UNIVERSITY OF BRISTOL

School of Mathematics

APPLIED PARTIAL DIFFERENTIAL EQUATIONS 2

MATH–20402 (Paper code MATH–20402)

Summer 2020 2 hours 30 minutes

This paper contains FOUR questions

All FOUR answers will be used for assessment.
Calculators are not permitted in this examination.

On this examination, the marking scheme is indicative and is intended only as a guide to the
relative weighting of the questions.

Do not turn over until instructed.

Page 1 of 5
1. Consider a uniform string of length `. Its vertical displacement u = u(x, t) above its
equilibrium at position x and time t obeys the equation

utt = c2 uxx , 0 < x < ` , t > 0 ,

where c > 0 is a constant.

(a) (3 marks) What is the physical interpretation of the constant c ?

(b) (4 marks) Show that, for every pair of sufficiently smooth functions F and G of one
variable,
u(x, t) = F (x + ct) + G(x − ct) (∗)
solves the partial differential equation.
(c) For the remainder of the question, it will be assumed that the string is tied down at
the ends, so that
u(0, t) = u(`, t) = 0 , t > 0 .
The string is pulled upward at the middle, raised to a height h and then released from
rest so that (
2h
`
x if 0 < x ≤ `/2
u(x, 0) = u0 (x) := 2h
`
(` − x) if `/2 ≤ x < `
and
ut (x, 0) = 0 , 0 < x < ` .
i. (8 marks) Compute the coefficients bn in the expansion
∞
X  π 
u0 (x) = bn sin n x .
n=1
`

ii. (5 marks) Hence express the vertical displacement u(x, t) of the string, for 0 <
x < ` and t > 0, as a sine series in x, with t-dependent coefficients.
iii. (5 marks) Find F and G such that the vertical displacement is given, for 0 <
x < ` and t > 0, by Equation (∗).

Page 2 of 5
2. (a) (5 marks ) What does it mean to say that a function u = u(x, y) is harmonic in some
open subset of R2 ? Give an example of a non-constant function harmonic in the unit
disk.
(b) (5 marks) In the polar coordinates
p y
2 2
r := x + y and θ := arctan ,
x
Laplace’s equation on the unit disk for the unknown function u = u(r, θ) takes the
form
∂ 2 u 1 ∂u 1 ∂ 2u
+ + = 0 , 0 < r < 1 , 0 ≤ θ < 2π .
∂r2 r ∂r r2 ∂θ2
Find the particular solution that satisfies the boundary condition

u(1, θ) = cos θ , 0 ≤ θ < 2π .

(c) The coordinates (r, θ, z), where r and θ are the polar coordinates in the xy–plane, and
z is the usual Cartesian coordinate, are called the cylindrical coordinates. In these
coordinates, the Laplacian takes the form

∂2 1 ∂ 1 ∂2 ∂2
∆= + + + .
∂r2 r ∂r r2 ∂θ2 ∂z 2
Let
Ω := (x, y, z) ∈ R3 : x2 + y 2 < 1, 0 < z < 1


be an open bounded cylinder in three-dimensional space and consider the following

eigenvalue problem: Find λ ∈ R such that there exists a non-zero solution u =
u(r, θ, z) of
∆u = λu , 0 < r < 1 , 0 ≤ θ < 2π , 0 < z < 1 , (†)
subject to homogeneous Dirichlet boundary conditions on ∂Ω.
i. (5 marks) Suppose that Equation (†) admits a solution of the form

u(r, θ, z) = R(r) Θ(θ) Z(z) .

Show that such solutions may be found only if there exist constants µ and ν such
that
Z 00 = µZ , 0 < z < 1 ,
Θ00 = νΘ , 0 ≤ θ < 2π .
ii. (5 marks) Show that, for the corresponding solution u to be non-zero, smooth
and satisfy the boundary condition, µ and ν cannot be arbitrary but must take
very particular values; find all these values and then solve the equation for Θ and
the equation for Z for these values.
iii. (5 marks) Write down the resulting equation for R. What auxiliary conditions
should its solution satisfy? (You need not solve the equation for R.)

Continued...

Page 3 of 5
3. The Fourier transform fb(k) of a function f (x) is defined by
Z ∞
fb(k) ≡ F{f } = f (x)e−ikx dx.
−∞

(a) i. (2 marks) Given fb(k), write down the formula (the inverse Fourier transform)
for f (x).
ii. (2 marks) Show that fb0 (k) = ik fˆ(k).
iii. (4 marks) Show that if g(x) = f (x + a), then ĝ(k) = eika fˆ(k).
(b) (7 marks) Show that F{δ} = 1, where δ(x) is the δ-function. Z ∞ Show that the inverse
2
formula F −1 {1} = δ is correct by calculating the limit of eikx−k dk as  → 0; you
Z ∞ −∞
2 √
are given that e−x dx = π.
−∞
(c) (5 marks) Now consider the PDE

ut + cux = −αu, −∞ < x < ∞, t > 0

for constants c and α, with initial condition

u(x, 0) = f (x), −∞ < x < ∞, f (x) → 0, as x → ±∞

Take Fourier transforms in x to show that

b(k, t) = fb(k)e−ikct−αt .
u

(d) (5 marks) By taking an inverse transform, show that

u(x, t) = e−αt f (x − ct).

Continued...

Page 4 of 5
4. (a) (6 marks)
Say, giving reasons, whether each of the following equations is linear, semilinear,
quasilinear, or none of those types.
i. ut − u2 ux = 0
ii. vt + xvx = ev
iii. zx + zzy2 = 0
(b) (6 marks) Consider the initial value problem

ut + g(u)ux = 0, u(x, 0) = f (x). (1)

Show, using the method of characteristics, that if u satisfies the equation

u(x, t) = f (x − g(u)t), (2)

then it is a solution of (1).

(c) (6 marks) Consider the initial condition

0 for |x| ≥ 1
f (x) = .
1 − x2 for|x| < 1

For g(u) = 1 + u, give a solution of (1) in parametric form. Calculate the time ts and
position xs where a shock first forms.
(d) (7 marks) Now consider the equation

ut + (1 + u)ux = u2 , u(x, 0) = f (x),

with the same initial condition as before. Calculate the time ts where a shock first
forms.

End of examination.

Page 5 of 5