0% found this document useful (0 votes)
51 views

$R3MB5VL

This document provides solutions to exercises from a practice midterm exam in math 425. The first exercise involves solving a boundary value problem using comparison principles and showing that the solution converges to another function as time approaches infinity. The second exercise solves the same boundary value problem using separation of variables and shows uniqueness. The third exercise discusses properties of harmonic functions in three dimensions using the mean value property and integrability conditions.

Uploaded by

husseinmarwan589
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
51 views

$R3MB5VL

This document provides solutions to exercises from a practice midterm exam in math 425. The first exercise involves solving a boundary value problem using comparison principles and showing that the solution converges to another function as time approaches infinity. The second exercise solves the same boundary value problem using separation of variables and shows uniqueness. The third exercise discusses properties of harmonic functions in three dimensions using the mean value property and integrability conditions.

Uploaded by

husseinmarwan589
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

MATH 425, PRACTICE MIDTERM EXAM 2, SOLUTIONS.

Exercise 1. Suppose that u solves the boundary value problem:



ut (x, t) − uxx (x, t) = 1, for 0 < x < 1, t > 0

(1) u(x, 0) = 0, for 0 ≤ x ≤ 1

u(0, t) = u(1, t) = 0, for t > 0.

a) Find a function v = v(x) which solves:


(
−vxx (x) = 1, for 0 < x < 1
v(0) = v(1) = 0.
b) Show that:
u(x, t) ≤ v(x)
for all x ∈ [0, 1], t > 0.

c) Show that:
u(x, t) ≥ (1 − e−2t )v(x)
for all x ∈ [0, 1], t > 0.

d) Deduce that, for all x ∈ [0, 1]:


u(x, t) → v(x)
as t → ∞.
Solution:

a) We need to solve v 00 (x) = −1 with boundary conditions v(0) = v(1) = 0. The ODE implies
that v(x) = − 21 x2 + Ax + B for some constants A, B. We get the system of linear equations:
(
B=0
− 12 + A + B = 0
from where it follows that:
1
A= and B = 0.
2
Hence:
1
v(x) = x · (1 − x).
2
b) Let us now think of v as a function of v as a function of (x, t) which doesn’t depend on x. By
construction, we know that:

vt (x, t) − vxx (x, t) = 1, for 0 < x < 1, t > 0

v(x, 0) ≥ 0, for 0 ≤ x ≤ 1

v(0, t) = v(1, t) = 0, for t > 0.

Here, we used the fact that 21 x · (1 − x) ≥ 0 for 0 ≤ x ≤ 1. By using the Comparison principle for
the heat equation (Exercise 3 on Homework Assignment 4), it follows that:
u(x, t) ≤ v(x, t) = v(x)
1
2 MATH 425, PRACTICE MIDTERM EXAM 2, SOLUTIONS.

for all x ∈ [0, 1], t > 0.

c) Let us define:
1
w(x, t) := (1 − e−2t )v(x) = · (1 − e−2t ) · x(1 − x)
2
We compute:
wt (x, t) = e−2t · x(1 − x)
wxx (x, t) = −(1 − e−2t ) = −1 + e−2t .
Hence:  
wt (x, t) − wxx (x, t) = 1 − e−2t 1 − x(1 − x) .
We know that for x ∈ [0, 1], one has: x(1 − x) ∈ [0, 1]. Hence, it follows that:
wt (x, t) − wxx (x, t) ≤ 1
for all 0 ≤ x ≤ 1, t > 0. In particular, we deduce that:

wt (x, t) − wxx (x, t) = 1, for 0 < x < 1, t > 0

w(x, 0) = 0, for 0 ≤ x ≤ 1

w(0, t) = w(1, t) = 0, for t > 0.

By using the comparison principle, it follows that, for all x ∈ [0, 1], t > 0, the following holds:
1
· (1 − e−2t ) · x(1 − x) = (1 − e−2t )v(x).
u(x, t) ≥ w(x, t) =
2
d) Combining the results of parts b) and c), it follows that, for all x ∈ [0, 1], t > 0, it holds that:
(1 − e−2t )v(x) ≤ u(x, t) ≤ v(x).
Letting t → ∞, it follows that:
u(x, t) → v(x)
as t → ∞. 
Exercise 2. a) Find the function u solving (1) of the previous exercise by using separation of
variables. Leave the Fourier coefficients in the form of an integral. [HINT: Consider the function
w := u − v for u, v as in the previous exercise.]

b) Show that this is the unique solution of the problem (1).

c) By using the formula from part a), give an alternative proof of the fact that u(x, t) → v(x)
as t → ∞. In this part, one is allowed to assume that the Fourier coefficients at time zero are
absolutely summable without proof.
Solution:

a) Let ũ(x, t) := u(x, t) − 21 x(1 − x). Then the function ũ solves:



ũt (x, t) − ũxx (x, t) = 0, for 0 < x < 1, t > 0

ũ(x, 0) = − 21 x(1 − x), for 0 ≤ x ≤ 1

ũ(0, t) = ũ(1, t) = 0, for t > 0.

We look for ũ in the form of a Fourier sine series with coefficients which depend on t.

X
ũ(x, t) = An (t) sin(nπx).
n=1
MATH 425, PRACTICE MIDTERM EXAM 2, SOLUTIONS. 3

We first set t = 0 to deduce that:



1 X 1
ũ(x, 0) = − x(1 − x) = An (0) sin(nπx) = − x(1 − x).
2 n=1
2

Hence, An (0) equals the n-th Fourier sine series coefficient of the function − 12 x(1 − x) on [0, 1]. In
particular,
Z 1
1 
An (0) = 2 − x(1 − x) sin(nπx) dx.
0 2
In order for ũ to solve the heat equation, we need:
A0n (t) − n2 π 2 An (t) = 0.
Hence:
2
π2 t
An (t) = An (0) · e−n .
Consequently:

2
π2 t
X
ũ(x, t) = An (0) · e−n · sin(nπx).
n=1
We then deduce that:

1 X 2 2
u(x, t) = x(1 − x) + An (0) · e−n π t · sin(nπx).
2 n=1

b) Uniqueness of the problem (1) was shown in class by using the maximum principle and by using
the energy method.

c) We note that:
∞ ∞ ∞
2
π2 t 2
π2 t 2
X X X
|u(x, t) − v(x)| = An (0) · e−n · sin(nπx) ≤ |An (0)| · e−n ≤ e−π t · |An (0)|.
n=1 n=1 n=1
1
As is noted in the problem, we are allowed to assume that

X
|An (0)| < ∞.
n=1

The claim now follows. 


Exercise 3. Suppose that u : R3 → R is a harmonic function.

a) By using the Mean Value Property (in terms of averages over spheres), show that, for all x ∈ R3 ,
and for all R > 0, one has:
Z
3
u(x) = u(y) dy.
4πR3 B(x,R)

R
b) Suppose, moreover, that R3
|u(y)| dy < ∞. Show that then, one necessarily obtains:
u(x) = 0
for all x ∈ R3 .

1We can integrate by parts twice in the definition of A (0) and use the fact that − 1 x(1 − x) vanishes at x = 0
n
P 2
and x = 1 in order to deduce that: |An (0)| ≤ nC2 from where it indeed follows that ∞n=1 |An (0)| < ∞.
4 MATH 425, PRACTICE MIDTERM EXAM 2, SOLUTIONS.

Solution:

a) Let us fix x ∈ R3 . The Mean Value Property, proved in Exercise 1 of Homework Assignment 7,
implies that, for all r > 0:
Z
1
(2) u(x) = u(y) dS(y).
4πr2 ∂B(x,r)
We note that:
Z Z R Z
3 3 
u(y) dS(y) = u(y) dS(y) dr.
4πR3 B(x,R) 4πR3 0 ∂B(x,r)

By the Mean Value Property (2), it follows that this expression equals:
Z R Z R
3 2 3
4πr u(x) dr = u(x) · · 4πr2 dr = u(x).
4πR3 0 4πR3 0
b) We note that, by part a), it follows that:
Z Z
3 3
|u(x)| ≤ |u(y)| dy ≤ |u(y)| dy.
4πR3 B(x,R) 4πR3 R3
R
Since R3 |u(y)| dy < ∞, we can let R → ∞ to deduce that |u(x)| = 0. It follows that u is identically
equal to zero. 
Exercise 4. Suppose that u : B(0, 2) → R is a harmonic function on the open ball B(0, 2) ⊆ R2 ,
which is continuous on its closure B(0, 2). Suppose that, in polar coordinates:
u(2, θ) = 3 sin 5θ + 1
for all θ ∈ [0, 2π].
a) Find the maximum and minimum value of u in B(0, 2) without explicitly solving the Laplace
equation.

b) Calculate u(0) without explicitly solving the Laplace equation.


Solution:

a) By using the weak maximum principle for solutions to the Laplace equation, we know that
the maximum of the function u on B(0, 2) is achieved on ∂B(0, 2). We observe that the function
u(2, θ) = 3 sin 5θ + 1 takes values in [−2, 4]. It equals −2 when sin 5θ = −1, which happens at
θ = 3π π
10 (for example). Moreover u(2, θ) = 4 when sin 5θ = 1, which happens at θ = 10 (for exam-
ple). Hence, the maximum value of u on B(0, 2) is 4 and the minimum value of u on B(0, 2) is
−2.

b) We use the Mean Value Property to deduce that u(0) equals the average of u over the circle
∂B(0, 2). Since the average of the 3 sin 5θ term equals zero, it follows that u(0) = 1. 

You might also like