國立陽明交通大學 112 學年度微積分 II 課號：564011 測驗 ♯2 詳解

單選題

 xy − x − 2y + 2
, if (x, y) ̸= (2, 1);
1. Let f (x, y) = x2 + y 2 − 4x − 2y + 5
0, if (x, y) = (2, 1).

(i) g y (x) = f (x, y) is a continuous function of the x-variable for all fixed y ∈ R.
(ii) ∇f (2, 1) = ⟨0, 0⟩
(iii) Du f (2, 1) = 0 for all unit vector u ∈ R2 .
(iv) f is continuous at (2, 1).

Among the above three statements, how many of them are true?

A. Only one. B. Only two. C. Three of them. D. All of them.

Solution: B

 (x − 2)(y − 1)
, if (x, y) ̸= (2, 1);
We note that f (x, y) = (x − 2)2 + (y − 1)2 The detailed argument can be seen as the function

0, if (x, y) = (2, 1).

 xy
2 + y2
, if (x, y) ̸= (0, 0);
f (x, y) = x with (0, 0) replaced by (2, 1).
0, if (x, y) = (0, 0).

 3 2
 x y , if (x, y) ̸= (0, 0);
2. Let f (x, y) = x6 + y 4 Which of the following statements is false?

0, if (x, y) = (0, 0).

A. f is not continuous at (0, 0).

B. The limit of f (x, y) is 0, as (x, y) → (0, 0) along any line L through (0, 0).
C. There is some unit vector u, Du f (0, 0) fails to exists.
1
D. There is curve C through (0, 0) such that the limit of f (x, y) is , as (x, y) → (0, 0) along C.
2

Solution: C
 2
 x y , if (x, y) ̸= (0, 0);
This is modified from f (x, y) = x4 + y 2 See the lecture note for details.

0, if (x, y) = (0, 0).

3. Consider the following vector-valued functions F defined on R2 by

(i) ⟨x + 4y, 3x − y⟩. (ii) ⟨2x + 3y, 3x − y⟩.
2 2
(iii) ⟨2xy−2y + (y + 1) , (x−1) + 2yx + 2x⟩. (iv) ⟨x2 y, y + 5⟩.
Among the above four vector-valued functions F, how many of them are the gradient vector of an f : R2 −→ R?

A. Only one. B. Only two. C. Three of them. D. All of them.

Solution: B .
If so F = ⟨fx , fy ⟩. We note further that all the component functions of above four vector-valued functions F are
differentiable on R2 .
(i) is from Exercise ♯95 of Section 14.3. fxy = (x + 4y)y = 4 ̸= 3 = (3x − y)x = fyx . So no such an f exists by Clairaut’s
theorem in this case.
For (iv) fxy = (x2 y)y = x2 ̸= 0 = (y Z+ 5)x = fyx . So no such an f exists by Clairaut’s theorem in this case.
For (ii), fx = 2x + 3y =⇒ f (x, y) = (2x + 3y) dx = x2 + 3xy + g(y) =⇒ fy = 3x + g ′ (y) = 3x − y =⇒ g ′ (y) = −y =⇒
 國立陽明交通大學 112 學年度微積分 II 課號：564011 測驗 ♯2 詳解

y2
f (x, y) = x2 + 3xy − + C, for any constant C is such an f .
2 Z
2 2
For (iii), fx = 2xy−2y + (y + 1) =⇒ f (x, y) = (2xy−2y + (y + 1) ) dx = x2 y − 2xy + x(y + 1)2 + g(y) =⇒ fy (x, y) =
x2 − 2x + 2x(y + 1) + g ′ (y) = (x−1) + 2yx + 2x =⇒ g ′ (y) = 1 =⇒ f (x, y) = x2 y − 2xy + x(y + 1)2 + y + C, for any
2

constant C is such an f .

p
4. Let f (x, y) = 3
x3 + y 3 .
(i) f is continuous on R.
(ii) ∇f (0, 0) = ⟨1, 1⟩ attains its global maximum and minimum values in D.
(iii) Du f (0, 0) exists for all unit vector u.
(iv) f is not differentiable at (0, 0).
Among the above four statements, how many of them are false?
A. Only one. B. Only two. C. Three of them. D. None of them.

Solution: D (The function f is from Exercise ♯100 of Section 14.3.)

(i) is true, because it is a composite function of√two continuous ones.
f (h, 0) − f (0, 0) h3 − 0
3

For (ii) fx (0, 0) = lim = lim = 1 and fy (0, 0) = 1 can be computed similarly. So (ii) is true.
h→0 h h→0 h
For (iii), with θ ∈ [0, 2π) fixed, let u⟨cos θ, sin θ⟩,
p
t cos3 θ + sin3 θ − 0
3 p
3
Du f (0, 0) = lim = cosθ + sin3 θ.
t→0 t
So (iii) is true. p
For (iv), if f is differentiable at√(0, 0),rthen Du f (0, 0) = ∇f (0, 0) · u = cos θ + sin θ = cos3 θ + sin3 θ for all θ ∈ [0, 2π).
3

π π √ √ 2 π π
But cos + sin = 2 ̸= 2 = 3 cos3 + sin3 , so (iv) is true.
3

4 4 2 4 4

   t2 √ √ du
5. If u = ln sin xy − 2 , when x = , y = t2 − 1. At t = 2,
1
=
2 dt
√
2π √ √ √
A. . B. 3 2 tan 1. C. 3 2 cot 1. D. 2 cot
2
1
.
2

Solution: D  
√
2, x = y = 1 =⇒ y − 2 = 1 and xy − 2 = 1. Using chain rule
1 1
We first note that at t =

   du ∂u dx ∂u dy  − 1       1 3  t
u = ln sin xy − 2 = y 2 cot xy − 2 t + x cot xy − 2 − y− 2 √
1 1 1
=⇒ = +
dt ∂x dt ∂y dt 2 t2 − 1
√ √
du √ (1 cot 1) 2 2 cot 1
=⇒ = (1 cot 1) 2 − = .
dt √ 2 2
t= 2
 國立陽明交通大學 112 學年度微積分 II 課號：564011 測驗 ♯2 詳解

−2x
6. Let f (x, y) = for (x, y) ∈ D = {(x, y) |0 ≤ x ≤ 2, −x ≤ y ≤ x}, which of the following is true:
x2 + y 2 + 1
A. The global maximum value of f on D is only attained at one point.
B. The global minimum value of f on D is only attained at one point.
4
C. The global minimum value of f on D is − .
9
D. f has a saddle point in the interior of D.

Solution: A
We note that D is the triangular region with vertices (0, 0), (2, −2) and (2, 2) which lies on the right half plane, and we
further note that f (x, y) is a rational function of x and y with positive denominator. So global maximum value of f
is 0 = f (0, 0) > f (x, y) for (x, y) ∈ D \ {(0, 0)}.

7. Let f : R2 −→ R so that all its second order partial derivatives are continuous.
(i) If f (a, b) is a saddle point on R2 , then fxx (a, b)fyy (a, b) − fxy
2
(a, b) < 0.
(ii) Both fx and fy are differentiable on R2 .
(iii) If f (a1 , b1 ), f (a2 , b2 ) are local minimum values of f , then there is an (a3 , b3 ) ∈ R2 , f attains its local maximum value
there.
(iv) If (a, b) is the unique local minimum value of f on R2 , then f (a, b) must also be the global minimum value on R2 .
Among the above four statements, how many of them are true?
A. Only one. B. Only two. C. Three of them. D. None of them.

Solution: A .
(i) is false, check 填空題 ♯10.
(ii) is true. Since fxx = (fx )x , fxy = (fx )y are continuous, fx has continuous first order partial derivatives. So fx is
differentiable. Similar argument shows that fy is also differentiable.
(iii) is false. For example, f (x, y) = (x2 − 1)2 + (x2 y − x − 1)2 (Exercise ♯41 of Section §14.7 of the text. You may check
that (−1, 0) and (1, 2) are the only 2 critical points of f where f attains its local minimum value 0. But if there is an
(a3 , b3 ) ∈ R2 , f attains its local maximum value there, it must also be a critical point of f .
(iv) is false. For example, f (x, y) = x3 + y 3 + 3x2 + 3y 2 . You may check f (−2, −2) = 8 is the local maximum value and
f (0, 0) = 0 is the local minimum value. lim f (x, y) = −∞, so there is no global minimum value on R2
x=y→∞

8. The two particles

√ move √ x2 and x − y = 1 respectively. How far apart are they when they are closest?
freely on the√curves y = √
3 2 5 2 7 2 9 2
A. . B. . C. . D. .
8 8 8 8

Solution: C .
Let particle A(x, y) on the curve y = x2 and particle B(u, v) on the curve u − v = 1. Then, we try to minimize
f (x, y, u, v) = (x − u)2 + (y − v)2 subject to the constrains g(x, y, u, v) = y − x2 = 0 and h(x, y, u, v) = u − v − 1 = 0.
By the method of Lagrange multipliers, we solve
 

 2(x − u) = −2λx, 
 x − u = −λx,

 


 2(y − v) = λ, 
 2(y − v) = λ,

 

−2(x − u) = µ, −2(x − u) = µ,
⇐⇒

 −2(y − v) = −µ, 
 2(y − v) = µ,

 


y − x2 = 0, 
y − x2 = 0,

 

 
u−v−1=0 u−v−1=0

By the third and the fourth equations, x − u = −(y − v). This together with the first and the second equations,
1 1
x = =⇒ y = , by the fifth equation, and λ = µ.
2 4
 國立陽明交通大學 112 學年度微積分 II 課號：564011 測驗 ♯2 詳解

1
By adding fifth equation the sixth equation, we get (y − v) − x2 + u = 1 Replacing y − v by −(x − u) and x = , we
2
1 1 3 5
get + u − + u+ = 1 ⇐⇒ u = =⇒ v = −
2 4 8 8
   
1 1 3 5
In conclusion, the point , 2
on the curve y = x and the point ,− on the curve x − y = 1, respectively,
2 4 8 8
r √
4 49 53
are closest together. So, the distance is + = .
64 64 8

填空題
x2 + y 2
1. lim p = (1) .
(x,y)→(0,0) x2 + y 2 + 1 − 1

Solution:
" p #
x2 + y 2 x2 + y 2 x2 + y 2 + 1 + 1
lim p = lim p ·p
(x,y)→(0,0) x2 + y 2 + 1 − 1 (x,y)→(0,0) x2 + y 2 + 1 − 1 x2 + y 2 + 1 + 1
 p
2  p
(x
 + y 2 )( x2 + y 2 + 1 + 1)
= lim 22 = lim ( x2 + y 2 + 1 + 1) = 2 .
(x,y)→(0,0) 
x +y (x,y)→(0,0)

e−(x +y ) − 1
2 2

2. Let f (x, y) = , if (x, y) ̸= (0, 0). Is it possible to define f (0, 0) so that f is continuous at (0, 0)?
x2 + y 2
Answer: (2) .

Solution:
e−(x +y ) − 1 e−r − 1 l’H re−r
2 2 2 2
2
−
lim = lim = lim = −1 .
(x,y)→(0,0) x2 + y 2 r→0+ r2 r→0+ 2
r

Z y

3. Let f (x, y) = cos et dt. Compute fx (x, y) = (3) .
x

Solution: By fundamental theorem of calculus,

Z y

f (x, y) = cos et dt =⇒ − cos (ex ) .
x

Z xy 2
sin (xyt)
4. Let f (x, y) = dt. Compute fx (x, y) = (4) .
1 t

Solution: By the formula,

Z ! Z
f (x,y) f (x,y)
∂ ∂g
g(x, y, t) dt = g(x, y, f (x, y))fx (x, y) + (x, y, t) dt.
∂x a a ∂x

Z 2  
sin xy · xy 2 xy
y t · cos (xyt)
=⇒ fx (x, y) = · y
2 + ds
xy 2
1 t

Z xy2
 t=xy2
sin x2 y 3 sin x2 y 3 sin(xyt)
= +y cos(xyt) dt = +
x 1 x x t=1
2 3


2 sin x y sin(xy)
= − .
x x
 國立陽明交通大學 112 學年度微積分 II 課號：564011 測驗 ♯2 詳解

√
5. Let f (x, y, z) = xy 2 z 3 + arcsin(x z), fxzy (x, y, z) = (5) .

Solution: By Clairaut’s theorem,

 
fxzy (x, y, z) = fyxz (x, y, z) = (fy )xz (x, y, z) = (2xyz 3 )x z = (2yz 3 )z = 6yz 2 .

 3
 x y − 2xy ,
3

2 2
if (x, y) ̸= (0, 0); ∂2f
6. Let f (x, y) = 3x + y (0, 0) = (6) .
 ∂x∂y
0, if (x, y) = (0, 0).

Solution:
 
∂2f ∂ ∂f fy (h, 0) − fy (0, 0)
(0, 0) = (0, 0) = lim .
∂x∂y ∂x ∂y h→0 h

Thus, we first compute fy (x, 0) explicitely.

f (x, k) − f (x, 0)
fy (x, 0) = lim
k→0 k
 x3 k−2xk3 3 3

 − x3x0−2x0 x3 − 2xk 2 x
 lim 3x +k
2 2 2 +02
= lim = , if x ̸= 0; x
k→0 k k→0 3x2 + k 2 3
= = .


03 k−20y 3
− 0 3
 lim 30 +k2 2
= 0, if x = 0
k→0 k

∂2f 1
Hence, (0, 0) = .
∂x∂y 3

7. Suppose you need to know an equation of the tangent plane to a surface S at the point P (2, 1, 3). You don’t have an equation
for S but you know that the curves

r1 (t) = ⟨2 + 3t, 1 − t2 , 3 − 4t + t2 ⟩
r2 (u) = ⟨1 + u2 , 2u3 − 1, 2u + 1⟩

both lie on S. Find an equation of the tangent plane at P . Answer: (7) .

Solution: We first check that r1 (0) = ⟨2, 1, 3⟩ and r2 (1) = ⟨2, 1, 3⟩, so both curves pass through the point P . Their
tangent vectors

r′1 (t) = ⟨3, −2t, −4 + 2t⟩ =⇒ r1 (0) = ⟨3, 0, −4⟩

r′2 (u) = ⟨2u, 6u2 , 2u⟩ =⇒ r2 (1) = ⟨2, 6, 2⟩

are parallel to the tangent plane. Thus, r′1 (0) × r′2 (1) = ⟨3, 0, −4⟩ × ⟨2, 6, 2⟩ = ⟨24, −14, 18⟩ is a normal direction.
Hence, the tangent plane equation is 12(x − 2) − 7(y − 1) + 9(z − 3) = 0 ⇐⇒ 12x − 7y + 9z = 44 .

8. Let f (x, y) = x2 + xy 3 and P (2, 1) be a point on R2 . How many directions u such that Du f (2, 1) = 2 hold true?
Answer: (8) .

Solution: We first note that f is differentiable everywhere on R2 . Thus, Du f (2, 1) = ∇f (2, 1) · u for any unit vector
u = ⟨a, b⟩ ⇐⇒ a2 + b2 = 1. Compute ∇f (x, y) = ⟨2x + y 3 , 3xy 2 ⟩ =⇒ ∇f (2, 1) = ⟨5, 6⟩. Hence,
  2
2 = Du f (2, 1) = 5a + 6b ⇔ b = 1
− 5a 2 − 5a
3 6 =⇒ a2 + = 1 ⇐⇒ 61a2 −20a−32 = 0
a 2 + b2 = 1 6
 國立陽明交通大學 112 學年度微積分 II 課號：564011 測驗 ♯2 詳解

* √ √ !+ * √ √ !+
10 + 6 57 1 5 10 + 6 57 10 − 6 57 1 5 10 − 6 57
⇐⇒ u = , − or , − .
61 3 6 61 61 3 6 61

9. Find the direction in which the function f (x, y) = x2 y + exy sin y increases most rapidly at the point (1, 0).
Answer: (9) .

Solution:
∇f (x, y) = 2xy + yexy sin y, x2 + xexy sin y + exy cos y =⇒ ∇f (1, 0) = ⟨0, 2⟩ .

10. How many local maximum values does the function f (x, y) = 10xye−(x
2
+y 2 )
have? Answer: (10) .

Solution: Compute
∇f (x, y) = ⟨fx , fy ⟩ = 10e−(x
2
+y 2 )
⟨y − 2x2 y, x − 2xy 2 ⟩.
Then the critical points of f are (
y(1 − 2x2 ) = 0;
∇f (x, y) = ⟨0, 0⟩ ⇐⇒
x(1 − 2y 2 ) = 0.
By the first equation, we have two possibilities:

1. y = 0, then by the second equation, x = 0. So (0, 0) is a critical point of f in this case.

±1 ±1
2. y ̸= 0 =⇒ x = √ then by the second equation, y = √ .
2 2
       
1 1 −1 1 1 −1 −1 −1
So √ , √ , √ ,√ , √ ,√ , √ ,√ are the critical points of f in this case.
2 2 2 2 2 2 2 2

The discriminant of f is

fxx fxy −4xy − 2x(y − 2x2 y) 1 − 2x2 − 2y(y − 2x2 y)

= 100e−2(x
2
+y 2 )
Df (x, y) =
fyx fyy 1 − 2y 2 − 2x(x − 2xy 2 ) −4xy − 2y(x − 2xy 2 )
−6xy + 4x3 y 1 − 2x2 − 2y 2 + 4x2 y 2
= 100e−2(x
2
+y 2 )
.
1 − 2y 2 − 2x2 + 4x2 y 2 −6xy + 4xy 3

It then follows that

0 0
Df (0, 0) = 100 = 0 =⇒ (0, 0, 0) is a saddle point for the graph of f,
0 0

since in any disc of radius r centered at (0, 0)

r r 5    r r   −r r 
2 −r 2 /4 −r r 5 2 −r2 /4
f , = r e > 0 = f (0, 0) > f , =− r e and , , , lie in the disc.
2 2 2 2 2 2 2 2 2 2
   
1 1 −2 0 1 1
Df √ , √ = 100e−2 > 0 =⇒ f √ , √ = 5e−1 is a local maximum value of f.
2 2 0 −2 2 2
   
−1 1 2 0 −1 1
Df √ , √ = 100e−2 > 0 =⇒ f √ , √ = −5e−1 is a local minimum value of f.
2 2 0 2 2 2
   
1 −1 2 0 1 −1
Df √ , √ = 100e−2 > 0 =⇒ f √ , √ = −5e−1 is a local minimum value of f.
2 2 0 2 2 2
   
−1 −1 −2 0 −1 −1
Df √ , √ = 100e−2 > 0 =⇒ f √ , √ = 5e−1 is a local maximum value of f.
2 2 0 −2 2 2

11. The maximum value of the function f (x, y) = x2 y defined on the ellipse x2 + 2y 2 = 6 is (11) .
 Solution:


解 1. Using the method of Lagrange multiplier for F (x, y, λ) = x2 y + λ x2 + 2y 2 − 6 , we solve


Fx (x, y, λ) = 2xy + 2xλ = 0
Fy (x, y, λ) = x2 + 4yλ = 0 ⇐⇒ (x, y) = (2, 1), (2, −1), (−2, 1)or (−2, −1) .


Fλ (x, y, λ) = x + 2y − 6 = 0
2 2

=⇒ f (±2, 1) = 4 > f (±2, −1) = −4.

√ √ √
解 2. x2 + 2y 2 = 6 ⇐⇒ x2 = 6 − 2y 2 , so we try to maximize g(y) = (6 − 2y 2 )y for y ∈ [− 3, 3]. g(± 3) = 0,
g ′ (y) = 6 − 6y 2 = 0 ⇐⇒ y = ±1 =⇒ x = ±2.

=⇒ f (±2, 1) = 4 > f (±2, −1) = −4.

解 3. Using the inequality of arithmetic-geometric means,

x2 x2 r p
+ + 2y 2 2 x2 3
x4 y 2
2 2 3 x
2= ≥ · · 2y 2 = √ 3
⇐⇒ 16 ≥ x4 y 2 = (x2 y)2
3 2 2 2

with equality holds true when x2 = 4y 2 .

∂w
12. Find = (12) at the point (x, y, z) = (2, −1, 1) if w = x2 + y 2 + z 2 and z 3 − xy + yz + y 3 = 1.
∂x

Solution: First, check that the point (x, y, z) = (2, −1, 1) satisfies the equation z 3 − xy + yz + y 3 = 1. By the equation
z 3 − xy + yz + y 3 = 1, we may view z as a function of x and y. By implicit differentiation and differentiating with
respect to x,
y 1
z 3 − xy + yz + y 3 = 1 =⇒ 3z 2 zx − y + yzx = 0 =⇒ zx = =⇒ zx (2, −1) = − .
3z 2 + y 2
By chain rule,
∂w ∂w
= wx = 2x + 2zzx =⇒ (2, −1, 1) = 4 + 2zx (2, −1) = 3 .
∂x ∂x