SEHH2250 2024-2025 Semester One Exam Revision Solution

Question 1

(a)
From the following table
z z arg ( z ) Euler Form
π
( 3)
2 π
1 + 3i 12 + =
2 2e
i
3
3
π
(2 3)
π
+ ( −2 ) =4
2
2 3 − 2i
2
− 4e
−i
6
6
3π 3π
−1 + i ( −1)
2
+ 12 =2 2e
i
4
4
we have
8
 i π3 
(1 + 3i )  2e 
8

=  
( 2 3 − 2i ) ( −1 + i )
6 π 6 3π 4
4
 −i 6   i 
 4e   2e 4 
   
1  π  π  3π 
8− 2⋅6 − ⋅4 i  ⋅8 − −  6 − ⋅4 
=2 2 3  6 
e 4 

1 i 23π
= e .
64
(b)
3π
−i
The Euler form of −3 − 3i is 18e 4
. Thus, the roots of the given equation are
  3π 
 2 kπ +  − 4 
i  
 3 

( 18 )
1
 
zk = 3
e  
for k = 0,1, 2,
i.e.,
1 π 1 5π 1 11π
−i i −i
z = 18 e 6 4
, 18 e 6 12
or 18 e 6 12
.
(c)
Write z= x + yi. Then

1
 x + yi − 2i 1
=
i ( x + yi ) − 2 2

22 x + ( y − 2 ) i =− y − 2 + xi
2 2

2 2
4  x2 + ( y − 2)  =  ( − y − 2) + x2 
2 2

   
( )
4 x2 + y 2 − 4 y + 4 = y 2 + 4 y + 4 + x2
3 x 2 + 3 y 2 − 20 y + 12 =
0.
2
 10  64
The locus of z is the circle x +  y −  = .
2

 3 9
(d)
 π  i 3t − π6  
Since cos  3t − =Re  e    and sin ( 3= t ) Re ( −iei 3t ) , we have
 6  
 
 π  i 3t − π6  
4 cos  3t −  − =
 6 
8sin ( 3t ) 4 Re  e    − 8 Re −iei 3t
  ( )
 
 i 3t − π6  
= Re  4e   + 8iei 3t 
 
 
  −i π

= Re  4ei 3t  e 6 + 2i   .
 
  
With
π
−i  π  π
e 6
+ 2i = cos  −  + i sin  −  + 2i
 6  6
3 1
= − i + 2i
2 2
3 3
= + i
2 2
π
i
= 3e 3 ,
it follows that
 π  i 3t π
i   i  3t +  
 π
 π
4 cos  3t −  − 8sin (=
3t ) Re  4e ⋅ 3e=3
 Re  4 3e =
 3
 4 3 cos  3t +  .
 6     3
 
Thus,
π
A = 4 3 and ϕ = − .
3

2
Question 2

(a)
Let f ( x, y ) = 3 x + 3 y − x − 3 xy . Then,
2 2 3 2

f x =6 x − 3 x 2 − 3 y 2 , f=
y 6 y − 6 xy, f xx= f yy = 6 − 6 x and f xy = −6 y.
Setting f=
x f=
y 0, i.e.,
6 x − 3 x − 3 y =
2 2
0,

 6 y (1 − x ) =
0,
we obtain from the second equation that x = 1 or y = 0.
If x = 1 (resp. y = 0 ), then the first equation becomes 3 − 3 y 2 = 0 ), so that y = −1 or
0 (resp. 6 x − 3 x 2 =
y = 1 (resp. x = 0 or x = 2 ). Thus, the stationary points of f are
(1, −1) , (1,1) , ( 0, 0 ) and ( 2, 0 ) .
( )
2
Let D = f xx ⋅ f yy − f xy . We classify the nature of the above stationary points as follows.
( x, y ) f xx ( x, y ) D ( x, y ) Nature
(1, −1) 0 −36 ( < 0 ) saddle point
(1,1) 0 −36 ( < 0 ) saddle point
( 0, 0 ) 6 ( > 0) 36 ( > 0 ) local minimum point
( 2, 0 ) −6 ( < 0 ) 36 ( > 0 ) local maximum point
(b)
Let f ( x, y ) = 2 xy − x 2 − 3 y 2 and g ( x, y ) =x 2 + 3 y 2 − 6. By the method of Lagrange multipliers, we
consider the system of equations
∇f = λ∇g ,

 g = 0,
i.e.,
 2 y − 2x = 2λ x ,

 2x − 6 y = 6λ y,
 x2 + 3 y 2 − 6 =
 0.
From the first and second equations in the above system,
3 y ( 2 y − 2 x ) − x ( 2 x − 6 y ) = 6λ xy − 6λ xy
x2 = 3 y 2 .
This, together with the third equation in the system, gives
3y2 + 3y2 − 6 = 0
y = −1 or 1.
Then
3 ( ±1)
2
x=
2

x = − 3 or 3.

3
When ( x, y ) = ( ) ( )
3,1 or ( x, y ) =− 3, −1 , we have

f ( ± 3, ±1) = 2 ( ± 3 ) ( ±1) − ( ± 3 ) − 3 ( ±1) = 2

2 2
3 − 6.

When ( x=
, y) ( 3, −1) or ( x, y ) = ( − 3,1) , we have
f ( ± 3, 1) =−2 ( 3 ) (1) − ( ± 3 ) − 3 ( 1) =−2
2 2
3 − 6.
Hence the minimum and maximum values of f subject to g = 0 are −2 3 − 6 and 2 3 − 6 respectively.

Question 3

(a)
A unit vector u in the direction of 4i + 3 j is
1 1 4 1 4
=u ( 4=
i + 3 j) =    
4i + 3 j 42 + 32  3  5  3 
and the gradient (vector) of f at P is
 fx  6 x + 2 y 2 − ye xy  6
∇f (1, 0 )=   =   =  −1 .
f
  (1,0) 
y 4 xy − xe xy
 (1,0)  
The required rate of change is the directional derivative of f at P in the direction of 4i + 3 j, which is given
by
 6  1  4  21
∇f (1, 0 ) ⋅ u
=  ⋅  =  .
 −1 5  3  5
(b)
Since the maximum rate of change of f at P is
6
∇f (1, 0=
) 62 + ( −1)=
2
 −1= 37 < 8,
 
there is no direction in which the rate of change of f at P equals 8.
(c)
Let g ( =
x, y, z ) f ( x, y ) − z. Then
 gx   fx  6
g 
∇g (1, 0, 2 ) = f 
=  −1 .
=
 y  y  
 g z   −1  −1
(1,0,2 ) (1,0,2 )
The equation of the required tangent plane is
6 ( x − 1) + ( −1)( y − 0 ) + ( −1)( z − 2 ) =0
6x − y − z − 4 =0.

4
Question 4

(a)
The Maclaurin series of f ( x ) = e x up to degree 8 is given by
2

f ( x ) = ex
2

(x ) + (x ) + (x )
2 3 4
2 2 2

=
1+ x 2
+ +
2!
3! 4!
4 6
x x x8
≈ 1 + x2 + + + .
2 6 24
If x ≠ 0, then
x 4 x 6 x8
1 + x2 + + + −1
ex −1
2

2 6 24 x3 x5 x7
≈ =x + + + .
x x 2 6 24
Thus,
1
ex −1 1 x3 x5 x7   x2 x4 x8 
2
1 x6
∫0 x dx ≈ ∫0  2 6 24   2 4 ⋅ 2 6 ⋅ 6 8 ⋅ 24  ≈ 0.6580.
x + + + dx = + + +
0
(b)
x2 − y3
Let w = . Then,
z +2
2x 3y2 x2 − y3
wx = , wy = − and wz = − .
z +2 z +2 ( )
2
2 z z +2
With ∆x =2.98 − 3 =−0.02, ∆y= 1.01 −=
1 0.01 and ∆=
z 4.03 −=
4 0.03, we have
∆w ≈ dw
= wx ∆x + wy ∆y + wz ∆z
2x 3y2 x2 − y3
= ∆x − ∆y − ∆z
z +2 z +2 ( )
2
2 z z +2
2⋅3 3 ⋅12 32 − 13
= ( −0.02 ) − ⋅ 0.01 − ⋅ 0.03
4+2 4+2 ( )
2
2 4 4+2
= −0.04125.
Hence
( 2.98) − (1.01)
2 3
32 − 13
= − 0.04125 ≈ 1.9588.
4.03 + 2 4+2
x2 − y3
Alternative Solution Let g ( x, y, z ) = . With
z +2

5
 2x 3y2 x2 − y3
gx = , gy = − and g z = − ,
z +2 z +2 ( )
2
2 z z +2
it follows from the linear approximation given by Taylor’s theorem that
( 2.98) − (1.01)
2 3

= g ( 2.98,1.01, 4.03)
4.03 + 2
≈ g ( 3,1, 4 ) + g x ( 3,1, 4 )( 2.98 − 3) + g y ( 3,1, 4 )(1.01 − 1) + g z ( 3,1, 4 )( 4.03 − 4 )
32 − 13 2⋅3 3 ⋅12 32 − 13
= + ( −0.02 ) − ⋅ 0.01 − ⋅ 0.03
4+2 4+2 4+2 ( )
2
2 4 4+2
≈ 1.9588.

Question 5

(a)
The given (non-linear) equation is homogeneous, since
4
3y4  y
4 3 
=
3y
= x 4
x ,
3 xy − x
3 4
3 xy − x
3 4
 y
3

x4
3   −1
x
y y
which is a function of only. Let v = . Then,
x x
dy dv
y = xv and = x + v.
dx dx
The equation now becomes
dv 3v 4
x +v = 3 ,
dx 3v − 1
so that
dv v
x = 3 .
dx 3v − 1
The above equation is separable, for we may express it as
 2 1 1
 3v −  dv = dx.
 v x
Integrating both sides gives
v3 − ln v = ln x + c,
where c is any constant. Hence the solution is
3
 y
  − ln y =
c.
x

6
(b)
An integrating factor of this equation is
2
∫ − x dx 1
e= e −=
2ln x ln x −2
e= ( since x > 0 ) .
x2
Then
d  1  1
 2 ⋅ y  = 2 ⋅ 2x e ,
3 x2

dx  x  x
so that

∫ e d ( x =)
y
= ∫ 2 xe = e x + c,
x2 x2 2 2
dx
x2
i.e.,
=y x 2 e x + cx 2 ,
2

where c is any constant.

(c)
The given equation is exact, since
∂ ∂
∂y
( )
2 x + sin y − e − x y = cos y − e − x =
∂x
(
x cos y + cos y + e − x . )
Let ψ be a function such that
 ∂ψ
 ∂x = 2 x + sin y − e − x y,
 ∂ψ
 = x cos y + cos y + e − x .
 ∂y
From the first equation on ψ ,
ψ ( x, y ) =∫ ( 2 x + sin y − e − x y ) dx =x 2 + x sin y + e − x y + g ( y ) ,
where g is a function of y. Moreover, from the second equation on ψ ,
ψ ( x, y )= ∫ ( x cos y + cos y + e − x ) dy= x sin y + sin y + e − x y + h ( x ) ,
where h is a function of x. It follows that
g ( y ) = sin y and h ( x ) = x 2 .
Hence the general solution is
x 2 + e − x y + x sin y + sin y =
c,
where c is any constant.

7
Question 6

(a)
The auxiliary equation of the corresponding homogeneous equation is
m 2 + 2m − 8 =0.
Its two roots are m = −4 or m = 2.
Two linearly independent solutions are e −4 x and e 2 x , and the general solution of the associated
homogeneous equation is
( x ) c1e−4 x + c2e2 x .
yh=
Let
y p ( x=
) Axe2 x + Bx 2 + Cx + D
be a particular solution of the given inhomogeneous equation. Then,
dy p d 2 yp
= 2 Axe 2 x + Ae 2 x + 2 Bx + C and 2
= 4 Axe 2 x + 4 Ae 2 x + 2 B.
dx dx
Substituting these into the inhomogeneous equation gives
( ) (
4 Axe 2 x + 4 Ae 2 x + 2 B + 2 2 Axe 2 x + Ae 2 x + 2 Bx + C − 8 Axe 2 x + Bx 2 + Cx + D )
= 3e 2 x − x 2 + 1.
Grouping like terms, we obtain
6 Ae 2 x − 8 Bx 2 + ( 4 B − 8C ) x + 2 B + 2C − 8 D= 3e 2 x − x 2 + 1.
It follows that
 −8 B =−1,

6 A = 3 and  4 B − 8C = 0,
2 B + 2C − 8 D =1.

1 1 1 5
Solving gives A = , B = , C = and D = − . A particular solution is
2 8 16 64
1 2x 1 2 1 5
y p ( x )= xe + x + x − .
2 8 16 64
Hence the general solution of the equation is
1 2x 1 2 1 5
y ( x ) = yh ( x ) + y p ( x ) = c1e −4 x + c2 e 2 x + xe + x + x − ,
2 8 16 64
where c1 and c2 are any constants.
(b)
The auxiliary equation of the corresponding homogeneous equation is
m 2 − 2m + 1 =0.
Its only root is m = 1.
Two linearly independent solutions are y1 ( x ) = e x and y2 ( x ) = xe x , and the general solution of the
associated homogeneous equation is
yc (=
x ) c1e x + c2 xe x ,
where c1 and c2 are any constants. Moreover,

8
 y y2 ex xe x
W ( y1 , y2 ) = 1 =x =( x + 1) e2 x − xe2 x =e2 x .
y1 ' y2 ' e ( )
x + 1 e x

Put R ( x ) = e x tan −1 x. By the method of variation of parameters, a particular solution y p is given by

 y ( x) R ( x)   y ( x) R ( x) 
=y p ( x ) y2 ( x )  ∫ 1 dx  − y1 ( x )  ∫ 2 dx 
 W ( y1 , y2 )   W ( y1 , y2 ) 
e x ⋅ e x tan −1 x x xe ⋅ e tan
x x −1
x
= xe x ∫ 2x
dx − e ∫ 2x
dx
e e
= xe x ∫ tan −1 x dx − e x ∫ x tan −1 x dx.
Using integration by parts, we have

∫ tan
−1
( )
x dx= x tan −1 x − ∫ x d tan −1 x = x tan −1 x −
1 2x
∫
2 1+ x 2
1
dx= x tan −1 x − ln 1 + x 2
2
( )
and
 x2 
∫ x tan x dx = ∫ tan x d  2 
−1 −1

x2 x2
=
2
tan −1 x − ∫ d tan −1 x
2
( )
2
x 1 x2
= tan −1 x − ∫ dx
2 2 1 + x2
x2 1  1 
= tan −1 x − ∫ 1 −  dx
2 2  1 + x2 
x2
=
2
1
(
tan −1 x − x − tan −1 x
2
)
(we omit the constants of integration in the above two integrals). Thus,
   x2 
=

1
(

) 1
y p ( x ) xe x  x tan −1 x − ln 1 + x 2  − e x  tan −1 x − x − tan −1 x  ( )
2 2 2 
=
1 2 x
2
1
2
( 1
)
x e tan −1 x − xe x ln 1 + x 2 + e x x − tan −1 x .
2
( )
The general solution of the given equation is
y ( x ) = yc ( x ) + y p ( x ) = c1e x + c2 xe x +
1 2 x
2
1
2
( 1
)
x e tan −1 x − xe x ln 1 + x 2 + e x x − tan −1 x .
2
( )

Question 7

(a)
Since
cos ( 4 ( t − x ) ) dx =
et ∗ cos ( 4t ) ,
t
∫e
x
0
it follows from the convolution theorem that

9
 L {∫ e cos ( 4 (t − x )) dx} =L{e ∗ cos ( 4t )} =L{e } L{cos ( 4t )}
t

0
x t t

1 s s
= ⋅ 2 = .
s −1 s + 4 2
(
( s − 1) s 2 + 16 )
(b)
Write
s+4 s + 3 +1 s+3 1 6
= = + ⋅ .
s + 6 s + 45 ( s + 3) + 62 ( s + 3) + 6 6 ( s + 3) + 62
2 2 2 2 2

Then

−1  s+4  
−1  s+3  1 −1  6 
L=  2  L   + L  
 s + 6 s + 45   ( s + 3) + 6   ( s + 3) + 6 
2 2 2 2
6
1
= e −3t cos ( 6t ) + e −3t sin ( 6t ) .
6
(c)
Take the Laplace transforms on both sides of the given equation:
 t  
L  ∫ ( t − x )
2 dy
+ y ( x )  dx  = { }
L t2
 0 dx  
 t
L ∫ (t − x )
0
2 dy 
{
dx  + L ∫ y ( x ) dx =
dx 
t

0 }2
s3
Since
L {∫ y ( x ) dx} = 1s L { y}
t

and the convolution theorem yields

 t 2 dy   dy 
L ∫ (t − x ) dx  = L t 2 ∗ 
0 dx   dt 
 dy 
{ }
= L t2 L  
 dt 
2
= 3  sL { y} − y ( 0 ) 
s
= =
2
s2
L { y} ( as y ( 0 ) 0 ) ,
it follows that
2 1 2
2
L { y} + L { y} = ,
s s s3
i.e.,
2 s2 2
L { y} =3 ⋅ = .
s s + 2 s ( s + 2)
Hence

10
  2  
−1  s + 2 − s 

) L−1 
y ( t= =  L  = −1  1 
 L  −L 
−1  1 
= −2 t
 1− e .
 (
s s + 2 )   (
s s + 2 )   
s  s + 2 

Question 8

(a)
With
f (t ) =
t + u ( t − 2 )( 2 − t ) ,
we have
1 e −2 s
L { f ( t )} =L {t} - L {u ( t − 2 )( t − 2 )} =L {t} − e −2 s L {t} = − .
s2 s2
(b)
Taking the Laplace transforms on both sides of the given equation:
d2 y 
L  2 + 16 y  = L { f ( t )}
 dt 
 2 dy  1 e −2 s
 s Y ( s ) − sy ( 0 ) − dt ( 0 )  + 16Y ( s ) = −
s2 s2
dy
Since y ( 0 ) = 1 and ( 0 ) = 0, it follows that
dt
1 e −2 s
( )
s 2 + 16 Y ( s ) = 2 − 2 + s,
s s
i.e.,
1 e −2 s s
Y (s) = − + 2
( ) (
s s + 16 s s + 16 s + 16
2 2 2 2
)
1 s 2 + 16 − s 2 1 −2 s  s 2 + 16 − s 2  s
= ⋅ 2 2 − e  2 2 + 2
(
16 s s + 16 16 ) (
 s s + 16  s + 16 )
11 1  1 −2 s  1 1  s
=  2− 2 − e  2 − 2 + 2 .
16  s s + 16  16 s s + 16  s + 16
Hence
1 1 s 
) L−1   2 − 2
1  1 −2 s  1 1 
y ( t= − e  2 − 2 + 2 
16  s s + 16  16 s s + 16  s + 16 
1  1  1  
t − sin ( 4t )  − u ( t − 2 ) t − 2 − sin ( 4 ( t − 2 ) )  + cos ( 4t )
1
= 
16  4  16  4 
 1 1 
  t − sin ( 4t ) + cos ( 4t ) if 0 ≤ t < 2, 
 = 16 64 
  1 − 1 sin 4 + cos 4 + 1 sin 4 − 8 if ≥ 2. 
  ( t) ( t) ( t ) t 
  8 64 64 

11
Question 9

12
Question 10

The set T is not a subspace of  2 . For example, it does not contain the zero vector [ 0 0] because
T

02 + 02 =0 ≠ 1.
The set U is also not a subspace of  2 . For example, both vectors [1 0] and [ 0 1] are in U because
T T

12 + 02 =1 ≤ 1 and 02 + 12 =1 ≤ 1. However, the vector [1 0] + [ 0 1] =

[1 1] is not in U for
T T T

12 + 12 = 2 > 1.

Question 11

(a)
Reduce the matrix A by elementary row operations as follows.
 1 −2 3 2  −3R
2R1 + R 2 → R 2  1
+ R →R
−2 3 2 
3R + R → R
1 −2 3 2 
 2 −3 3 1  1 3 3 0 1 −3 −3 2 3 3 0 1 −3 −3
     
 −3 3 0 3  0 −3 9 9  0 0 0 0 
From the rightmost matrix, the leading non-zero entries of the non-zero rows occur in the first and second
columns. Thus, a basis for Col ( A ) is

{[1 2 −3] , [ −2 −3 3] .
T T
}
An equivalent system of the homogeneous system Ax = 0 is

13
  x1 − 2 x2 + 3 x3 + 2 x4 = 0,

 x2 − 3 x3 − 3 x4 = 0.
Solving gives x=
1 3s + 4t , x=
2 3s + 3t , x3 = s and x4 = t , where s and t are any scalars. Thus, a basis for
Nul ( A ) is

{[3 3 1 0] , [ 4 3 0 1] .
T
}
(b)
From (a),
rank ( A )= 2= dim ( Nul ( A ) ) .
With
rank ( A ) + dim ( Nul ( A ) ) = 2 + 2 = 4 = number of columns of A,
this verifies the fundamental theorem of linear algebra.
(c)
Let s and t be scalars such that
[5 6] s [1 2 −3] + t [ −2 −3 3] ,
T T T
c=
i.e.,
 s − 2t =
5,

 2 s − 3t =
c,
−3s + 3t =
 6.
Solving the first and third equations, we have
s = −9 and t = −7.
Thus, c = 2 ( −9 ) − 3 ( −7 ) = 3.

14