11/1/2017 Second Hourly Practice 6 Math 21a, Fall 2017

Name:

• Start by printing your name in the above box

MWF 9 Jameel Al-Aidroos and check your section in the box to the
left.
MWF 9 Dennis Tseng
• Do not detach pages from this exam packet
MWF 10 Yu-Wei Fan
or unstaple the packet.
MWF 10 Koji Shimizu
• Please write neatly. Answers which are illeg-
MWF 11 Oliver Knill ible for the grader cannot be given credit.
MWF 11 Chenglong Yu • Show your work. Except for problems 1-
MWF 12 Stepan Paul 3,8, we need to see details of your computa-
tion.
TTH 10 Matt Demers
• All functions can be differentiated arbitrarily
TTH 10 Jun-Hou Fung
often unless otherwise specified.
TTH 10 Peter Smillie
• No notes, books, calculators, computers, or
TTH 11:30 Aukosh Jagannath other electronic aids can be allowed.
TTH 11:30 Sebastian Vasey • You have 90 minutes time to complete your
work.

1 20

2 10

3 10

4 10

5 10

6 10

7 10

8 10

9 10

10 10

Total: 110
Problem 1) True/False questions (20 points), no justifications needed

There is a function f (x, y) for which the linearization at (0, 0) is L(x, y) =

1) T F x2 + y 2.

Solution:
The linearization is a linear function and not quadratic.

For any two functions f, g and unit vector ~u we have

2) T F D~u (f + g) = D~u f + D~u g.

Solution:
This follows directly from the definition.

R 2 R √4−x2 R 2 R π/2
3) T F 0 0 (x2 + y 2 ) dydx = 0 0 r 2 dθdr.

Solution:
The r factor is forgotten.

4) T F If we solve sin(y) − xy 2 = 0 for y, then y ′ = −y 2 /(cos(y) − 2xy).

Solution:
This is an application of the implicit differentiation formula. But the sign is wrong.

If f (x, 0) R= 0R for all x and f (0, y) = 0 for all y, then

5) T F g(x, y) = 0x 0y f (s, t) dtds solves gxy (x, y) = f (x, y).

Solution:
Fundamental theorem of calculus. The assumption f (x, 0) = 0 and f (0, y) = 0 is not
necessary.

If |∇f | = 1 at (0, 0), then there exists a direction in which the slope of the
6) T F graph of f at (0, 0) is 1.
 Solution:
It is the direction of the gradient

The function f (x, y) = x2 + y 2 satisfies the partial differential equation

7) T F 2
fxx fyy − fxy = 4.

Solution:
Yes, this is a computation.

The height of Mount Wachusett is f (x, y) = 4 − 2x2 − y 2. On the trail

8) T F x2 + y 2 = 1, the point (1, 0) is a maximum.

Solution:
It is a local minimum.

Mount Wachusett has height f (x, y) = 4−2x2 −y 2 . Except at the maximum

9) T F (0, 0), the gradient vector is perpendicular to the graph of the function.

Solution:
The gradient vector is a vector with two components, not a vector in space.

Solution:
The gradient vector of f is a vector with two components and not in space.

If fx (a, b) > 0 and fy (a, b) > 0 then for any unit vector ~u we must have
10) T F D~u f (a, b) > 0.

Solution:
Take a unit vector h−1, 0i for example. The directional derivative in this direction is
zero.

If f (x, y) has two local minima, then f must have at least one local maxi-
11) T F mum.
 Solution:
Take a function like −x2 exp(−x2 − y 2). It has two local minima and a saddle point.

If ~r(t) is a curve on the surface g(x, y, z) = x2 + y 2 − z 2 = 6 then

12) T F ′
∇g(~r(t)) · ~r (t) = 0.

Solution:
This is a recurring theme. The gradient vector is perpendicular to the surface. This fact
is based on the chain rule and the fact that g(r(t)) is constant so that d/dtg(r(t)) is zero.

13) T F If f and g have the same trace {x = 5} then fx (5, y) = gx (5, y) for all y.

Solution:
We know f (5, y) = g(5, y) but the x derivative can be different. Take f (x, y) = x − 5 and
g(x, y) = y(x − 5) then fx = 1 and gx = y.

14) T F If f and g have the same trace {x = 5} then fy (5, y) = gy (5, y) for all y.

Solution:
Because f (5, y) = g(5, y), also the derivatives are the same.

2
The
√ surface√ area of ~r1 (u, v) = hu cos(v), u sin(v), u i and ~r2 (u, v) =
15) T F h u cos(v), u sin(v), ui defined on {0 ≤ u, v ≤ 1 } are the same.

Solution:
Surface area does not depend on parametrization.

If ~r(t) is a curve on a graph z = f (x, y) of a function f (x, y), then the

16) T F velocity vector of ~r is perpendicular to the vector hfx , fy , −1i.

Solution:
The vector hfx , fy , −1i is the gradient of the function g(x, y, z) = f (x, y) − z which has
as a level surface the graph g = 0 of f .
 A continuous function f (x, y) on the closed disc R = {x2 + y 2 ≤ 512 }
17) T F (of course, R is called “area 51π”) has a global maximum on R.

Solution:
We know that a continuous function has a maximum on a closed bounded domain. Inter-
esting corollary: take for f the probability density that an alien has landed there. Then
there is a point where the probability density is largest, proving so that aliens are likely
in area 51.

Any continuous function f (x, y) has a global minimum and maximum on

18) T F the curve y = x2 .

Solution:
The curve is an unbounded parabola. The function f (x, y) = x for example is unbounded
on it.

Rb Rd Rb Rd
19) T F Fubini’s theorem assures that a c f (x, y) dydx = a c f (x, y) dxdy.

Solution:
The integrals have not been switched.

RR
20) T F R sin(x + y) dxdy = 0 for R = {−π ≤ x ≤ π, −π ≤ y ≤ π }.

Solution:
Directly integrate. One can see this also by symmetry. The integral has an interpretation
as a volume, with exactly the same amount below than above the plane.

Problem 2) (10 points)

a) (6 points) Match the integration regions with the integrals. Each integral matches ex-
actly one region A − F .
A B C

D E F

Enter A-F Integral

b) (4 points) Fill in one word names (like
R 1 R x2 “Heat”,“Wave” etc) for the partial differen-
−1 −x2 f (x, y) dydx.
tial equations:
R 1 R y2
−1 −y 2 f (x, y) dxdy. Enter one word PDE
R 1 R 1−y2 gx = gy
−1 y 2 −1 f (x, y) dxdy.
√ gxx = gyy
R1 R 1−y 2
−1 0 f (x, y) dxdy.
gxx = −gyy
R 1 R 1−x2
−1 x2 −1 f (x, y) dydx.
gx = gyy
R1 R0
√ f (x, y) dydx.
−1 − 1−x2

Solution:
a) C,D,E,B,A,F
b) Transport, Wave, Laplace, Heat.
Problem 3) (10 points)
(10 points) A function f (x, y) of two variables has level curves as shown in the picture.
We also see a constraint in the form of a curve g(x, y) = 0 which has the shape of the
graph of the cos function. The arrows show the gradient. In this problem, each of the 10
letters A, B, C, D, E, F, G, H, K, M appears exactly once.

Enter A-P Description

a local maximum of f (x, y).
a local minimum of f (x, y).
a saddle point of f (x, y) where fxx < 0.
a saddle point of f (x, y) where fxx > 0.
a saddle point of f (x, y) where fxx is close to zero
a point, where fx = 0 and fy 6= 0
a point, where fy = 0 and fx 6= 0
the point, where |∇f | is largest
a local maximum of f (x, y) under the constraint g(x, y) = 0.
a local minimum of f (x, y) under the constraint g(x, y) = 0.

K
G
A D
B
F
H C

E M
Solution:
D,C,B,H,M,E,A,K,G,F.

Problem 4) (10 points)

Find and classify all the extrema of the function f (x, y) = x5 + y 3 − 5x − 3y. This function
measures “eat temptation” in the x=Easy-y=Tasty plane. Is there a global minimum or
global maximum?

The “Easy-Tasty plane” was introduced in the XKCD cartoon titled

“F&#% Grapefruits”.

Solution:
Find the critical points ∇f (x, y) = h0, 0i, then find the discriminant D and fxx .

ne]hline point D fxx type

(−1, −1) 120 -20 maximum
(−1, 1) -120 -20 saddle
(1, −1) -120 20 saddle
(1, 1) 120 20 minimum
Problem 5) (10 points)

After having watched the latest Disney movie “Tangled”, we want to build a hot air
balloon with a cuboid mesh of dimension x, y, z which together with the top and bottom
fortifications uses wires of total length g(x, y, z) = 6x + 6y + 4z = 32. Find the balloon
with maximal volume f (x, y, z) = xyz.

Solution:
The gradients are ∇f (x, y, z) = hyz, xz, xyi and ∇g(x, y, z) = h6, 6, 4i. The Lagrange
equations are
yz = λ · 6
xz = λ · 6
xy = λ · 4

Getting rid of λ and using that x, y, z and so λ must all be positive for having
a positive volume gives x = y = 2z/3. Plugging this into the constraints gives
x = 16/9, y = 16/9, z = 8/3 .

Problem 6) (10 points)

a) (8 points) Find the tangent plane to the surface f (x, y, z) = x2 − y 2 + z = 6 at the point
(2, 1, 3).

b) (2 points) A curve ~r(t) on that tangent plane of the function f (x, y, z) in a) has constant
speed |~r ′ | = 1 and passes through the point (2, 1, 3) at t = 0. What is dtd f (~r(t)) at t = 0?
Solution:
a) The gradient is ∇f (x, y, z) = h2x, −2y, 1i. At the point (2, 1, 3) this is h4, −2, 3i. The
plane has the form 4x − 2y + z = d where d can now be obtained by plugging in the point,
which is 8 − 2 + 3 = 9. The plane is 4x − 2y + z = 9.
b) Since the curve is on the surface and f does not change, we have d/dtf (~r(t)) = 0. We
can also see it from the fact that ~r′ (t) is perpendicular to the gradient. The answer is
0.

Problem 7) (10 points)

q
a) (5 points) Estimate sin(0.0004) + 1.0012 using linear approximation.

b) (5 points) We know f (0, 0) = 1, Dh 3 , 4 i f (0, 0) = 2 and Dh− 4 , 3 i f (0, 0) = −1. If L(x, y)

5 5 5 5
is the linear approximation to f (x, y) at the point (0, 0), find L(0.06, 0.08).

Solution: q q
a) fx = cos(x)/(2 sin(x) + y 2 ) and fy = 2y/(2 sin(x) + y 2) at the point (0, 1) we get
∇f (0, 1) = h1/2, 1i. The linearization is
1
L(x, y) = 1 + · 0.0004 + 1 · 0.001 = 1 + 0.0002 + 0.001 = 1.0012 .
2
The answer is 1.0012 .
b) Write ∇f = ha, bi. We get a · (3/5) + b · (4/5) = 2, (−4/5) · a + (3/5) · b = −1. We can
solve for a, b to get a = 2, b = 1. We have therefore the linear approximation
L(x, y) = f (0, 0) + 2 · 0.06 + 1 · 0.08 = 1.2 .

The answer is 1.2 .

Problem 8) (10 points)

a) (5 points) Find the following double integral

√
1 x π sin(πy)
Z Z
√ dydx .
0 x2 y2 − y

b) (5 points) Evaluate the following double integral

√
sin(π x2 + y 2 )
Z Z
√ 2 dxdy
R x + y2
over the region
R = {x2 + y 2 ≤ 1, x > 0 } .

Solution:
a) Make a picture and change the order of integration.
√
1 y π sin(πy) 1
Z Z Z
√ dxdy = − sin(πy) dy = −2 .
0 y2 (y 2 − y) 0

The answer is −2 .
b) We use polar coordinates:
1 π/2 sin(πr)
Z Z
r dθ dr = 2π/π = 2 .
0 −π/2 r

The answer is 2 .
Remark. In a), Mathematica 8 can not compute the integral. We have to help it to
change the order of integration. This example shows that Mathematica does not have the
change of order of integration trick implemented.

Problem 9) (10 points)

a) (8 points) Find the surface area of the surface parametrized as

~r(u, v) = hu − v, u + v, (u2 − v 2 )/2i ,

where (u, v) is in the unit disc R = {u2 + v 2 ≤ 1 }.

b) (2 points) Give a nonzero vector ~n normal to the surface at ~r(4, 2) = h2, 6, 6i.
Solution:
a) We have ru = h1, √ 1, ui
√ and rv = h−1, 1, −vi so that ru × rv = h−u − v, v − u, 2i. Its
length is |ru × rv | = 2 u2 + v 2 + 2. Use polar coordinates to integrate this over R:
Z Z 2π Z 1 q
|ru × rv | dudv = 2(r 2 + 2)r drdθ .
R 0 0

Use substitution u = r 2 + 2 to solve the inner integral:

Z 2π √
2 2
(r + 2)3/2 |10 dθ .
0 3
√
Evaluate this to get the final answer (π/3)(6 6 − 8) .
b) We have computed ~ru × ~rv already in a). Just take (u, v) = (4, 2) to get h−6, −2, 2i .
Of course any vector parallel to this is also correct.

Problem 10) (10 points)

a) (6 points) Integrate
π/2 π/2 cos(y)
Z Z
dydx
0 x y
b) (4 points) Find the moment of inertia
ZZ
(x2 + y 2) dydx ,
R

where R is the ring 1 ≤ x2 + y 2 ≤ 9.

Solution:
a) To change the order of integration, make a figure. The integration region is the upper
left triangle in the square [0, π/2] × [0, π/2]. We get
Z π/2 Z y Z π/2
π/2
cos(y)/y dxdy = cos(y) dy = sin(y)0 =1.
0 0 0

The answer is 1 .
b) This is a polar integration problem
Z Z 3 Z 2π r4 3 Z 2π
02π r 2 r dr dθ = |1 dθ = 20 dθ = 40π .
0 1 0 4 0

The answer is 40π .