IFEM Solution Ch17

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3
At a glance
Powered by AI
The document discusses solutions to exercises related to isoparametric quadrilaterals and membrane elements. It also examines the use of different order Gauss integration rules for computing element stiffness matrices.

Isoparametric elements are elements where the interpolation functions used to describe geometry are the same as those used for the field variables. This allows the element to better represent curved geometries. They are commonly used in finite element analysis.

Higher order Gauss integration rules integrate the element stiffness matrix integrand more accurately, capturing higher order terms. However, for some elements like isoparametric quadrilaterals, lower order rules can still produce the exact same stiffness matrix as higher order rules under certain conditions.

1719

Solutions to Exercises

Homework Exercises for Chapter 17: Isoparametric Quadrilaterals


Solutions
EXERCISE 17.1 This exercise can be done using the Mathematica script listed in Figure E17.5. Running

that code produces the results shown in Figure E17.6.


ClearAll[Em,,a,b,h]; Em=48; h=1; a=4; b=2; =0;
ncoor={{0,0},{a,0},{a,b},{0,b}};
Emat=Em/(1-^2)*{{1,,0},{,1,0},{0,0,(1-)/2}};
For [p=1, p<=4, p++,
Ke= Quad4IsoPMembraneStiffness[ncoor,Emat,h,{True,p}];
Print["Gauss integration rule: ",p," x ",p];
Print["Ke=",Chop[Ke]//MatrixForm];
Print["Eigenvalues of Ke=",Chop[Eigenvalues[N[Ke]]]]
];

Figure E17.5. Script to do Exercise 17.1.


Gauss integration rule: 1 x 1
18.
6.
6.
6.
18.
6.
6.
27.
6.
21.
6.
27.
6.
6.
18.
6.
6.
6.

6.
21.

6.
27.
6.

21.
Ke =
18.
6.
6.
6.
18.
6.
6.
27.
6.
21.
6.
27.
6.
6.
18.
6.
6.
6.
6.
21.
6.
27.
6.
21.
Eigenvalues of Ke= {896., 60., 24., 0, 0, 0, 0, 0}
Gauss integration rule: 2 x 2
24.
6.
0
6.
12.
6.
6.
36.
6.
12.
6.
18.
0
6.
24.
6.
12.
6.
6.
12.
6.
36.
6.
30.
12.
6.
12.
6.
24.
6.
6.
18.
6.
30.
6.
36.
12.
6.
12.
6.
0
6.
6.
30.
6.
18.
6.
12.
Eigenvalues of Ke= {896., 60., 36., 24., 24., 0, 0, 0}
Gauss integration rule: 3 x 3
24.
6.
0
6.
12.
6.
6.
36.
6.
12.
6.
18.
0
6.
24.
6.
12.
6.
6.
12.
6.
36.
6.
30.
Ke =
12.
6.
12.
6.
24.
6.
6.
18.
6.
30.
6.
36.
12.
6.
12.
6.
0
6.
6.
30.
6.
18.
6.
12.
Eigenvalues of Ke= {896., 60., 36., 24., 24., 0, 0, 0}
Gauss integration rule: 4 x 4
24.
6.
0
6.
12.
6.
6.
36.
6.
12.
6.
18.
0
6.
24.
6.
12.
6.
6.
12.
6.
36.
6.
30.
Ke =
12.
6.
12.
6.
24.
6.
6.
18.
6.
30.
6.
36.
12.
6.
12.
6.
0
6.
6.
30.
6.
18.
6.
12.
Eigenvalues of Ke= {896., 60., 36., 24., 24., 0, 0, 0}

6.
6.
18.
6.
6.
6.
18.
6.

6.
21.
6.
27.
6.
21.
6.
27.

12.
6.
12.
6.
0
6.
24.
6.

6.
30.
6.
18.
6.
12.
6.
36.

12.
6.
12.
6.
0
6.
24.
6.

6.
30.
6.
18.
6.
12.
6.
36.

12.
6.
12.
6.
0
6.
24.
6.

6.
30.
6.
18.
6.
12.
6.
36.

Figure E17.6. Results from running the script of Figure E17.5.

1719

1720

Chapter 17: ISOPARAMETRIC QUADRILATERALS

The Ke given by the 1 1 Gauss rule has a rank deciency of two because it has ve zero eigenvalues instead
of three. This behavior is explained in Chapter 19.
The 2 2, 3 3 and 4 4 rules produce the same stiffness matrix. This matrix has three zero eigenvalues,
which correspond to the three independent rigid body modes in two dimensions.
The reason for the repeating stiffness matrices is that the integrand hBT EB J is at most quadratic in and
because h and E are constant, B is linear in and , and for a rectangle J is constant over the element. A
2-point product Gauss rule is exact for up to cubic polynomials in the and directions, so it does quadratics
exactly.
EXERCISE 17.2 The verication can be done in several ways: (1) doing it all by hand, (2) using the suggested

Mathematica script of Figure E17.3 and visually comparing to (E17.5) entry by entry, or (3) doing a fully
automatic verication. The script shown in the top cell of Figure E17.7 takes the latter approach. (Any of the
3 ways gets credit, but the last one is quicker.)
ClearAll[Em,,a,b, ,h,1,2,3,4,5,6,7,8];
1=(1+)* ; 2=(1-3*)* ; 3=2+(1-)* ^2;
4=2* ^2+(1-); 5=(1-)* ^2-4; 6=(1-)* ^2-1;
7=4* ^2-(1-); 8= ^2-(1-); b=a/ ;
kfac=Em*h/(24* *(1-^2));
Kr=kfac*{{43, 31, 25,-32,-23,-31,-46, 32},
{0,
44, 32, 48,-31,-24,-32,-27},
{0,0,
43,-31,-46,-32,-23, 31},
{0,0,0,
44, 32,-27, 31,-24},
{0,0,0,0,
43, 31, 25,-32},
{0,0,0,0,0,
44, 32, 48},
{0,0,0,0,0,0,
43,-31},
{0,0,0,0,0,0,0,
44}};
For[i=2,i<=8,i++,For[j=1,j<=i-1,j++,Kr[[i,j]]=Kr[[j,i]]]];
Emat=Simplify[Em/(1-^2)*{{1,,0},{,1,0},{0,0,(1-)/2}}];
ncoor={{0,0},{a,0},{a,b},{0,b}};
Ke=Quad4IsoPMembraneStiffness[ncoor,Emat,h,{False,2}];
dK=Simplify[Kr-Ke]; Print["This should be the null matrix:",
dK//MatrixForm];

This should be the null matrix:

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0

Figure E17.7. Script to do Exercise 17.2 and results.

The code of Figure E17.7 builds the alleged exact stiffness (E17.5) in Kr (the 2 For loops symmetrize that
matrix), then invokes Quad4IsoPMembraneStiffness with symbolic inputs placing the returned stiffness in
Ke. The difference Ke-Kr is printed upon simplication. If the two matrices are identical for any symbolic
input the difference should be the null matrix of order 8, which can be easily inspected. As shown in the
bottom cell of Figure E17.7, the matrices match.
EXERCISE 17.3

The verication (a) is immediate. The script shown in Figure E17.8 does (b), computes the energy ratio r for
(c), and plots it for Poissons ratios = 0, 14 , 12 . The plots show that r never exceeds 1.

1720

1721

Solutions to Exercises

ClearAll[Em,,a,b,h,Iz,M]; b= *a; Iz=h*b^3/12; kappa=M/(Em*Iz);


ncoor={{-a/2,-b/2},{a/2,-b/2},{a/2,b/2},{-a/2,b/2}};
Emat=Em/(1-^2)*{{1,,0},{,1,0},{0,0,(1-)/2}};
Ke= Quad4IsoPMembraneStiffness[ncoor,{Emat,0,0},{h},{False,2}];
Ke=Simplify[Ke]; u=kappa*a*b*{-1,0,1,0,-1,0,1,0}/4;
Uquad=Simplify[u.Ke.u/2]; Ubeam=Simplify[M*kappa*a/2];
Print["Uquad=",Uquad, ", Ubeam=",Ubeam];
r=Simplify[Ubeam/Uquad]; Print["r=Ubeam/Uquad=",r];
Plot[{r/.->0,r/.->1/4,r/.->1/2},{ ,0.1,10}];
3 M2 1  2 2  
 ,
Uquad 
a2 Em h 5 1  2 

6 M2

Ubeam 
a2 Em h 3

2 2 1  2 
rUbeamUquad 

1  2 2 
1
0.8
0.6
0.4
0.2
4

10

Figure E17.8. Script to do Exercise 17.3 and results.

EXERCISE 17.4 Not assigned.


EXERCISE 17.5 Not assigned.
EXERCISE 17.6 Not assigned.
EXERCISE 17.7 Not assigned.

1721

You might also like