Ilovepdf.com (3)

Multiple-Choice Test
Eulers Method
Ordinary Differential Equations
COMPLETE SOLUTION SET

1. To solve the ordinary differential equation
( ) 5 0 , sin 5 3
2
= = + y x y
dx
dy

by Eulers method, you need to rewrite the equation as
(A) ( ) 5 0 , 5 sin
2
= = y y x
dx
dy

(B) ( ) ( ) 5 0 , 5 sin
3
1
2
= = y y x
dx
dy

(C) ( ) 5 0 ,
3
5
cos
3
1
3
=
= y
y
x
dx
dy

(D) ( ) 5 0 , sin
3
1
= = y x
dx
dy

Solution
The correct answer is (B).

To solve ordinary differential equations by Eulers method, you need to rewrite the equation in
the following form
( ) ( )
0
0 , , y y y x f
dx
dy
= =
Thus,
( ) 5 0 , sin 5 3
2
= = + y x y
dx
dy

( )
( ) ( ) 5 0 , 5 sin
3
1
5 0 , 5 sin 3
2
2
= =
= =
y y x
dx
dy
y y x
dx
dy

2. Given
( ) 5 3 . 0 , sin 5 3
2
= = + y x y
dx
dy

and using a step size of 3 . 0 = h , the value of ( ) 9 . 0 y using Eulers method is most nearly
(A) 318 . 35
(B) 458 . 36
(C) 91 . 658
(D) 05 . 669

Solution
The correct answer is (A).

First rewrite the differential equation in the proper form.

( )
( ) ( )
2
2
5 sin
3
1
,
5 sin
3
1
y x y x f
y x
dx
dy
=
=

Eulers method is given by
( )h y x f y y
i i i i
,
1
+ =
+

where
3 . 0 = h
For 0 = i , 3 . 0
0
= x , 5
0
= y

( )
( )
( ) ( ) ( )
( )
4704 . 7
470 . 12 5
3 . 0 5 5 3 . 0 sin
3
1
5
3 . 0 5 , 3 . 0 5
,
2
0 0 0 1
=
+ =
+ =
+ =
+ =
f
h y x f y y

1
y is the approximate value of y at
h x x x + = =
0 1
3 . 0 3 . 0 + = 6 . 0 =
For 1 = i , 6 . 0
1
= x , 4704 . 7
1
= y

( )
( )
( ) ( ) ( )
318 . 35
847 . 27 4704 . 7
3 . 0 4704 . 7 5 6 . 0 sin
3
1
4704 . 7
3 . 0 4704 . 7 , 6 . 0 4704 . 7
,
2
1 1 1 2
=
=
+ =
+ =
+ =
f
h y x f y y

2
h x x x + = =
1 2
3 . 0 6 . 0 + = 9 . 0 =
( ) 318 . 35 9 . 0 y
3. Given
( ) 5 3 . 0 , 3
1 . 0
= = + y e y
dx
dy
x

and using a step size of 3 . 0 = h , the best estimate of ( ) 9 . 0
dx
dy
using Eulers method is most
nearly
(A) 37319 . 0
(B) 36288 . 0
(C) 35381 . 0
(D) 34341 . 0

Solution

First rewrite the differential equation in the proper form.

( )
( ) ( ) y e y x f
y e
dx
dy
x
x
=
=
1 . 0
1 . 0
3
1
,
3
1

( )h y x f y y
i i i i
,
1
+ =
+

where
3 . 0 = h
For 0 = i , 3 . 0
0
= x , 5
0
= y

( )
( )
( )
( )
8794 . 4
12056 . 0 5
3 . 0 5
3
1
5
3 . 0 5 , 3 . 0 5
,
3 . 0 1 . 0
0 0 0 1
=
+ =
+ =
+ =
+ =
e
f
h y x f y y

1
h x x x + = =
0 1
3 . 0 3 . 0 + = 6 . 0 =
For 1 = i , 6 . 0
1
= x , 8794 . 4
1
= y

( )
( )
( )
( ) 11471 . 0 8794 . 4
3 . 0 8794 . 4
3
1
8794 . 4
3 . 0 8794 . 4 , 6 . 0 8794 . 4
,
6 . 0 1 . 0
1 1 1 2
+ =
+ =
+ =
+ =
e
f
h y x f y y

2
h x x x + = =
1 2
3 . 0 6 . 0 + = 9 . 0 =

( ) 7647 . 4 9 . 0 y

Thus

( )
( ) ( )
36288 . 0
7647 . 4
3
1
9 . 0
3
1
9 . 0 1 . 0
1 . 0
=

=
e
dx
dy
y e
dx
dy
x

4. The velocity (m/s) of a body is given as a function of time (seconds) by
( ) ( ) 0 , 1 ln 200 + = t t t t v
Using Eulers method with a step size of 5 seconds, the distance traveled in meters by the body
from 2 = t to 12 = t seconds is most nearly
(A) 3133.1
(B) 3939.7
(C) 5638.0
(D) 39397

Solution

( ) ( )
( )
( ) ( ) t t S t f
t t
dt
dS
t t t v
+ =
+ =
+ =
1 ln 200 ,
1 ln 200
1 ln 200

( )h S t f S S
i i i i
,
1
+ =
+

where
5 . 0 = h
For 0 = i , s 2
0
= t , m 0
0
= S (assuming m 0
0
= S would make
2
S the value of the distance
covered, as the distance covered is
0 2
S S )
( )
( )
( ) ( )
7
5 2
m 6 . 1088
5 2 2 1 ln 200 0
5 0 , 2 0
,
0 1
0 0 0 1
=
+ =
+ =
=
+ + =
+ =
+ =
h t t
f
h S t f S S

For 1 = i , s 7
1
= t , m 61 . 1088
1
= S
( )
( )
( ) ( )
m 1 . 3133
4 . 2044 6 . 1088
5 7 7 1 ln 200 6 . 1088
5 6 . 1088 , 7 6 . 1088
,
1 1 1 2
=
+ =
+ + =
+ =
+ =
f
h S t f S S

m 1 . 3133
) 2 ( ) 12 (
0 2
=
S S S S
Note to the student:
You do not have to assume m 0
0
= S . Instead, let it be some unknown constant, that is, C S =
0
.
In that case, if you follow Eulers method as above, you would get
( )
( )
( ) ( )
7
5 2
m 6 . 1088
5 2 2 1 ln 200
5 0 , 2
,
0 1
0 0 0 1
=
+ =
+ =
+ =
+ + =
+ =
+ =
h t t
C
C
f C
h S t f S S

For 1 = i , s 7
1
= t , m 61 . 1088
1
+ = C S
( )
( )
( ) ( )
m 1 . 3133
4 . 2044 6 . 1088
5 7 7 1 ln 200 6 . 1088
5 6 . 1088 , 7 6 . 1088
,
1 1 1 2
+ =
+ + =
+ + + =
+ + =
+ =
C
C
C
f C
h S t f S S

m 1 . 3133
1 . 3133
) 2 ( ) 12 (
0 2
=
+ =

C C
S S S S
5. Eulers method can be derived by using the first two terms of the Taylor series of writing the
value of
1 + i
y , that is the value of y at
1 + i
x , in terms of
i
y and all the derivatives of y at
i
x . If
i i
x x h =
+1
, the explicit expression for
1 + i
y if the first three terms of the Taylor series are
chosen for the ordinary differential equation
( ) 7 0 , 3 2
5
= = +

y e y
dx
dy
x

would be
(A) ( )h y e y y
i
x
i i
i
3
2
1
5
1
+ =

+

(B) ( )
2 5 5
1
2
5
2
1
3
2
1
h e h y e y y
i i
x
i
x
i i

+ =

+

(C) ( )
2 5 5
1
4
9
4
13
2
1
3
2
1
h y e h y e y y
i
x
i
x
i i
i i
+ + + =

+

(D) ( )
2 5
1
2
3
3
2
1
h y h y e y y
i i
x
i i
i
+ =

+

Solution
The correct answer is (C).

The differential equation
( ) 7 0 , 3 2
5
= = +

y e y
dx
dy
x

is rewritten as

( ) ( )
( ) ( ) y e y x f
y y e
dx
dy
x
x
3
2
1
,
7 0 , 3
2
1
5
5
=
= =

The Taylor series is given by
( ) ( ) ( ) ...
! 3
1
! 2
1
3
1
,
3
3
2
1
,
2
2
1
,
1
+ + + + =
+ + + + i i
y x
i i
y x
i i
y x
i i
x x
dx
y d
x x
dx
y d
x x
dx
dy
y y
i i i i
i i

( ) ( ) ( ) ... ) , ( ' '
! 3
1
) , ( '
! 2
1
) , (
3
1
2
1 1 1
+ + + + =
+ + + + i i i i i i i i i i i i i i
x x y x f x x y x f x x y x f y y
If we look at the first three terms of the Taylor series

( ) ( )
( ) ( )
2
2
1 1 1
, '
! 2
1
,
) , ( '
! 2
1
) , (
h y x f h y x f y
x x y x f x x y x f y y
i i i i i
i i i i i i i i i i
+ + =
+ + =
+ + +

where
i i
x x h =
+1

( )
( ) ( )
y e
y e e
dx
dy
y
f
x
f
y x f
x
x x
4
9
4
13

3
2
1
2
3
5
2
1

,
5
5 5
+ =
+ =

then the value of
1 + i
y is given by

( )
2 5 5
1
4
9
4
13
2
1
3
2
1
h y e h y e y y
i
x
i
x
i i
i i
+ + + =

+
6. A homicide victim is found at 6:00 PM in an office building that is maintained at 72 F.
When the victim was found, his body temperature was at 85 F. Three hours later at 9:00 PM,
his body temperature was recorded at 78 F. Assume the temperature of the body at the time
of death is the normal human body temperature of 98.6 F.
The governing equation for the temperature of the body is
) (
a
k
dt
d

=
where
= temperature of the body, F

a
= ambient temperature, F
t = time, hours
k = constant based on thermal properties of the body and air
The estimated time of death most nearly is
(A) 2:11 PM
(B) 3:13 PM
(C) 4:34 PM
(D) 5:12 PM
Solution

a
a
k k
dt
d
k
dt
d

= +
= ) (

The characteristic equation of the above ordinary differential equations is
0 = + k r
The solution to this equation is

( )
a
kt
H
k k D
Ae
k r

= +
=
=

The particular solution is of the form
B
P
=
Substituting this solution in the ordinary differential equation,

a
a
B
k kB

=
= + 0

The complete solution is

a
kt
P H
Ae

+ =
+ =

Given
72 =
a

and using 12 noon as the reference time of 0 = t ,
( )
( )
( ) 6 . 98
78 9
85 6
=
=
=
B

where
B = time of death
we get
72 85
6
+ =
k
Ae (1)
72 78
9
+ =
k
Ae (2)
72 6 . 98 + =
B k
Ae (3)
Use Equations (1) and (2) to find A and k.

13
72 85
6
6
=
+ =

k
k
Ae
Ae
(4)
6
72 78
9
9
=
+ =

k
k
Ae
Ae
(5)
Dividing Equation (4) by Equation (5) gives
( ) ( )
hours
1
25773 . 0
1667 . 2 ln
3
1
1667 . 2
6
13
3
9
6
=
=
=
=
k
e
Ae
Ae
k
k
k

Knowing the value of k, from Equation (5)
028 . 61 = A F
Substitute k and A into Equation (3) to find B.
( )
hours 2220 . 3
83042 . 0 25773 . 0
25773 . 0 028 . 61 ln 6 . 26 ln
028 . 61 6 . 26
72 028 . 61 6 . 98
72 6 . 98
25773 . 0
25773 . 0
=
=
+ =
=
+ =
+ =

B
B
B
e
e
Ae
B
B
B k

Note to the student:
You can also do the problem by assuming that the initial time reference is zero, and that the
temperature then is 6 . 98 ) 0 ( = . Then the temperature is given at the time the body was found
as 85 ) ( = C F, and that 78 ) 3 ( = + C F. You can now find k, A and C just like as given
above. The value of C in fact is the time between the body was found and the time of death.
You will get 7780 . 2 = C hrs.

The time of death is 3.2220 hrs from 12 noon, that is = PM ) 60 2220 . 0 ( : 3 3:13 PM.

Runge-Kutta 2
nd
Order Method

( ) 5 0 , sin 3
2
= = + y x xy
dx
dy

by the Runge-Kutta 2
nd
order method, you need to rewrite the equation as
(A) ( ) 5 0 , sin
2
= = y xy x
dx
dy

(B) ( ) ( ) 5 0 , sin
3
1
2
= = y xy x
dx
dy

(C) ( ) 5 0 ,
3
cos
3
1
3
=
= y
xy
x
dx
dy

(D) ( ) 5 0 , sin
3
1
= = y x
dx
dy

Solution

To solve ordinary differential equations by the Runge-Kutta 2
nd
order method, you need to
rewrite the equation in the following form
( ) ( )
0
0 , , y y y x f
dx
dy
= =
Thus,
( )
5 ) 0 ( , sin 3
5 0 , sin 3
2
2
= =
= = +
y xy x
dx
dy
y x xy
dx
dy

( ) ( ) 5 0 , sin
3
1
2
= = y xy x
dx
dy

2. Given
( ) 5 3 . 0 , sin 5 3
2
= = + y x y
dx
dy

and using a step size of , the value of 3 . 0 = h ( ) 9 . 0 y using the Runge-Kutta 2
nd
order Heun
method is most nearly
(A) 4297.4
(B) 4936.7
(C) 0.21336
14
10
(D) 0.24489
14
10

Solution
The correct answer (A).

( ) 5 3 . 0 , sin 5 3
2
= = + y x y
dx
dy

is rewritten as

( ) ( )
( ) ( )
2
2
5 sin
3
1
,
, 5 sin
3
1
y x y x f
y x f y x
dx
dy
=
= =

In Huens method
2
1
2
= a is chosen, giving
2
1
1
= a
1
1
= p
1
11
= q
resulting in
h k k y y
i i

+ + =
+ 2 1 1
2
1
2
1

where
(
i i
y x f k ,
1
= )
( ) h k y h x f k
i i 1 2
, + + =
3 . 0 = h

for 5 , 3 . 0 , 0
0 0
= = = y x i

( )
( )
( ) ( ) ( )
5682 . 41
5 5 3 . 0 sin
3
1
5 , 3 . 0
,
2
0 0 1
=
=
=
=
f
y x f k

( )
( ) ( )
( )
( ) ( )
( )
824 . 92
04 . 279 56464 . 0
3
1
4704 . 7 5 ) 6 . 0 sin(
3
1
4704 . 7 , 6 . 0
3 . 0 5682 . 41 5 , 3 . 0 3 . 0
,
2
1 0 0 2
=
=
=
=
+ + =
+ + =
f
f
h k y h x f k

( ) ( )
( ) ( )
( )
159 . 15
3 . 0 196 . 67 5
3 . 0 824 . 92
2
1
5682 . 41
2
1
5
2
1
2
1
2 1 0 1
=
+ =
+ + =
+ + = h k k y y

6 . 0
3 . 0 3 . 0
0 1
=
+ =
+ = h x x

( ) 159 . 15 6 . 0
1
= y y
for 159 . 15 , 6 . 0 , 1
1 1
= = = y x i

( )
( )
( ) ( ) ( )
80 . 382
159 . 15 5 6 . 0 sin
3
1
159 . 15 , 6 . 0
,
2
1 1 1
=
=
=
=
f
y x f k

( )
( ) ( )
( )
( ) ( )
( )
28166
84500 78333 . 0
3
1
00 . 130 5 ) 9 . 0 sin(
3
1
00 . 130 , 9 . 0
3 . 0 80 . 382 159 . 15 , 3 . 0 6 . 0
,
2
1 1 1 2
=
=
=
=
+ + =
+ + =
f
f
h k y h x f k

( ) ( )
( ) ( )
( )
4 . 4297
3 . 0 14274 159 . 15
3 . 0 28166
2
1
80 . 382
2
1
159 . 15
2
1
2
1
2 1 1 2
=
+ =
+ + =
+ + = h k k y y

9 . 0
3 . 0 6 . 0
1 2
=
+ =
+ = h x x

( ) 4 . 4297 9 . 0
2
= y y
3. Given
( ) 5 3 . 0 , 5 3
1 . 0
= = + y e y
dx
dy
x

and using a step size of , the best estimate of 3 . 0 = h ( 9 . 0
dx
dy
) using the Runge-Kutta 2
nd
order
midpoint method most nearly is
(A) 2.2473
(B) 2.2543
(C) 2.6188
(D) 3.2045

Solution

x
e y
dx
dy
1 . 0
5 3 = +
is rewritten as

( ) ( )
( ) ( ) y e y x f
y x f y e
dx
dy
x
x
5
3
1
,
, 5
3
1
1 . 0
1 . 0
=
= =

In the midpoint method is chosen, giving 1
2
= a
0
1
= a

2
1
1
= p

2
1
11
= q
resulting in
h k y y
i i 2 1
+ =
+
where
(
i i
y x f k ,
1
= )

+ + = h k y h x f k
i i 1 2
2
1
,
2
1

3 . 0 = h

for 5 , 3 . 0 , 0
0 0
= = = y x i

( )
( )
( )
( )
3833 . 3
180 . 11 0305 . 1
3
1
5 5
3
1
5 , 3 . 0
,
3 . 0 1 . 0
0 0 1
=
=
=
=
=
e
f
y x f k

( )
( )
( )
( )
1839 . 3
598 . 10 0460 . 1
3
1
4925 . 4 5
3
1
4925 . 4 , 45 . 0
3 . 0 3833 . 3
2
1
5 ,
2
3 . 0
3 . 0
2
1
,
2
1
45 . 0 1 . 0
1 0 0 2
=
=
=
=
+ + =
+ + =
e
f
f
h k y h x f k

h k y y
2 0 1
+ =
( )
0448 . 4
3 . 0 1839 . 3 5
=
+ =

6 . 0
3 . 0 3 . 0
0 1
=
+ =
+ = h x x

( ) 1839 . 3 6 . 0
1
= y y
for 0448 . 4 , 6 . 0 , 1
1 1
= = = y x i

( )
( )
( )
( )
9980 . 2
056 . 10 0618 . 1
3
1
0448 . 4 5
3
1
0448 . 4 , 6 . 0
,
6 . 0 1 . 0
1 1 1
=
=
=
=
=
e
f
y x f k

( )
( )
( )
8008 . 2
5951 . 3 5
3
1
5951 . 3 , 75 . 0
3 . 0 9980 . 2
2
1
0448 . 4 ,
2
3 . 0
6 . 0
2
1
,
2
1
075 . 0
1 1 1 2
=
=
=
+ + =
+ + =
e
f
f
h k y h x f k

h k y y
2 1 2
+ =
( )
2046 . 3
3 . 0 8008 . 2 04483 . 4
=
+ =

9 . 0
3 . 0 6 . 0
1 2
=
+ =
+ = h x x
( ) 2046 . 3 9 . 0
2
= y y
Thus

( ) ( )
( )
( )
( )
( )
6188 . 2
9507 . 8 0942 . 1
3
1
2046 . 3 5
3
1
2046 . 3 , 9 . 0
,
, 9 . 0
9 . 0 1 . 0
2 2
9 . 0
=
=
=
=
=
e
f
y x f
y x f
dx
dy
x

4. The velocity (m/s) of a body is given as a function of time (seconds) by
( ) ( ) 0 , 1 ln 200 + = t t t t v
Using the Runge-Kutta 2
nd
order Ralston method with a step size of 5 seconds, the distance in
meters traveled by the body from to 2 = t 12 = t seconds is estimated most nearly as
(A) 3904.9
(B) 3939.7
(C) 6556.3
(D) 39397

Solution

( ) ( ) 0 , 1 ln 200 , + = = t t t S t f
dt
dS

In the Ralston method
3
2
2
= a is chosen, giving

3
1
1
= a

4
3
1
= p

4
3
11
= q
resulting in
h k k S S
i i

+ + =
+ 2 1 1
3
2
3
1

where
(
i i
S t f k ,
1
= )

+ + = h k S h t f k
i i 1 2
4
3
,
4
3

5 = h
for we are assuming , m 0 , s 2 , 0
0 0
= = = S t i ( ) 0 2 = S

( )
( )
( )
72 . 217
2 2 1 ln 200
0 , 2
,
0 0 1
=
+ =
=
=
f
S t f k

( )
( )
( ) ( )
16 . 376
75 . 5 91 . 381
75 . 5 75 . 5 1 ln 200
46 . 816 , 75 . 5
5 72 . 217
4
3
0 , 5
4
3
2
4
3
,
4
3
1 0 0 2
=
=
+ =
=
+ + =
+ + =
f
f
h k S h t f k

( ) ( )
m 7 . 1616
5 16 . 376
3
2
72 . 217
3
1
0
3
2
3
1
2 1 0 1
=
+ + =
+ + = h k k S S

7
5 2
0 1
=
+ =
+ = h t t
( ) m 7 . 1616 7
1
= S S
for m 7 . 1616 , s 7 , 1
1 1
= = = S t i

( )
( )
( )
89 . 408
7 7 1 ln 200
7 . 1616 , 7
,
1 1 1
=
+ =
=
=
f
S t f k

( )
( )
( ) ( )
02 . 482
75 . 10 77 . 492
75 . 10 75 . 10 1 ln 200
1 . 3150 , 75 . 10
5 89 . 408
4
3
7 . 1616 , 5
4
3
7
4
3
,
4
3
1 1 1 2
=
=
+ =
=
+ + =
+ + =
f
f
h k S h t f k

( ) ( )
( )
m 9 . 3904
5 64 . 457 7 . 1616
5 02 . 482
3
2
89 . 408
3
1
7 . 1616
3
2
3
1
2 1 1 2
=
+ =
+ + =
+ + = h k k S S

12
5 7
1 2
=
+ =
+ = h t t
( ) m 9 . 3904 12
2
= S S
Hence the distance covered between 2 = t and 12 = t seconds is

( ) ( )
m 9 . 3904
0 9 . 3904
2 12
0 2
=
=

=
S S
S S d

5. The Runge-Kutta 2
nd
order method can be derived by using the first three terms of the Taylor
series of writing the value of (that is the value of at ) in terms of (that is the value
of at ) and all the derivatives of at . If
1 + i
y y
1 + i
x
i
y
y
i
x y
i
x
i i
x x h =
+1
, the explicit expression for if
the first three terms of the Taylor series are chosen for solving the ordinary differential equation
1 + i
y
( ) 7 0 , 3 5
2
= = +

y e y
dx
dy
x

would be
(A) ( )
2
5 5 3
2
2
1
h
h y e y y
i
x
i i
i
+ + =

+

(B) ) ( ) (
2
21 5 3
2
2 2
1
h
e h y e y y
i
x
i
x
i i
i i
+ + + =

+
25y
(C) ) ( ) (
2
6 5 3
2
2
1
h
h y e y y
i
i
x
i i

+
+ + =
2
e
i
x

(D) ) ( ) (
2
6 5 3
2
2 2
1
e h y e y y
i i
x
i
x
i i
+ + =

+
5
h
+
Solution

The first three terms of the Taylor series are as follows
( ) ( )
2
1
,
! 2
1
, h y x f h y x f y y
i i i i i i
+ + =
+

Our ordinary differential equation is rewritten as
( ) 7 0 , 3 5
2
= = +

y e y
dx
dy
x

( ) ( ) 7 0 , 5 3 ,
2
= =

y y e y x f
x
Now since y is a function of x,
( )
( ) ( )
dx
dy
y
y x f
x
y x f
y x f
=
, ,
,
( ) ( ) [ ]( ) y e y e
y
y e
x
x x x
5 3 5 3 5 3
2 2 2

=

( ) y e e
x x
5 3 ) 5 ( 6
2 2
+ =

y e
x
25 21
2
+ =

The 2
nd
order formula for the above ordinary differential equation would be
( ) ( )
2
1
,
! 2
1
, h y x f h y x f y y
i i i i i i
+ + =
+

( ) ( )
2 2 2
25 21
2
1
5 3 h y e h y e y
i
x
i
x
i
i i
+ + + =

6. A spherical ball is taken out of a furnace at 1200 K and is allowed to cool in air. You are given
the following

radius of ball = 2 cm
specific heat of ball = 420
K kg
J

density of ball = 7800
3
m
kg

convection coefficient = 350
K m s
J
2

ambient temperature = 300 K

The ordinary differential equation that is given for the temperature of the ball is
( )
8 4 13
10 81 10 20673 . 2 =

dt
d

if only radiation is accounted for. The ordinary differential equation if convection is accounted
for in addition to radiation is
(A) ( ) ( ) 300 10 6026 . 1 10 81 10 20673 . 2
2 8 4 13
=

dt
d

(B) ( ) ( ) 300 10 3982 . 4 10 81 10 20673 . 2
2 8 4 13
=

dt
d

(C) ( ) 300 10 6026 . 1
2
=

dt
d

(D) ( ) 300 10 3982 . 4
2
=

dt
d

Solution

The rate of heat loss due to convection
Rate of heat loss due to convection = ( )
a
hA `
where
convection coefficient = 350 = h
K m s
J
2

A = surface area of the ball, m
2

The energy stored by mass is
Energy stored by mass = mC
where

m = mass of the ball , kg
C = specific heat of the ball,
K kg
J

2
2
2
m 0050265 . 0
02 . 0 4
4
=
=
=
r A

kg 26138 . 0
02 . 0
3
4
7800
3
4
3
3
=
=
=
r
V m

From the energy balance
(Rate at which heat is gained) (Rate at which heat is lost) = (Rate at which heat is stored)
we get

( ) ( ) ( )
( ) ( )
( )
( ) ( )
( )( )( )
( ) ( )
( ) ( ) 300 10 6026 . 1 10 81 10 20673 . 2
300 016026 . 0 10 81 10 20673 . 2
420 26138 . 0
300 0050265 . 0 350
10 81 10 20673 . 2
10 81 10 20673 . 2
10 81 10 20673 . 2
2 8 4 13
8 4 13
8 4 13
8 4 13
8 4 13
=
=

=
=
=

mC
hA
dt
d
hA mC
dt
d
mC
a
a

Runge-Kutta 4
th
Order Method

( ) 5 0 , sin 3
2
= = + y x xy
dx
dy
,
by Runge-Kutta 4
th
order method, you need to rewrite the equation as
(A) ( ) 5 0 , sin
2
= = y xy x
dx
dy

(B) ( ) ( ) 5 0 , sin
3
1
2
= = y xy x
dx
dy

(C) ( ) 5 0 ,
3
cos
3
1
3
=
= y
xy
x
dx
dy

(D) ( ) 5 0 , sin
3
1
= = y x
dx
dy

Solution
The correct answer is (B)

( ) 5 0 , sin 3
2
= = + y x xy
dx
dy

is rewritten as

( )
( ) ( ) 5 0 , sin
3
1
5 0 , sin 3
2
2
= =
= =
y xy x
dx
dy
y xy x
dx
dy

2. Given ( ) 5 3 . 0 , sin 5 3
2
= = + y x y
dx
dy
and using a step size of 3 . 0 = h , the value of using
Runge-Kutta 4
th
order method is most nearly
( 9 . 0 y )
(A) 0.25011
40
10
(B) 4297.4
(C) 1261.5
(D) 0.88498

Solution
The correct answer is (C)

( ) 5 3 . 0 , sin 5 3
2
= = + y x y
dx
dy

is rewritten as
( ) ( ) 5 3 . 0 , 5 sin
3
1
2
= = y y x
dx
dy

the Runge-Kutta 4
th
order method is
( h k k k k y y
i i 4 3 2 1 1
2 2
6
1
+ + + + =
+
)
where
( )
i i
y x f k ,
1
=

+ + = h k y h x f k
i i 1 2
2
1
,
2
1

+ + = h k y h x f k
i i 2 3
2
1
,
2
1

( ) h k y h x f k
i i 3 4
, + + =
for 5 , 3 . 0 , 0
0 0
= = = y x i

( )
( )
( ) ( )
5682 . 41
5 5 3 . 0 sin
3
1
5 , 3 . 0
,
2
0 0 1
=
=
=
=
f
y x f k

( )
( ) ( )
2
1 0 0 2
23523 . 1 5 45 . 0 sin
3
1
23523 . 1 , 45 . 0
3 . 0 5682 . 41
2
1
5 , 3 . 0
2
1
3 . 0
2
1
,
2
1
=
=
+ + =
+ + =
f
f
h k y h x f k

398 . 2 =
( )
( ) ( )
7423 . 35
6403 . 4 5 45 . 0 sin
3
1
6403 . 4 , 45 . 0
3 . 0 398 . 2
2
1
5 , 3 . 0
2
1
3 . 0
2
1
,
2
1
2
2 0 0 3
=
=
=
+ + =
+ + =
f
f
h k y h x f k

( )
( )
( )
( ) ( )
3938 . 54
72269 . 5 5 6 . 0 sin
3
1
72269 . 5 , 6 . 0
3 . 0 7423 . 35 5 , 3 . 0 3 . 0
,
3 0 0 4
=
=
=
+ + =
+ + =
f
f
h k y h x f k

( )
( )
61213 . 3
3 . 0 3938 . 54 7423 . 35 2 398 . 2 2 5682 . 41
6
1
5
2 2
6
1
4 3 2 1 0 1
=
+ =
+ + + + = h k k k k y y

6 . 0
3 . 0 3 . 0
0 1
=
+ =
+ = h x x

for 61213 . 3 , 6 . 0 , 1
1 1
= = = y x i

( )
( )
( ) ( )
5576 . 21
61213 . 3 5 6 . 0 sin
3
1
61213 . 3 , 6 . 0
,
2
1 1 1
=
=
=
=
f
y x f k

( )
( ) ( )
2
1 1 1 2
84577 . 6 5 75 . 0 sin
3
1
84577 . 6 , 75 . 0
3 . 0 5576 . 21
2
1
61213 . 3 , 3 . 0
2
1
6 . 0
2
1
,
2
1
=
=
+ + =
+ + =
f
f
h k y h x f k

8804 . 77 =
( )
( ) ( )
627 . 389
2942 . 15 5 75 . 0 sin
3
1
2942 . 15 , 75 . 0
3 . 0 8804 . 77
2
1
61213 . 3 , 3 . 0
2
1
6 . 0
2
1
,
2
1
2
2 1 1 3
=
=
=
+ + =
+ + =
f
f
h k y h x f k

( )
( )
( )
( ) ( )
2 . 24200
5 . 120 5 9 . 0 sin
3
1
5 . 120 , 9 . 0
3 . 0 627 . 389 61213 . 3 , 3 . 0 6 . 0
,
3 0 0 4
=
=
=
+ + =
+ + =
f
f
h k y h x f k

( )h k k k k y y
4 3 2 1 1 2
2 2
6
1
+ + + + =
( )
45 . 1261
3 . 0 2 . 24200 627 . 389 2 8804 . 77 2 5576 . 21
6
1
61213 . 3
=
+ =

( )
( ) 45 . 1261 9 . 0
45 . 1261
9 . 0
3 . 0 6 . 0
2
1 2

=
+ =
+ =
y
x y
h x x

3. Given ( ) 5 3 . 0 , 3
2
= = + y e y
dx
dy
x
, and using a step size of 3 . 0 = h , the best estimate of ( ) 9 . 0
dx
dy

Runge-Kutta 4
th
order method is most nearly
(A) -1.6604
(B) -1.1785
(C) -0.45831
(D) 2.7270

Solution
The correct answer is (A)

( ) 5 3 . 0 , 3
2
= = + y e y
dx
dy
x

is rewritten as
( ) ( ) 5 3 . 0 ,
3
1
2
= = y y e
dx
dy
x

the Runge-Kutta 4
th
order method is
( h k k k k y y
i i 4 3 2 1 1
2 2
6
1
+ + + + =
+
)
where
( )
i i
y x f k ,
1
=

+ + = h k y h x f k
i i 1 2
2
1
,
2
1

+ + = h k y h x f k
i i 2 3
2
1
,
2
1

( ) h k y h x f k
i i 3 4
, + + =
for 5 , 3 . 0 , 0
0 0
= = = y x i

( )
( )
( ) ( )
8833 . 7
5
3
1
5 , 3 . 0
,
2 3 . 0
0 0 1
=
=
=
=
e
f
y x f k

( )
( ) ( )
2 45 . 0
1 0 0 2
81751 . 3
3
1
81751 . 3 , 45 . 0
3 . 0 8833 . 7
2
1
5 , 3 . 0
2
1
3 . 0
2
1
,
2
1
=
=
+ + =
+ + =
e
f
f
h k y h x f k

33502 . 4 =
( )
( )
784 . 5
34975 . 4
3
1
34975 . 4 , 45 . 0
3 . 0 33502 . 4
2
1
5 , 3 . 0
2
1
3 . 0
2
1
,
2
1
2 45 . 0
2 0 0 3
=
=
=
+ + =
+ + =
e
f
f
h k y h x f k

( )
( )
( )
( )
9456 . 2
2648 . 3
3
1
2648 . 3 , 6 . 0
3 . 0 784 . 5 5 , 3 . 0 3 . 0
,
2 6 . 0
3 0 0 4
=
=
=
+ + =
+ + =
e
f
f
h k y h x f k

( )
( )
44665 . 3
3 . 0 9456 . 2 784 . 5 2 33502 . 4 2 88338 . 7
6
1
5
2 2
6
1
4 3 2 1 0 1
=
+ =
+ + + + = h k k k k y y

for 44665 . 3 , 6 . 0 , 1
1 1
= = = y x i

( )
( )
( )
35243 . 3
44665 . 3
3
1
44665 . 3 , 6 . 0
,
2 6 . 0
1 1 1
=
=
=
=
e
f
y x f k

( )
( ) ( )
2 75 . 0
1 1 1 2
94378 . 2
3
1
94378 . 2 , 75 . 0
3 . 0 35243 . 3
2
1
44665 . 3 , 3 . 0
2
1
6 . 0
2
1
,
2
1
=
=
+ + =
+ + =
e
f
f
h k y h x f k

18295 . 2 =
( )
( ) ( )
53749 . 2
11921 . 3
3
1
11921 . 3 , 75 . 0
3 . 0 18295 . 2
2
1
44665 . 3 , 3 . 0
2
1
6 . 0
2
1
,
2
1
2 75 . 0
2 1 1 3
=
=
=
+ + =
+ + =
e
f
f
h k y h x f k

( )
( )
( )
( ) ( )
58392 . 1
6854 . 2
3
1
6854 . 2 , 9 . 0
3 . 0 53749 . 2 44665 . 3 , 3 . 0 6 . 0
,
2 9 . 0
3 0 0 4
=
=
=
+ =
+ + =
e
f
f
h k y h x f k

( )h k k k k y y
4 3 2 1 1 2
2 2
6
1
+ + + + =
( )
72779 . 2
3 . 0 58392 . 1 53749 . 2 2 18295 . 2 2 35243 . 3
6
1
44665 . 3
=
+ =

9 . 0
3 . 0 6 . 0
1 2
=
+ =
+ = h x x

( )
( ) ( ) ( ) ( )
( ) ( )
6604 . 1
72779 . 2
3
1
9 . 0
3
1
9 . 0
3
1
2 9 . 0
2 9 . 0
2

=
e
y e
dx
dy
y e
dx
dy
x

4. The velocity (m/s) of a parachutist is given as a function of time (seconds) by
( ) 0 ), 17 . 0 tanh( 8 . 55 = t t t
Using Runge-Kutta 4
th
order method with a step size of 5 seconds, the distance traveled by the
body from to seconds is estimated most nearly as 2 = t 12 = t
(A) 341.43 m
(B) 428.97 m
(C) 429.05 m
(D) 703.50 m

Solution
The correct answer is (C)

( ) 0 ), 17 . 0 tanh( 8 . 55 = t t t
the Runge-Kutta 4
th
order method is
( h k k k k S S
i i 4 3 2 1 1
2 2
6
1
+ + + + =
+
)
where
( )
i i
S t f k ,
1
=

+ + = h k S h t f k
i i 1 2
2
1
,
2
1

+ + = h k S h t f k
i i 2 3
2
1
,
2
1

( ) h k S h t f k
i i 3 4
, + + =
for 0 , 2 , 0
0 0
= = = S t i

( )
( )
2732 . 18
) 2 17 . 0 tanh( 8 . 55
0 , 2
,
0 0 1
=
=
=
=
f
S t f k
( )
( )
9359 . 35
5 . 4 17 . 0 tanh 8 . 55
683 . 45 , 5 . 4
5 2732 . 18
2
1
0 , 5
2
1
2
2
1
,
2
1
1 0 0 2
=
=
=
+ + =
+ + =
f
f
h k S h t f k

+ + =
+ + =
5 9359 . 35
2
1
0 , 5
2
1
2
2
1
,
2
1
2 0 0 3
f
h k S h t f k

( )
( )
9359 . 35
5 . 4 17 . 0 tanh 8 . 55
8398 . 89 , 5 . 4
=
=
= f

( )
( )
( )
( )
3463 . 46
7 17 . 0 tanh 8 . 55
68 . 179 , 7
5 9359 . 35 0 , 5 2
,
3 0 0 4
=
=
=
+ + =
+ + =
f
f
h k S h t f k

( )
( )
636 . 173
5 3463 . 46 9359 . 35 2 9359 . 35 2 2732 . 18
6
1
0
2 2
6
1
4 3 2 1 0 1
=
+ + + + =
+ + + + = h k k k k S S

7
5 2
0 1
=
+ =
+ = h t t
for 636 . 173 , 7 , 1
1 1
= = = S t i

( )
( )
3463 . 46
) 7 17 . 0 tanh( 8 . 55
636 . 173 , 7
,
1 1 1
=
=
=
=
f
S t f k
( )
( )
5534 . 51
5 . 9 17 . 0 tanh 8 . 55
502 . 289 , 5 . 9
5 3463 . 46
2
1
636 . 173 , 5
2
1
7
2
1
,
2
1
1 1 1 2
=
=
=
+ + =
+ + =
f
f
h k S h t f k

+ + =
+ + =
5 5534 . 51
2
1
636 . 173 , 5
2
1
7
2
1
,
2
1
2 1 1 3
f
h k S h t f k

( )
( )
5534 . 51
5 . 9 17 . 0 tanh 8 . 55
52 . 302 , 5 . 9
=
=
= f

( )
( )
( )
( )
9445 . 53
12 17 . 0 tanh 8 . 55
403 . 431 , 12
5 5534 . 51 636 . 173 , 5 7
,
3 1 1 4
=
=
=
+ + =
+ + =
f
f
h k S h t f k
( )
( )
m
h k k k k S S
05 . 429
5 9445 . 53 5534 . 51 2 5534 . 51 2 3463 . 46
6
1
636 . 173
2 2
6
1
4 3 2 1 1 2
=
+ + + + =
+ + + + =

12
5 7
1 2
=
+ =
+ = h t t
The distance traveled from t = 2 to t = 12 is
Distance traveled
0 2
S S =

( ) ( )
( ) ( )
m
S S
t S t S
05 . 429
0 05 . 429
2 12
0 2
=
=
=
=
5. Runge-Kutta method can be derived from using first three terms of Taylor series of writing
the value of , that is the value of at , in terms of and all the derivatives of at .
If , the explicit expression for if the first five terms of the Taylor series are
chosen for the ordinary differential equation
1 + i
y y
1 + i
x
i
y y
i
x
i i
x x h =
+1 1 + i
y
( ) 7 0 , 3 5
2
= = +

y e y
dx
dy
x
,
would be
(A) ( )
2
5
5 3
2
2
1
h
h y e y y
i
x
i i
i
+ + =

+

(B)
( ) ( )
( ) ( )
24
390625 300909
6
625 483
2
25 21 5 3
4
2
3
2
2
2 2
1
h
y e
h
y e
h
y e h y e y y
i
x
i
x
i
x
i
x
i i
i i
i i
+ + + +
+ + + =

+

(C)
( ) ( ) ( )
( )
24
24
6
12
2
6 5 3
4
2
3
2
2
2 2
1
h
e
h
e
h
e h y e y y
i
i i i
x
x x
i
x
i i

+
+
+ + + =

(D)
( ) ( ) ( )
( )
24
24
6
12
2
5 6 5 3
4
2
3
2
2
2 2
1
h
e
h
e
h
e h y e y y
i
i i i
x
x x
i
x
i i

+
+
+ + + + =

Solution
The correct answer is (B)

The first five terms of the Taylor series are as follows
( ) ( ) ( ) ( )
4 ' ' ' 3 ' ' 2 '
1
,
! 4
1
,
! 3
1
,
! 2
1
, h y x f h y x f h y x f h y x f y y
i i i i i i i i i i
+ + + + =
+

our ordinary differential equation
( ) 7 0 , 3 5
2
= = +

y e y
dx
dy
x

( ) y e y x f
x
5 3 ,
2
=

Now since y is a function of x,
( )
( ) ( )
dx
dy
y
y x f
x
y x f
y x f
=
, ,
,
( ) ( ) [ ]( ) y e y e
y
y e
x
x x x
5 3 5 3 5 3
2 2 2

=

( ) y e e
x x
5 3 ) 5 ( 6
2 2
+ =

y e
x
25 21
2
+ =

( )
( ) ( )
2
2
, ,
,
dx
y d
y
y x f
x
y x f
y x f

=
( ) ( ) [ ]( ) y e y e
y
y e
x
x x x
25 21 25 21 25 21
2 2 2
+ +
+ +
=

( ) y e e
x x
25 21 ) 25 ( 42
2 2
+ + =

y e
x
625 483
2
+ =

( )
( ) ( )
3
3
, ,
,
dx
y d
y
y x f
x
y x f
y x f

=
( ) ( ) [ ]( ) y e y e
y
y e
x
x x x
625 483 625 483 625 483
2 2 2
+ +
+ +
=

( ) y e e
x x
625 483 ) 625 ( 966
2 2
+ + =

y e
x
390625 3000909
2
+ =

The 4
th
order formula for the above example would be
( ) ( ) ( ) ( )
4 ' ' ' 3 ' ' 2 '
1
,
! 4
1
,
! 3
1
,
! 2
1
, h y x f h y x f h y x f h y x f y y
i i i i i i i i i i
+ + + + =
+

( ) ( )
( ) ( )
24
390625 300909
6
625 483
2
25 21 5 3
4
2
3
2
2
2 2
1
h
y e
h
y e
h
y e h y e y y
i
x
i
x
i
x
i
x
i i
i i
i i
+ + + +
+ + + =

+

6. A hot solid cylinder is immersed in an cool oil bath as part of a quenching process. This
process makes the temperature of the cylinder,
c
, and the bath,
b
, change with time. If the
initial temperature of the bar and the oil bath is given as 600 C and 27C, respectively, and
Length of cylinder = 30 cm
Radius of cylinder = 3 cm
Density of cylinder = 2700
3
m
kg

Specific heat of cylinder = 895
K kg
J

Convection heat transfer coefficient = 100
K m
W
2

Specific heat of oil = 1910
K kg
J

Mass of oil = 2 kg
the coupled ordinary differential equation giving the heat transfer are given by

Oil
Cylinder

(A)
b c
c
dt
d

= + 4 . 362
c b
b
dt
d

= + 5 . 675
(B)
b c
c
dt
d

= 4 . 362
c b
b
dt
d

= 5 . 675
(C)
b c
c
dt
d

= + 5 . 675
c b
b
dt
d

= + 4 . 362
(B)
b c
c
dt
d

= 5 . 675
Solution
The correct answer is (A)

For the cylinder the rate of heat lost due to convection = ( ) ( )
b c
A h .
where
h( ) = the convective cooling coefficient, W/(m
2
-K) and is a function of temperature
= A surface area of the cylinder
The energy stored in the mass is given by
Energy stored by mass = m.C.
where
m = mass of the cylinder, kg
C = specific heat of the cylinder, J/(kg-K)
From an energy balance,
Rate at which heat is gained - Rate at which heat is lost =
= Rate at which heat is stored
gives
( ) ( )
dt
d
mC A h
b c

=
where
( ) = h 100
K m
W
2

( )
( )
( )
kg
m L r
m
kg
V m
m
rL r A
2902 . 2
3 . 0 03 . 0 2700
2700
062204 . 0
3 . 0 03 . 0 2 03 . 0 2
2 2
2
3 2
3
2
2
2
=
=
=
=
=
+ =
+ =

Thus

( ) ( )
( )
b c
c
c
b c
c
b c
dt
d
dt
d
dt
d
mC A h

= +
=
=
517 . 329
895 2902 . 2 062204 . 0 100
Similarly for the oil

( ) ( )
( )
c b
b
b
c b
b
c b
dt
d
dt
d
dt
d
mC A h

= +
=
=
108 . 614
1910 2 062204 . 0 100

Gauss-Seidel Method of Solving
Simultaneous Linear Equations

1. A square matrix [ ] is diagonally dominant if
n n
A

(A) ,
1
n
j i
j
ij ii
a a n i ,..., 2 , 1 =
(B) ,
1
n
j i
j
ij ii
a a n i ,..., 2 , 1 and = ,
1
=
>
n
j i
j
ij ii
a a for any n i ,..., 2 , 1 =
(C) ,
1
n
j
ij ii
a a n i ,..., 2 , 1 and = ,
1
=
>
n
j
ij ii
a a for any n i ,..., 2 , 1 =
(D) ,
1
n
j
ij ii
a a n i ,..., 2 , 1 =
Solution
A nn square matrix [A] is a diagonally dominant matrix if ,
1
n
j i
j
ij ii
a a for all i = 1, 2, , n
and
=
>
n
j i
j
ij ii
a a
1
for at least one i, that is, for each row, the absolute value of the diagonal
element is greater than or equal to the sum of the absolute values of the rest of the elements of
that row, and that the inequality is strictly greater than for at least one row. Diagonally dominant
matrices are important in ensuring convergence in iterative schemes of solving simultaneous
linear equations.
[ ]
=
6 2 3
2 4 2
7 6 15
A is a diagonally dominant matrix as

13 7 6 15 15
13 12 11
= + = + = = a a a
4 2 2 4 4
23 21 22
= + = + = = a a a
5 2 3 6 6
32 31 33
= + = + = = a a a
and for at least one row, that is Rows 1 and 3 in this case, the inequality is a strictly greater than
inequality.
2. Using as the initial guess, the values of after three iterations in
the Gauss-Seidel method for
] 5 , 3 , 1 [ ] , , [
3 2 1
= x x x ] , , [
3 2 1
x x x

6
5
2
11 7 2
1 5 1
3 7 12
3
2
1
x
x
x
are
(A) [-2.8333 -1.4333 -1.9727]
(B) [1.4959 -0.90464 -0.84914]
(C) [0.90666 -1.0115 -1.0243]
(D) [1.2148 -0.72060 -0.82451]

Solution

Rewriting the equations gives

11
7 2 6
5
5
12
3 7 2
2 1
3
3 1
2
3 2
1

=

=

=
x x
x
x x
x
x x
x

Iteration #1
Given the initial guess of the solution vector as

5
3
1
3
2
1
x
x
x
we get

( ) ( )
( ) ( )
( ) ( )
9727 . 1
11
4333 . 1 7 8333 . 2 2 6
4333 . 1
5
5 8333 . 2 5
8333 . 2
12
5 3 3 7 2
3
2
1
=

=
=

=
=

=
x
x
x

At the end of the first iteration, the estimate of the solution vector is

9727 . 1
4333 . 1
8333 . 2
3
2
1
x
x
x
Iteration #2
Since the estimate of the solution vector at the end of Iteration #1 is

9727 . 1
4333 . 1
8333 . 2
3
2
1
x
x
x
we get

( ) ( )
( ) ( )
( ) ( )
84915 . 0
11
90464 . 0 7 4960 . 1 2 6
90465 . 0
5
9727 . 1 4960 . 1 5
4960 . 1
12
9727 . 1 3 4333 . 1 7 2
3
2
1
=

=
=

=
=

=
x
x
x

At the end of the second iteration, the estimate of the solution vector is

84915 . 0
90465 . 0
4960 . 1
3
2
1
x
x
x
Iteration #3
Since the estimate of the solution vector at the end of Iteration #2 is

84915 . 0
90465 . 0
4960 . 1
3
2
1
x
x
x
we get

( ) ( )
( ) ( )
( ) ( )
0243 . 1
11
0115 . 1 7 90666 . 0 2 6
0115 . 1
5
84914 . 0 90666 . 0 5
90666 . 0
12
84915 . 0 3 90465 . 0 7 2
3
2
1
=

=
=

=
=

=
x
x
x

At the end of the third iteration, the estimate of the solution vector is

0243 . 1
0115 . 1
90666 . 0
3
2
1
x
x
x

3. To ensure that the following system of equations,

17 2 5 7
5 2
6 11 7 2
3 2 1
3 2 1
3 2 1
= + +
= + +
= +
x x x
x x x
x x x
converges using the Gauss-Seidel method, one can rewrite the above equations as follows:
(A)

17
5
6
2 5 7
1 2 1
11 7 2
3
2
1
x
x
x
(B)
6
5
17
11 7 2
1 2 1
2 5 7
3
2
1
x
x
x
(C)
17
5
6
11 7 2
1 2 1
2 5 7
3
2
1
x
x
x
(D) The equations cannot be rewritten in a form to ensure convergence.

Solution

A system of equations will converge using the Gauss-Seidel method if the coefficient matrix is
diagonally dominant. Thus, rewriting the system of equations as

6
5
17
11 7 2
1 2 1
2 5 7
3
2
1
x
x
x
results in the coefficient matrix being diagonally dominant.

7 2 5 7 7
13 12 11
= + = + = = a a a
2 1 1 2 2
23 21 22
= + = + = = a a a
9 7 2 11 11
32 31 33
= + = + = = a a a
4. For and using
2
7
22
11 7 2
1 5 1
3 7 12
3
2
1
x
x
x
[ ] [ ] 1 2 1
3 2 1
= x x x as the initial
guess, the values of [ are found at the end of each iteration as ]
3 2 1
x x x

Iteration #
1
x
2
x
3
x
1 0.41667 1.1167 0.96818
2 0.93990 1.0184 1.0008
3 0.98908 1.0020 0.99931
4 0.99899 1.0003 1.0000

At what first iteration number would you trust at least 1 significant digit in your solution?
(A) 1
(B) 2
(C) 3
(D) 4

Solution

The absolute relative approximate error at the end of the first iteration is
100
41667 . 0
0000 . 1 41667 . 0
1

=
a

= 140.00%
100
1167 . 1
0000 . 2 1167 . 1
2

=
a

= 79.104%
100
96818 . 0
0000 . 1 96818 . 0
3

=
a

= 3.2864%
The maximum absolute relative approximate error is 140%.
Since, , . There are no significant digits correct. We are looking for 1
significant digit, so we must continue to conduct iterations.
m

2
10 5 . 0 % 140 1 = m

The absolute relative approximate error at the end of the second iteration is

100
0.93990
41667 . 0 0.93990
1

=
a

= 55.669%
100
1.0184
1167 . 1 1.0184
2

=
a

= 9.6509%
100
1.0008
96818 . 0 1.0008
3

=
a

= 3.2564%
The maximum absolute relative approximate error is 55.669%.
Since, , . There are no significant digits correct. We are looking for
1 significant digit so we must continue to conduct more iterations.
m

2
10 5 . 0 % 669 . 55 1 = m

The absolute relative approximate error at the end of the third iteration is

100
0.98908
0.9399 - 0.98908
1
=
a

= 4.9727%
100
1.0020
1.0184 - 1.0020
2
=
a

= 1.6322%
100
0.99930
1.0007 - 0.99930
3
=
a

= 0.14660%
The maximum absolute approximate error is 4.9733%.
Since, , . There is at least one significant digit correct. Since we
were looking for 1 correct significant digit, we need not conduct any more iterations.
m

2
10 5 . 0 % 9733 . 4 1 = m
5. The algorithm for the Gauss-Seidel method to solve [ ][ ] [ ] C X A = is given as follows when
using iterations. The initial value of max n [ ] X is stored in [ ] X .

(A) Sub Seidel ) nmax , , , , ( rhs x a n
For To nmax 1 = k
For 1 = i To n
For To n 1 = j
If ( j i <> ) Then
Sum = Sum + ) ( * ) , ( j x j i a
endif
Next j
) , ( / ) ) ( ( ) ( i i a Sum i rhs i x =
Next i
Next j
End Sub

(B) Sub Seidel nmax) , , , , ( rhs x a n
For 1 = k To nmax
For To n 1 = i
Sum = 0
For To n 1 = j
If ( j i <> ) Then
Sum = Sum + ) ( * ) , ( j x j i a
endif
Next j
) , ( / ) ) ( ( ) ( i i a Sum i rhs i x =
Next i
Next k
End Sub

(C) Sub Seidel nmax) , , , , ( rhs x a n
For To 1 = k nmax
For To 1 = i n
Sum = 0
For To 1 = j n
Sum = Sum + ) ( * ) , ( j x j i a
Next j
) , ( / ) ) ( ( ) ( i i a Sum i rhs i x =
Next i
Next k
End Sub

(D) Sub Seidel ) nmax , , , , ( rhs x a n
For To 1 = k nmax
For To n 1 = i
Sum = 0
For To n 1 = j
If ( j i <> ) Then
Sum = Sum + ) ( * ) , ( j x j i a
endif
Next j
) , ( / ) ) ( ( ) ( i i a Sum i rhs i x =
Next i
Next k
End Sub

Solution

Sub Seidel(n, a, x, rhs, nmax)
For k = 1 To nmax
For i = 1 To n
Sum = 0
For j = 1 To n
If (i <> j) Then
Sum = Sum + a(i, j) * x(j)
endif
Next j
x(i) = (rhs(i) - Sum) / a(i, i)
Next i
Next k
End Sub
Choice (A) is incorrect because the value of the variable Sum needs to be reset on each row.
Choice (C) is incorrect because it does not include an if statement that only adds the sum when
the value of i is not equal to the value of j. The value of Sum would include the value of
a(i,i)*x(i), which would give an incorrect answer for the value of x(i). Choice (D) is incorrect
because it does not subtract the value of Sum from rhs(i), which again, would give the
incorrect answer for x(i).
6. Thermistors measure temperature, have a nonlinear output and are valued for a limited range.
So when a thermistor is manufactured, the manufacturer supplies a resistance vs. temperature
curve. An accurate representation of the curve is generally given by
( ) { } ( ) { }
3
3
2
2 1 0
ln ln ) ln(
1
R a R a R a a
T
+ + + =
where T is temperature in Kelvin, R is resistance in ohms, and are constants of the
calibration curve. Given the following for a thermistor
3 2 1 0
, , , a a a a
R T
ohm C
1101.0
911.3
636.0
451.1
25.113
30.131
40.120
50.128
the value of temperature in for a measured resistance of 900 ohms most nearly is C
(A) 30.002
(B) 30.473
(C) 31.272
(D) 31.445

Solution

Given
( ) { } ( ) { }
3
3
2
2 1 0
ln ln ) ln(
1
R a R a R a a
T
+ + + =
Then from the data in the table we can write four equations

( ) { } ( ) { }
( ) { } ( { } )
( ) { } ( ) { }
( ) { } ( { }
3
3
2
2 1 0
3
3
2
2 1 0
3
3
2
2 1 0
3
3
2
2 1 0
1 . 451 ln 1 . 451 ln ) 1 . 451 ln(
128 . 50
1
636 ln 636 ln ) 636 ln(
120 . 40
1
3 . 911 ln 3 . 911 ln ) 3 . 911 ln(
131 . 30
1
1101 ln 1101 ln ) 1101 ln(
113 . 25
1
a a a a
a a a a
a a a a
a a a a
+ + + =
+ + + =
+ + + =
+ + + =
)

which reduce to

3 2 1 0
3 2 1 0
3 2 1 0
3 2 1 0
29 . 228 353 . 37 1117 . 6 019949 . 0
99 . 268 670 . 41 4552 . 6 024925 . 0
50 . 316 442 . 46 8149 . 6 033188 . 0
58 . 343 056 . 49 0040 . 7 039820 . 0
a a a a
a a a a
a a a a
a a a a
+ + + =
+ + + =
+ + + =
+ + + =

In matrix form the equations can be rewritten as

019949 . 0
024925 . 0
03318 . 0
039820 . 0
1 1117 . 6 353 . 37 29 . 228
1 4552 . 6 670 . 41 99 . 268
1 8149 . 6 442 . 46 5 . 316
1 0040 . 7 056 . 49 58 . 343
0
1
2
3
a
a
a
a
Using any method gives
5964 . 2
2605 . 1
20448 . 0
0.011173
0
1
2
3
=
=
=
=
a
a
a
a

( ) { } ( ) { }
( ) { } ( { } )
C 473 . 30
032816 . 0
1
032816 . 0
900 ln 011173 . 0 900 ln 20448 . 0 ) 900 ln( 2605 . 1 5964 . 2
ln 011173 . 0 ln 20448 . 0 ) ln( 2605 . 1 5964 . 2
1
3 2
3 2
=
=
=
+ + + =
+ + + =
T
R R R
T

Background
Interpolation

1. The number of different polynomials that can go through two fixed data points ( )
1 1
, y x and
( )
2 2
, y x is
A) 0
B) 1
C) 2
D) infinite

Solution
The correct answer is (D).

The number of polynomials that can go through two fixed data points ( )
1 1
, y x and is
infinite. For example the polynomials
( )
2 2
, y x
6 5 = x y and go through the two data points
(2,4) and (3,9). So does every polynomial of the form , where m and n are any
positive integers of your choice.
2
x y =
n m
bx ax y + =

2. Given n+1 data pairs, a unique polynomial of degree __________________ passes
through 1 + n data points.
(A) 1 + n
(B) 1 + n or less
(C) n
(D) n or less

Solution

A unique polynomial of degree n or less passes through 1 + n data points. If the polynomial is
not unique, then at least two polynomials of order n or less pass through the data points.
Assume two polynomials and
1 + n
( ) x P
n
( ) x Q
n
go through 1 + n data points,
( ) ( ) (
n n
y x y x y x , , , , , ,
1 1 0 0
)
Then
( ) ( ) ( ) x Q x P x R
n n n
=
Since and pass through all the ( ) x P
n
( ) x Q
n
1 + n data points,
( ) ( ) n i x Q x P
i n i n
, , 0 , = =
Hence
( ) ( ) n i x Q x P x R
i n i n i n
, , 0 , 0 ) ( = = =
The n
th
order polynomial has ( ) x R
n
1 + n zeros. A polynomial of order n can have zeros
only if it is identical to a zero polynomial, that is,
1 + n
( ) 0 x R
n

Hence
( ) ( ) x Q x P
n n

Extra Notes for the Student:
How can one show that if a second order polynomial has three zeros, then it is zero everywhere?
If , then if it has three zeros at x
1
, x
2
, and x
3
, then ( )
2
2 1 0 2
x a x a a x R + + =
( )
( )
( ) 0
0
0
2
3 2 3 1 0 3 2
2
2 2 2 1 0 2 2
2
1 2 1 1 0 1 2
= + + =
= + + =
= + + =
x a x a a x R
x a x a a x R
x a x a a x R

which in matrix form gives

0
0
0
1
1
1
2
1
0
2
3 3
2
2 2
2
1 1
a
a
a
x x
x x
x x
The above set of equations has a trivial solution, that is, 0
3 2 1
= = = a a a . But is this the only
solution? That is true if the coefficient matrix is invertible.

The determinant of the coefficient matrix can be found symbolically with the forward
elimination steps of nave Gauss elimination to give
) )( )( (
1
1
1
det
1 3 3 2 2 1
2
2
1
2
2 1 3
2
1
2
3 1 3
2
2
2
3 2
2
3 3
2
2 2
2
1 1
x x x x x x
x x x x x x x x x x x x
x x
x x
x x
=
+ + =

Since

3 2 1
x x x
the determinant is non-zero. Hence the coefficient matrix is invertible. is the
only solution, that is,
0
3 2 1
= = = a a a
( ) . 0
2
x R

3. The following function(s) can be used for interpolation:
(A) polynomial
(B) exponential
(C) trigonometric
(D) all of the above

Solution

Polynomials are often used for interpolation because they are easy to evaluate, differentiate and
integrate. However, other functions such as trigonometric and exponential can be used for
interpolation. How is a polynomial easy to evaluate as compared to a trigonometric function?
Because terms such as x
m
in a polynomial involve multiplication of x to itself times.
However, trigonometric and exponential functions include the use of computationally more
involved calculations via a Taylor series.
1 m

4. Polynomials are the most commonly used functions for interpolation because they are
easy to
(A) evaluate
(B) differentiate
(C) integrate
(D) evaluate, differentiate and integrate

Solution

Polynomials are often used for interpolation because they are easy to evaluate, differentiate and
integrate. However, other functions such as trigonometric and exponential can be used for
interpolation. How is a polynomial easy to evaluate as compared to a trigonometric function?
Because terms such as x
m
in a polynomial involve multiplication of x to itself times.
However, trigonometric and exponential functions include the use of computationally more
involved calculations via a Taylor series.
1 m

5. Given 1 + n data points ( ) ( ) ( ) ( )
n n n n
y x y x y x y x , , , ,......, , , ,
1 1 1 1 0 0
, assume you pass a
function ) (x f through all the data points. If now the value of the function ) (x f is
required to be found outside the range of the given x-data, the procedure is called
(A) extrapolation
(B) interpolation
(C) guessing
(D) regression

Solution

If x falls outside the range of x for which the data is given, it is no longer interpolation but
instead is called extrapolation.

6. Given three data points (1,6), (3,28), and (10, 231), it is found that the function
1 3 passes through the three data points. Your estimate of y at 2 = x is
most nearly
2
2
+ + = x x y
(A) 6
(B) 15
(C) 17
(D) 28

Solution

15
1 6 8
1 2 3 2 2 ) 2 (
1 3 2 ) (
2
2
=
+ + =
+ + =
+ + =
y
x x x y

Direct Method
Interpolation

1. Given n+1 data pairs, a unique polynomial of degree ________________ passes through 1 + n
data points.
(A) 1 + n
(B) 1 + n or less
(C) n
(D) n or less

Solution

A unique polynomial of degree n or less passes through 1 + n data points. If the polynomial is
not unique, then at least two polynomials of order n or less pass through the data points.
Assume two polynomials and
1 + n
( ) x P
n
( ) x Q
n
go through 1 + n data points,
( ) ( ) (
n n
y x y x y x , , , , , ,
1 1 0 0
)
Then
( ) ( ) ( ) x Q x P x R
n n n
=
n
( ) x Q
n
1 + n data points,
( ) ( ) n i x Q x P
i n i n
, , 0 , = =
Hence
( ) ( ) n i x Q x P x R
i n i n i n
, , 0 , 0 ) ( = = =
The n
th
n
1 + n
( ) 0 x R
n

Hence
( ) ( ) x Q x P
n n

Extra Notes for the Student:
If , then if it has three zeros at , , and , then ( )
2
2 1 0 2
x a x a a x R + + =
1
x
2
x
3
x
( )
( )
( ) 0
0
0
2
3 2 3 1 0 3 2
2
2 2 2 1 0 2 2
2
1 2 1 1 0 1 2
= + + =
= + + =
= + + =
x a x a a x R
x a x a a x R
x a x a a x R

which in matrix form gives

0
0
0
1
1
1
2
1
0
2
3 3
2
2 2
2
1 1
a
a
a
x x
x x
x x
The above set of equations has a trivial solution, that is, 0
3 2 1
= = = a a a . But is this the only
solution? That is true if the coefficient matrix is invertible.

The determinant of the coefficient matrix can be found symbolically with the forward
elimination steps of nave Gauss elimination to give
) )( )( (
1
1
1
det
1 3 3 2 2 1
2
2
1
2
2 1 3
2
1
2
3 1 3
2
2
2
3 2
2
3 3
2
2 2
2
1 1
x x x x x x
x x
x x
x x
=
+ + =

Since

3 2 1
x x x
the determinant is non-zero. Hence the coefficient matrix is invertible. is the
only solution, that is,
0
3 2 1
= = = a a a
( ) . 0
2
x R

2. The data of the velocity of a body as a function of time is given as follows.
Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
The velocity in m/s at 16 using linear polynomial interpolation is most nearly s
(A) 27.867
(B) 28.333
(C) 30.429
(D) 43.000

Solution

For the first order polynomial, we choose the velocity given by
( ) t a a t v
1 0
+ =

Since we want to find the velocity at 16 = t , we choose the two data points that are closest to
and that also bracket . Those two points are 16 = t 16 = t 15
0
= t and 18
1
= t . Then
( ) 24 , 15
0 0
= = t v t
( ) 37 , 18
1 1
= = t v t
gives
( ) ( ) 24 15 15
1 0
= + = a a v
( ) ( ) 37 18 18
1 0
= + = a a v
Writing the equations in matrix form
37
24
18 1
15 1
1
0
a
a

and solving the above two equations gives
41
0
= a
3333 . 4
1
= a
Hence
( ) t a a t v
1 0
+ =
18 15 , 3333 . 4 41 + = t t
( ) ( ) 16 3333 . 4 41 16 + = v
m/s 333 . 28 =

3. The following data of the velocity of a body as a function of time is given as follows.
Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
The velocity in m/s at 16 using quadratic polynomial interpolation is most nearly s
(A) 27.867
(B) 28.333
(C) 30.429
(D) 43.000

Solution

For second order polynomial interpolation (also called quadratic interpolation), we choose the
velocity given by
( )
2
2 1 0
t a t a a t v + + =
Since we want to find the velocity at 16 = t , we need to choose the three data points that are
closest to and that also bracket 16 = t 16 = t . These three points are 15
0
= t , , and
.
18
1
= t
22
2
= t
( ) 24 , 15
0 0
= = t v t
( ) 37 , 18
1 1
= = t v t
( ) 25 , 22
2 2
= = t v t
gives
( ) ( ) ( ) 24 15 15 15
2
2 1 0
= + + = a a a v
( ) ( ) ( ) 37 18 18 18
2
2 1 0
= + + = a a a v
( ) ( ) ( ) 25 22 22 22
2
2 1 0
= + + = a a a v
Writing the three equations in matrix form
25
37
24
484 22 1
324 18 1
225 15 1
2
1
0
a
a
a

and the solution of the above three equations gives
86 . 323
0
= a
905 . 38
1
= a
0476 . 1
2
= a
Hence
( ) 22 15 , 0476 . 1 905 . 38 86 . 323
2
+ = t t t t v
At , 16 = t
( ) ( ) ( )
2
16 0476 . 1 16 905 . 38 86 . 323 16 + = v
m/s 429 . 30 =
4. The following data of the velocity of a body is given as a function of time
Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
Using quadratic interpolation, the interpolant
( ) 24 18 , 3523 67 . 349 667 . 8
2
+ = t t t t v
approximates the velocity of the body. From this information, one of the times in seconds at
which the velocity of the body is 35 m/s during the above time interval of to is s 18 = t s 24 = t
(A) 18.667
(B) 20.850
(C) 22.200
(D) 22.294

Solution

Using the interpolant, set the velocity equal to 35 m/s and solve for time.
3488 67 . 349 667 . 8 0
3523 67 . 349 667 . 8 35
2
2
+ =
+ =
t t
t t

Using the quadratic equation solution
334 . 17
1249 . 1347 67 . 349
667 . 8 2
3488 667 . 8 4 ) 67 . 349 ( ) 67 . 349 (
2

= t

gives
s 055 . 18
s 294 . 22
=
=
t
t

5. The following data of the velocity of a body is given as a function of time
Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
One of the interpolant approximations for the velocity from the above data is given as
24 18 , 3523 67 . 349 6667 . 8 ) (
2
+ = t t t t v
Using the above interpolant, the distance in meters covered by the body between and
is most nearly
s 19 = t
s 22 = t
(A) 10.337
(B) 88.500
(C) 93.000
(D) 168.00

Solution

Since
dt
dx
v = , taking the integral of the velocity will give the location, x. By taking the integral
over the interval of to we can find the distance traveled, , over that interval. s 19 = t s 22 = t s
( )
[ ]
( ) (
m 337 . 10
23637 23647
) 19 ( 3523 ) 19 ( 84 . 174 ) 19 ( 8889 . 2 ) 22 ( 3523 ) 22 ( 84 . 174 ) 22 ( 8889 . 2
3523 84 . 174 8889 . 2
3523 67 . 349 6667 . 8
2 3 2 3
22
19
2 3
22
19
2
=
=
+ + =
+ =
+ =

t t t
dt t t s
)

6. The following data of the velocity of a body is given as a function of time.

Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123

If you were going to use quadratic interpolation to find the value of the velocity at
seconds, what three data points of time would you choose for interpolation?
9 . 14 = t
(A) 0, 15, 18
(B) 15, 18, 22
(C) 0, 15, 22
(D) 0, 18, 24

Solution

We need to choose the three points closest to s 9 . 14 = t that also bracket . Although the
data points in choice (B) are closest to 14.9, they do not bracket it. This would be performing
extrapolation, not interpolation. Choices (C) and (D) both bracket
s 9 . 14 = t
s 9 . 14 = t but they are not the
closest three data points.
Time (s) Velocity (m/s) How far is s 9 . 14 = t
0 22
9 . 14 0 9 . 14 =
15 24
1 . 0 15 9 . 14 =
18 37
1 . 3 18 9 . 14 =
22 25
1 . 7 22 9 . 14 =
24 123
1 . 9 24 9 . 14 =

Lagrange Method
Interpolation

1. Given n+1 data pairs, a unique polynomial of degree ______________ passes through the
1 + n data points.
(A) 1 + n
(B) n
(C) n or less
(D) 1 + n or less

Solution

A unique polynomial of degree n or less passes through 1 + n data points. Assume two
polynomials and go through ( ) x P
n
( ) x Q
n
1 + n data points,
( ) ( ) (
n n
y x y x y x , , , , , ,
1 1 0 0
)
Then
( ) ( ) ( ) x Q x P x R
n n n
=
n
( ) x Q
n
1 + n data points,
( ) ( ) n i x Q x P
i n i n
, , 0 , = =
Hence
( ) ( ) n i x Q x P x R
i n i n i n
, , 0 , 0 ) ( = = =
The n
th
n
1 + n
( ) 0 x R
n

Hence
( ) ( ) x Q x P
n n

2
2 1 0 2
x a x a a x R + + =
1
x
2
x
3
x
( )
( )
( ) 0
0
0
2
3 2 3 1 0 3 2
2
2 2 2 1 0 2 2
2
1 2 1 1 0 1 2
= + + =
= + + =
= + + =
x a x a a x R
x a x a a x R
x a x a a x R

Which in matrix form gives

(
(
(
=
(
(
(
(
(
(
0
0
0
1
1
1
2
1
0
2
3 3
2
2 2
2
1 1
a
a
a
x x
x x
x x
The final solution exists if the coefficient matrix is invertible. The determinant
of the coefficient matrix can be found symbolically with the forward elimination steps of nave
Gauss elimination to give
0
3 2 1
= = = a a a
) )( )( (
1
1
1
det
1 3 3 2 2 1
2
2
1
2
2 1 3
2
1
2
3 1 3
2
2
2
3 2
2
3 3
2
2 2
2
1 1
x x x x x x
x x
x x
x x
=
+ + =
(
(
(

Since

3 2 1
x x x = =
the determinant is non-zero. Hence, the coefficient matrix is invertible. Therefore,
is the only solution, that is, 0
3 2 1
= = = a a a ( ) . 0
2
x R

2. Given the two points ( ) | | ( ) | | b f b a f a , , , , the linear Lagrange polynomial ( ) x f
1
that passes
through these two points is given by
(A) ( ) ( ) ( ) b f
b a
a x
a f
b a
b x
x f
=
1

(B) ( ) ( ) ( ) b f
a b
x
a f
a b
x
x f
=
1

(C) ( ) ( )
( ) ( )
( ) a b
a b
a f b f
a f x f
+ =
1

(D) ( ) ( ) ( ) b f
a b
a x
a f
b a
b x
x f
=
1

Solution

( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) b f x L a f x L
x f x L x f x L
x f x L x f
x f x L x f
i
i i
n
i
i i n
1 0
1 1 0 0
1
0
1
0
) ( ) ( ) (
) ( ) ( ) (
+ =
+ =
=
=
=
=

a b
a x
x x
x x
x x
x x
x L
b a
b x
x x
x x
x x
x x
x L
x x
x x
x L
j
j j
j
j
j j
j
n
i j
j j i
j
i
=
[
[
[
=
=
=
=
=
=
0 1
0
1
1
0 1
1
1 0
1
1
0
0 0
0
0
) (
) (
) (

( ) ( ) ( ) ( )
( ) ( ) b f
a b
a x
a f
b a
b x
b f x L a f x L x f
=
+ =
1 0 1
) (

3. The Lagrange polynomial that passes through the 3 data points is given by
x
15 18 22
y
24 37 25
( ) ( )( ) ( )( ) ( )( ) 25 37 24
2 1 0 2
x L x L x L x f + + =
The value of at is ( ) x L
1
16 = x
(A) 0.071430
(B) 0.50000
(C) 0.57143
(D) 4.3333

Solution

[
[
=
=
=
=
=
2
1
0 1
1
0
) (
) (
j
j j
j
n
i j
j j i
j
i
x x
x x
x L
x x
x x
x L

50000 . 0
4
6
3
1
22 18
22 16
15 18
15 16
2 1
2
0 1
0
=
|
.
|
\
|
|
.
|
\
|
=
|
.
|
\
|
|
.
|
\
|
=
|
|
.
|
\
|
|
|
.
|
\
|
=
x x
x x
x x
x x

Time (s) 10 15 18 22 24
Velocity (m/s) 22 24 37 25 123
A quadratic Lagrange interpolant is found using three data points, 15 = t , 18 and 22. From
this information, at what of the times given in seconds is the velocity of the body during
the time interval of to seconds.
m/s 26
15 = t 22 = t
(A) 20.173
(B) 21.858
(C) 21.667
(D) 22.020

Solution

=
=
n
i
i i n
t v t L t v
0
) ( ) ( ) (
where
( ) 24 , 15
0 0
= = t v t
( ) 37 , 18
1 1
= = t v t
( ) 25 , 22
2 2
= = t v t
gives

[
=
=

=
n
i j
j j i
j
i
t t
t t
t L
0
) (
[
=
=

=
2
0
0 0
0
) (
j
j j
j
t t
t t
t L

|
|
.
|
\
|
|
|
.
|
\
|
=
2 0
2
1 0
1
t t
t t
t t
t t

[
=
=

=
2
1
0 1
1
) (
j
j j
j
t t
t t
t L

|
|
.
|
\
|
|
|
.
|
\
|
=
2 1
2
0 1
0
t t
t t
t t
t t

[
=
=

=
2
2
0 2
2
) (
j
j j
j
t t
t t
t L

|
|
.
|
\
|
|
|
.
|
\
|
=
1 2
1
0 2
0
t t
t t
t t
t t

) ( ) ( ) ( ) (
2
1 2
1
0 2
0
1
2 1
2
0 1
0
0
2 0
2
1 0
1
2
t v
t t
t t
t t
t t
t v
t t
t t
t t
t t
t v
t t
t t
t t
t t
t v
|
|
.
|
\
|
|
|
.
|
\
|
+
|
|
.
|
\
|
|
|
.
|
\
|
+
|
|
.
|
\
|
|
|
.
|
\
|
=
( ) ( )
( )
( )
278 . 15 and 858 . 21
0476 . 1 2
86 . 349 0476 . 1 4 905 . 38 905 . 38
86 . 349 905 . 38 0476 . 1 0
86 . 323 905 . 38 0476 . 1 26
07 . 241 464 . 29 89286 . 0
5 . 1017 08 . 114 0833 . 3 57 . 452 714 . 45 1429 . 1 26
25
28
270 33
37
12
330 37
24
21
396 40
26
25
4
18
7
15
37
4
22
3
15
24
7
22
3
18
26
25
18 22
18
15 22
15
37
22 18
22
15 18
15
24
22 15
22
18 15
18
26
) 22 (
18 22
18
15 22
15
) 18 (
22 18
22
15 18
15
) 15 (
22 15
22
18 15
18

2
2
2
2
2 2
2 2 2
=

=
+ =
+ =
+ +
+ + + =
|
|
.
|
\
| +
+
|
|
.
|
\
|
+
+
|
|
.
|
\
| +
=
|
.
|
\
|
|
.
|
\
|
+
|
.
|
\
|
|
.
|
\
|
+
|
.
|
\
|
|
.
|
\
|
|
.
|
\
|
|
.
|
\
|
+
|
.
|
\
|
|
.
|
\
|
+
|
.
|
\
|
|
.
|
\
|
=
|
.
|
\
|
|
.
|
\
|
+
|
.
|
\
|
|
.
|
\
|
+
|
.
|
\
|
|
.
|
\
|
=
t
t t
t t
t t
t t t t
t t t t t t
t t t t t t
t t t t t t
v
t t
v
t t
v
t t

5. The path that a robot is following on a x-y plane is found by interpolating four data points as

x 2 4.5 5.5 7
y 7.5 7.5 6 5
( ) 9000 . 3 6048 . 9 2571 . 2 15238 . 0
2 3
+ = x x x x y
The length of the path from to 2 = x 7 = x is
(A) ( ) ( ) ( ) ( ) ( ) ( )
2 2 2 2 2 2
5 . 5 7 6 5 5 . 4 5 . 5 5 . 7 6 2 5 . 4 5 . 7 5 . 7 + + + + +
(B) dx x x x
}
+ +
7
2
2 2 3
) 9000 . 3 6048 . 9 2571 . 2 15238 . 0 ( 1
(C) dx x x
}
+ +
7
2
2 2
) 6048 . 9 5142 . 4 45714 . 0 ( 1
(D) dx x x x
}
+
7
2
2 3
) 9000 . 3 6048 . 9 2571 . 2 15238 . 0 (

Solution

The length S of the curve from a to b is given by ( ) x y
dx
dx
dy
S
b
a
}
|
.
|
\
|
+ =
2
1
where
7
2
=
=
b
a

giving
dx
dx
dy
S
}
|
.
|
\
|
+ =
7
2
2
1
( )
6048 . 9 5142 . 4 45714 . 0
9000 . 3 6048 . 9 2571 . 2 15238 . 0
2
2 3
+ =
+ =
x x
dx
dy
x x x x y

Thus,
( ) dx x x S
}
+ + =
7
2
2
2
6048 . 9 5142 . 4 45714 . 0 1

Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
seconds, what three data points of time would you choose for interpolation?
9 . 14 = t
(A) 0, 15, 18
(B) 15, 18, 22
(C) 0, 15, 22
(D) 0, 18, 24

Solution

s 9 . 14 = t
0 22
9 . 14 0 9 . 14 =
15 24
1 . 0 15 9 . 14 =
18 37
1 . 3 18 9 . 14 =
22 25
1 . 7 22 9 . 14 =
24 123
1 . 9 24 9 . 14 =

Newtons Divided Difference Polynomial Method
Interpolation

1. If a polynomial of degree n has more than n zeros, then the polynomial is
(A) oscillatory
(B) zero everywhere
(C) quadratic
(D) not defined

Solution

A unique polynomial of degree n or less passes through 1 + n data points. Assume two
polynomials and go through ( ) x P
n
( ) x Q
n
1 + n data points,
( ) ( ) (
n n
y x y x y x , , , , , ,
1 1 0 0
)
Then
( ) ( ) ( ) x Q x P x R
n n n
=
n
( ) x Q
n
1 + n data points,
( ) ( ) n i x Q x P
i n i n
, , 0 , = =
Hence
( ) ( ) n i x Q x P x R
i n i n i n
, , 0 , 0 ) ( = = =
The n
th
n
1 + n
( ) 0 x R
n

Hence
( ) ( ) x Q x P
n n

2
2 1 0 2
x a x a a x R + + =
1
x
2
x
3
x
( )
( )
( ) 0
0
0
2
3 2 3 1 0 3 2
2
2 2 2 1 0 2 2
2
1 2 1 1 0 1 2
= + + =
= + + =
= + + =
x a x a a x R
x a x a a x R
x a x a a x R

Which in matrix form gives

0
0
0
1
1
1
2
1
0
2
3 3
2
2 2
2
1 1
a
a
a
x x
x x
x x
The final solution exists if the coefficient matrix is invertible. The determinant
of the coefficient matrix can be found symbolically with the forward elimination steps of nave
Gauss elimination to give
0
3 2 1
= = = a a a

( )( )( )
1 3 3 2 2 1
2
2
1
2
2 1 3
2
1
2
3 1 3
2
2
2
3 2
2
3 3
2
2 2
2
1 1
1
1
1
det
x x x x x x
x x
x x
x x
=
+ + =
Since

3 2 1
x x x
the determinant is non-zero. Hence, the coefficient matrix is invertible. Therefore,
is the only solution, that is, 0
3 2 1
= = = a a a ( ) . 0
2
x R
2. The following x - data is given. y
x 15 18 22
y
24 37 25
The Newtons divided difference second order polynomial for the above data is given by
( ) ( )( ) 18 15 15 ) (
2 1 0 2
+ + = x x b x b b x f
The value of is most nearly
1
b
(A) 1.0480
(B) 0.14333
(C) 4.3333
(D) 24.000

Solution

Given

( )
( ) 37
24
18
15
1 2
0 2
1
0
=
=
=
=
x f
x f
x
x
we have
( ) ( )( )
1 0 2 0 1 0 2
) ( x x x x b x x b b x f + + =
Then
( ) ( )( )
( ) ( )(
( )
( )
0 1 1 0 2
0 1 1 0
1 1 0 1 2 0 1 1 0 1 2
0
1 0 0 0 2 0 0 1 0 0 2
) (

) (

) (
x x b x f
x x b b
x x x x b x x b b x f
b
x x x x b x x b b x f
+ =
+ =
+ + =
=
)
+ + =

Thus,
3333 . 4
15 18
24 37
) ( ) (
0 1
0 2 1 2
1
=

=
x x
x f x f
b

3. The polynomial that passes through the following x - data y

x 18 22 24
y
? 25 123

is given by
24 18 , 3237 75 . 324 125 . 8
2
+ x x x
The corresponding polynomial using Newtons divided difference polynomial is given by

( ) ( )( ) 22 18 18 ) (
2 1 0 2
+ + = x x b x b b x f
The value of is
2
b

(A) 0.25000
(B) 8.1250
(C) 24.000
(D) not obtainable with the information given

Solution

Expanding,
( ) ( )( )
( )
( ) ( )
2
2 2 1 2 1 0
2
2 1 1 0
2 1 0 2
40 396 18
396 40 18
22 18 18 ) (
x b x b b b b b
x x b b x b b
x x b x b b x f
+ + + =
+ + + =
+ + =

This needs to be the same as
3237 75 . 324 125 . 8
2
+ x x
Hence
125 . 8
2
= b
4. Velocity vs. time data for a body is approximated by a second order Newtons divided
difference polynomial as

( ) ( ) ( )( ) 20 10 , 15 20 5540 . 0 20 622 . 39
0
+ + = t t t t b t v

The acceleration in at is
2
m/s 15 = t
(A) 55400 . 0
(B) 622 . 39
(C) 852 . 36
(D) not obtainable with the given information

Solution

( ) ( ) ( )( ) 20 10 , 15 20 5540 . 0 20 622 . 39
0
+ + = t t t t b t v

( ) ( ) ( )
( ) ( ) ( )
( ) ( )
( ) ( ) ( )
2
m/s 852 . 36
7700 . 2 622 . 39
15 15 5540 . 0 20 15 5540 . 0 622 . 39 15
20 10 , 15 5540 . 0 20 5540 . 0 622 . 39
15 5540 . 0 20 5540 . 0 1 622 . 39
=
=
+ + =
+ + =
+ + =
=
a
t t t
t t
t v
dt
d
t a
5. The path that a robot is following on a x-y plane is found by interpolating four data points as

x 2 4.5 5.5 7
y 7.5 7.5 6 5
( ) 9000 . 3 6048 . 9 2571 . 2 15238 . 0
2 3
+ = x x x x y
The length of the path from to 2 = x 7 = x is
(A) ( ) ( ) ( ) ( ) ( ) ( )
2 2 2 2 2 2
5 . 5 7 6 5 5 . 4 5 . 5 5 . 7 6 2 5 . 4 5 . 7 5 . 7 + + + + +
(B) dx x x x
+ +
7
2
2 2 3
) 9000 . 3 6048 . 9 2571 . 2 15238 . 0 ( 1
(C) dx x x
+ +
7
2
2 2
) 6048 . 9 5142 . 4 45714 . 0 ( 1
(D) dx x x x
+
7
2
2 3
) 9000 . 3 6048 . 9 2571 . 2 15238 . 0 (

Solution

The length S of the curve from a to b is given by ( ) x y
dx
dx
dy
S
b
a
+ =
2
1
where
7
2
=
=
b
a

giving
dx
dx
dy
S

+ =
7
2
2
1
( )
6048 . 9 5142 . 4 45714 . 0
9000 . 3 6048 . 9 2571 . 2 15238 . 0
2
2 3
+ =
+ =
x x
dx
dy
x x x x y

Thus,
( ) dx x x S

+ + =
7
2
2
2
6048 . 9 5142 . 4 45714 . 0 1

Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
seconds, the three data points of time you would choose for interpolation are
9 . 14 = t
(A) 0, 15, 18
(B) 15, 18, 22
(C) 0, 15, 22
(D) 0, 18, 24

Solution

s 9 . 14 = t
0 22
9 . 14 0 9 . 14 =
15 24
1 . 0 15 9 . 14 =
18 37
1 . 3 18 9 . 14 =
22 25
1 . 7 22 9 . 14 =
24 123
1 . 9 24 9 . 14 =

Spline Method
Interpolation

1. The following n data points, ( )
1 1
, y x , ( )
2 2
, y x , .. ( )
n n
y x , , are given. For conducting
quadratic spline interpolation the x-data needs to be
(A) equally spaced
(B) placed in ascending or descending order of x -values
(C) integers
(D) positive

Solution

The following n data points, ( )
1 1
, y x , ( )
2 2
, y x , .. ( )
n n
y x , , are given. For conducting
quadratic spline interpolation the x-data needs to be arranged in ascending or descending order.

2. In cubic spline interpolation,
(A) the first derivatives of the splines are continuous at the interior data
points
(B) the second derivatives of the splines are continuous at the interior data
points
(C) the first and the second derivatives of the splines are continuous at the
interior data points
(D) the third derivatives of the splines are continuous at the interior data
points

Solution

In cubic spline interpolation, the first and the second derivatives of the splines are continuous at
the interior data points. In quadratic spline interpolation, only the first derivatives of the splines
are continuous at the interior data points.

3. The following incomplete y vs. x data is given.

x 1 2 4 6 7
y
5 11 ???? ???? 32
The data is fit by quadratic spline interpolants given by
( ) 1 = ax x f , 2 1 x
( ) 4 2 , 9 14 2
2
+ = x x x x f
( ) 6 4 ,
2
+ + = x d cx bx x f
( ) 7 6 , 928 303 25
2
+ = x x x x f
where d c b a and , , , are constants. The value of c is most nearly
(A) 00 . 303
(B) 50 . 144
(C) 0.0000
(D) 14.000

Solution

Solution Method 1:
Since the first derivatives of two quadratic splines are continuous at the interior points, at x = 4
( ) ( )
4
2
4
2
9 14 2
= =
+ = +
x x
x x
dx
d
d cx bx
dx
d

14 ) 4 ( 4 ) 4 ( 2
14 4 2
4 4
+ = +
+ = +
= =
c b
x c bx
x x

2 8 = + c b (1)

and at x = 6
( ) ( )
6
2
6
2
928 303 25
= =
+ = +
x x
x x
dx
d
d cx bx
dx
d

303 ) 6 ( 50 ) 6 ( 2
303 50 2
6 6
= +
= +
= =
c b
x c bx
x x

3 12 = + c b (2)

Equations (1) and (2) in matrix form
(
=
(
3
2
1 12
1 8
c
b

Solving these equations gives
0
25 . 0
=
=
c
b

Solution Method 2:
The third spline d cx bx + +
2
goes through 4 = x . However, so does the second spline. We can
use this latter knowledge to find the value of y at . 4 = x
, 9 14 2 ) (
2
+ = x x x f 4 2 x
9 ) 4 ( 14 ) 4 ( 2 ) 4 (
2
+ = f
15 =
2
goes through 4 = x .
Hence
15 ) 4 ( ) 4 (
2
= + + d c b
15 4 16 = + + d c b (1)
2
goes through 6 = x . However, so does the fourth spline. We can use
this latter knowledge to find the value of y at 6 = x .
, 928 303 25 ) (
2
+ = x x x f 7 6 x
928 ) 6 ( 303 ) 6 ( 25 ) 6 (
2
+ = f
10 =
2
goes through 6 = x .
Hence
10 ) 6 ( ) 6 (
2
= + + d c b
10 6 36 = + + d c b (2)
Since the first derivatives of second and third quadratic splines are continuous at the interior
points, at x = 4
( ) ( )
4
2
4
2
9 14 2
= =
+ = +
x x
x x
dx
d
d cx bx
dx
d

14 ) 4 ( 4 ) 4 ( 2
14 4 2
4 4
+ = +
+ = +
= =
c b
x c bx
x x

2 8 = + c b (3)

Equations (1), (2) and (3) are then
15 4 16 = + + d c b
10 6 36 = + + d c b
2 0 8 = + + d c b
Putting these equations in matrix form gives

(
(
(
0 1 8
1 6 36
1 4 16
(
(
(
d
c
b
=
(
(
(
2
10
15

Solving the above equations gives

19
0
25 . 0
=
=
=
d
c
b


x 1 2 4 6 7
y
5 11 ???? ???? 32
( ) 2 1 , 1 = x ax x f ,
( ) 4 2 , 9 14 2
2
+ = x x x x f
( ) 6 4 ,
2
+ + = x d cx bx x f
( ) 7 6 ,
2
+ + = x g fx ex x f
where g f e d c b a and , , , , , , are constants. The value of
dx
df
at 6 . 2 = x most nearly is
(A) 50 . 144
(B) 0000 . 4
(C) 3.6000
(D) 12.200

Solution

Since the spline
( ) 9 14 2
2
+ = x x x f
is valid in the interval 4 2 x , the derivative at x = 2.6 is
( )
( )
6000 . 3
14 4 . 10
14 6 . 2 4 6 . 2
14 4
=
+ =
+ =
+ =
dx
df
x x
dx
df


x 1 2 4 6 7
y
5 11 ???? ???? 32
( ) 2 1 , 1 = x ax x f ,
( ) 4 2 , 9 14 2
2
+ = x x x x f
( ) 6 4 ,
2
+ + = x d cx bx x f
( ) 7 6 , 928 303 25
2
+ = x x x x f
Where d c b a and , , , are constants. What is the value of ( )
5 . 3
5 . 1
dx x f ?
(A) 23.500
(B) 25.667
(C) 25.750
(D) 28.000
Solution

To find ( )
5 . 3
5 . 1
dx x f we must take ( ) ( )dx x x dx ax

+ +
5 . 3
2
2
2
5 . 1
9 14 2 1 but first we have to find the
value of the constant a. Since at x = 2, y = 11
6
2
1 11
11 1 2
=
+
=
=
a
a

Thus,
( ) ( ) ( )
( ) ( ) | |
( ) ( ) | |
( ) ( ) | |
| | | |
75 . 25
21 75 . 4
18 28 3333 . 5 5 . 31 75 . 85 583 . 28
5 . 1 75 . 6 2 12
2 9 2 7 2
3
2
5 . 3 9 5 . 3 7 5 . 3
3
2
5 . 1 5 . 1 3 2 2 3
9
2
14
3
2
2
6
9 14 2 1 6
2 3 2 3
2 2
5 . 3
2
2 3
2
5 . 1
2
5 . 3
2
2
2
5 . 1
5 . 3
5 . 1
=
+ =
+ +
+ =
(
|
.
|
\
|
+
|
.
|
\
|
+
+ =
(
+ +
(
=
+ + =

x x x x x
dx x x dx x dx x f

6. A robot needs to follow a path that passes consecutively through six points as shown
in the figure. To find the shortest path that is also smooth you would recommend
which of the following?
(A) Pass a fifth order polynomial through the data
(B) Pass linear splines through the data
(C) Pass quadratic splines through the data
(D) Regress the data to a second order polynomial
Path of a Robot
0
1
2
3
4
5
6
7
8
0 5 10 15
x
y

Solution

Using linear splines (Choice B) would create a straight-line path between consecutive points.
Although this will be the shortest path it will not be smooth. Regressing the data to a second
order polynomial (Choice D) will result in a smooth path but it will not pass through all the
points. As demonstrated in the following figure, using polynomial interpolation such as choice
(A) is a bad idea and will result in a long path. By using quadratic spline interpolation (choice
C), the path will be short as well as smooth.

0
2
4
6
8
10
0 2 4 6 8 10 12
x

y

Cubic Spline Fifth Order Polynomial

Romberg Rule
Integration

1. If
n
I is the value of ( )
b
a
dx x f using the n-segment trapezoidal rule, a better estimate
of the integral can be found using Richardsons extrapolation as
(A)
15
2
2
n n
n
I I
I

+
(B)
3
2
2
n n
n
I I
I

+
(C)
n
I
2

(D)
n
n n
n
I
I I
I
2
2
2

+

Solution

Error in Multiple-Segment Trapezoidal Rule
The true error obtained when using the multiple segment trapezoidal rule with n segments to
approximate an integral
( )
=
b
a
dx x f I (1)
is given by
( )
( )
n
f
n
a b
E
n
i
i
t
=
1
2
3
12
(2)
where for each i ,
i
is a point somewhere in the domain ( ) | | ih a h i a + + , 1 , and
the term
( )
n
f
n
i
i
=

1

can be viewed as an approximate average value of ( ) x f in | | b a, . This
leads us to say that the true error
t
E in Equation (2) is approximately proportional to
2
1
n
E
t
(3)
for the estimate of ( )
b
a
dx x f using the n -segment trapezoidal rule.
Richardsons Extrapolation Formula for Trapezoidal Rule
The true error,
t
E , in the n -segment trapezoidal rule is estimated as
2
1
n
E
t

2
n
C
E
t
(4)
where C is an approximate constant of proportionality.
Since
n t
I TV E = (5)
where
TV = true value
n
I = approximate value using n -segments
Then from Equations (4) and (5),
n
I TV
n
C

2
(6)
If the number of segments is doubled from n to n 2 in the trapezoidal rule,
( )
n
I TV
n
C
2
2
2
(7)
Equations (6) and (7) can be solved simultaneously to get
3
2
2
n n
n
I I
I TV

+ (8)

2. The estimate of ( )
19
3
dx x f is given as 1860.9 using the 1-segment trapezoidal rule.
Given ( ) 27 . 20 7 = f , ( ) 125 . 45 11 = f , and ( ) 23 . 82 14 = f , the value of the integral
using the 2-segment trapezoidal rule would most nearly be
(A) 787.32
(B) 1072.0
(C) 1144.9
(D) 1291.5

Solution

The 1-segment trapezoidal rule is
(
+

2
) ( ) (
) (
b f a f
a b I
( )
( )
(
+

2
) 19 ( 3
) 16 ( 9 . 1860
2
) 19 ( 3
) 3 19 ( 9 . 1860
f f
f f

The 2-segment trapezoidal rule is
( ) | |
| | ( ) ( )
( ) ( )
( )
5 . 1291
125 . 45 2
4
16
2
9 . 1860
11 2
4
16
2
9 . 1860
11 2
4
16
) 19 ( ) 3 (
4
16
) 19 ( 11 2 ) 3 (
2 2
3 19
8
2
3 19

) ( ) ( 2 ) (
2
1
1
=
+ =
+ =
+ + =
+ +
=
(
+
)
`
+ +
=
f
f f f
f f f I
n
a b
h
b f ih a f a f
n
a b
I
n
i

3. The value of ( )
b
a
dx x f using the 1-, 2-, and 4-segment trapezoidal rule is given as
5.3460, 2.7708, and 1.7536, respectively. The best estimate of the integral you can
find using Romberg integration is most nearly
(A) 1.3355
(B) 1.3813
(C) 1.4145
(D) 1.9124

Solution

3460 . 5
1 , 1
= I
7708 . 2
2 , 1
= I
7536 . 1
3 , 1
= I

where the above three values correspond to using the 1-, 2- and 4-segment trapezoidal rule,
respectively. To get the first order extrapolation values,
2 ,
1 4
1
, 1 1 , 1
1 , 1 ,

+ =

+
+
k
I I
I I
k
j k j k
j k j k

3
1 , 1 2 , 1
2 , 1 1 , 2
I I
I I

+ =

3
3460 . 5 7708 . 2
7708 . 2

+ =
9124 . 1 =
Similarly

4145 . 1
3
7708 . 2 7536 . 1
7536 . 1
3
2 , 1 3 , 1
3 , 1 2 , 2
=
+ =
+ =
I I
I I

For the second order extrapolation value,

15
1 , 2 2 , 2
2 , 2 1 , 3
I I
I I

+ =

15
9124 . 1 4145 . 1
4145 . 1

+ =
3813 . 1 =
4. Without using the formula for the 1-segment trapezoidal rule for estimating ( )
b
a
dx x f ,
the true error
t
E can be found directly as well as exactly by using the formula

( )
( ) f
a b
E
t

=
12
3
, b a
for
(A) ( )
x
e x f =
(B) ( ) x x x f 3
3
+ =
(C) ( ) 3 5
2
+ = x x f
(D) ( )
x
e x x f + =
2
5

Solution

For

( )
( )
( ) 10
10
3 5
2
=
=
+ =
x f
x x f
x x f

Hence ( ) 10 = f irrespective of the value of . Hence, for this function, the true error can be
found exactly, that is,

( )
( )
3
3
6
5
10
12
a b
a b
E
t
=
=

5. For ( )
b
a
dx x f , the true error
t
E in the 1-segment trapezoidal rule is given by
( )
( ) f
a b
E
t

=
12
3
, b a
The value of for dx e
x
2 . 7
5 . 2
2 . 0
3 is most nearly
(A) 2.7998
(B) 4.8500
(C) 4.9601
(D) 5.0327

Solution

The estimation for the 1-segment trapezoidal rule is

( ) ( )
379 . 41
2
3 3
) 5 . 2 2 . 7 (
2
) (
2 . 7 2 . 0 5 . 2 2 . 0
=
(
+
=
(
+

e e
b f a f
a b I

The true value of the integral is

580 . 38
2 . 0
3
2 . 0
3
2 . 0
3
3
5 . 2 2 . 0 2 . 7 2 . 0
2 . 7
5 . 2
2 . 0 2 . 7
5 . 2
2 . 0
=
=
(
e e
e
dx e
x
x

Thus,

7998 . 2
379 . 41 580 . 38
=
=
t
E

( )
( )
( )
( )

2 . 0
2 . 0
2 . 0
2 . 0
12 . 0
12 . 0
6 . 0
3
e f
e x f
e x f
e x f
x
x
x
=
=
=
=

From

( )
( ) f
a b
E
t

=
12
3

we get
( )
( )
( ) ( )
9601 . 4
2 . 0 99202 . 0
ln 69671 . 2 ln
6968 . 2
12 . 0 32360 . 0
12 . 0 65192 . 8 7998 . 2
12
5 . 2 2 . 7
7998 . 2
2 . 0
2 . 0
2 . 0
2 . 0
3
=
=
=
=
=
=

e
e
e
e
f
t (s) 2 4 6 8 10 25
(m/s) 0.166 0.55115 1.8299 6.0755 20.172 8137.5
The best estimate for the distance in meters covered between 2 = t s and 10 = t s by
using the Romberg rule based on trapezoidal rule results would be
(A) 33.456
(B) 36.877
(C) 37.251
(D) 81.350

Solution

The estimate for the 1-segment trapezoidal rule is

352 . 81
2
172 . 20 166 . 0
) 2 10 (
2
) ( ) (
) (
1 , 1
=
(
+
=
(
+

b f a f
a b I


( ) | |
996 . 47
172 . 20 8299 . 1 2 166 . 0
2 2
2 10
2 , 1
=
+ +
= I


( ) | |
251 . 37
172 . 20 0755 . 6 8299 . 1 55115 . 0 2 166 . 0
4 2
2 10
3 , 1
=
+ + + +
= I

To get the first order extrapolation values,

3
1 , 1 2 , 1
2 , 1 1 , 2
I I
I I

+ =

3
352 . 81 996 . 47
996 . 47

+ =
m 877 . 36 =
Similarly

3
2 , 1 3 , 1
3 , 1 2 , 2
I I
I I

+ =

3
996 . 47 251 . 37
251 . 37

+ =
m 670 . 33 =
For the second order extrapolation values,

15
1 , 2 2 , 2
2 , 2 1 , 3
I I
I I

+ =

m 456 . 33
15
877 . 36 670 . 33
670 . 33
=
+ =

Simpsons 1/3 Rule
Integration

1. The highest order of polynomial integrand for which Simpsons 1/3 rule of integration is
exact is
(A) first
(B) second
(C) third
(D) fourth

Solution

Simpsons 1/3 rule of integration is exact for integrating polynomials of third order or less.
Although Simpsons 1/3 rule is derived by approximating the integrand by a second order
polynomial, the area under the curve is exact for a third order polynomial. Without proof it can
be shown that the truncation error in Simpsons 1/3 rule is b a f
a b
E
t
< <
= ), (
2880
) (
) 4 (
5
.
Since the fourth derivative of a third order polynomial is zero, the truncation error would be zero.
Hence Simpsons 1/3 rule is exact for integrating polynomials of third order or less.

2. The value of
by using 2-segment Simpsons 1/3 rule most nearly is

2 . 2
2 . 0
dx e
x
(A) 7.8036
(B) 7.8423
(C) 8.4433
(D) 10.246

Solution

The multiple segment equation for Simpsons 1/3 rule is

+ + +

=
=
=
=
) ( ) ( 2 ) ( 4 ) (
3
) (
2
even
2
1
odd
1
0 n
n
i
i
i
n
i
i
i
b
a
x f x f x f x f
n
a b
dx x f
Using two-segments gives

2 . 2
1 2 . 1
2 . 1
1 2 . 0
2 . 0
1
2
2 . 0 2 . 2
2
2 . 2
2 . 0
2
0 1
0
=
+ =
=
+ =
+ =
=
=
=
=
=
=
x
h x x
x
n
a b
h
n
b
a

( )
[ ]
[ ]
[ ]
8423 . 7
527 . 23 33333 . 0
4 33333 . 0
) 2 . 2 ( ) 2 . 1 ( 4 ) 2 . 0 (
2 3
2 . 0 2 . 2
) 2 . 2 ( ) ( 2 ) ( 4 ) 2 . 0 (
2 3
2 . 0 2 . 2
) 2 . 2 ( 2 ) ( 4 ) 2 . 0 (
2 3
2 . 0 2 . 2
2 . 2 2 . 1 2 . 0
0
even
2
1
odd
1
2 2
even
2
1 2
odd
1
2 . 2
2 . 0
=
=
+ + =
+ +
+ + +
+ + +
=
=
=
=
=
=
=
=
e e e
f f f
f x f x f f
f x f x f f dx e
i
i
i
i
i
i
i
i
i
i
i
i
x

3. The value of
by using 4-segment Simpsons 1/3 rule most nearly is

2 . 2
2 . 0
dx e
x
(A) 7.8036
(B) 7.8062
(C) 7.8423
(D) 7.9655

Solution


+ + +

=
=
=
=
) ( ) ( 2 ) ( 4 ) (
3
) (
2
even
2
1
odd
1
0 n
n
i
i
i
n
i
i
i
b
a
x f x f x f x f
n
a b
dx x f

Using 4 segments gives

5 . 0
4
2 . 0 2 . 2
4
2 . 2
2 . 0
=
=
=
=
=
n
a b
h
n
b
a

( ) ( ) ( ) ( ) ( ) [ ] ) ( 2 4 ) (
4 3
2 . 0 2 . 2
) ( ) ( 2 ) ( 4 ) (
4 3
2 . 0 2 . 2
) ( ) ( 2 ) ( 4 ) (
4 3
2 . 0 2 . 2
4 2 3 1 0
4
2
even
2
3
odd
1
0
4
2 4
even
2
1 4
odd
1
0
2 . 2
2 . 0
x f x f x f x f x f
x f x f x f x f
x f x f x f x f dx e
i
i
i
i
i
i
i
i
i
i
i
i
x
+ + + +
+ + +
+ + +
=
=
=
=
=
=
=
=

So
( )
( ) ( )
( ) ( ) ( )
( ) 0138 . 2 7 . 0
7 . 0 5 . 0 2 . 0
2214 . 1 2 . 0
7 . 0
1
2 . 0
0
= =
= + =
= = =
=
e f
f f x f
e f x f
e x f
x

( ) ( ) ( )
( )
( ) ( ) ( )
( ) 4739 . 5 7 . 1
7 . 1 5 . 0 2 . 1
3201 . 3 2 . 1
2 . 1 5 . 0 7 . 0
7 . 1
3
2 . 1
2
= =
= + =
= =
= + =
e f
f f x f
e f
f f x f

( ) ( ) 0250 . 9 2 . 2
2 . 2
4
= = = e f x f

( ) ( ) ( ) ( ) ( ) [ ]
( ) ( ) [ ]
[ ]
8062 . 7
0250 . 9 6402 . 6 951 . 29 2214 . 1 16667 . 0
0250 . 9 3201 . 3 2 4739 . 5 0138 . 2 4 2214 . 1
4 3
2 . 0 2 . 2
) ( 2 4 ) (
4 3
2 . 0 2 . 2
4 2 3 1 0
2 . 2
2 . 0
=
+ + + =
+ + + +
=
+ + + +
x f x f x f x f x f dx e
x

4. The velocity of a body is given by
5 1 , 2 ) ( = t t t v
14 5 , 3 5
2
< + = t t
where t is given in seconds, and is given in m/s. Using two-segment Simpsons
1/3 rule, the distance in meters covered by the body from
v
2 = t to seconds most
nearly is
9 = t
(A) 949.33
(B) 1039.7
(C) 1200.5
(D) 1442.0

Solution


+ + +

=
=
=
=
) ( ) ( 2 ) ( 4 ) (
3
) (
2
even
2
1
odd
1
0 n
n
i
i
i
n
i
i
i
b
a
x f x f x f x f
n
a b
dx x f

5 . 3
2
2 9
2
9
2
=
=
=
=
=
n
a b
h
n
b
a

( ) ( ) [ ] ) ( 4 ) (
2 3
2 9
) ( ) ( 2 ) ( 4 ) (
2 3
2 9
) ( ) ( 2 ) ( 4 ) (
2 3
2 9
) (
2 1 0
2
0
even
2
1
odd
1
0
2
2 2
even
2
1 2
odd
1
0
9
2
t v t v t v
t v t v t v t v
t v t v t v t v dt t v
i
i
i
i
i
i
i
i
i
i
i
i
+ +
+ + +
+ + +
=
=
=
=
=
=
=
=

So
5 1 , 2 ) ( = t t t v
14 5 , 3 5
2
< + = t t

( ) ( )
( ) ( ) ( )
( )
( ) ( ) m/s 408 3 9 5 9
m/s 25 . 154 3 5 . 5 5 5 . 5
5 . 5 5 . 3 2
m/s 4 2 2 2
2
2
2
1
0
= + = =
= + =
= + =
= = =
v t v
v
v v t v
v t v

( ) ( ) [ ]
( ) [ ]
[ ]
m 5 . 1200
408 25 . 154 4 4 1667 . 1
) 9 ( 5 . 5 4 ) 2 (
2 3
2 9
) ( 4 ) (
2 3
2 9
2 1 0
9
2
=
+ + =
+ +
=
+ +
v v v
t v t v t v dt t v

5. The value of by using 2-segment Simpsons 1/3 rule is estimated as
702.039. The estimate of the same integral using 4-segment Simpsons 1/3 rule most
nearly is
19
3
) ( dx x f
(A) ( ) ( ) ( ) [ ] 15 2 11 7 2
3
8
039 . 702 f f f + +
(B) ( ) ( ) ( ) [ ] 15 2 11 7 2
3
8
2
039 . 702
f f f + +
(C) ( ) ( ) [ ] 15 2 7 2
3
8
039 . 702 f f + +
(D) ( ) ( ) [ ] 15 2 7 2
3
8
2
039 . 702
f f +
Solution

Using 2-segment Simpsons 1/3 rule gives
( ) [ ] ) ( 4 ) (
2 3
3 19
) ( ) ( 4 ) (
2 3
3 19
) ( ) ( 2 ) ( 4 ) (
2 3
3 19
) (
2 1 0
2
1
odd
1
0
2
2 2
even
2
1 2
odd
1
0
19
3
x f x f x f
x f x f x f
x f x f x f x f dx x f
i
i
i
i
i
i
i
i
i
+ +
+ +
+ + +
=
=
=
=
=
=

( ) [ ] ) 19 ( 11 4 ) 3 (
2 3
3 19
039 . 702 f f f + +

Using 4-segment Simpsons 1/3 rule gives
( ) ( ) ( ) ( ) ( ) [ ]
( ) ( ) ( ) ( ) ( ) [ ] ) 19 ( 11 2 15 7 4 ) 3 (
4 3
3 19
) ( 2 4 ) (
4 3
3 19
) ( ) ( 2 ) ( 4 ) (
4 3
3 19
) ( ) ( 2 ) ( 4 ) (
4 3
3 19
) (
4 2 3 1 0
4
2
even
2
3
odd
1
0
4
2 4
even
2
1 4
odd
1
0
19
3
f f f f f
x f x f x f x f x f
x f x f x f x f
x f x f x f x f dx x f
i
i
i
i
i
i
i
i
i
i
i
i
+ + + +
=
+ + + +
+ + +
+ + +
=
=
=
=
=
=
=
=

( ) ( ) ( ) ( ) [ ]
( ) [ ] ( ) ( ) ( ) [ ]
( ) ( ) ( ) [ ]
( )
( ) ( ) ( ) [ ]
( ) ( ) ( ) [ ] 11 15 2 7 2
3
8
2
039 . 702
11 15 2 7 2
4 3
3 19 2
2
039 . 702
11 2 15 4 7 4
4 3
3 19
2
039 . 702
11 2 15 4 7 4
4 3
3 19
) 19 ( 11 4 ) 3 (
4 3
3 19
) 19 ( 11 2 15 4 11 4 7 4 ) 3 (
4 3
3 19
f f f
f f f
f f f
f f f f f f
f f f f f f
+ + =
+
+ =
+
+ =
+
+ + +
=
+ + + +
=

Time (s) 4 7 10 15
Velocity (m/s) 22 24 37 46
The best estimate of the distance in meters covered by the body from to 4 = t 15 = t
using combined Simpsons 1/3 rule and the trapezoidal rule would be
(A) 354.70
(B) 362.50
(C) 368.00
(D) 378.80

Solution

15 , 10 , 7 , 4
3 2 1 0
= = = = t t t t
We can use Simpsons 1/3 rule from 4 = t to 10 = t as we have three equidistant points,
. 10 , 7 , 4 = t

+ + +

=
=
=
=
) ( ) ( 2 ) ( 4 ) (
3
) (
2
even
2
1
odd
1
0 n
n
i
i
i
n
i
i
i
b
a
t v t v t v t v
n
a b
dt t v
where
2
4 10
3
3

0 1
1 2
2
0
=
=
=
=
=
=
=
n
n
n
a b
h
t t
t t h
t b
t a

Thus, using 2-segment Simpsons 1/3 rule
( ) ( ) ( ) [ ]
2 1 0
0 2
4
2 3
) (
2
0
t v t v t v
t t
dt t v
t
t
+ +

Using the trapezoidal rule with unequal segments from 10 = t to 15 = t
( ) ( )
( ) ( )
2
3 2
2 3
3
2
t v t v
t t dt t v
t
t

Thus,
( ) ( ) ( ) [ ]
( ) ( ) ( ) [ ]
[ ]
[ ] [ ]
m 5 . 362
5 . 41 ) 5 ( 833 . 25 ) 6 (
2
46 37
) 10 15 ( 37 24 4 22
2 3
4 10
2
) 15 ( ) 10 (
) 10 15 ( 10 7 4 4
2 3
4 10
) (
2
) ( ) (
) ( 4
2 3
) (
14
4
3 2
2 3 2 1 0
0 2
3
0
=
+ =

+
+ + +
+
+ + +
+
+ + +
v v
v v v dt t v
t v t v
t t t v t v t v
t t
dt t v
t
t

07.08.1


Chapter 07.08
Simpson 3/8 Rule For Integration

1. Simpson 3/8 rule for integration is mainly based upon the idea of
(A) approximating ) (x f in
=
b
a
dx x f I ) ( by a cubic polynomial
(B) approximating ) (x f in
=
b
a
dx x f I ) ( by a quadratic polynomial
(C) Converting the limit of integral limits ] , [ b a into ] 1 , 1 [ +
(D) Using similar concepts as Gauss quadrature formula

2. The exact value of
4
1
2 2
) 8 4 ( dx x e
x
most nearly is
(A) 6.0067
(B) 5.7606
(C) 60.0675
(D) 67.6075

3. The approximate value of
4
1
2 2
) 8 4 ( dx x e
x
by a single application of Simpsons
3/8 rule is
(A) 61.3740
(B) 60.0743
(C) 59.3470
(D) 58.8992

4
1
2 2
) 8 4 ( dx x e
x
by a multiple-segment Simpsons 3/8
rule with n=6 segments is most nearly
(A) 60.8206
(B) 60.6028
(C) 61.0677
(D) 60.0675

07.08.2 Chapter 07.08

07.08.2
4
1
2 2
) 8 4 ( dx x e
x
by combination of Simpsons 1/3 rule
(n=6 segments) and Simpsons 3/8 rule (n=3 segments) most nearly is
(A) 60.0677
(B) 59.0677
(C) 61.0677
(D) 59.7607

6. Comparing Simpsons 3/8 rule truncated error formula
( )
) 4 (
5
6480
) (
f
a b
E
t

= , b a ,
with Simpsons 1/3 rule truncated error formula
b a f
a b
E
t
< <
= ), (
2880
) (
) 4 (
5

the following conclusion can be made.
(A) Simpsons 3/8 rule is significantly more accurate than Simpsons 1/3 rule
(B) It is worth it in terms of computational efforts versus accuracy to use
Simpsons 3/8 rule instead of Simpsons 1/3 rule.
(C) It is worth it in terms of computational efforts versus accuracy to use
Simpsons 3/8 rule instead of Simpsons 1/3 rule.
(D) Simpsons 3/8 rule is less accurate than Simpsons 1/3 rule.

Trapezoidal Rule
Integration

1. The two-segment trapezoidal rule of integration is exact for integrating at most
________ order polynomials.
(A) first
(B) second
(C) third
(D) fourth
Solution

The single segment trapezoidal rule is exact for at most a first order polynomial. The two
segment trapezoidal rule is also exact only for the same order of polynomial, that is, a first order
polynomial.

2. The value of
by the using one-segment trapezoidal rule is most nearly

2 . 2
2 . 0
dx xe
x
(A) 11.672
(B) 11.807
(C) 20.099
(D) 24.119
Solution

( )

2
) ( ) (
) (
b f a f
a b dx x f
b
a

where
2 . 0
2 . 0
=
=
b
a

( )
( )
( )
( )
099 . 20
050 . 10 2
2
855 . 19 24428 . 0
2 . 0 2 . 2
855 . 19
2 . 2 2 . 2
24428 . 0
2 . 0 2 . 0
2 . 2
2 . 0
2 . 2
2 . 0
=
=

+

=
=
=
=
=
dx xe
e f
e f
xe x f
x
x

3. The value of
by using the three-segment trapezoidal rule is most nearly

2 . 2
2 . 0
dx xe
x
(A) 11.672
(B) 11.807
(C) 12.811
(D) 14.633
Solution

( )

+ +

=
) ( ) ( 2 ) (
2
1
1
b f ih a f a f
n
a b
dx x f
n
i
b
a

where

x
xe x f
n
a b
h
n
b
a
=
=
=
=
=
=
) (
66667 . 0
3
2 . 0 2 . 2
3
2 . 2
2 . 0

Thus
( )

+ +

=
) 2 . 2 ( ) 66667 . 0 2 . 0 ( 2 ) 2 . 0 (
3 2
2 . 0 2 . 2
1 3
1
f i f f dx x f
i
b
a

+ + =

=
) 2 . 2 ( ) 66667 . 0 2 . 0 ( 2 ) 2 . 0 (
6
2

2
1
f i f f
i

( ) [ ] ) 2 . 2 ( 5333 . 1 2 ) 86667 . 0 ( 2 ) 2 . 0 (
3
1
f f f f + + + =
where
( )
( )
( )
( )
855 . 19
2 . 2 2 . 2
1048 . 7
5333 . 1 5333 . 1
0618 . 2
86667 . 0 86667 . 0
24428 . 0
2 . 0 2 . 0
2 . 2
5333 . 1
86667 . 0
2 . 0
=
=
=
=
=
=
=
=
e f
e f
e f
e f

Hence

[ ]
[ ]
811 . 12
433 . 38 33333 . 0
855 . 19 1048 . 7 2 0618 . 2 2 24428 . 0 33333 . 0
2 . 2
2 . 0
=
=
+ + +
dx xe
x
4. The velocity of a body is given by
5 1 , 2 ) ( = t t t v
14 5 , 3 5
2
< + = t t
where t is given in seconds, and v is given in m/s. Use the two-segment trapezoidal
rule to find the distance covered by the body from 2 = t to 9 = t seconds.
(A) 935.0 m
(B) 1039.7 m
(C) 1260.9 m
(D) 5048.9 m
Solution

( )

+ +

=
) ( ) ( 2 ) (
2
1
1
b v ih a v a v
n
a b
dt t v
n
i
b
a

where
14 5 , 3 5
5 1 , 2 ) (
5 . 3
2
2 9

2
9
2
2
< + =
=
=
=
=
=
=
t t
t t t v
n
a b
h
n
b
a

Thus
( )
( ) [ ]
( )
( )
( )
m/s 408
3 9 5 9
m/s 25 . 154
3 5 . 5 5 5 . 5
m/s 4
2 2 2
) 9 ( 5 . 5 2 ) 2 (
4
7
) 9 ( ) 5 . 3 2 ( 2 ) 2 (
2 2
2 9
2
2
1 2
1
9
2
=
+ =
=
+ =
=
=
+ + =
+ +

=
v
v
v
v v v
v i v v dt t v
i

( ) [ ]
[ ]
m 9 . 1260
5 . 720 75 . 1
408 25 . 154 2 4 75 . 1
9
2
=
=
+ +
dt t v

5. The shaded area shows a plot of land available for sale. Your best estimate of the
area of the land is most nearly
(A) 2500 m
2

(B) 4775 m
2

(C) 5250 m
2

(D) 6000 m
2

Solution

60
75
45
25
100 60
A
B
C
60
75
45
25
60 100
2
2
2
2
m 4775
500 675 3600
m 500
25 20
m 675
15 45
m 3600
60 60
=
+ + =
+ +
=
=
=
=
=
=
C B A Area
C
B
A

Time (s) 0 15 18 22 24
Velocity (m/s) 22 24 37 25 123
The distance in meters covered by the body from s 12 = t to s 18 = t calculated using
the trapezoidal rule with unequal segments is
(A) 162.90
(B) 166.00
(C) 181.70
(D) 436.50
Solution

Use the trapezoidal rule with unequal segments.

( ) ( ) ( )
( )
( ) m/s 37 18
m/s 24 15
18
12
15
12
18
15
=
=
+ =

v
v
dt t v dt t v dt t v

To find the value of the velocity at 12 s, we will use linear interpolation.
( ) 15 0 ,
1 0
+ = t t a a t v
At s 0 = t
0 22
1 0
a a + =
At s 15 = t
15 24
1 0
a a + =
which gives
13333 . 0
22
1
0
=
=
a
a

Hence,
( )
( )
( ) ( ) ( ) ( )
m 90 . 162
2
37 24
) 15 18 (
2
24 6 . 23
) 12 15 (
2
18 15
) 15 18 (
2
15 12
) 12 15 ( ) (
m/s 600 . 23
12 13333 . 0 22 12
15 0 , 13333 . 0 22
18
12
=

+
+
+
=
+
+
+

=
+ =
+ =
v v v v
dt t v
v
t t t v

Problem Set#1
Multiple Choice Test
Chapter 01.02 Measuring Errors

1. True error is defined as
(A) Present Approximation Previous Approximation
(B) True Value Approximate Value
(C) abs (True Value Approximate Value)
(D) abs (Present Approximation Previous Approximation)

Solution

True error is defined as the difference between the exact (true) value and the approximate
value.
2. The expression for true error in calculating the derivative of ( ) x 2 sin at
4
= x by
using the approximate expression ( )
( ) ( )
h
x f h x f
x f
+
is
(A)
( )
h
h h 1 2 cos

(B)
( )
h
h h 1 cos

(C)
( )
h
h 2 cos 1

(D)
( )
h
h 2 sin

Solution
Exact answer
) 2 sin( ) ( x x f =
) 2 cos( 2 ) ( x x f =
4
2 cos 2
4

f
=0
Approximate Solution
( )
( ) ( )
h
x f h x f
x f
+

) 2 sin( ) ( x x f =

h
x h x
x f
) 2 sin( )) ( 2 sin(
) (
+

h
x h x h x ) 2 sin( ) 2 sin( ) 2 cos( ) 2 cos( ) 2 sin( +
=

h
h h
f

4
2 sin ) 2 sin(
4
2 cos ) 2 cos(
4
2 sin
4

h
h h

=
2
sin ) 2 sin(
2
cos ) 2 cos(
2
sin

h
h h 1 ) 2 sin( ) 0 ( ) 2 cos( ) 1 ( +
=

h
h 1 ) 2 cos(
=

=
t
E True Value Approximate Value

h
h 1 ) 2 cos(
0

=
h
h) 2 cos( 1
=
4. The relative approximate error at the end of an iteration to find the root of an equation
is % 004 . 0 . The least number of significant digits we can trust in the solution is
(A) 2
(B) 3
(C) 4
(D) 5

Solution

If
m
a

2
10 5 . 0 , then at least m significant digits are correct in the answer.
m

2
10 5 . 0 004 . 0
m is at least 1, as , 10 5 . 0 004 . 0
1 2
that is, , 5 004 . 0 is true,
m is at least 2, as
2 2
10 5 . 0 004 . 0

, that is, , 5 . 0 004 . 0 is true,
m is at least 3, as
3 2
10 5 . 0 004 . 0

, that is, , 05 . 0 004 . 0 is true,
m is at least 4, as
4 2
10 5 . 0 004 . 0

, that is, , 005 . 0 004 . 0 is true,
m is at not at least 5, as , 10 5 . 0 004 . 0
5 2
that is, , 0005 . 0 004 . 0 is not true,
So the least number of significant digits correct in my answer is 4.

Alternative solution

m

2
10 5 . 0 004 . 0

m
2
10
5 . 0
004 . 0

m
2
10 008 . 0
Taking log of both sides
( ) ( )
m
2
10 10
10 log 008 . 0 log
m 2 0969 . 2
0969 . 2 2 + m
0969 . 4 m
Since m can only be an integer, 4 m .
So the least number of significant digits correct in my answer is 4.
5. The number
3
10 01850 . 0 has ________ significant digits
(A) 3
(B) 4
(C) 5
(D) 6

Solution

The number
3
10 01850 . 0 has 4 significant digits. The trailing zero in the number is
significant, while all the leading zeros are not significant.
6. The following gas stations were cited for irregular dispensation by the Department of
Agriculture. Which one cheated you the most?
Station Actual
Gasoline
dispensed
Gasoline
Reading at
pump
Ser
Cit
Hus
She
9.90
19.90
29.80
29.95
10.00
20.00
30.00
30.00

(A) Ser
(B) Cit
(C) Hus
(D) She

Solution

The conclusion for this question will be based on relative true errors.
% 0101 . 1 100
90 . 9
00 . 10 90 . 9
=
=
t
(Ser)
% 50251 . 0 100
90 . 19
00 . 20 90 . 19
=
=
t
(Cit)
% 67114 . 0 100
80 . 29
00 . 30 80 . 29
=
=
t
(Hus)
% 16694 . 0 100
95 . 29
00 . 30 95 . 29
=
=
t
(She)

The gas station that cheated you the most is Ser as it has the largest magnitude of a
negative relative true error of 1.0101%.
6. The number of significant digits in the number 219900 is
(E) 4
(F) 5
(G) 6
(H) 4 or 5 or 6

Solution

In this case, the number of significant digits can be 4, 5 or 6. The trailing zeros may or
may not be significant. For example, if someone asked you what is the population of
Tampa, you generally will reply as 325,000. However, if someone was refunding taxes
to every citizen of Tampa, you will need the exact number that very well could be exactly
325,000 or 325,200 or 325,221 or 325,214.
Problem Set#1
Chapter 01.06 Propagation Errors

1. If and 05 . 0 56 . 3 = A 04 . 0 25 . 3 = B , the values of B A+ are
(A) 90 . 6 81 . 6 + B A
(B) 90 . 6 72 . 6 + B A
(C) 81 . 6 81 . 6 + B A
(D) 91 . 6 71 . 6 + B A

Solution

05 . 0 50 . 3 = A
Hence
05 . 0 56 . 3 05 . 0 56 . 3 + A
61 . 3 51 . 3 A

04 . 0 25 . 3 = B
Hence
04 . 0 25 . 3 04 . 0 25 . 3 + B
29 . 3 21 . 3 B
Hence
29 . 3 61 . 3 21 . 3 51 . 3 + + + B A
90 . 6 72 . 6 + B A
2. A number A is correctly rounded to 3.18 from a given number B. Then C B A ,
where C is
(A) 0.005
(B) 0.01
(C) 0.18
(D) 0.09999

Solution

Since A is rounded off to 3.18, the number can be where XYZ 17 . 3 X is a number
between 5 and 9 or where XYZ 18 . 3 X is a number between 0 and 4. Hence,
C B A makes 005 . 0 = C
3. Two numbers A and B are approximated as C and D, respectively. The relative error
in is given by D C
(A)

B
D B
A
C A

(B)

B
D B
A
C A
B
D B
A
C A

(C)

B
D B
A
C A
B
D B
A
C A

(D)

B
D B
A
C A

Solution

Rel ( )
B A
D C B A
D C

=

True Error = True Value Approximate Value
Approximate Value = True Value True Error
= A C
= B D
Where and are the true errors in the representation of A and B, respectively.
Rel ( )
AB
B A AB
CD
) )( (
=

AB
A B AB AB + +
=
AB cancels which yields,
Rel
B A B A
CD

+ = ) (
=Rel(A) + Rel(B) + Rel(A)Rel(B)

=
B
D B
A
C A
B
D B
A
C A

4. The formula for normal strain in a longitudinal bar is given by
AE
F
= where
F = normal force applied
A = cross-sectional area of the bar
E = Youngs modulus
If , , and , the maximum
error in the measurement of strain is
N 5 . 0 50 = F
2
m 002 . 0 2 . 0 = A Pa 10 1 10 210
9 9
= E
(A)
12
10

(B)
11
10 95 . 2

(C)
9
10 22 . 1

(D)
9
10 19 . 1

Solution

The total error for strain is given by
E
E
A
A
F
F

= | |
The partial derivatives are then

AE F
1
=

,
E A
F
A
2
=

,
2
AE
F
E
=

The total error for strain is then
E
AE
F
A
E A
F
F
AE

=
2 2
1
| |

) 10 1 (
) 10 210 )( 2 . 0 (
50
) 002 . 0 (
) 10 210 ( ) 2 . 0 (
50
) 5 . 0 (
) 10 210 )( 2 . 0 (
1
| |
9
9 9 2 9

=

11 12 11 11
10 95 . 2 10 67 . 5 10 19 . 1 10 19 . 1 | |

= + + =
5. A wooden block is measured to be 60mm by a ruler and the measurements are
considered to be good to 1/4
th
of a millimeter. Then in the measurement 60mm, we have
. . significant digits
(A) 0
(B) 1
(C) 2
(D) 3

Solution

We are given the uncertainty is within 1/4
th
of a millimeter so at least 2 significant digits
are accurate.
6. In the calculation of the volume of a cube of nominal size , the uncertainty in the
measurement of each side is 10%. The uncertainty in the measurement of the volume
would be
" 5
(A) 5.477%
(B) 10.00%
(C) 17.32%
(D) 30.00%

Solution

For this problem, where a is the length of the side of the cube.
3
a V =
a
da
dV
V =
a a =
2
3
a
a
a
V =
3
3
a
a
V
V = 3

a
a
V
V
=
3
Plugging in numbers yields

V
V
1 . 0 3 =

% 30
3 . 0
=
=
Problem Set#1
Chapter 01.03 Sources of Error

1. Truncation error is caused by approximating
(A) irrational numbers
(B) fractions
(C) rational numbers
(D) exact mathematical procedures.

Solution

Truncation error is related to approximating mathematical procedures. Examples include
using a finite number of terms of a Taylor series to approximate transcendental and
trigonometric functions, the use of a finite number of areas to find the integral of a
function, etc.
2. A computer that represents only 4 significant digits with chopping would calculate
66.666*33.333 as
(A) 2220
(B) 2221
(C) 2221.17778
(D) 2222

Solution

66 . 66 666 . 66
33 . 33 333 . 33

7778 . 2221 33 . 33 66 . 66 =
2221
3. A computer that represents only 4 significant digits with rounding would calculate
66.666*33.333 as
(A) 2220
(B) 2221
(C) 2221.17778
(D) 2222

Solution

67 . 66 666 . 66
33 . 33 333 . 33

1111 . 2222 33 . 33 67 . 66 =
2222
4. The truncation error in calculating ( ) 2 f for ( )
2
x x f = by ( )
( ) ( )
h
x f h x f
x f
+

with 2 . 0 = h is
(A) -0.2
(B) 0.2
(C) 4.0
(D) 4.2

Solution
The approximate value of ) 2 ( f is

h
x f h x f
x f
) ( ) (
) (
+

, 2 = x 2 . 0 = h

2 . 0
) 2 ( ) 2 . 0 2 (
) 2 (
f f
f
+

2 . 0
) 2 ( ) 2 . 2 ( f f
=

2 . 0
2 2 . 2
2 2
=
2 . 4 =
The true value of ( ) 2 f is
( )
( )
( )
4
2 2 2
2
2
=
=
=
=
f
x x f
x x f

Thus, the true error is
2 . 0
2 . 4 4
Value e Approximat - Value True
=
=
=
t
E

5. The truncation error in finding
9
3
3
dx x using LRAM (left end point Riemann
approximation method) with equally portioned points 9 6 3 0 3 < < < < is
(A) 648
(B) 756
(C) 972
(D) 1620

Solution

Graph of f(x)=x
3
for LRAM Approximation
-100
0
100
200
300
400
500
600
700
800
-3 0 3 6 9
x
f
(
x
)

3 ) 6 ( 3 ) 3 ( 3 ) 0 ( 3 ) 3 ( + + + = f f f f LRAM
3 ) 6 ( 3 ) 3 ( 3 ) 0 ( 3 ) 3 (
3 3 3 3
+ + + =

648
648 81 0 81
=
+ + + =

1620
4
) 3 ( 9
4
4 4
9
3
4
9
3
3
=
x
x

Truncation Error = True Value Approximate Value (if there is no round-off error)
= 648 1620
= 972
6. The number
10
1
is registered in a fixed 6 bit-register with all bits used for the
fractional part. The difference is accumulated every
10
1
th of a second for one day. The
magnitude of the accumulated difference is
(A) 0.082
(B) 135
(C) 270
(D) 5400

Solution

Number Number after decimal Number before decimal
2 1 . 0 0.2 0.2 0
2 2 . 0 0.4 0.4 0
2 4 . 0 0.8 0.8 0
2 8 . 0 1.6 0.6 1
2 6 . 0 1.2 0.2 1
2 2 . 0 0.4 0.4 0
2 4 . 0 0.8 0.8 0
2 8 . 0 1.6 0.6 1
2 6 . 0 1.2 0.2 1

( ) ( )
2 10
000110011 . 0 1 . 0
Hence
( ) ( )
2 10
000110 . 0 1 . 0 in a six bit fixed register.
( )
6 5 4 3 2 1
2
2 0 2 1 2 1 2 0 2 0 2 0 000110 . 0

+ + + + + =
09375 . 0 =
The difference (true error) between 0.1 and 0.09375 is

00625 . 0
09375 . 0 1 . 0
=
=

The accumulated difference in a day is then
24 60 60 10 00625 . 0 =
5400 =
Problem Set#1
Chapter 01.07 Taylors Series Revisited

1. The coefficient of the
5
x term in the Maclaurin polynomial for ( ) x 2 sin is
(A) 0
(B) 0.0083333
(C) 0.016667
(D) 0.26667

Solution

The Maclaurin series for ) 2 sin( x is

( ) ( )
+ + =
! 5
2
! 3
2
2 ) 2 sin(
5 3
x x
x x
+ + =
120
32
6
8
2
5 3
x x
x
+ + =
5 3
26667 . 0 3333 . 1 2 x x x
Hence, the coefficient of the
5
x term is 0.26667.
2. Given ( ) 6 3 = f , ( ) 8 3 = f , ( ) 11 3 = f , and all other higher order derivatives of ( ) x f
are zero at 3 = x , and assuming the function and all its derivatives exist and are
continuous between 3 = x and 7 = x , the value of ( ) 7 f is
(A) 38.000
(B) 79.500
(C) 126.00
(D) 331.50

Solution

The Taylor series is given by
( ) ( ) ( )
( ) ( )
+

+

+ + = +
3 2
! 3 ! 2
h
x f
h
x f
h x f x f h x f
3 = x , 4 3 7 = = h
( ) ( ) ( )
( ) ( )
+

+

+ + = +
3 2
4
! 3
3
4
! 2
3
4 3 3 4 3
f f
f f f
( ) +

+

+ + =
3 2
4
! 3
) 3 (
4
! 2
) 3 (
4 ) 3 ( ) 3 ( 7
f f
f f f
Since all the derivatives higher than second are zero,
( ) ( ) ( )
( )
2
4
! 2
3
4 3 3 7
f
f f f

+ + =

2
4
! 2
11
4 8 6 + + =
126 =

3. Given that ( ) x y is the solution to 2
3
+ = y
dx
dy
, 3 ) 0 ( = y the value of ( ) 2 . 0 y from a
second order Taylor polynomial around x=0 is
(A) 4.400
(B) 8.800
(C) 24.46
(D) 29.00

Solution

The second order Taylor polynomial is
( ) ( ) ( )
( )
2
! 2
h
x y
h x y x y h x y

+ + = +
0 = x , 2 . 0 0 2 . 0 = = h
( ) ( ) ( )
( )
2
2 . 0
! 2
0
2 . 0 0 0 2 . 0 0
y
y y y

+ + = +
( ) ( ) ( ) ( ) 02 . 0 0 2 . 0 0 0 2 . 0 + + = y y y y
( ) 3 0 = y
( ) 2
3
+ = y x y
( ) 2 3 0
3
+ = y
29 =
( )
dx
dy
y x y
2
3 =
( ) 2 3
3 2
+ = y y
( ) ( ) ( ) 2 3 3 3 0
3 2
+ = y
783 =
( ) 02 . 0 783 2 . 0 29 3 2 . 0 + + = y
46 . 24 =

4. The series ( )
=

0
2
4
)! 2 (
1
n
n
n
n
n
x
is a Maclaurin series for the following function
(A) ( ) x cos
(B) ( ) x 2 cos
(C) ( ) x sin
(D) ( ) x 2 sin

Solution
( ) ( )
( )
=
=
0
2
! 2
1 cos
n
n
n
n
x
x
( ) ( )
( )
( )
=
=
0
2
! 2
2
1 2 cos
n
n
n
n
x
x
( )
( )
=
=
0
2 2
! 2
2
1
n
n n
n
n
x

( )
( )
=
=
0
2
! 2
4
1
n
n n
n
n
x

5. The function ( )

=
x
t
dt e x erf
0
2 2
is called the error function. It is used in the field of

probability and cannot be calculated exactly. However, one can expand the integrand as
a Taylor polynomial and conduct integration. The approximate value of ( ) 0 . 2 erf using
the first three terms of the Taylor series around 0 = t is
(A) -0.75225
(B) 0.99532
(C) 1.5330
(D) 2.8586

Solution

Rewrite the integral as
( )

=
x
t
dt e x erf
0
2 2

The first three terms of the Taylor series for ( )
2 2
t
e t f

=
around 0 = t are
( )
2 2
t
e t f

=

( )
2
0
2
0

= e f
2
=
( ) ( ) t e t f
t
2
2 2
=

( ) ( ) ( ) 0 2
2
0
2
=
t
e f

0 =
( ) ( )( ) ( ) 2
2
2 2
2 2 2
+ =
t t
e t t e t f

( ) ( ) ( ) ( ) ( ) ( ) 2
2
0 2 0 2
2
0
2 2
0 0
+ =

e e f

4
=
The first three terms of the Taylor series are

2
! 2
) (
) ( ) ( ) ( h
x f
h x f x f h x f

+ + = +

2
! 2
) 0 (
) 0 ( ) 0 ( ) 0 ( h
f
h f f h f

+ + = +

2
! 2
) 0 (
) 0 ( ) 0 ( ) ( h
f
h f f h f

+ + =

! 2
4
) ( 0
2
2
h
h

+ =

! 2
4 2
2
h

=

2
2 2
h

=

2 2
h
e
2
2 2
h

, or

2 2
x
e
2
2 2
x

Hence
( )
|
|
.
|
\
|

x
dt t x erf
0
2
2 2

x
t
t
0
3
3
2 2
(
=

3
2 2
3
x
x

=
( ) ( )
3
2 2
2
2
2
3

= erf
75225 . 0 =
Note: Compare with the exact value of ( ) 2 erf

6. Using the remainder of Maclaurin polynomial of
th
n order for ( ) x f defined as
( )
( )
( )
( ) x c n c f
n
x
x R
n
n
n

+
=
+
+
0 , 0 ,
! 1
1
1

the order of the Maclaurin polynomial at least required to get an absolute true error of at
most
6
10
in the calculation of ( ) 1 . 0 sin is (do not use the exact value of ( ) 0.1 sin or
( ) 0.1 cos to find the answer, but the knowledge that 1 ) sin( | x | and 1 | ) cos( | x ).
(A) 3
(B) 5
(C) 7
(D) 9

Solution
( )
( )
( ) c f
n
x
x R
n
n
n
1
1
)! 1 (
+
+
+
= , 0 n , x c 0
( )
( )
( )
( ) 1 . 0 0 , 0 ,
)! 1 (
1 . 0
1 . 0
1
1

+
=
+
+
c n c f
n
R
n
n
n

Since derivatives of ( ) x f are simply ( ) x sin and ( ) x cos , and
( ) 1 sin x and ( ) 1 cos x

( )
( ) 1
1
+
c f
n

( )
( )
( ) 1
)! 1 (
1 . 0
1 . 0
1
+
+
n
R
n
n

( )
)! 1 (
1 . 0
1
+
=
+
n
n

So when is
( )
6
10 1 . 0

<
n
R

( )
6
1
10
)! 1 (
1 . 0

+
<
+ n
n

4 n
But since the Maclaurin series for ( ) x sin only includes odd terms, 5 n .

Bisection Method
Nonlinear Equations

1. The bisection method of finding roots of nonlinear equations falls under the category of a (an)
_________ method.
(A) open
(B) bracketing
(C) random
(D) graphical

Solution

The bisection method is a bracketing method since it is based on finding the root between two
guesses that bracket the root, that is, where the real continuous function ( ) x f in the equation
changes sign between the two guesses. ( ) 0 = x f
2. If for a real continuous function ( ) x f , ( ) 0 ) ( < b f a f , then in the range of [ for ] b a, ( ) 0 = x f ,
there is (are)
(A) one root
(B) an undeterminable number of roots
(C) no root
(D) at least one root

Solution

If , then and have opposite signs. Since is continuous between a
and b, the function needs to cross the x-axis. The point where the function crosses the x-
axis is the root of the equation
( ) 0 ) ( < b f a f ) (a f ) (b f ) (x f
) (x f
. 0 ) ( = x f
3. Assuming an initial bracket of [ , the second (at the end of 2 iterations) iterative value of
the root of using the bisection method is
] 5 , 1
0 3 . 0 =
t
te
(A) 0
(B) 1.5
(C) 2
(D) 3

Solution

3 . 0 ) ( =
t
te t f
If the initial bracket is [1,5] then
1
5
=
=
t
t
u

Check to see if the function changes sign between and
t
u
t
0679 . 0
3 . 0 1 ) (
2663 . 0
3 . 0 5 ) (
1
5
=
=
=
=
e t f
e t f
u

Hence,
0 0181 . 0
) 0679 . 0 )( 2663 . 0 (
) 1 ( ) 5 ( ) ( ) (
< =
=
= f f t f t f
u

So there is at least one root between and .
t
u
t
Iteration 1
The estimate of the root is
1506 . 0
3 . 0 3 ) (
3
2
5 1
2
3
=
=
=
+
=
+
=
e t f
t t
t
m
u
m

Thus,
0 0102 . 0
) 1506 . 0 )( 0679 . 0 (
) 3 ( ) 1 ( ) ( ) (
< =
=
= f f t f t f
m

The root lies between and , so the new upper and lower guesses for the root are
t
m
t
3
1
= =
= =
m u
t t
t t

Iteration 2
2
2
3 1
2
=
+
=
+
=
u
m
t t
t

4. To find the root of f(x) = 0, a scientist is using the bisection method. At the beginning of an
iteration, the lower and upper guesses of the root are and . At the end of the iteration, the
absolute relative approximate error in the estimated value of the root would be
l
x
u
x
(A)
x x
x
u
u
+

(B)
x x
x
u
+

(C)
x x
x x
u
u
+

(D)
x x
x x
u
u
+

Solution

The absolute relative approximate error is

new
m
old
m
new
m
a
x
x x
=
where

2
u new
m
x x
x
+
=

If

x x
old
m
=

x x
x x
x x
x x x
x x
x
x x
u
u
u
u
u
u
a
+
=
+
+
=
+

+
=
2 ) (
2
2

If

u
old
m
x x =

x x
x x
x x
x x x
x x
x
x x
u
u l
u
u u
u
u
u
a
+
=
+
+
=
+

+
=
2 ) (
2
2

The answer is the same whether as x
m
is exactly in the middle of and x
u
.
u
old
m
x or x x

=

x
5. For an equation like , a root exists at 0
2
= x 0 = x . The bisection method cannot be adopted
to solve this equation in spite of the root existing at 0 = x because the function ( )
2
x x f =
(A) is a polynomial
(B) has repeated roots at 0 = x
(C) is always non-negative
(D) has a slope equal to zero at 0 = x

Solution

Since will never be negative, the statement
2
) ( x x f = 0 ) ( ) ( <
x f x f
u
will never be true.
Therefore, no interval [ will contain the root of x
2
= 0. ]
u
x x ,
f (x )
x

6. The ideal gas law is given by
RT pv =
where is the pressure, is the specific volume, R is the universal gas constant, and T is the
absolute temperature. This equation is only accurate for a limited range of pressure and
temperature. Vander Waals came up with an equation that was accurate for larger ranges of
pressure and temperature given by
p v
( ) RT b v
v
a
p =
+
2

where a and b are empirical constants dependent on a particular gas. Given the value of
, , , 08 . 0 = R 592 . 3 = a 04267 . 0 = b 10 = p and 300 = T (assume all units are consistent), one is
going to find the specific volume, , for the above values. Without finding the solution from the
Vander Waals equation, what would be a good initial guess for v ?
v
(A) 0
(B) 1.2
(C) 2.4
(D) 3.6

Solution

From the physics of the problem, the initial guess can be found from the linear relationship of the
ideal gas law

4 . 2
10
300 08 . 0
=
=
=
=
p
RT
v
RT pv

03.06.1


Chapter 03.06
False-Position Method of Solving a Nonlinear
Equation

1. The false-position method for finding roots of nonlinear equations belongs to a class
of a (an) ____________ method.
(A) open
(B) bracketing
(C) random
(D) graphical

2. The newly predicted root for false-position and secant method can be respectively
given as
( ){ }
( ) ( )
L U
L U U
U r
x f x f
x x x f
x x

=
and
( ){ }
( ) ( )
1
1
1

=
i i
i i i
i i
x f x f
x x x f
x x ,
While the appearance of the above 2 equations look essentially identical, and both
methods require two initial guesses, the major difference between the above two
formulas is
(A) false-position method is not guaranteed to converge.
(B) secant method is guaranteed to converge
(C) secant method requires the 2 initial guesses
i i
x and x
1
to satisfy
( ) ( ) 0
1
<
i i
x f x f
(D) false-position method requires the 2 initial guesses
U L
x and x to satisfy
( ) ( ) 0 <
U L
x f x f
03.06.2 Chapter 03.06

03.06.2

3. Given are the following nonlinear equation
0 36 4
2 2
= +
x e
x

two initial guesses, 1 =
L
x and 4 =
U
x , and a pre-specified relative error tolerance of
0.1%. Using the false-position method, which of the following tables is correct
r
x ( =
predicted root)?
(A)
Iteration
L
x
U
x
r
x
1 1 4 ?
2 ? ? 2.939
(B)
Iteration
L
x
U
x
r
x
1 1 4 ?
2 ? ? 2.500

(C)
Iteration
L
x
U
x
r
x
1 1 4 ?
2 ? ? 1.500

(D)
Iteration
L
x
U
x
r
x
1 1 4 ?
2 ? ? 2.784

03.06.3 Chapter 03.06

4. Given are the following nonlinear equation
0 36 4
2 2
= +
x e
x

two initial guesses, 1 =
L
x and 4 =
U
0.1%. Using the false-position method, which of the following tables is correct
r
x ( =
predicted root, | |
a
= percentage absolute relative approximate error).
(A)
Iteration
L
x
U
x
r
x |
a
| %
1 1 4 ? ?
2 ? ? ? 11.63

(B)
Iteration
L
x
U
x
r
x |
a
| %
1 1 4 ? ?
2 ? ? ? 6.11

(C)
Iteration
L
x
U
x
r
x |
a
| %
1 1 4 ? ?
2 ? ? ? 5.14

(D)
Iteration
L
x
U
x
r
x |
a
| %
1 1 4 ? ?
2 ? ? ? 4.15

5. The root of ( ) ( ) 0 2 4
2
= + x x was found using false-position method with initial
guesses of 5 . 2 =
L
x and 0 . 1 =
U
6
10
%. The final converged root was found as 9999997 . 1 ==

r
x , and the
corresponding percentage absolute relative approximate error was found as
% 10 7610979 . 8 | |
5
=
a
. Based on the given information, the number of significant
digits of the converged root
r
x that can be trusted at least are
(A) 3
(B) 4
(C) 5
(D) 6

03.06.4 Chapter 03.06

03.06.2

6. The false-position method may have difficulty in finding the root of
( ) 0 69 . 13 4 . 7
2
= + = x x x f because
(A) ( ) x f is a quadratic polynomial
(B) ) (x f a straight line
(C) one cannot find initial guesses
L
x and
U
x that satisfy ( ) ( ) 0 <
U L
x f x f
(D) the equation has two identical roots.

Newton-Raphson Method
Nonlinear Equations

1. The Newton-Raphson method of finding roots of nonlinear equations falls under the category
of _____________ methods.
(A) bracketing
(B) open
(C) random
(D) graphical

Solution

The Newton-Raphson method is an open method since the guess of the root that is needed to get
the iterative method started is a single point. Other open methods such as the secant method use
two initial guesses of the root but they do not have to bracket the root.
2. The Newton-Raphson method formula for finding the square root of a real number R from the
equation 0
2
= R x is,
(A)
2
1
i
i
x
x =
+

(B)
2
3
1
i
i
x
x =
+

(C)
+ =
+
i
i i
x
R
x x
2
1
1

(D)
=
+
i
i i
x
R
x x 3
2
1
1

Solution

The Newton-Raphson method formula for solving f(x) = 0 is
) (
) (
1
i
i
i i
x f
x f
x x
=
+

where
x x f
R x x f
2 ) (
) (
2
=
=

Thus,
+ =
+ =
+ =
=
+
i
i
i
i
i
i
i
i
i
i i
x
R
x
x
R
x
x
R x
x
x
R x
x x
2
1
2 2
1
2 2
2
2
1

3. The next iterative value of the root of 0 4
2
= x using the Newton-Raphson method, if the
initial guess is 3, is
(A) 1.5
(B) 2.067
(C) 2.167
(D) 3.000

Solution

) (
) (
1
i
i
i i
x f
x f
x x
=
+

Chose i = 0,
6
3 2
2 ) (
5
4 3
4 ) (
3
0 0
2
2
0 0
0
=
=
=
=
=
=
=
x x f
x x f
x

Thus,
166 . 2
6
5
3
) (
) (
0
0
0 1
=
=
=
x f
x f
x x

4. The root of the equation 0 ) ( = x f is found by using the Newton-Raphson method. The initial
estimate of the root is 3
0
= x , ( ) 5 3 = f . The angle the line tangent to the function ) (x f makes
at 3 = x is 57 with respect to the x-axis. The next estimate of the root,
1
x most nearly is
(A) 3.2470
(B) 0.24704
(C) 3.2470
(D) 6.2470

Solution

f(x)
f(x
0
)

f(x
1
)
x
1

x
0

x

[x
0
,f(x
0
)]

x

Since,
5 ) (
and , 3
, 57
0
0
=
=
=
x f
x

1
1 0
1 0
3
0 5
) 57 tan(
) ( ) (
) 57 tan(
) tan(
x
x x
x f x f
run
rise
=
=

24704 . 0
5399 . 1
5399 . 1 3 5
) 57 tan(
)) 57 (tan( 3 5
1
=

=

= x

5. The root of 4
3
= x is found by using the Newton-Raphson method. The successive iterative
values of the root are given in the table below.
Iteration
Number
Value of Root
0 2.0000
1 1.6667
2 1.5911
3 1.5874
4 1.5874

The iteration number at which I would first trust at least two significant digits in the answer
is
(A) 1
(B) 2
(C) 3
(D) 4

Solution

The absolute relative approximate error for the first iteration is
% 999 . 19
100 19999 . 0
100
6667 . 1
2 6667 . 1
=
=
=
a

Since,
m

2
10 5 . 0 % 99 . 19 , 0 = m . There are no significant digits correct.
The absolute relative approximate error for the second iteration is
% 7514 . 4
100 047514 . 0
100
5911 . 1
6667 . 1 5911 . 1
=
=
=
a

Since,
m

2
10 5 . 0 % 7514 . 4 , 1 = m . There is at least one significant digit correct.
The absolute relative approximate error for the third iteration is
% 23308 . 0
100 0023308 . 0
100
5874 . 1
5911 . 1 5874 . 1
=
=
=
a

Since,
m

2
10 5 . 0 % 23308 . 0 , 2 = m . There are at least two significant digits correct.
Thus, the third iteration is the first iteration in which at least two significant digits are correct.
6. The ideal gas law is given by
RT pv =
where p is the pressure, v is the specific volume, R is the universal gas constant, and T is the
absolute temperature. This equation is only accurate for a limited range of pressure and
temperature. Vander Waals came up with an equation that was accurate for larger range of
pressure and temperature given by
( ) RT b v
v
a
p =
+
2

where a and b are empirical constants dependent on a particular gas. Given the value of
08 . 0 = R , 592 . 3 = a , 04267 . 0 = b , 10 = p and 300 = T (assume all units are consistent), one is
going to find the specific volume, v , for the above values. Without finding the solution from the
Vander Waals equation, what would be a good initial guess for v ?
(A) 0
(B) 1.2
(C) 2.4
(D) 3.6

Solution

From the physics of the problem, the initial guess can be found from the linear relationship of the
ideal gas law

4 . 2
10
300 08 . 0
=
=
=
=
p
RT
v
RT pv

Secant Method
Nonlinear Equations

1. The secant method of finding roots of nonlinear equations falls under the category of
_____________ methods.
(A) bracketing
(B) graphical
(C) open
(D) random

Solution

The secant method of finding roots of nonlinear equations falls under the category of open
methods. The secant method uses two initial guesses of the root but unlike the bisection method,
they do not have to bracket the root.

f(x)
f(x
i
)
f(x
i-1
)
x
i+1
x
i-1
x
i

x
B
C
E D A

Figure 1 Geometrical representation of the secant method.

2. The secant method formula for finding the square root of a real number R from the equation
is 0
2
= R x
(A)
1
1
+
+
i i
i i
x x
R x x

(B)
1
1
+
i i
i i
x x
x x

(C)
+
i
i
x
R
x
2
1

(D)
1
1
2
2
+
+
i i
i i i
x x
R x x x

Solution
The correct solution is (A).

The secant method formula for finding the root of f(x) = 0 is
) ( ) (
) )( (
1
1
1

=
i i
i i i
i i
x f x f
x x x f
x x
where
R x x f =
2
) (
Thus

( ) ( )
1
1
1
2
1
2
1
2
1
1
2
1 1
1
2
2
1
2
1
2
2
1
2
1
2
1
) )( (
) )( (
) )( (
) ( ) (
) )( (
+
+
+
=
+
+ +
=
+
+
=
+
=
+

=

=

=
i i
i i
i i
i i i i
i i
i i i i
i i
i
i
i i i i
i i i
i
i i
i i i
i
i i
i i i
i i
x x
R x x
x x
R x x x x
x x
R x x x x
x x
R x
x
x x x x
x x R x
x
x x
x x R x
x
R x R x
x x R x
x x

3. The next iterative value of the root of using secant method, if the initial guesses
are 3 and 4, is
0 4
2
= x
(A) 2.2857
(B) 2.5000
(C) 5.5000
(D) 5.7143

Solution
The correct solution is (A).

The first iteration is

( )( )
( ) ( )
1
1
1

=
i i
i i i
i i
x f x f
x x x f
x x
for 3 , 4 , 0
1 0
= = =

x x i

=
1
x
( )( )
( ) ( )
1 0
1 0 0
0
x f x f
x x x f
x
( )( )
( ) ( )
( )( )
( ) ( )
( )( )
( ) ( )
2857 . 2
7
12
4
5 12
1 12
4
4 3 4 4
3 4 4 4
4
4 4
4
2 2
2
2
1
2
0
1 0
2
0
0
=
=
=

=

=
x x
x x x
x

4. The root of the equation is found by using the secant method. Given one of the
initial estimates is , , and the angle the secant line makes with the x-axis is ,
the next estimate of the root, , is
( ) 0 = x f
3
0
= x ( ) 5 3 = f 57
1
x
(A) 3.2470
(B) 0.24704
(C) 3.247
(D) 6.2470

Solution

f(x)
f(x
0
)
x
1
x
0

x
B
C 57
A
secant line

( )
( )
( )
( )
( )
24704 . 0
3 2470 . 3
3
5399 . 1
5
57 tan
3
3
3
3
57 tan
) tan(
1
1
1
1 0
0
=
+ =
+ =
=
=
x
f
x
x
f
x x
x f

5. For finding the root of by the secant method, the following choice of initial guesses
would not be appropriate.
0 sin = x
(A)
4
and
2

(B)
4
and
4
3

(C)
2
and
2

(D)
3
and
2

Solution

4
and
4
3
are not good initial guesses because
2
1
4
3
sin
4
sin = =

. This would result in
division by zero in the formula

( )( )
( ) ( )
1
1
1

=
i i
i i i
i i
x f x f
x x x f
x x
6. When drugs are given orally to a patient, the drug concentration c in the blood stream at time
is given by a formula t
at
Kte c

=
where K is dependent on parameters such as the dose administered while a is dependent on the
absorption and elimination rates of the drug. If 2 = K and 25 . 0 = a , and t is in seconds and c is
in mg/ml, the time at which the maximum concentration is reached is given by the solution of the
equation,
(A) 0 2
25

. 0
=
t
te
(B) 0 2 2
25 . 0 25 . 0
=
t t
te e
(C) 0 5 . 0 2
25 . 0 25 . 0
=
t t
te e
(D) 2 2
25

. 0
=
t
te

Solution

at
Kte c

=
Given

t
te c
a K
25 . 0
2
25 . 0 , 2
=
= =
To find the time of maximum concentration, we set
( )
0 5 . 0 2
0 2
0
25 . 0 25 . 0
25 . 0
=
=
=

t t
t
te e
te
dt
d
dt
dc

Note: The solution of the above equation will only give a local maximum or minimum. We need
to do the second derivative test after finding the root of the equation.

Background

1. Given [ = then [ is a (an) ______________ matrix. ] A
6 0 0 0
5 4 0 0
3 2 1 0
9 3 2 6
] A

(A) diagonal
(B) identity
(C) lower triangular
(D) upper triangular

Solution

A square matrix [ is an upper triangular matrix if ] A 0 =
ij
a when j i > , that is, all the elements
below the diagonal are zero. Note that the statement 0 =
ij
a when j i > implies that the matrix
elements are zero for all elements where the row number is strictly greater than the column
number.
2. A square matrix [ is lower triangular if ] A
(A) i j a
ij
> = , 0
(B) j i a
ij
> = , 0
(C) j i a
ij
> , 0
(D) i j a
ij
> , 0

Solution

A matrix [ is lower triangular if n n ] A 0 =
ij
a for i j > . That is, all elements above the
diagonal are zero. Note that all the elements of [ ] A for which the column number is greater than
the row numbers are zero. An example of a lower triangular matrix is
=
2 . 3 3 9 2
0 0 5 4
0 0 3 6
0 0 0 2
] [ A

3. Given
,
=
3 . 12 3 . 11 3 . 10
3 . 11 3 . 10 3 . 11
3 . 20 3 . 12 3 . 12
] [ A
=
20 11
6 5
4 2
] [B

then if
[ ] [ ][ ] B A C = , then
=
31
c _____________________

(A) 2 . 58
(B) 6 . 37
(C) 219.4
(D) 259.4

Solution

The i
th
row and j
th
column of the [ matrix in ] C [ ] [ ][ ] B A C = is calculated by multiplying the i
th

row of by the j
th
column of , that is, [ ] A [ ] B
[ ]
pj ip j i j i
pj
j
j
ip i i ij
b a b a b a
b
b
b
a a a c
2 2 1 1
2
1
2 1
........ + + + =

=
=
p
k
kj ik
b a
1
Therefore,
[ ]
2 . 58
3 . 135 5 . 56 6 . 20
) 11 3 . 12 ( ) 5 3 . 11 ( ) 2 3 . 10 (
11
5
2
3 . 12 3 . 11 3 . 10
31
=
+ + =
+ + =

= c

4. The following system of equations has ____________ solution(s).
2 = + y x
12 6 6 = + y x
(A) infinite
(B) no
(C) two
(D) unique

Solution

The system of equations
2 = + y x (1)
12 6 6 = + y x (2)
has an infinite number of solutions because the two equations are the same. Equation (2) is a
multiple of 6 of Equation (1).

5. Consider there are only two computer companies in a country. The companies are named
Dude and Imac. Each year, Dude keeps 1/5
th
of its customers, while the rest switch to Imac.
Each year, Imac keeps 1/3
rd
of its customers, while the rest switch to Dude. If in 2003, Dude had
1/6
th
of the market and Imac had 5/6
th
of the market, what will be the share of Dude computers
when the market becomes stable?
(A) 37/90
(B) 5/11
(C) 6/11
(D) 53/90

Solution

If D is the current market if Dude computers and M is the current market of Imac computers, and
if D
n
is the next years Dude market and M
n
is the next years Imac market, then since we want
when the market is stable, the market share should not change from year to year.
+ =
+ =
0
0
3
2
5
4
3
2
5
4
3
1
5
4
3
2
5
1
3
1
5
4
3
2
5
1
n
n
n n n
n n n
n
n
n
n
M
D
M D M
M D D
M
D
M
D

This has a trivial solution if D
n
= 0, M
n
= 0, but we know that 1 = +
n n
M D . So we are looking for
a non-trivial solution. Note also that the coefficient matrix is singular.
1
3
2
5
4
0
= +
=
n n
n n
M D
M D

gives,
11
5
=
n
D
Extra notes for the student:
If one was going to find what the market share would be in 2004
90
37
90
53
6
5
6
1
3
1
5
4
3
2
5
1
n
n
M
D

One would use this number to find the market share in 2005 and so on. Eventually the market
share would stabilize. But that would be a lenthier way to solve the problem.
6. Three kids - Jim, Corey and David receive an inheritance of $2,253,453. The money is put in
three trusts but is not divided equally to begin with. Corey's trust is three times that of David's
because Corey made an A in Dr. Kaws class. Each trust is put in an interest generating
investment. The three trusts of Jim, Corey and David pays an interest of 6%, 8%, 11%,
respectively. The total interest of all the three trusts combined at the end of the first year is
$190,740.57. The equations to find the trust money of Jim (J), Corey (C) and David (D) in a
matrix form is
(A)
57 . 740 , 190
0
453 , 253 , 2
11 . 0 08 . 0 06 . 0
1 3 0
1 1 1
D
C
J
(B)
57 . 740 , 190
0
453 , 253 , 2
11 . 0 08 . 0 06 . 0
3 1 0
1 1 1
D
C
J
(C)
57 . 740 , 190
0
453 , 253 , 2
11 8 6
3 1 0
1 1 1
D
C
J
(D)
057 , 074 , 19
0
453 , 253 , 2
11 8 6
1 3 0
1 1 1
D
C
J

Solution

Let J, C, and D be the inheritance portions of Jim, Corey and David, respectively.

The total inheritance is $2,253,453 gives
453 , 253 , 2 $ = + + D C J

Corey's trust is three times that of David's
D C 3 =
gives
0 3 = D C

The three trusts of Jim, Corey and David pay an interest of 6%, 8%, 11%, respectively. The total
interest of all the three trusts combined at the end of the first year is $190,740.57.
The total interest is
57 . 740 , 190 $
100
11
100
8
100
6
= + + D C J
gives
57 . 740 , 190 $ 11 . 0 08 . 0 06 . 0 = + + D C J
Three equations can be made from the information given
57 . 740 , 190 $ 11 . 0 08 . 0 06 . 0
0 3
453 , 253 , 2 $
= + +
=
= + +
D C J
D C
D C J

Setting the three equations in matrix form is as follows
57 . 740 , 190
0
453 , 253 , 2
11 . 0 08 . 0 06 . 0
3 1 0
1 1 1
D
C
J

Gaussian Elimination

1. The goal of forward elimination steps in the Nave Gauss elimination method is to reduce the
coefficient matrix to a (an) _____________ matrix.
(A) diagonal
(B) identity
(C) lower triangular
(D) upper triangular

Solution

By reducing the coefficient matrix to an upper triangular matrix, starting from the last equation,
each equation can be reduced to one equation-one unknown to be solved by back substitution.

2. Division by zero during forward elimination steps in Nave Gaussian elimination of the set of
equations [ ][ ] [ ] C X A = implies the coefficient matrix [ ] A
(A) is invertible
(B) is nonsingular
(C) may be singular or nonsingular
(D) is singular

Solution

Division by zero during forward elimination does not relate to whether or not the coefficient
matrix is singular or nonsingular. For example
12
12
5
6 5 1
7 3 2
3 2 0
3
2
1
x
x
x

would give a division by zero error in the first step of forward elimination. However, the
coefficient matrix in this case is nonsingular.
In another example
12
12
5
14 6 4
7 3 2
3 2 0
3
2
1
x
x
x

would also give a division by zero error in the first step of forward elimination. In this case the
coefficient matrix is singular.
3. Using a computer with four significant digits with chopping, the Nave Gauss elimination
solution to
23 . 47 123 . 7 239 . 6
12 . 58 23 . 55 0030 . 0
2 1
2 1
=
= +
x x
x x

is
(A) ; 66 . 26
1
= x 051 . 1
2
= x
(B) ; 769 . 8
1
= x 051 . 1
2
= x
(C) ; 800 . 8
1
= x 000 . 1
2
= x
(D) ; 771 . 8
1
= x 052 . 1
2
= x
Solution
Writing all the entries with 4 significant digits
23 . 47
12 . 58
123 . 7 239 . 6
23 . 55 003000 . 0
2
1
x
x

Forward Elimination: Divide Row 1 by 0.003000 and multiply it by 6.239, giving the multiplier
as . 2079
003000 . 0
239 . 6
=
[Row 1] 2079 gives Row 1 as
[ ] 12 . 58 | 23 . 55 003000 . 0
[ ]
5 5
10 208 . 1 | 10 148 . 1 239 . 6
Subtract the above result from Row 2 changes Row 2 to
] 10 207 . 1 | 10 148 . 1 0 [
5 5

and hence giving the set of equations at the end of the 1
st
step of forward elimination as

5
2
1
5
10 207 . 1
12 . 58
10 148 . 1 0
23 . 55 0030 . 0
x
x

This is also the end of all steps of forward elimination.

Back substitution: From the second equation
( )
051 . 1
10 1.148
10 207 . 1
10 207 . 1 10 1.148
5
5
2
5
2
5
=

=
=
x
x

From the first equation,
12 . 58 23 . 55 003000 . 0
2 1
= + x x

003000 . 0
23 . 55 12 . 58
2
1
x
x

=
( )

0030 . 0
051 . 1 23 . 55 12 . 58
=
003000 . 0
04 . 58 12 . 58

=
003000 . 0
08000 . 0
=
66 . 26 =
4. Using a computer with four significant digits with chopping, the Gaussian elimination with
partial pivoting solution to
23 . 47 123 . 7 239 . 6
12 . 58 23 . 55 0030 . 0
2 1
2 1
=
= +
x x
x x

is
(A) ; 66 . 26
1
= x 051 . 1
2
= x
(B) ; 769 . 8
1
= x 051 . 1
2
= x
(C) ; 800 . 8
1
= x 000 . 1
2
= x
(D) ; 771 . 8
1
= x 052 . 1
2
= x
Solution

Writing all the entries with 4 significant digits
23 . 47
12 . 58
123 . 7 239 . 6
23 . 55 003000 . 0
2
1
x
x

Forward elimination:
Now for the first step of forward elimination, the absolute value of first column elements is
003000 . 0 , 239 . 6
or
0.003000, 6.239
So we need to switch Row 1 with Row 2, to get

12 . 58
12 . 47
23 . 55 003000 . 0
123 . 7 239 . 6
2
1
x
x

Divide Row 1 by 6.239 and multiply it by 0.00300, gives the multiplier as
4
10 808 . 4
239 . 6
003000 . 0

=
[Row 1]
4
10 808 . 4

gives Row 1 as
] 10 265 . 2 | 10 424 . 3 10 999 . 2 [
2 3 3

Subtract the above result from Row 2 changes Row 2 to
] 09 . 58 | 22 . 55 0 [
and hence giving the set of equations at the end of the 1
st
step of forward elimination as

09 . 58
23 . 47
22 . 55 0
123 . 7 239 . 6
2
1
x
x

Back substitution: From the second equation
051 . 1
22 . 55
09 . 58
09 . 58 22 . 55
2
2
=
=
=
x
x

Substituting the value of x
2
in the first equation
( )
769 . 8
239 . 6
71 . 54
239 . 6
486 . 7 23 . 47

239 . 6
051 . 1 123 . 7 23 . 47
239 . 6
123 . 7 23 . 47
23 . 47 123 . 7 239 . 6
2
1
2 1
=
=
+
=
+
=
+
=
=
x
x
x x

5. At the end of the forward elimination steps of the Nave Gauss elimination method on the
following equations

0
007 . 0
0
10 887 . 7
10 6057 . 3 10 2857 . 4 0 0
15384 . 0 5 . 6 15384 . 0 5 . 6
10 4619 . 5 10 2857 . 4 10 4619 . 5 10 2857 . 4
0 0 10 2307 . 9 10 2857 . 4
3
4
3
2
1
5 7
5 7 5 7
5 7
c
c
c
c

the resulting equations in matrix form are given by
4
2
3
3
4
3
2
1
5
5 7 5
5 7
10 90336 . 1
10 19530 . 1
10 887 . 7
10 887 . 7
10 62500 . 5 0 0 0
579684 . 0 9140 . 26 0 0
10 4619 . 5 10 2857 . 4 10 7688 . 3 0
0 0 10 2307 . 9 10 2857 . 4
c
c
c
c

The determinant of the original coefficient matrix is
(A) 0.00
(B)
7
10 2857 . 4
(C)
19
10 486 . 5
(D)
20
10 445 2 .
Solution

If a matrix is upper triangular, lower triangular or diagonal, then the determinant is
=
=
n
i
ii nn
a a a a
1
22 11
.........
Thus, the determinant, D, of the matrix is
( ) ( ) ( ) ( )
20
5 5 7
10 445 . 2
10 62500 . 5 9140 . 26 10 7688 . 3 10 2857 . 4
=
= D

6. The following data is given for the velocity of the rocket as a function of time. To find the
velocity at s 21 = t , you are asked to use a quadratic polynomial, c bt at t v + + =
2
) ( to
approximate the velocity profile.
t ) s ( 0 14 15 20 30 35
) (t v ) s / m (
0 227.04 362.78 517.35 602.97 901.67
The correct set of equations that will find a, b and c are
(A)
35 517
78 362
04 227
1 20 400
1 15 225
1 14 176
.
.
.
c
b
a

(B)
97 . 602
35 . 517
78 . 362
1 30 900
1 20 400
1 15 225
c
b
a

(C)
35 517
78 362
0
1 20 400
1 15 225
1 0 0
.
.
c
b
a

(D)
67 . 901
97 . 602
35 . 517
1 35 1225
1 30 900
1 20 400
c
b
a

Solution

First choose the three points closest to s 21 = t that also bracket it.
m/s 97 . 602 ) ( , s 30
m/s 35 . 517 ) ( , s 20
m/s 78 . 362 ) ( , s 15
2 2
1 1
0 0
= =
= =
= =
t v t
t v t
t v t

Such that
c b a v
c b a v
c b a v
+ + = =
+ + = =
+ + = =
) 30 ( ) 30 ( 97 . 602 ) 30 (
) 20 ( ) 20 ( 35 . 517 ) 20 (
) 15 ( ) 15 ( 78 . 362 ) 15 (
2
2
2

This expands to
+ +
+ +
+ +
= + +
= + +
= + +
97 . 602
35 . 517
78 . 362
30 900
20 400
15 225
97 . 602 30 900
35 . 517 20 400
78 . 362 15 225
c b a
c b a
c b a
c b a
c b a
c b a

This can be rewritten as
97 . 602
35 . 517
78 . 362
1 30 900
1 20 400
1 15 225
c
b
a

Ilovepdf.com (3)

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Ilovepdf.com (3)

Transféré par

Droits d'auteur :

Formats disponibles

Multiple-Choice Test

by using 2-segment Simpsons 1/3 rule most nearly is

by using 4-segment Simpsons 1/3 rule most nearly is

by the using one-segment trapezoidal rule is most nearly

by using the three-segment trapezoidal rule is most nearly

is called the error function. It is used in the field of

%. The final converged root was found as 9999997 . 1 ==

Vous aimerez peut-être aussi