Department of Mathematics and Philosophy of Engineering

MHZ4553-Engineering Mathematics III

Assignment No.01

Due date: 07/07/2023 Academic Year: 2022/2023.

1. a) Let
cos y sin x
(
x ̸= 0
f (x, y) = x
cos y x=0

Is f continuous at (0, 0)? Justify your answer.

We see that f (0, 0) = cos 0 = 1.

Consider
cos y sin x
lim = lim cos y × 1 = lim cos y = cos 0 = 1.
(x,y)→ (0,0) x (x,y)→ (0,0) (x,y)→ (0,0)

sin x
Since, we know that lim(x,y)→ (0,0) = 1.
x
Then,
cos y sin x
f (0, 0) = lim =1
(x,y)→ (0,0) x
Hence, f is continuous at (0, 0). We can show that f is continuous at all points
in R2 . As long as x ̸= 0, we can evaluate the limit directly. Also when x = 0, a
similar analysis shows that the limit is cos y. Thus we can say that f is continuous
everywhere.
b) Find the following limits if they exist:

8xy 5x2 y 2
(i). lim (ii). lim .
(x,y)→ (0,0) x2 + y 2 (x,y)→ (0,0) x2 + y 2

8xy
(i) Evaluating lim(x,y)→ (0,0) along the lines y = mx
x2 + y2

8xy 8mx2 8m
lim = lim = lim .
(x,y)→ (0,0) x2 + y 2 (x,y)→ (0,0) x2 + m2 x2 (x,y)→ (0,0) 1 + m2

We can get a different limit for each choice of m. i.e. we get differing limiting values
along different lines. Then limit does not exist.
(ii) It is relatively to see that along the line y = mx, for any m ∈ R, the limit is 0.
This is not enough to prove that the limit exists.
To prove the limit is 0, we need show that for any ϵ > 0 there exist a δ > 0 such that
for ||x − y|| < δ, |f (x, y) − 0| < ϵ.
 Let ϵ > 0, p p p
Take δ = 5ϵ . Then if (x − 0)2 + (y − 0)2 < δ, then x2 + y 2 < 5ϵ and hence
p
ϵ
x2 + y 2 < , since x2 ≤ x2 + y 2 , x2 ≤ δ 2 .
5
5x2 y 2 2 y2 2
2 x +y
2
|g(x, y) − 0| = 2 2
− 0 = 5|x | 2 2
≤ 5|x | 2 2
= 5x2 < 5δ 2 < ϵ.
x +y x +y x +y

Thus
5x2 y 2
lim = 0.
(x,y)→ (0,0) x2 + y 2

c) A scalar field is given by f (x, y, z) = cos[π(x + 2y + 3z)].

(i). Evaluate the gradient of the function given above.

 
∂f ∂f ∂f
∇f = , ,
∂x ∂y ∂z
∇f = (−π sin[(x + 2y + 3z)π], −2π sin[(x + 2y + 3z)π], −3π sin [(x + 2y + 3z)π])

(ii). At the point P located at (0.5, 0, 0), what is the unit vector pointed in the
direction of maximum change for the function f ?
At point P,
 π π π
∇f = −π sin , −2π sin , −3π sin = (−π, −2π, −3π)
2 2 2
p √
|∇f | = (−π)2 + (−2π)2 + (−3π)2 = 14π
∇f (−π, −2π, −3π)
Unit vector = = √ .
|∇f | 14π

(iii). What is the directional derivative of the function at the point P along a line
from P to the origin of the coordinate system ?
 π π π
∇f = −π sin , −2π sin , −3π sin = (−π, −2π, −3π) .
2 2 2
Vector from point to origin= (−0.5, 0, 0) = S

(−0.5, 0, 0)
⃗as = p = (−1, 0, 0)
(−0.5)2 + 0 + 0

Du = ∇f.⃗as
Du = π.

2. Define f : R2 → R by f (x, y) = xy 2 + xy − 3x2 y − 2x.

a) Find and classify the critical points of f (say X0 ).

The critical points of f can be obtained when fx = 0, fy = 0.

fx = y 2 + y − 6xy − 2 (1)

fy = 2xy + x − 3x2 (2)

By (2),
x(2y + 1 − 3x) = 0
2y + 1
x = 0, x= .
3
By (1), when x = 0,
y2 + y − 2 = 0
(y + 2)(y − 1) = 0
y = −2, y=1
Then, (0, −2), and (0, 1) are critical points of f.

△ = fxx fyy − (fxy )2

△ = −4y 2 − 1 − 36x2 − 4y + 12x + 12xy

△(0, −2) = −9 < 0
△(0, 1) = −9 < 0
(0, −2) and (0, 1) are saddle points.

b) Use Taylor series approximation to find a quadratic approximation for f at X0 (say

fq (x, y)).

Consider the critical point as X0 = (0, −2).

a0 = f (x0 , y0 ) = 0

∂f ∂f
ax = |(x ,y ) = 0, ay = |(x ,y ) = 0
∂x 0 0 ∂y 0 0
1 ∂ 2f 1 ∂ 2f
axx = |(x ,y ) = 6, ayy = |( x0 , y0 ) = 0
2! ∂x2 0 0 2! ∂y 2
∂ 2f ∂ 2f
axy = |(x ,y ) = −3, ayx = |(x ,y ) = −3,
∂x∂y 0 0 ∂y∂x 0 0
fq (x, y) = a0 +ax (x−x0 )+ay (y −y0 )+axx (x−x0 )2 +ayy (y −y0 )2 .+ayx (y −y0 )(x−x0 )
fq (x, y) = −3xy + 6x2 − 6x
Consider the critical point as X0 = (0, 1)

a0 = f (x0 , y0 ) = 0

∂f ∂f
ax = |(x ,y ) = 0, ay = |(x ,y ) = 0
∂x 0 0 ∂y 0 0
 1 ∂ 2f 1 ∂ 2f
axx = |(x ,y ) = −3, ayy = |( x0 , y0 ) = 0
2! ∂x2 0 0 2! ∂y 2
∂ 2f ∂ 2f
axy = |(x ,y ) = 3, ayx = |(x ,y ) = 3,
∂x∂y 0 0 ∂y∂x 0 0
fq (x, y) = a0 +ax (x−x0 )+ay (y −y0 )+axx (x−x0 )2 +ayy (y −y0 )2 +ayx (y −y0 )(x−x0 )
Substitute x0 = 0, y0 = −2

fq (x, y) = 3xy − 3x2 − 3x.

c) Use the change of variables to find the quadratic form (f¯q ) of fq at each X0 .
X0 = (0, −2), Change the variable x and y by getting x = x1 and y = y1 − 2.

f¯q (x1 , y1 ) = −3x1 (y1 − 2) + 6x21 − 6x1

f¯q (x1 , y1 ) = 6x21 − 3x1 y1 .

X0 = (0, 1), Change the variable x and y by getting x = x1 and y = y1 + 1.

f¯q (x1 , y1 ) = 3x1 (y1 + 1) − 3x21 − 3x1

f¯q (x1 , y1 ) = 3x1 y1 − 3x21 .

d) Find a symmetric matrix A such that XAX T = f¯q (X).

  
 a b x1
= 6x21 − 3x1 y1

x1 y 1
b c y1
ax21 + 2bx1 y1 + cy12 = 6x21 − 3x1 y1
By comparing both sides, we can get

a=6
3
c = 0, and b = −
2
For Then,  
3
 6 −2
A= 3 .
− 0
2
Again for (0, 1)   
 a b x1
= 3x1 y1 − 3x21

x1 y 1
b c y1
ax21 + cx1 y1 + bx1 y1 = 3x1 y1 − 3x21
By comparing both sides, we can get

a = −3
 3
c = 0, and b =
2
Then,  
3
−3
A =  3 2 .

0
2

e) Determine whetherf¯q (X) is Positive definite, Positive semidefine, Negative define,

Negative semidefinite or Indefinite.

3 √
The eigen values of matrix A are λ = (2 ± 5) when X0 = (0, −2). The eigen values
2
3 √
of matrix A are (−1 ± 2) when X0 = (0, 1). We can see that both matrices A have
2
positive and negative eigen values. Then by theorem, both f¯q (X) are indefinite.
f) Use part (e) to determine whether f¯q has maximum, minimum and saddle.

Since both matrices A have positive and negative eigen values. Then by the theorem,
fq has a saddle at (0, −2) and (0, 1).
g) Find and classify the critical points of each fq .
The critical points of each fq can be obtained when (fq )x = 0, (fq )y = 0.
Consider first fq = −3xy + 6x2 − 6x

(fq )x = −3y + 12x − 6 = 0 (3)

(fq )y = −3x = 0 (4)

By (4),
x=0
By (3),
y = −2.
△ = (fq )xx (fq )yy − ((fq )xy )2
△(0, −2) = −36 < 0.
Then, (0, −2) is a saddle point.
Consider second fq = 3xy − 3x2 − 3x

(fq )x = 3y − 6x − 3 = 0 (5)

(fq )y = 3x = 0 (6)

By (6),
x=0
By (5),
y = 1.
△ = (fq )xx (fq )yy − ((fq )xy )2
△(0, 1) = −36 < 0.
Then, (0, 1) is a saddle point.
 h) Compare the optimality of fq with optimality of f.

Both f and fq have same critical points as (0, −2), (0, 1).
i) Find and classify the critical points of f (x, y, z) = y 2 ex + xz + x2 + 2z 2 subject to
the constrain x + 2z = −2.
Here is the system that we need to solve, ∇f = λ∇g
y 2 ex + z + 2x = λ (7)

2yex = 0 (8)

x + 4z = 2λ (9)

x + 2z + 2 = 0 (10)
7 1 3
By solving above equatiobs we can get λ = − , x = − , y = 0, z = − . So, f
4 2 4
1 3 (x + 2)
has only one critical point and it is (− , 0, − ). By substituting z = − , in
2 4 2
to f (x, y, z) we can get f (x, y) = y 2 ex + x2 + x + 2.
fx = y 2 ex + 2x + 1, fxx = y 2 ex + 2
fy = 2yex fyy = 2ex fxy = 2yex
1 1 1 2 1
△(− , 0) = fxx (− , 0)fyy (− , 0) − fxy (− , 0) = 2.426 > 0
2 2 2 2
1 3
Since the second derivative test implies that f has a local minimum at (− , 0, − ).
2 4

−y i + x j R
3. Let F (x, y) = p . Compute C F · dr where C is the circle of radius r, centered
x2 + y 2
at the origin and traversed counterclockwise.
Let the parameterization be given by
r(θ) = x(θ) i + y(θ) j
x(θ) = r cos θ, y(θ) = r sin θ.
Because the curve is a circle, we parameterize it with the angle θ. Thus, we need an
expression relating the x and y coordinates of each point on the circle to θ:
r(θ) = r cos θ i + r sin θ j.
Traversing the circle counterclockwise means θ ranges from 0 to 2π. For convenience, we
sketch the curve C and its parameterization:
Expressing F in terms of θ,
  −r sin θ i + r cos θ j
F x(θ), y(θ) = √ = − sin θ i + cos θ j
r2 cos2 θ + r2 sin2 θ
Expressing dr in terms of θ:
 
dr = − r sin θ i + r cos θ j dθ

Expressing F.dr in terms of θ:

 
2 2
F.dr = r sin θ + r cos θ dθ = r.dθ

Noting that the bounds of integration are from θ = 0 to θ = 2π, we write the integral as:
Z Z 2π
F.dr = r.dθ = 2πr.
C 0