MS3271906F1PV1

MS327/E

Module Examination 2019

Deterministic and stochastic dynamics

Wednesday 12 June 2019 10.00 am – 1.00 pm

Time allowed: 3 hours

Instructions
There are FOUR parts to this paper. In each part the questions are
arranged, as far as possible, in the order in which they appear in the
module.
Part 1 consists of 6 shorter questions each worth 10 marks, taken from the
entire module. The marks from your best four answers will be added
together to give a maximum of 40 marks for this part.
Parts 2, 3 and 4 consist of 2 longer questions each worth 20 marks from
Books 1, 2 and 3 respectively. The marks from your best answer to each
part will be added together to give a maximum of 60 marks for these parts.
To achieve full marks you must correctly answer four questions from
part 1, and one question from each of parts 2, 3 and 4.
Write your answers in pen in the answer book(s) provided. The marks
allocated to each part of each question are given in square brackets in the
margin. Unless you are directed otherwise in the question, you may use
any formula or other information from the Handbook in your answers. Do
not cross out any answers unless you have a better alternative answer –
everything not crossed out may receive credit.
At the end of the examination: Check that you have written your personal
identifier and examination number on each answer book used. Failure to
do so may mean that your work cannot be identified. Attach your signed
desk record to the front of your answer book(s) using the paper fastener.

Copyright 
c 2019 The Open University
Part 1
These questions cover the entire module and each question is worth
10 marks. To achieve full marks for this part you must answer FOUR
of the following six questions correctly.

Question 1
An asymmetric coupled oscillator is described by the differential
equations
ẍ1 = −2x1 + 3x2 ,
ẍ2 = 2x1 − 7x2 ,
where x1 and x2 denote the displacements of each mass from
equilibrium.
(a) Write down the coefficient matrix of the system and deduce its
eigenvalues. [4]
T T
Hint: v1 = (3 1) and v2 = (1 − 2) are eigenvectors of the
coefficient matrix.
(b) Write down the general solution describing the motion of the
system and identify any normal mode oscillations, stating their
angular frequencies and whether they are in-phase or
phase-opposed. [6]

Question 2
Find and classify all the fixed points of the following nonlinear system:
ẋ = x + y, ẏ = y 2 − 2x − 3. [10]

Question 3
Consider the functional
 2
2
S[y] = y  ey dx, y(1) = 2 ln 2, y(2) = 2 ln 3.
1

Use the first integral of the Euler-Lagrange equation to show that the
stationary path is given by
y(x) = 2 ln(x + 1). [10]

2 MS327 June 2019

Question 4
Consider the one-dimensional map xn+1 = f (xn ) where f (x) is defined
by

7x, x ≤ 12 ,
f (x) =
7 − 7x, x > 12 .

(a) Draw a graph of the function f (x) over the interval 0 ≤ x ≤ 1. [2]
(b) Consider the sets Sn of points which remain inside the interval
[0, 1] after n applications of the map. Each of these sets consists of
a finite number of intervals of equal length. Without giving a
detailed proof, list the intervals in S1 and S2 . How many intervals
comprise Sn , and how long is each of these intervals? [5]
(c) The set of points which never escape from the interval [0, 1] is a
Cantor set. What is the box-counting dimension of this Cantor
set? [3]

Question 5
Calculate the Fourier transform of
1
f (x) = (x − 2) e− 2 |x−2| . [10]

Question 6
The probability density function of a random walk in 0 ≤ x ≤ 1 is:
 πx 
−π 2 t
P (x, t) = αe cos ,
2
where α is a constant.
(a) Find α by imposing that P (x, t) is normalised at t = 0. [2]
(b) Find the diffusion constant, D, of this process by imposing that
P (x, t) satisfies the diffusion equation. [3]
(c) Find the first-passage probability density function of this random
walk and the mean exit time. [5]

MS327 June 2019 TURN OVER 3

Part 2
These questions cover the material in Book 1 and each question is
worth 20 marks. To achieve full marks for this part you must answer
ONE of the following two questions correctly.

Question 7
Consider the function

et for − π ≤ t < 0,
f (t) = −t
(1)
e for 0 ≤ t < π,
f (t) = f (t + 2π).
(a) State the fundamental period and sketch the function over the
interval −3π < t < 3π. [4]
(b) Show that the exponential Fourier series of this function is
∞
1 − (−1)n e−π int
f (t) = 2 + 1)
e . [8]
n=−∞
π(n

(c) A particle with position x(t) undergoes driven and damped

harmonic motion and is described by the differential equation
d2 x dx
2
+ 2Γ + 4x = f (t)
dt dt
where Γ > 0 is the damping parameter and the periodic driving
term f (t) is given by equation (1). Find the Fourier series of a
particular solution to this motion.
Describe the nature of the motion, as the damping Γ gets smaller
(Γ → 0), and in particular state which angular frequency will
dominate the motion. [8]

4 MS327 June 2019

Question 8
Consider the continuous random walk described by the recurrence
relation
xn+1 = xn + rn , x0 = 0,
where xn is the position of the walker at step n and rn are independent
continuous random variables each with the same probability density
function as the random variable r which is given by
⎧
⎪
⎨a for − 3 ≤ r ≤ −2,
ρ(r) = 2a(2 − r) for 1 ≤ r ≤ 2,
⎪
⎩
0 otherwise,
where a is a constant.
(a) Show that the probability that rn > 0 is equal to that of rn < 0
and use normalization to determine the value of a. [3]
(b) Show that the probability that |rn | > 3/2 is 5/8. [4]
(c) Find the mean, r, and variance, Var(r), of the random variable r. [8]
(d) Let X be the position of the random walker at step n = N.
Determine the mean and variance of X in terms of N and show
that, for large N, the probability density function for X is well
approximated by
12 (12X + 7N)2
ρN (X)  √ exp − . [5]
7 22πN 1078N

MS327 June 2019 TURN OVER 5

Part 3
These questions cover the material in Book 2 and each question is
worth 20 marks. To achieve full marks for this part you must answer
ONE of the following two questions correctly.

Question 9
A bead of mass m moves on a wire with equation z = 12 r 2 , where z
measures the height of the bead and r is the horizontal distance in the
plane of the wire.
The plane in which the wire lies is free to rotate about a fixed vertical
axis passing through x = 0, so that its angle relative to the xz plane
is θ.

θ
x

(a) At any point of the motion the bead has Cartesian position
(x, y, 12 (x2 + y 2 )). Express this position in terms of polar
coordinates (r, θ) where x = r cos θ and y = r sin θ. [1]
(b) Calculate the velocities ẋ, ẏ and ż in terms of the polar coordinates
and their derivatives. [3]
(c) Show that the Lagrangian of the system, in terms of the polar
coordinates and their derivatives, is given by

L = 12 m (1 + r 2 )ṙ 2 + r 2 θ̇2 − 12 mgr 2 . [4]

(d) Identify any ignorable coordinates and state which physical

quantities will be conserved in the system. [3]
(e) Use the Euler-Lagrange equation to deduce the equations of motion
for r and θ. [4]
(f) Show that the equation for θ can be integrated to give r 2 θ̇ = K,
where K is a constant, and state (without proof) the physical
meaning of this quantity. [2]
(g) Deduce an equation of motion for r which does not contain θ̇ and
show that it has a solution r(t) = R, where R is a constant.
Deduce the value for R in terms of K and g. [3]

6 MS327 June 2019

Question 10
(a) Consider the following two-dimensional map:
xn+1 = xn + yn ,
yn+1 = yn + axn (1 − xn ) ,
where a is a constant with 0 < a < 1.
(i) Determine the Jacobian matrix of the map at an arbitrary
point (x, y). Is this map area-preserving? [4]
(ii) Determine all the fixed points of this map. [4]
(iii) Classify all the fixed points. [4]
(b) A closely related map is
xn+1 = xn + yn+1 ,
yn+1 = yn + axn (1 − xn ) ,
also with the constant a having the same restriction 0 < a < 1.
Is this map area-preserving? Does it share any fixed points with
the map considered in part (a)? If so, classify them. [8]

MS327 June 2019 TURN OVER 7

Part 4
These questions cover the material in Book 3 and each question is
worth 20 marks. To achieve full marks for this part you must answer
ONE of the following two questions correctly.

Question 11
The concentration of particles that diffuse in a three-dimensional
spherical region depends only on the distance, r, from the origin, and at
time t > 0 is given by:
2 /t
c(r, t) = f (t)e−r ,
for some function f (t). Assume that the diffusion coefficient is D.
(a) Use Fick’s law to find an expression for the flux J. [4]
(b) Using Gauss’ theorem show that

8πDR3 2
∇ · J dV = f (t)e−R /t ,
V t
where V is a sphere of radius R centered at the origin. [6]
(c) If f (t) = t−3/2 , is there any value of D for which the continuity
equation for the concentration of particles is satisfied? [10]

8 MS327 June 2019

Question 12
The temperature distribution Θ(x, t) along an insulated metal rod of
length L is described by the differential equation
∂2Θ 1 ∂Θ
= (0 < x < L, t > 0),
∂x2 D ∂t
where D = 0 is a constant. The rod is held at a fixed temperature of
0◦ C at one end and is insulated at the other end, which gives rise to
the boundary conditions Θ(0, t) = 0 and Θx (L, t) = 0, for t > 0.
The initial temperature distribution in the rod is given by
7πx
Θ(x, 0) = 0.5 sin (0 ≤ x ≤ L).
2L
(a) Use the method of separation of variables, with
Θ(x, t) = X(x) T (t), to show that the function X(x) satisfies the
differential equation
X  − μX = 0 (2)
for some constant μ. Write down the corresponding differential
equation that T (t) must satisfy. [4]
(b) Write down the boundary conditions that X(x) must satisfy.
Consider the functions
(2n − 1)π
Xn (x) = sin(kn x), where kn = and n = 1, 2, 3, . . . .
2L
Show that function Xn (x) satisfies the boundary conditions you
found. Show that Xn (x) satisfies differential equation (2) for some
constant μ (which you should specify). [6]
(c) Showing your working, solve the differential equation found in
part (a) that the function T (t) must satisfy. [4]
(d) Use your answers to write down a family of product solutions
Θn (x, t) = Xn (x) Tn (t) that satisfy the first two boundary
conditions. Hence show that the general solution of the partial
differential equation may be expressed as

∞
D(2n − 1)2 π 2 t (2n − 1)πx
Θ(x, t) = Cn exp − sin . [2]
n=1
4L2 2L

(e) Find the particular solution that satisfies the given initial
temperature distribution. [4]

[END OF QUESTION PAPER]

MS327 June 2019 9