National University of Computer and Emerging Sciences

Instructor: M. Usman Rashid MT1006

6 Differential Equations
Spring 2023 BS CS & BS DS
Practice Question

Heat Equation

= , 0< < , >0. (1)

((0, ) = 0, ( , ) = 0, >0. (2)

( , 0) = , 0 < < . (3)
Solution:
First note that the given equation is heat equation. Compare the given equation with general heat equation

= ,

to notice/observe that = 1 in your case which will provide an ease in the computations for the solution.
solution Furthermore, in
the given boundary conditions, is precisely given to be as which simply means that in general solutions derived during
lectures, one must use = .
With the usual assumption that ( , ) = ( ) ( ) .
⇒ = , = .
⇒ = .

⇒ = =− . (4)

Remember that is a constant and one imposes a negative sign for the sake of calculation convenience, nothing more or
less.
+ =0, (5)
and
+ =0. (6)
Boundary Conditions
The boundary conditions translate into (0 = 0 and
(0) ( ) = 0.
( , )= ( ) ( ).
⇒ ( , )= ( ) ( ).
(0, ) = (0) ( ) .
⇒ (0) ( ) = 0 .
⇒ (0) = 0 .
( , )=0,
⇒ ( ) ( )=0.
⇒ ( )=0.
The regular SL problem is then written as
+ = 0, (0) = 0, ( )=0. (7)
One seeks the values of for which the boundary-value problem has nontrivial, that is, nonzero, solutions.
One shall consider three cases: = 0, < 0, and > 0.
Case I: When = .
=0.
⇒ =0.
⇒ = 0, =0.
⇒ ( )= + .
Observe that the conditions are on derivatives, that is, . Implies ( )= . Using (0) = 0, gives
′(0) = .
⇒ =0.
⇒ ( )= .
Using ( ) = 0, gives
( )= , ⇒ =0.
This means the solution to the BVP in first case when = 0 is,
( )= .
In this case, the eigenfunction is = , corresponding to the eigenvalue = .
Case II: When < 0.
For λ < 0 it is convenient to write =− , where α denotes a positive number.
− =0.
⇒ − =0.
⇒ = , =− .
⇒ ( )= cosh + sinh .

Note that sinh = cosh , cosh = sinh . In our case sinh = cosh , cosh = sinh .

⇒ ( )= sinh + cosh .
Using (0) = 0, gives
(0) = sinh( ∙ 0) + cosh( ∙ 0) = ∙0+ ∙1= .
=0.
⇒ =0.
⇒ =0, ≠0.
⇒ ( )= cosh .
Using ( ) = 0, gives
( )= sinh , ( )= sinh .
0= sinh .
Either = 0 or sinh . For ≠ 0, sinh ≠ 0; consequently, one is forced to choose = 0.
=0.
The only solution of the BVP is the trivial solution ( ) = 0, and consequently ( , ) = .
Case III: When > 0.
For > it is convenient to write = , where denotes a positive number.
+ =0.
⇒ + =0 =− .
⇒ , =± .
⇒ ( )= cos + sin .
⇒ ( )=− sin + cos .
Using (0) = 0, gives
(0) = − sin( ∙ 0) + cos( ∙ 0) = .
⇒ =0,
⇒ = 0, ≠0.
⇒ ( )= cos . (8)
Using ( ) = 0, gives
( )=− sin , ( )=− sin .
sin =0.
⇒ =0 sin , ≠0.
⇒ sin =0.
⇒ = , ℎ = 0, ±1, ±2, ±3, ±4, …
This gives the value of to be = , = 0, ±1, ±2, ±3, ±4, … . So
= = , , = .
Such values of are called the eigenvalues of the problem. The eigenfunctions are
( ) = cos . (9)
Since solution occurred in the case > 0. The equation for ( ) can be written as
+ =0,
⇒ + =0.
⇒ =− =− .
( )= . (10)
Plugging the expressions of ( ) and ( ) in ( , ) = ( ) ( ), gives
( , )= cos ∙ = ∙ cos .
( , )= cos , = . (11)
Now by the superposition principle

( , )= cos = + cos . (12)

( )

What about the initial condition ( , ) = ?

Using ( , ) = from (3) by substituting = 0 into (12) implies that

( , 0) = + cos = + cos = + cos .

 ⏟ = + cos .
( )

2
= ( ) .

In our case = and ( ) = . This means that

2 2 2 0 2
= = = − = = .
2 2 2 2

2 2 2
= ( ) cos = cos = cos .

The above integral can be solved through integration by parts.

Taking = , = cos , gives = , = . By parts gives

= [ ∙ ]| − .

That is,
sin sin sin sin ∙ 0 1 cos
cos = ∙ − = ∙ − 0∙ − − .

1 1
cos =0+ [cos ∙ − cos ∙ 0] = [cos − 1] .

Since cos = (−1) , implies

1 1
cos = [cos − 1] = [(−1) − 1] .

2 2 1 2
= cos = ∙ [(−1) − 1] = [(−1) − 1] .

2
= , , = [(−1) − 1] .

Substituting and in ( , ) = +∑ cos , gives

2
( , )= + ((−1) − 1) cos .

One concludes that a solution of the boundary-value problem described in (1), (2), and (3) is given by the infinite series

2
( , )= + [(−1) − 1] cos .

∎
Good Luck

© M. Usman Rashid