Assuming N=1, use the simplest polynomial basis function to obtain an approximation for the buckling
load for this problem (i.e., the critical value of P for which the deflection w is not zero for zero transverse
load
q x
( )
=0
).
Question 2: Assembling (2 points)
Consider the simple 2-D truss structure shown in
Figure 2. It is composed of 4 nodes and 5
elements.
The connectivity table linking nodes to elements
is as follows:
Element #
Node a
Node b
1
1
3
2
4
2
3
3
2
4
3
4
5
4
1
E, I, L
P
q(x)
Figure 2
P
pf2

Partial preview of the text

Download 3 Questions on Finite Element Analysis - Midterm Exam 1 | AE 420 and more Exams Aerospace Engineering in PDF only on Docsity!

AE420/ME471 – Spring 2007 – First Midterm Friday, February 23, 2007 50 minutes – Closed notes/closed books/no calculator Question 1: Rayleigh-Ritz Method (7 points) Consider the simple beam buckling problem described in Figure 1.

Figure 1 The beam is of length L , stiffness E and moment of inertia I. It is simply supported at both ends, x=0 and x=L. It is subjected to a compressive axial force P applied at both ends and to a distributed transverse load q(x) (as shown in Figure 1). Note: we only solve for the transverse deflection w , not for the axial displacement. The boundary conditions for this problem are simply

w ( 0 ) = w! !( 0 ) = w ( L ) = w! !( L ) = 0.

The potential energy for this problem can be shown to be

! =

EI

d^2 w dx^2

2 (

P

dw dx

#$^

2 ( qw

dx 0

L

/^.

a) Denoting the approximate deflection solution as

w! ( x ) = ai fi ( x )

i = 1

N

!^ ,

where the ai are the unknown coefficients and fi are the basis functions, derive the general expression of the

linear system

[ K ] { A } =^ { R }

used to solve the N-component vector A = a 1 a 2 ... aN. Write the matrix [ K ] and vector { R } both in

“matrix/vector” (i.e., [ K ]=…, { R }=…) and component (i.e., Kij =…, Ri =…) forms. What are the

conditions on the basis functions to make them admissible?

b) Assuming N =1, use the simplest polynomial basis function to obtain an approximation for the buckling load for this problem (i.e., the critical value of P for which the deflection w is not zero for zero transverse

load q ( x ) = 0 ).

Question 2: Assembling (2 points) Consider the simple 2-D truss structure shown in Figure 2. It is composed of 4 nodes and 5 elements. The connectivity table linking nodes to elements is as follows: Element # Node a Node b 1 1 3 2 4 2 3 3 2 4 3 4 5 4 1

E, I, L

P

q ( x )

Figure 2

P

Denoting the 16 components of the local stiffness matrix of element m by ki ( j^^ m ), with i and j ranging from 1 to

4, indicate where (i.e., at what row and column) in the stiffness matrix the following components are assembled:

k 12 (^1 ) k 33 (^2 ) k 24 (^4 ) k 23 (^5 )

Question 3: Finite element formulation (11 points) Consider the simple structural problem shown in Figure 3. It consists of a string of length L , with tension T , on an elastic foundation (with distributed stiffness ks ) and subjected to a transverse load q(x). It is attached

at both ends, (i.e., w ( 0 ) = w ( L ) = 0 ) with w denoting the transverse displacement of the string.

a) Starting from the expression of the potential energy for this problem

! =

T

dw dx

#$^

2

  • ksw^2

0

L

( dx^ )^ qw^ dx

0

L

(^ ,

derive the finite element formulation (i.e., the expression of the local stiffness matrix [ k ] and local load vector { r }) for a generic M - node string element of length l.

b) If M = 3, what is the expression of the shape functions for this element (assuming the second node is in the middle of the element)? Sketch the corresponding shape functions.

c) Knowing that the Principle of Virtual Work (PVW) for this structural problem is

! T dw dx

d " w dx

+ ksw " w

$%^

L

) dx^ +^ q^ " w^ dx

0

L

) =^0 for^ any^ " w^ ,

derive the finite element formulation using the PVW approach, and compare with the solution found in a).

Figure 3.