Basic Concepts of PDEs

• A partial differential equation (PDE) is an equation involving one

or more partial derivatives of an (unknown) function, call it u, that
depends on two or more variables, often time t and one or several
variables in space. The order of the highest derivative is called the
order of the PDE. Just as was the case for ordinary differential
equations (ODEs), second-order PDEs will be the most important
ones in applications.

• PDEs: linear, nonlinear, homogeneous, non-homogeneous

Important Second-Order PDEs

 2u 2  2
u
2
c One-dimensional wave equation
t x 2
2
u 2  u One-dimensional heat equation
c
t x 2
 2u  2u
2
 2 0 Two-dimensional Laplace equation
x y
 2u  2u
2
 2  f ( x, y ) Two-dimensional Poisson equation
x y
 2u 2  2
u  2
u
2
c ( 2  2) Two-dimensional wave equation
t x y
 2u  2u  2u
2
 2  2 0 Three-dimensional Laplace equation
x y z
• Additional conditions: “boundary conditions”, “initial conditions”
• We know that if an ODE is linear and homogeneous, then from
known solutions we can obtain further solutions by superposition. For
PDEs the situation is quite similar:

Fundamental Theorem on Superposition

If u1 and u2 are solutions of a homogeneous linear PDE in some

region R,uthen
 c1u1  c2 u2 c1
with any constants c2 and is also a
solution of that PDE in the region R.
Wave Equation
 Modeling: Vibrating String, Wave Equation

Physical Assumptions

1. The mass of the string per unit length is constant (“homogeneous

string”). The string is perfectly elastic and does not offer any resistance
to bending.
2. The tension caused by stretching the string before fastening it at the
ends is so large that the action of the gravitational force on the string
(trying to pull the string down a little) can be neglected.
3. The string performs small transverse motions in a vertical plane; that
is, every particle of the string moves strictly vertically and so that the
deflection and the slope at every point of the string always remain small
in absolute value.
Derivation of the PDE of the Model (“Wave Equation”) from Forces

Forces acting on a small portion (PQ) of the string

Derivation of the PDE of the Model (“Wave Equation”) from Forces

T1cos  T2 cos =T=const

2u
Newton’s second law T2 sin   T1 sin   x 2
t

2u 2
2  u
T
• One-dimensional wave equation: c c2 
t 2
x 2 
 Solution by Separating Variables
Use of Fourier Series

Problem of the one dimensional Solution

wave equation:
Step 1. By the “method of separating
variables” or product method,
2u 2  2
u
 c
t 2 x 2 Step 2. We determine solutions of
these ODEs that satisfy the boundary
conditions (2).
u (0, t )  0, u(L,t)=0, for all t  0
u(x,0)=f(x), u t ( x, 0)  g ( x) (0  x  L) Step 3. Finally, using Fourier series
 Step 1. Two ODEs from the Wave Equation
In the method of separating variables, or product method, we determine
solutions of the wave equation of the form
u ( x, t )  F ( x).G(t )

2
u 2 F ''  kF  0
2  u
 c
t 2 x 2
  c 2 kG  0
G

Here, the separation constant k is still arbitrary

 Step 2. Satisfying the Boundary Conditions
Use of the boundary conditions
u (0, t )  F (0)G(t )  0, u(L,t)=F(L)G(t)=0, for all t  0

F ''  kF  0 F (0)  0, F(L)=0

For k=0, F=ax  b, a=b=0 so that F  0 and u=FG  0

For positive k= 2 , F=Ae  x  Be   x , F  0 and u=FG  0
Which is of no interest
Hence we are left with the possibility of choosing k negative, say, k   2
F ( x)  Acos x  B sin  x
F (0)  A  0 and then F(L)=Bsin L=0
n
Fn ( x)  sin x, (n integer)
L
 Step 2. Satisfying the Boundary Conditions (2)
with k=- 2  (n / L)2
   2 G  0 cn
G n where n  c  
L
A general solution is Gn (t )  Bn cosn t  B*n sin n t

* n
un ( x, t )  ( Bn cosn t  B sin n t )sin
n x
L

These functions are called the eigenfunctions, or characteristic functions

The values are n  cn / L called the eigenvalues, or characteristic values
The set 1 , 2 ,..... is called the spectrum.
 Step 3. Solution of the Entire Problem.
Fourier Series
 
n
un ( x, t )   un ( x, t )   ( Bn cosn t  B*n sin n t )sin x
n 1 n 1 L

Satisfying Initial Condition u  n 

 (Bnn sinnt  Bnncosnt)sin x
*

t t0 n1 L t0


n
u(x,0)  Bn sin x  f (x) n

(0  x  L)
L   B  sin x  g(x)
*
n n
n1
n1 L
L L
2 n x 2 n x
Bn   f ( x) sin

*
dx Bn  g ( x) sin dx
L0 L cn 0 L
 Solution Established
We consider only the case when the initial velocity is identically zero.

n cn
u ( x, t )   Bn cosn t sin x, n 
n 1 L L
cn n x 1   n   n 
cos t sin  sin  ( x  ct )   sin  ( x  ct ) 
L L 2  L   L 
1   n  1   n 
u ( x, t )   Bn sin  ( x  ct )    Bn sin  ( x  ct ) ,
2 n1  L  2 n 1  L 
1 *
u ( x, t )   f ( x  ct )  f * ( x  ct ) 
2

Odd periodic extension of f(x)

EXAMPLE : Vibrating String if the Initial Deflection is Triangular
 2k L
 L x if 0 x 
2
f ( x)  
 2k ( L  x ) if
L
x L
 L 2
and initial velocity zero
Solution
 D’Alembert’s Solution of the Wave Equation.
Characteristics
1 *
u ( x, t )   f ( x  ct )  f * ( x  ct ) 
2

It is interesting that the solution can be immediately obtained by

transforming in a suitable way, namely, by introducing the new
independent variables v  x  ct , w  x  ct
uxx  (uv  uw ) x  (uv  uw )v vx  (uv  uw )w w x  uvv  2uvw  uww

utt  c 2 (uvv  2uvw  uww )

2u
uvw  0
wv

u ( x, t )   ( x  ct )   ( x  ct ) This is known as d’Alembert’s

solution of the wave equation
 D’Alembert’s Solution Satisfying
the Initial Conditions
u ( x, 0)  f ( x), u t ( x, 0)  g ( x)
ut ( x, t )  c ' ( x  ct )  c ' ( x  ct )
u ( x, 0)   ( x)   ( x)  f ( x)
ut ( x, 0)  c ' ( x)  c ' ( x)  g ( x)
x
1
 ( x)   ( x)  k ( x0 )   g ( s) ds, k ( x0 )   ( x0 )   ( x0 )
c x0
x
1 1 1
 ( x)  f ( x)   g ( s) ds  k ( x0 )
2 2c x0 2
x
1 1 1
 ( x)  f ( x)   g ( s )ds  k ( x0 )
2 2c x0 2
x  ct
1 1
u ( x, t )   f ( x  ct )  f ( x  ct )    g ( s)ds
2 2c x ct
 Modeling: Membrane,
Two-Dimensional Wave Equation
Physical Assumptions
1. The mass of the membrane per unit area is constant (“homogeneous
membrane”). The membrane is perfectly flexible and offers no resistance to
bending.
2. The membrane is stretched and then fixed along its entire boundary in the xy-
plane. The tension per unit length T caused by stretching the membrane is the
same at all points and in all directions and does not change during the motion.
3. The deflection u ( x, y , t ) of the membrane during the motion is small
compared to the size of the membrane, and all angles of inclination are small.
Derivation of the PDE of the Model (“Two-Dimensional Wave
Equation”) from Forces.

Vibrating membrane
 Horizontal Components of the Forces. We first consider the horizontal
components of the forces. These components are obtained by multiplying the
forces by the cosines of the angles of inclination. Since these angles are small,
their cosines are close to 1. Hence the horizontal components of the forces at
opposite sides are approximately equal. Therefore, the motion of the particles of
the membrane in a horizontal direction will be negligibly small. From this we
conclude that we may regard the motion of the membrane as transversal; that is,
each particle moves vertically.
Vertical Components of the Forces. These components along the right
side and the left side are , respectively T y sin and -Ty sin
Since the angles are small, we may replace their sines by their tangents.
Hence the resultant of those two vertical components is

T y (sin  sin )  T y (tan  tan  ) =T y ux ( x  x, y1 )  u x ( x, y2 ) 

 Newton’s Second Law Gives the PDE of the Model.
2u
xy 2 =T y u x ( x  x, y1 )  u x ( x, y2 )   T x u y ( x1 , y  y )  u y ( x 2 , y ) 
t
 2 u T  u x ( x  x, y1 )  u x ( x, y2 ) u y ( x1 , y  y )  u y ( x2 , y ) 
=   
t 2
 x y 
If we let x and y approach zero, we obtain the PDE of the model
2u 2   2
u  2
u
2
c  2  2 
t  x y 

This PDE is called the two-dimensional wave equation

The expression in parentheses is the Laplacian
2u 2 2
2
 c u
t
 Rectangular Membrane. Double Fourier Series
2u 2   2
u  2
u
2
c  2  2
t  x y 
u  0 on the boundary
u(x,y,0)=f(x,y)
u t ( x, y, 0)  g ( x, y ) Rectangular membrane

We shall solve this problem in three steps:

Step 1. separating variables
Step 2. (“eigenfunctions” umn ) that satisfy the boundary condition
Step 3. We compose the umn into a double series
 Step 1. Three ODEs From the Wave Equation
u ( x, y, t )  F ( x, y)G(t )

2u 2   2
u  2
u   c 2 ( F G  F G )
FG
2
c  2  2  xx yy
t  x y 
G 1 
 ( Fxx  Fyy ) G 1 2
2
cG F  ( Fxx  Fyy )  
c2G F

“time function”    2 G  0
G

“amplitude function” Fxx  Fyy   2 F  0, F(x,y)=H(x)Q(y)

d2H 2
2
 k H 0
dx
1 d2H 1 d 2Q 2 2
  (   Q )   k
H dx 2 Q dy 2 d 2Q 2
2
 p Q0
dy
 Step 2. Satisfying the Boundary Condition
General solutions H ( x)  A cos kx  B sin kx and Q(y)=C cos py  D sin py

H (0)  0, H (a)  0, Q(0)=0, Q(b)=0

m x n y
Fmn ( x, y )  H m ( x)Qn ( y )  sin sin , m=1,2,....;n=1,2,...
a b

Eigenfunctions and Eigenvalues. m2 n2

  mn  c 2
 2
a b

m x n y
umn ( x, y, t )   Bmn cos mn t  B mn sin mn t  sin
*
sin
a b
Step 3. Solution of the Model . Double Fourier Series
   
m x n y
u ( x, y, t )   umn ( x, y, t )    Bmn cos mn t  B mn sin mn t  sin
*
sin
m 1 n 1 m 1 n 1 a b

 
m x n y
u ( x, y, 0)   Bmn sin sin  f ( x, y )
m 1 n 1 a b
u  
m x n y
  B*mn mn sin sin  g ( x, y )
t t 0 m1 n1 a b

4 b a m x n y
Bmn 
ab 0 0
f ( x , y )sin
a
sin
b
dxdy

4 b a m x n y

*
B mn  g ( x, y ) sin sin dxdy
abmn 0 0 a b