ME 563 - Intermediate Fluid Dynamics - Su: Lecture 28 - Waves: The Basics

ME 563 - Intermediate Fluid Dynamics - Su
Lecture 28 - Waves: the basics
Reading: Acheson, §3.1-3.3.
We’re going to go back in the book briefly to touch on the subject of waves. The topic of waves is
very extensive and covers surface waves, interfacial waves, sound or pressure waves, shocks, solitons
and a variety of other phenomena that fill textbooks all by themselves. Our goal is just to cover
some fundamental ideas about waves, both to get a feel for the subject, and to learn enough to
look at the area of flow instability.
We will consider a simple harmonic surface wave, in which the position of the surface is defined
by
y = η(x, t) = A cos(kx − ωt) = η(θ), (1)
(A is a constant), where we define
θ(x, t) = kx − ωt.
The driving force behind the wave motion will be gravity. Sinusoidal waves are interesting because
Fourier theory tells us that we can decompose any function into sinusoidal components of different
wavelengths. So individual waves defined by (1) are the building blocks of general waves and
disturbances.
To determine the speed, c, of the wave, we plant ourselves on one particular point on the wave,
i.e. fixed θ. From time t to t + ∆t, that point on the wave will have moved from spatial position x
to x + ∆x, where x, t, ∆x and ∆t are related by
θ(x, t) = θ(x + ∆x, t + ∆t) = k(x + ∆x) − ω(t + ∆t).
The speed of the wave is then given by

∆x ω
c= = .
∆t k
As we will see later, ω is given by
p
ω= gk, (2)
where g is the gravitational acceleration, so

r
g
c= . (3)
k
This tells us that waves with different k travel at different speeds.

To illustrate the meaning of k, we will determine the wavelength, λ, of the wave defined by (1),
where λ is defined as the spatial distance between successive peaks at a given time. Recognizing
that in terms of θ, successive peaks are separated by ∆θ = 2π, we can write
θ(x, t) + 2π = θ(x + λ, t),
which gives us
kx − ωt + 2π = k(x + λ) − ωt.
1
Thus the wavelength of the wave can be written
2π
λ= .
k
(We can follow a similar analysis to find the period, T , of the wave, defined as the time between
successive peaks as seen by an observer at a fixed location.) Going back to (3), we see that waves
of longer wavelength (smaller k) travel faster. The property whereby individual waves of different
wavelengths travel at different speeds is known as dispersion.
Dispersion explains an interesting property of groups of waves. By ‘group of waves’ (some
authors use the term ‘wave packet’) we mean a pattern of waves that occupies a finite area in
space, not bounded by walls. Acheson points out that while an observer viewing a group of waves,
which consists of a multitude of different Fourier components (each defined by (1) with its own k
and ω), may count N peaks in that group at any particular time, that observer would count more
than N peaks as the group passed a fixed location. This is because the velocity of the group of
waves is different from the velocity of the individual peaks. The speed of the whole group of waves
is given by
dω
cg = ,
dk
which, using (2), becomes
r
1 g 1
cg = = c,
2 k 2
using (3). So, the group of waves travels half as fast as the individual wave peaks themselves.
1 Surface waves on deep water

While there’s a lot of interesting background material on waves in §3.1 in the text, we’re going to
skip the stuff on sound waves and solitary waves and go directly to the example of surface waves on
deep water. We will treat the water as an ideal fluid (negligible viscous forces). We are interested
in two-dimensional surface waves, so the velocity field is
u = [u(x, y, t), v(x, y, t), 0],
and we will assume that the flow is irrotational, so

∂v ∂u
ωz = − . (4)
∂x ∂y
The irrotationality of the flow is ensured if the fluid is initially at rest, i.e. initially has no vorticity;
we saw all the way back in the first chapter that in two-dimensional flow of ideal fluids subject
to gravity, the vorticity of each fluid element is conserved. Because the fluid is irrotational, there
exists a velocity potential, φ(x, y, t), such that
∂φ ∂φ
u= , and v = .
∂x ∂y
Also, because the fluid is incompressible, we have
∂u ∂v ∂2φ ∂2φ
∇·u= + = + 2 = ∇2 φ = 0. (5)
∂x ∂y ∂x2 ∂y
2
The position of the free surface will be given by
y = η(x, t),
and it is the deformation of this surface that will generate the fluid motion we’re interested in.
We will require that fluid particles on the surface remain there (this implicitly eliminates the
possibility of large disturbances, for example). If we define
F (x, y, t) = y − η(x, t),
then the free surface is defined as the surface where F = 0. Since fluid elements on the free surface
have to stay there, that means F is constant (and = 0) for those fluid elements, so in particular we
can write
DF ∂F
= + (u · ∇)F = 0 on the surface. (6)
Dt ∂t
Using the following relations,
∂F ∂η
=−
∂t ∂t
∂F ∂η
u = −u
∂x ∂x
∂F
v = v,
∂y
(6) becomes
∂η ∂η
+u =v on the surface. (7)
∂t ∂x
1.1 Small-amplitude approximations

Now, we’ll impose the requirement that η(x, t), the free surface displacement, and the velocity
components u and v and their derivatives be ‘small’. (The ‘smallness’ will amount to requiring
that the surface displacement be λ, the wavelength of the disturbances.) Then, in (7), we will
throw away terms that are the products of small quantities (this is called linearizing), leaving us
with
∂η
v(x, η, t) = , (8)
∂t
where the arguments for v establish that we’re talking about v at the surface. Since η is small, we
can expand v(x, η, t) around η = 0 in a Taylor series. In general, for a function f (x), the Taylor
expansion around x = 0 is
df x2 d2 f x3 d3 f
f (x) = f (0) + x (0) + (0) + (0) + · · · .
dx 2! dx2 3! dx3
For our problem, v(x, η, t) expands as
∂v
v(x, η, t) = v(x, 0, t) + η (x, 0, t) + · · · ,
∂y
and once again we toss out anything involving products of small quantities, so (8) becomes
∂φ ∂η
v(x, 0, t) = (x, 0, t) =
∂y ∂t
using v = ∂φ/∂y, where we can impose the condition y = 0 instead of y = η because of the
linearization.

ME 563 - Intermediate Fluid Dynamics - Su: Lecture 28 - Waves: The Basics

Uploaded by

Copyright:

Available Formats

ME 563 - Intermediate Fluid Dynamics - Su: Lecture 28 - Waves: The Basics

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ME 563 - Intermediate Fluid Dynamics - Su: Lecture 28 - Waves: The Basics

Uploaded by

Copyright:

Available Formats

ME 563 - Intermediate Fluid Dynamics - Su

Lecture 28 - Waves: the basics

Reading: Acheson, §3.1-3.3.

y = η(x, t) = A cos(kx − ωt) = η(θ), (1)

(A is a constant), where we define

θ(x, t) = θ(x + ∆x, t + ∆t) = k(x + ∆x) − ω(t + ∆t).

The speed of the wave is then given by

where g is the gravitational acceleration, so

This tells us that waves with different k travel at different speeds.

θ(x, t) + 2π = θ(x + λ, t),

1 Surface waves on deep water

u = [u(x, y, t), v(x, y, t), 0],

and we will assume that the flow is irrotational, so

1.1 Small-amplitude approximations

You might also like