MA 542 (2018) - Lec-8

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16

MA 542: Lecture - 8

Solution of one dimensional wave equation in


bounded domain using separation of variables

MA542(2018):PDE
Initial Boundary Value Problem (IBVP)

A physical problem governed by a PDE may contain both


boundary as well as initial conditions.

Equations that describe the behaviour of the solution on the


boundary of the region under consideration are called
boundary conditions.

Equations that describe the phenomenon at the initial values


of the solution at time t = 0 are called initial conditions.

When a physical problem seeks for solutions fulfilling both


initial and boundary conditions, the problem is called a initial
boundary value problems (IBVP).

MA542(2018):PDE
Separation of variables method for PDE

• A convenient and powerful solution technique of partial


differential equations.
• The basic idea of this method is that the solution is assumed
to consist of products of two or more functions, each function
being a function of one independent variable only.
• The number of functions involved depends on the number of
independent variables.
• For example to solve an equation for u = u(x, t) assume a
solution in the form u(x, t) = X (x)T (t), where X is a
function of x only and T is a function of t only.

MA542(2018):PDE
The Finite Vibrating String
Consider the transverse vibrations of a finite string described by
the following one dimensional wave equation:

utt − c 2 uxx = 0, 0 < x < L, 0 < t < ∞,


u(x, 0) = f (x), ut (x, 0) = g (x), 0 ≤ x ≤ L, (1)
u(0, t) = 0, u(L, t) = 0, 0 < t < ∞.

• In case of one dimensional wave equation for unbounded


domain (i.e. x ∈ R), we have found (D’Alembert’s solution)
solutions to be traveling waves, moving in opposite directions.
• But in case of one dimensional wave equation in bounded
domain (i.e. 0 < x < L), the waves no longer appear to be
moving due to their repeated interaction with the boundaries
(i.e. waves are standing waves).
• In this case the solutions appear to be vibrating waves.

MA542(2018):PDE
Remark:

If we know the shape Xn (x) of these standing waves and how each
one of them is vibrating Tn (t), we can write the solution as the
sum of the simple vibrations Xn (x)Tn (t), i.e, we can have a
separated solution given by
X
u(x, t) = cn Xn (x)Tn (t),
n∈N

where the coefficients cn ∈ R are choosen in such a way that


u(x, t) satifies the given initial and boundary conditions.

MA542(2018):PDE
Separated solution to the finite vibrating string

Consider the following initial/boundary value problem associated to


the vibrations of a finite string:

PDE : utt = c 2 uxx , 0 < x < L, 0 < t < ∞. (2)


u(x, 0) = f (x)
ICs ,0 ≤ x ≤ L (3)
ut (x, 0) = g (x)


u(0, t) = 0
BCs ,0 < t < ∞ (4)
u(L, t) = 0

MA542(2018):PDE
• Aim is to find standing-wave/separated solutions to the PDE
given in the form

u(x, t) = X (x)T (t)

• Substituting this in the PDE we get two ordinary differential


equations, viz.,
X 00 (x) T 00 (t)
= 2 . (5)
X (x) c T (t)
• As left hand side of (5) is a function of x and right side is
function of t, the equality holds only if

X 00 (x) T 00 (t)
= 2 = λ where λ ∈ R. (6)
X (x) c T (t)
• This can be rewritten as

X 00 (x) − λX (x) = 0 and T 00 (t) − λc 2 T (t) = 0. (7)

MA542(2018):PDE
• Since λ is any constant,
• it can be zero, or
• it can be positive, or
• it can be negative.
• Now as we are seeking for the nontrivial and bounded
solution, we look for those values of λ which correspond to
the nontrivial and bounded solutions X (x) and T (t) of the
above ODEs.

MA542(2018):PDE
Case I: λ = 0

• In this case the equations (7) reduce to

X 00 (x) = 0 and T 00 (t) = 0. (8)

• The general solutions of ODEs in (8) are given by

X (x) = Ax + B, T (t) = Ct + D.

• Now applying the BCs we have

u(0, t) = X (0)T (t) = 0, and u(L, t) = X (L)T (t) = 0.

• This implies T (t) = 0, ∀ t > 0 or X (0) = 0 = X (L) which


implies X (x) ≡ 0.
• Thus for λ = 0, there exists only trivial solution u(x, t) ≡ 0.

MA542(2018):PDE
Case II: λ > 0, (λ = k 2 ) k ∈ R

• In this case the equations (7) reduce to

X 00 − k 2 X = 0 and T 00 − c 2 k 2 T = 0.

• Giving rise to solutions

X (x) = Ae kx + Be −kx ,

T (t) = Ce ckt + De −ckt .


• Therefore

u(x, t) = (Ae kx + Be −kx )(Ce ckt + De −ckt ).

MA542(2018):PDE
• Using boundary condition,

X (0) = 0 we get A + B = 0.

• Also using the boundary condition,

X (L) = 0 we get Ae kL + Be −kL = 0.

• This is possible if A = 0 = B and hence X (x) ≡ 0.


• Thus for λ > 0, there exists only trivial solution u(x, t) ≡ 0.

MA542(2018):PDE
Case III: λ < 0 (λ = −k 2 , k ∈ R)

• In this case the equations (7) reduce to

X 00 + k 2 X = 0 and T 00 + c 2 k 2 T = 0.

• Giving rise to solutions

X (x) = Acos kx + Bsin kx,

T (t) = Ccos ckt + Dcos ckt.


• Hence u(x, t) = (Acos kx + Bsin kx)(Ccos ckt + Dsin ckt).

MA542(2018):PDE
• Using boundary condition

X (0) = 0 we get A = 0.

• Also using the boundary condition,

X (L) = 0 we get Bsin kL = 0.

• Now B 6= 0 as that will lead to a trivial solution.


• Thus we must have
sin kL = 0
which gives

k= , n = 1, 2, 3, ...
L
2
• Therefore λ = −k 2 = − nπL .
2
• Thus for λn = − L , there exists a nontrivial solution of

(2).

MA542(2018):PDE
• These λn s are called eigenvalues.
• Note that corresponding to each n there will be an eigenvalue.
• The function corresponding to each eigenvalue is called an
eigenfunction.
2
• The eigenfunction corresponding to eigenvalue λn = − nπ
L
is given by
nπx  ncπt ncπt 
un (x, t) = Xn (x)Tn (t) = sin Cn cos + Dn sin .
L L L

MA542(2018):PDE
• Since the wave equation is linear, therefore the superposition
of solutions un (x, t) is also a solution of (2). Hence the
solution exists in the following form:

X
u(x, t) = un (x, t)
n=1

X nπx  ncπt ncπt 
= sin Cn cos + Dn sin .
L L L
n=1

• The coefficients Cn and Dn are determined by using the initial


conditions (ICs) and the concept of Fourier series.

MA542(2018):PDE
Standing-wave solutions for different values of λ

Possible values of λ

λ=0 λ<0, (λ=k2)


λ<0, (λ=-k2)

T(t)= Aeckt+Be-ckt
T(t)=At+B
T(t)=A sin (ckt)+B cos (ckt) X(x)=Cekx+De-kx
X(x)=Cx+D
X(x)=C sin (kx)+D cos (kx)

u(x,t)=X(x)T(t)

MA542(2018):PDE

You might also like