UNIT – III- APPLICATIONS OF PARTIAL DIFFERENTIAL

EQUATIONS

CLASSIFICATION OF SECOND ORDER QUASI –LINEAR EQUATION:

Consider the second order linear homogeneous P.D.E

𝜕2𝑢 𝜕𝑢 𝜕𝑢
𝜕2𝑢 𝜕2𝑢 +𝐶 +𝐷 +𝐸 + 𝐹𝑢 = 0 ------ (1)
A 2
+ 𝐵
𝜕𝑥 𝜕𝑥𝜕𝑦 𝜕𝑦2 𝜕𝑥 𝜕𝑦

Where A, B, C, D, E are functions of x, y or they are real constants.

The P.D.E (1) is said to be

i. “parabolic equation” if B2-4AC = 0

ii. “hyperbolic equation” if B2-4AC> 0
iii. “elliptic equation” if B2-4AC < 0

Examples:

(i) Parabolic equation

𝜕𝑢 𝜕2𝑢
= 𝑐2 (One dimensional hear equation)
𝜕𝑡 𝜕𝑥2
𝜕𝑢 𝜕2𝑢 𝜕2𝑢
=𝑐 (2 + ) (Two dimensional hear equation)
𝜕𝑡 𝜕𝑥2 𝜕𝑦2

(ii) Hyperbolic equation

𝜕2𝑢 𝜕2𝑢
2
= 𝑐2 (One dimensional Wave equation)
𝜕𝑡 𝜕𝑥2
𝜕2𝑢 𝜕2𝑢 𝜕2𝑢
=𝑐 ( 2 + ) (Two dimensional Wave equation)
𝜕𝑡2 𝜕𝑥2 𝜕𝑦2

(iii) Elliptic equation

𝜕2𝑢 𝜕2𝑢
+ = 0 (Two dimensional Laplace equation)
𝜕𝑥2 𝜕𝑦2
Classify the following partial differential equations:

𝜕2𝑢 𝜕𝑢 𝜕𝑢
(i) =5 +
𝜕𝑥2 𝜕𝑥 𝜕𝑦
2𝑢 2
𝜕 𝑢
(ii) 𝜕 + =0
𝜕𝑥2 𝜕𝑦2
2𝑢 𝜕2𝑢 𝜕2𝑢
(iii) 𝜕 +3 + =0
𝜕𝑥2 𝜕𝑥𝜕𝑦 𝜕𝑦2

Solution:
𝜕2𝑢 𝜕𝑢 𝜕𝑢
(𝑖) −5 − =0
𝜕𝑥2 𝜕𝑥 𝜕𝑦

A=1, B=0, C=0

B2-4AC = 0

The given P.D.E is a parabola

2𝑢 𝜕2𝑢
(ii) 𝜕 + =0
𝜕𝑥2 𝜕𝑦2

A=1, B=0, C=1

B2-4AC < 0

The given P.D.E is an elliptic.

𝜕2𝑢 𝜕2𝑢 𝜕2𝑢

(𝑖𝑖𝑖) +3 + =0
𝜕𝑥2 𝜕𝑥𝜕𝑦 𝜕𝑦2
 A=1, B=3, C=1

B2-4AC > 0

The given P.D.E is a hyperbolic

Transverse Vibrations of a Stretched String (1-dimensional wave equation)

We make the following assumptions:

1. The motion takes place entirely in one plane.

2. n this plane, each particle moves at right angles to the equilibrium

position of the string.
3. The tension T caused by stretching the string before fixing it at the end
points is constant at all points of the deflecting string.
4. The tension T is considered to be very large compared with the weight of
the sting and hence the force of gravity is negligible.
5. The effect of friction is negligible.
6. The string is perfectly flexible.
7. The slope of the deflection curve at all points and at all instants is so
small.

Definition:
2𝑦 𝜕2𝑦
The P.D.E of 1-dimensional wave equation is given by 𝜕 = 𝑎2 ,
𝜕𝑡2 𝜕𝑥2

where a2 = 𝑇 . This is also called the P.D.E of the vibrating string.

𝑚
Solution of One Dimensional Wave Equation:

The 1-dimensional wave equation is given by

𝜕2𝑦 𝜕2𝑦
= 𝑎2 (1)
𝜕𝑡2 𝜕𝑥2

Assume that the solution of (1) is of the form

Y(x, t) = X(x).T (t) (2)

where X is a function of x only and T is a function of t only.

𝜕2𝑦 2𝑦 𝑑2𝑋 𝑑2𝑇

= 𝑋𝑇" and 𝜕 = 𝑋"𝑇, where 𝑋" = 𝑎𝑛𝑑𝑇" =
𝜕𝑡2 𝜕𝑥2 𝑑𝑥2 𝑑𝑡2
Hence (1) becomes, XT” =a2X”T

ie. X”/X = T”/a2T = k (say), where k is any constant.

Hence X”-k X=0 and T”-a2k T=0 (3)

Solution of these equations depends upon the nature of the value Of k.

Case1: Let k = p2, a positive number.

Now the equation (3) are X”- p2X=0 and T”- a2p2T=0.

Solving the ordinary differential equations we get,

X = A epx + B e-x and T = C epat + D e-pat.

Case 2: Let k=-p2, a negative number.

Then the equations (3) are X”+ p2X=0 and T”+ a2p2T=0.
Solving we get,
X = A cos px+ B sin px and T = C cos pat + D sin pat

Case 3: Let k=0.

Now the equations (3) are X”=0 and T”=0.

Integrating, we get,
X = Ax+B and T = Ct+D.

Thus the various possible solutions of the wave equation are

Y( x, t ) = ( A epx+B e-px )( Cepat+D e-pat) (I)

Y ( x, t ) = (A cos px + B sin px)( C cos pat + D sin pat) (II)
Y (x, t) = (Ax+B)( Ct+D) (III)

Out of these solutions, we have to select that particular solution which

suite the physical nature of the problem and the given boundary conditions. In
the case of vibration of string, it is evident that y must be a periodic function of
x and t. Hence we select the solution II as the probable solution of the wave

Page 130
equation. The constants are determined by using the boundary conditions in the
problem. In doing problems, we shall select the solution II directly.

Page 130
 One Dimensional Heat Flow Equation

Assumptions:

(i) Heat flow from a higher to lower temperature.

(ii) The amount of heat required for a change in temperature in a body
is proportional to the mass of the body and the change in
temperature.
(iii) The rate of flow of heat in an area is proportional to the area and
the temperature gradient normal.
Definition

The 1-dimensional heat flow equation is defined by

u  2u k
  2 2 , where 
2
is called the diffusivity of the material of the
t x c
bar.

Solution of 1-Dimensional Heat flow equation

We have to solve the equation

u  2u k
  2 2 , where 
2
(1)
t x c

Assume that the solution of (1) is of the form

u(x, t) = X(x).T (t) (2)

where X is a function of x only and T is a function of t only.

𝑑2𝑋 𝑑𝑇
Then (1) becomes, XT’ =2X”T where 𝑋" = 𝑎𝑛𝑑 𝑇′ =
𝑑𝑥2 𝑑𝑡

ie. X”/X = T’/2T = k (say), where k is any constant.

Hence X”-kX=0 and T’-2kT=0 (3)

Solution of these equations depends upon the nature of the value of k.

Case1: Let k=p2, a positive number.

Now the equation (3) are X”- p2X=0 and T’- 2p2T=0.
Solving the ordinary differential equations we get,
2 2
𝑝 𝑡
X=Aepx+Be-x and T= C𝑒𝛼

Case 2: Let k = -2, a negative number.

Then the equations (3) are X”+ p2X=0 and T’+ 2p2T=0.
Solving we get,
2 2
𝑝 𝑡
X=Acospx+Bsinpx and T= C𝑒−𝛼

Case 3: Let k=0.

Now the equations (3) are X”=0 and T’=0.

Integrating, we get,
X=Ax+B and T=C.

Thus the various possible solutions of the wave equation are

𝟐 𝟐
u = (Aepx+Be-px) C𝒆𝜶 𝒑 𝒕 (I)
𝟐 𝟐
u = (Acospx+Bsinpx) C𝒆−𝜶 𝒑 𝒕 (II)
u = (Ax+B) C (III)

Out of these solutions, we have to select that particular solution which

suite the physical nature of the problem and the given boundary conditions. As
we are concerned with heat conduction, u(x,t) must decrease with increase of
time. Therefore, out of the three solutions, we select the (II) solution to suit the
physical nature of the problem. In the steady state conditions, when the
temperature no longer varies with time, the solution of the diffusion equation (I)
will be the last solution(III).

PART A

Problem 1 In the one dimensional heat equation 𝒖𝒕= 𝒄𝟐 𝒖𝒙𝒙 , what

is 𝑐2?(May/June 2013)
𝑘
Solution : 𝑐2 = where k = thermal conductivity, 𝜌 = density and s =
𝜌.𝑠
Specific heat

Problem 2 An insulated rod of length 60 cm has its end at A and B maintained

at 20° C and 80° C respectively. Find the steady state solution of the rod.
(Nov/Dec 2012)

Solution : 𝑙 = 60 𝑐𝑚
𝟐𝒖
When steady state condition prevail the heat flow equation is 𝝏 =o
𝝏𝒙𝟐

𝑢(𝑥) = 𝑎𝑥 + 𝑏 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ → (𝑖)
When steady state conditions exists the boundary conditions are
𝑢(0) = 20 ; 𝑢(𝑙) = 80 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ → (𝑖𝑖)
Therefore solution can be got using the formula
𝐵−𝐴
𝑢(𝑥) = ( ) 𝑥 + 𝐴 where A = 20 and B = 80
𝑙
80−20
𝑢(𝑥) = ( ) 𝑥 + 20 ∵ 𝑙 = 60
𝑙
80−20
𝑢(𝑥) = ( )𝑥 + 20
60
60
⟹ 𝑢(𝑥) = (
) 𝑥 + 20
60
𝑨𝒏𝒔: 𝑢(𝑥) = 𝑥 + 20
Problem 3:What is the basic difference between the solutions of one
dimensional wave equation and one dimensional heat equation with respect to
the time? ( May/ June 2012 )

Solution :

Solution of the one dimensional wave equation is of periodic in nature. But

solution of the one dimensional heat equation is not of periodic in nature.
Problem 4 State the governing equation for one dimensional heat equation and
necessary to solve the problem. (Nov/Dec 2011)

Solution :

𝒖𝒕= 𝒄𝟐 𝒖𝒙𝒙
𝑘
𝑐2 = where k = thermal conductivity, 𝜌 = density and s = Specific heat
𝜌.𝑠

Boundary conditions

(i) 𝑢(0, 𝑡 ) = 0 ∀𝑡
(ii) 𝑢(𝑙, 𝑡 ) = 0 ∀𝑡

Initial conditions

(iii) 𝑢(𝑥, 0 ) = 0 ∀𝑥

Possible solutions
2 2
(i) 𝑢(𝑥, 𝑡 ) = (𝐴 𝑒𝑝𝑥 + 𝐵 𝑒−𝑝𝑥 ) 𝑒𝑐 𝑝 𝑡
2 2
(ii) 𝑢(𝑥, 𝑡 ) = (𝐴 𝑐𝑜𝑠 𝑝𝑥 + 𝐵 𝑠𝑖𝑛𝑝𝑥) 𝑒−𝑐 𝑝 𝑡
(iii) 𝑢(𝑥, 𝑡 ) = (𝐴 𝑥 + 𝐵)

Correct solution
2 2
𝑝 𝑡
⇒ 𝑢(𝑥, 𝑡 ) = (𝐴 𝑐𝑜𝑠 𝑝𝑥 + 𝐵 𝑠𝑖𝑛𝑝𝑥) 𝑒−𝑐

Problem 5 A rod 40 cm long with insulated sides has its ends A and B kept at
20° 𝐶 and 60° 𝐶 respectively. Find the steady state temperature at a location 15
cm from A. (April/ May 2011)

Solution :
𝐵−𝐴
𝑢(𝑥) = ( ) 𝑥 + 𝐴 where A = 20 and B = 60
𝑙
60−20
𝑢(𝑥) = ( 𝑥) + 20 ∵ 𝑙 = 40
𝑙
60−20
𝑢(𝑥) = ( ) 𝑥 + 20
40

𝑢(𝑥) = 𝑥 + 20

At x = 15 steady temperature 𝑢 = 15 + 20 = 35° 𝐶

Problem 6 :Write down the three possible solutions of one dimensional
heat equation (Nov/Dec 2010 , May/ June 2009)

Solution :

𝒖𝒕= 𝒄𝟐 𝒖𝒙𝒙
possible solutions are
2 2
(i) 𝑢(𝑥, 𝑡) = (𝐴 𝑒𝑝𝑥 + 𝐵 𝑒−𝑝𝑥) 𝑒𝑐 𝑝 𝑡
2 2
(ii) 𝑢(𝑥, 𝑡 ) = (𝐴 cos 𝑝𝑥 + 𝐵 sin 𝑝𝑥) 𝑒−𝑐 𝑝 𝑡
(iii) 𝑢(𝑥, 𝑦 ) = (𝐴 𝑥 + 𝐵)
Problem 7:A tightly stretched string with fixed end points x = 0 and x = l is
initially in a position given by 𝑦(𝑥, 0) = 𝑦 𝑠𝑖𝑛3 𝜋𝑥 . If it is released from
0 ( )
𝑙
rest in this position, write the boundary conditions. (April/ May 2010 )

Solution :

𝝏𝟐 𝒚 𝝏 𝟐𝒚
= 𝒄𝟐
𝝏𝒕𝟐 𝝏𝒙𝟐
(i) 𝑦(0, 𝑡) = 0 , 𝑡 >0
(ii) 𝑦(𝑙, 𝑡) = 0 , 𝑡 >0
(iii) 𝝏𝒚(𝒙,𝟎) = 𝟎 , 𝟎<𝑥<𝑙
𝝏𝒕
𝜋𝑥
(iv) 𝑦(𝑥, 0) = 𝑦0 𝑠𝑖𝑛3 ( ) , 𝟎<𝑥<𝑙
𝑙
𝟐𝒚
𝝏 𝝏𝒖
Problem 8: Classify the partial differential equation 4 =
𝝏𝒙𝟐 𝝏𝒕

Solution :
𝜕 2𝑦 𝜕𝑢
4 − =0
𝜕𝑥2 𝜕𝑡

A = 4 , B = 0 and C = 0

𝐵2 − 4 𝐴𝐶 = (0)2 − 4(4)(0)

=0
Therefore, Parabolic

Problem 9:Write sown all possible solutions of one dimensional wave

equation (Nov/Dec 2009)

Solution :

𝝏𝟐 𝒚 𝝏 𝟐𝒚
= 𝒄𝟐
Possible solutions
𝝏𝒕𝟐 𝝏𝒙𝟐

(i) 𝑦(𝑥, 𝑡) = (𝐴 𝑒𝑝𝑥 + 𝐵 𝑒−𝑝𝑥)(𝐶 𝑒𝑝𝑎𝑡 + 𝐷 𝑒−𝑝𝑎𝑡)

(ii) 𝑦(𝑥, 𝑡 ) = (𝐴 cos 𝑝𝑥 + 𝐵 sin 𝑝𝑥)(𝐶 cos 𝑝𝑎𝑡 + 𝐷 sin 𝑝𝑎𝑡)

(iii) 𝑦(𝑥, 𝑡 ) = (𝐴 𝑥 + 𝐵)(𝐶 𝑡 + 𝐷)

Problem 10: Verify that y = cos h (𝑥)𝑐𝑜𝑠 ℎ (−𝑎𝑡) is a solution of
𝜕2𝑦 𝜕2𝑦
= 𝛼2 (May/ June 2009)
𝜕𝑡2 𝜕𝑥2

Solution :

y = cos h (𝑥)𝑐𝑜𝑠 ℎ (−𝑎𝑡) → (1)

Differentiating (i) partially w.r.t x

𝝏𝒚
=  sin ℎ ( 𝑥) cosh(− 𝑎𝑡)
𝝏𝒙
𝝏𝟐𝒚 2
=  cosh( 𝑥) cosh(− 𝑎𝑡) → (2)
𝝏𝒙𝟐
𝝏𝒚
= −𝒂 cos ℎ ( 𝑥) sinh(− 𝑎𝑡)
𝝏𝒕
𝝏 𝟐𝒚
𝟐
= 𝑎22 cosh( 𝑥) cosh(− 𝑎𝑡)
𝝏𝒕
𝝏 𝟐𝒚
= 𝑎2 using (2)
𝝏𝒙𝟐
𝟐𝒚 𝝏 𝟐𝒚
Here y = cos h (𝑥)𝑐𝑜𝑠 ℎ (−𝑎𝑡) is a solution of 𝝏 = 𝑎2
𝝏𝒕𝟐 𝝏𝒙𝟐

Problem 16 A rod 50 cm long with insulated sides has its ends A and B kept at
20° C and 70° C respectively. Find the steady state temperature distribution of
the rod. (Nov/ Dec 2008)

Solution :
𝐵−𝐴
𝑢(𝑥) = ( ) 𝑥 + 𝐴 where A = 20 and B = 70
𝑙
70−20
𝑢(𝑥) = ( ) 𝑥 + 20 ∵ 𝑙 = 50 𝑐𝑚
𝑙
u(𝑥) = 70−20 𝑥 + 20
( )
50

u(𝑥) = 𝑥 + 20

Page 140
Problem 11. Classify the differential equation(Nov/ Dec 2008)
𝝏𝟐 𝒖 𝝏𝟐 𝒖 𝝏𝟐 𝒖 𝝏𝒖 𝝏𝒖
𝟑 𝟐+ 𝟒 +𝟔 −𝟐 + −𝒖=𝟎
𝝏𝒙 𝝏𝒙𝝏𝒚 𝝏𝒚 𝝏𝒙 𝝏𝒚

Solution :

A = 3 , B = 4 and C = 6

𝐵2 - 4 A C = 16 – 72 = - 56 < 0

𝐵2 - 4 A C < 0 Therefore the given differential equation is elliptic

Problem 12. The ends A and B of a rod of length 10cm long have their
temperature kept at 20° C and 70° C. Find the steady state temperature
distribution on the rod. (April/ May 2008)

Solution :
𝐵−𝐴
𝑢(𝑥) = ( ) 𝑥 + 𝐴 where A = 20 and B = 70
𝑙
70−20
𝑢(𝑥) = ( ) 𝑥 + 20 ∵ 𝑙 = 10 𝑐𝑚
𝑙
u(𝑥) = 70−20 𝑥 + 20
( )
10

u(𝑥) = 5𝑥 + 20

Problem 13. Classify the differential equation (April/ May 2008)

𝝏𝟐𝒖 𝝏𝟐𝒖 𝝏𝒖 𝝏𝒖
𝟑 𝟐+ 𝟒 −𝟐 +𝟑 =𝟎
𝝏𝒙 𝝏𝒙𝝏𝒚 𝝏𝒙 𝝏𝒚

Solution :
A = 3 , B = 4 and C = 0

𝐵2 - 4 A C = 16 > 0 . Therefore the equation is hyperbolic

 PART B

One Dimensional Wave Equation

Problems of Vibrating string with initial velocity:

Problem 1 A tightly stretched string of length l has its ends fastened at x=0 and x= l. The
midpoint of the string is then taken to height h and then released from rest in that position.
Find the lateral displacement of a point of the string at time t from the instant of
release.(April/ May 2010)
Solution:

 2y  2  y
2

The displacement of the string is given by a (1)

t 2 x 2

Take one end as the origin. The midpoint of the string has the height ‘h’.
h0

Equation of OA is y-0 = x  0
 
l/20 
 
2h
i.e. y = .x , when 0≤x≤l/2
l

 2h
Equation of AB is y = .x  l  , when l/2 ≤x≤l
l

Hence the boundary conditions are

(i) y(0,t) = 0, for all t≥0

(ii) y(l,t) = 0, for all t≥0
 y 
(iii)   0 , for all 0≤x≤ l
t
 att0
 2h
(iv) y(x,0) = f(x) = l .x,0  x  l / 2

 2h
 x  l , l / 2  x  l
l
Solve the equation (1) using the conditions (i), (ii), (iii) and (iv)
The most suitable solution of equation (1) is

Y(x,t) = (A cos px + B sin px) ( C cos pat + D sin pat ) (2)

Where A, B, C and D are arbitrary constants.

Using the boundary conditions (i), (ii) and (iii), we get

n
A = 0, p = and D=0
l

Equation (2) becomes,

nx nat
y(x,t) = B C sin cos (3)
l l

The most general solution of equation of (3) is

y(x,t) = ∑∞ 𝑛𝜋𝑥 𝑛𝜋𝑎𝑡 (4)

𝑛=1 𝐵𝑛𝑠𝑖𝑛 𝑙
𝑐𝑜𝑠
𝑙

Using boundary condition (iv) in (4), we get

 2h .x,0  x  l / 2
𝐵 𝑠𝑖𝑛 𝑛𝜋𝑥 = f(x) =  l
∞
y(x,0) = ∑
 2h
𝑛=1 𝑛 𝑙
 x  l , l / 2  x  l
 l

L.H.S is a half-range sine series.

2 1 𝑛𝜋𝑥
Bn = ∫ 𝑓(𝑥)𝑠𝑖𝑛 𝑑𝑥
𝑙 0 𝑙

Substituting f(x) value and integrating we get,

8ℎ 𝑛𝜋
Bn = 𝑠𝑖𝑛
𝑛 2𝜋 2 2

Substituting this in equation (4) , we get,

y(x,t) = 8ℎ 1 𝜋𝑥 𝜋𝑎𝑡 1 3𝜋𝑥 3𝜋𝑎𝑡

[ 𝑠𝑖𝑛 𝑐𝑜𝑠 − 32 𝑠𝑖𝑛 𝑐𝑜𝑠 +⋯]
𝜋2 𝑙2 𝑙 𝑙 𝑙 𝑙

which is the required displacement.

Problem 2. A tightly stretched string with fixed end points x= 0 and x = 2l is initially in
𝑘𝑥
,0<𝑥 <𝑙
𝑙
a position given by y(x,0) = { 𝑘(2𝑙−𝑥) . If it is released from rest from this
,𝑙 < 𝑥 <2
𝑙
𝑙

position, find the displacement function y(x,t) at any point of the string. (Nov/ Dec
2006, 2010, May/June 2008)
Solution:

 2y  2  y
2

The displacement of the string is given by a (1)

t 2 x 2

The boundary conditions are

(i) y(0,t) = 0, for all t≥0

(ii) y(2l,t) = 0, for all t≥0
 y 
(iii)   0 , for all 0≤x≤ l
t
 att0
𝑘𝑥
,0<𝑥<𝑙
𝑙
(iv) y(x,0) = {𝑘(2𝑙−𝑥)
, 𝑙 < 𝑥 < 2𝑙
𝑙

The most suitable solution of equation (1) is

y(x,t) = (Acospx+Bsinpx)(Ccospat+Dsinpat) (2)

Where A, B, C and D are arbitrary constants.

n
Using the boundary conditions (i), (ii) and (iii), we get A=0, p= and D=0
l

n x nat
∴ (2) becomes, y(x,t) =BC sin 𝑐𝑜𝑠
2l 2l
The most general equation is y(x,t) = ∑∞ 𝐵𝑛𝑠𝑖𝑛
𝑛𝜋𝑥
𝑐𝑜𝑠
𝑛𝜋𝑎𝑡
𝑛=1 2𝑙 2𝑙
𝑘𝑥
,0<𝑥<𝑙
∞ 𝑛𝜋𝑥
By applying the boundary condition (iv), y(x,0) = ∑𝑛=1 𝐵𝑛 𝑠𝑖𝑛 ={ 𝑙
2𝑙 𝑘(2𝑙−𝑥)
𝑙
, 𝑙 < 𝑥 < 2𝑙
1 2𝑙 𝑛𝜋𝑥 k 𝑙 𝑛𝜋𝑥 k 2𝑙 𝑛𝜋𝑥
𝐵 = ∫ f(x) 𝑠𝑖𝑛 𝑑𝑥 = ∫ x 𝑠𝑖𝑛 𝑑𝑥 + ∫ (2l − x) 𝑠𝑖𝑛 𝑑𝑥
𝑛𝑙0 2𝑙 𝑙2 0 2𝑙 𝑙2 l 2𝑙
 −𝑐𝑜𝑠𝑛𝜋𝑥 𝑠𝑖𝑛 𝑛𝜋𝑥 𝑙 −𝑐𝑜𝑠𝑛𝜋𝑥 𝑠𝑖𝑛𝑛𝜋𝑥 2𝑙
2𝑙 2𝑙 2𝑙 2𝑙
= 𝑘 [𝑥 ( )+( )] + 𝑘 [(2𝑙 − 𝑥) ( )−( )]
𝑛𝜋 𝑛2 𝜋2 𝑛𝜋 𝑛2 𝜋2
𝑙2 𝑙2
2𝑙 2𝑙
4𝑙 2 0 4𝑙 2 𝑙

8𝑘 𝑛𝜋
= 𝑛2𝜋2 𝑠𝑖𝑛
2

∴ y(x,t) = ∑∞ 8𝑘
𝑠𝑖𝑛
𝑛𝜋
𝑠𝑖𝑛
𝑛𝜋𝑥
𝑐𝑜𝑠
𝑛𝜋𝑎𝑡
𝑛=1 𝑛2𝜋 2 2 2𝑙 2𝑙

Problems with non-zero initial velocity:

Problem 3 A string of length l is initially at rest in its equilibrium position and

y
each of its points is given a velocity (x,0)  v sin 3 x ,0  x  l . Determine the
t
0
l
displacement y(x,t).(Nov/ Dec 2012, Nov/Dec 2004 )

Solution:

 2y  2  y
2

The wave equation is a

x 2 t 2

The solution is y = (A cos px + B sin px) ( C cos pat + D sin pat) (1)

We have the following boundary conditions

(i) y = 0 when x=0

(ii) y = 0 when x=l
(iii) y = 0 when t=0
y x
(iv)(iv) (x,0)  v sin 3 when t=0
t
0
l

Applying condition (i) , (ii) in (1), we get,

n nx  nat nat 
A=0, p = and y = Bsin C cos  D sin  (2)
 
l l  l l 

Using the condition (iii) in (2) we get C=0

Equation (2) becomes,

  bnsin nx sin nat
Y= (3)
n1 l l

Differentiating (4) w.r.t ‘t’ and using the boundary condition (iv), we get,

x 
nx
v sin 3
0
l
= 
n1
bn sin
l

To find the values of b1, b2 …we have to apply sin3θ formula and compare the
3v0l
like terms we get, b1= ,b  0,b   v0l ,b  b  0......
4a 12a
2 3 4 5

The solution is

3v0l x at
Y= sin sin  v0l sin 3x sin 3at
4a 12a


l l l l

Problem 4. A tightly stretched string with fixed end points x=0 and x=l is initially at rest in
its equilibrium position. If it is set vibrating by giving each point a velocity 3x(l-x). Find the
displacement of the string at any time. (Nov’03,May’08,N/Dec 2009, May/June 2013 if 3=
k)
Solution:
 2y  2  y
2

The displacement of the string is given by a (1)

t 2 x 2

The boundary conditions are

(i) y(0,t) = 0, for all t ≥ 0
(ii) y(l ,t) = 0, for all t ≥ 0
(iii) y(x ,0) = 0, for all x in (0,l)
y
(iv) (x,0) 3x(l  x) for all x in (0,l )
t
The most suitable solution of equation (1) is

The solution is y = (A cos px + B sin px ) ( C cos pat + D sin pat) (2)

 n
Applying the first three boundary conditions in (2) we get A=0, p = ,C=0 and
l
nx  nat 
y = B sin D sin  (3)
 
l  l 

The most general solution of equation of (3) is

y(x,t) = ∑∞ 𝑛𝜋𝑥 𝑛𝜋𝑎𝑡 (4)

𝑛=1 𝐵𝑛𝑠𝑖𝑛 𝑙
𝑠𝑖𝑛
𝑙

Differentiating (4) partially w.r.t ‘t’ we get

𝜕𝑦(𝑥,𝑡) 𝑛𝜋𝑎 𝑛𝜋𝑥 𝑛𝜋𝑎𝑡 (5)

= ∑∞𝑛=1 𝐵𝑛 𝑠𝑖𝑛 𝑐𝑜𝑠
𝜕𝑡 𝑙 𝑙 𝑙

Putting t = 0 in (5), we get

𝑛𝜋𝑥
𝜕𝑦(𝑥,0)
= ∑∞𝑛=1 𝐵
𝑛𝜋𝑎
𝑠𝑖𝑛
𝑛𝜋𝑥 = ∑∞ 𝑏 𝑠𝑖𝑛 = 3𝑥(𝑙 − 𝑥) where 𝑏 =𝐵 𝑛𝜋𝑎
𝜕𝑡 𝑛 𝑙 𝑙 𝑛=1 𝑛 𝑙 𝑛 𝑛𝑙

By expanding 3𝑥(𝑙 − 𝑥) in a half range Fourier sine series, we get

2 𝑙 𝑛𝜋𝑥 6 −𝑐𝑜𝑠𝑛𝜋𝑥 𝑛𝜋𝑥

) − (𝑙 − 2𝑥) (−𝑠𝑖𝑛 𝑙 ) +
𝑙
𝑏 = ∫ 3𝑥(𝑙 − 𝑥) 𝑠𝑖𝑛 𝑑𝑥 = [(𝑙𝑥 − 𝑥 2) (
𝑛 𝑙 0 𝑙
𝑙
𝑛𝜋 𝑛2 𝜋2
𝑙
𝑐𝑜𝑠 𝑙
𝑛𝜋𝑥 𝑙2
(−2) ( 𝑛3 𝜋3 )]
𝑙3
12𝑙2 𝑛] 0 , 𝑤ℎ𝑒𝑛 𝑛 𝑖𝑠 𝑒𝑣𝑒𝑛
= [ = { 24𝑙2
𝑛 3𝜋 3 1 − (−1)
𝑛3𝜋3
, 𝑤ℎ𝑒𝑛 𝑛 𝑖𝑠 𝑜𝑑𝑑

Therefore, 𝐵 =𝑏 𝑙 24𝑙3 and hence y(x,t) = ∑∞ 24𝑙3 𝑛𝜋𝑥 𝑛𝜋𝑎𝑡

𝑛=1,3,5,.. 𝑎𝑛4𝜋 4 𝑠𝑖𝑛 𝑙 𝑠𝑖𝑛 𝑙
𝑛 𝑛 𝑛𝜋𝑎 =
𝑎𝑛4𝜋 4

One Dimensional Heat Flow Equation

Problem 5. A rod of 30 cm long has its ends A and B kept at 20𝑜 c and 80 𝑜c
respectively, until steady state conditions prevail. The temperature at each end
is then suddenly reduced to 0𝑜 c and kept so. Find the resulting temperature
function u(x,t) taking x = 0 at A . (N/D 2009) ,(N/D 2008)
Solution:

The P.D.E satisfied by u(x, t) is

u 2 u
2

t x2 (1)

2 𝑝2 𝑡
u(x, t) = (Acospx+Bsinpx) 𝑒−𝛼 (2)
2𝑢
In steady-state, this equation reduces to 𝑑 = 0
𝑑𝑥2
Here l=30

Solving this, u = ax+b, where a and b are arbitrary constants.

u=20 at x=0 and u=80 at x=30.

 b=20 and 80= la+20

Solving we get a= 60
𝑙

Thus the temperature function in steady-state is

u(x) = 60x+20 (3)

𝑙

This temperature distribution reached at the steady state becomes initial temperature
distribution for the unsteady state. Then the temperature function u(x, t) satisfies (1).

The new boundary conditions are

(i) u(0, t) = 0  t>0

(ii) u(l, t) = 0  t>0
(iii) u(x, 0) = 60 x+20 for 0<x< l
𝑙

Using the boundary condition (i) in (2),

2𝑝 2𝑡
u(0, t) = A𝑒−𝛼 =0  t>0
A = 0

Using (ii) in (2),

 2𝑝2𝑡
B sin l p 𝑒−𝛼 =0,  t>0
Since B0, sin lp=0.

Hence lp = n

i.e. p = 𝑛, where n is any integer.
𝑙

Therefore, (2) becomes,

−𝛼 2𝑛 2 2 𝑡
𝑛𝑥
u(x, t) = 𝐵𝑛𝑠𝑖𝑛 𝑒 𝑙2
𝑙

The most general solution of (1) is,

−𝛼 2𝑛 2 2 𝑡
∞ 𝑛𝑥
u(x, t) = ∑𝑛=1 𝐵𝑛𝑠𝑖𝑛 𝑒 𝑙2 (4)
𝑙

Using the initial condition (iii) in (4),

𝑛𝑥
u(x, 0) = ∑∞ 𝐵 𝑠𝑖𝑛 = 60x+20, for 0<x< l.
𝑛=1 𝑛 𝑙 𝑙

Thus is itself Fourier half-range sine series for 60x+20 in 0<x<l if

𝑙

2 𝑙 60 𝑛𝑥
Bn = ∫ ( x + 20) 𝑠𝑖𝑛 𝑑𝑥
𝑙 0 𝑙 𝑙

= 40 [1 + 4(−1)𝑛+1]
𝑛 

Substituting this value of Bn in (4).

 −𝛼 2𝑛 2 2 𝑡
∞
40 𝑛𝑥
u(x, t) = ∑𝑛=1 [𝑛[1 + 4(−1)𝑛+1]] 𝑠𝑖𝑛 𝑒 𝑙2
𝑙

Problem 6. A rod of length 20 cm has its ends A and B kept at temperature

30𝑜c and 90𝑜c respectively until steady state conditions prevail. If the
temperature at each end is then suddenly reduced to 0𝑜 c and maintained so,
find the temperature distribution at a distance from A at time ‘t’(N/D 2005)
Solution:
The P.D.E satisfied by u(x, t) is
 u 2u
2 2
t x (1)
2𝑝 2𝑡
u(x, t) = (Acospx+Bsinpx) 𝑒−𝛼 (2)
2𝑢
In steady-state, this equation reduces to 𝑑 = 0
𝑑𝑥2
Here l=20

Solving this, u = ax+b, where a and b are arbitrary constants.

u=30 at x=0 and u=90 at x=20.

 b=30 and 90= la+30
Solving we get a= 60
𝑙

Thus the temperature function in steady-state is

u(x) = 60x+30 (3)
𝑙

This temperature distribution reached at the steady state becomes initial

temperature distribution for the unsteady state. Then the temperature function u(x,
t) satisfies (1).
The new boundary conditions are
(i) u(0, t) = 0  t>0
(ii) u(l, t) = 0  t>0
(iii) u(x, 0) = 60 x +30for 0<x< l
𝑙

Using the boundary condition (i) in (2),

2𝑝 2𝑡
u(0, t) = A𝑒−𝛼 =0  t>0
A = 0

Using (ii) in (2),

2𝑝2𝑡
B sin l p 𝑒−𝛼 =0,  t>0
Since B0, sin lp=0.

Hence lp = n

i.e. p=𝑛, where n is any integer.
𝑙

Page 150
Therefore, (2) becomes,

−𝛼 2𝑛 2 2 𝑡
𝑛𝑥
u(x, t) = 𝐵𝑛𝑠𝑖𝑛 𝑒 𝑙2
𝑙

The most general solution of (1) is,

−𝛼 2𝑛 2 2 𝑡
∞ 𝑛𝑥
u(x, t) = ∑𝑛=1 𝐵𝑛𝑠𝑖𝑛 𝑒 𝑙2 (4)
𝑙

Using the initial condition (3) in (4),

𝑛𝑥
u(x, 0) = ∑∞ 𝐵 𝑠𝑖𝑛 = 60x+30, for 0<x< l.
𝑛=1 𝑛 𝑙 𝑙

Thus is itself Fourier half-range sine series for 60x+30 in 0<x<l if

𝑙

2 𝑙 60 𝑛𝑥
Bn = ∫ ( x + 30) 𝑠𝑖𝑛 𝑑𝑥 = 0 , n is odd
𝑙 0 𝑙 𝑙

= −120 , n is even
𝑛 

Substituting this value of Bn in (4).

−120 −𝛼 2𝑛 2 2 𝑡
u(x, t) = ∑∞ 𝑠𝑖𝑛 𝑛𝑥
𝑒 𝑙2
𝑛=2,4,6⋯ 𝑛 𝑙

Problem 7. The ends A and B of a rod l cm long have their temperatures kept at
300C and 800C until steady state conditions prevails. The temperature of the
end B is suddenly reduced to 600C and that of A is increased to 400C. Find the
temperature distribution in the rod after time‘t’. (May’07)

Solution:

The P.D.E satisfied by u(x, t) is

u 2u
2 2
t x (1)
2𝑢
In steady-state, this equation reduces to 𝑑 = 0
𝑑𝑥2

Solving this, u = ax+b, where a and b are arbitrary constants.

u=30 at x=0 and u=80 at x= l.

  b=20 and 40=10a+b

Solving we get a=2

Thus the temperature function in steady-state is

u(x) = 2x+20 (2)

When the temperatures at A and B are changed, the state is no longer steady. Then the
temperature function u(x, t) satisfies (1).

The boundary conditions in the second state are

(iv) u(0, t) = 50  t>0

(v) u(10, t) = 10  t>0
(vi) u(x, 0) = 2x+20 for 0<x<10

Consider, u(x, t) = us(x)+ut(x, t) (3)

where us(x) is a solution of (1) involving x only and satisfying the boundary conditions (i)
and (ii), ut(x, t) is a function defined by (3) and satisfying the equation (1).

Thus, us(x) is a steady-state solution of (1) and ut(x, t) may then be regarded as a transient
solution which decreases with increase of t; us(x) satisfies (1).
𝑑2𝑢𝑠
= 0, where u (0) = 50 and u (10) = 10.
s s
𝑑𝑥2

Solving, us(x) = ax+b.

us(0) = b = 50, using(i)

and us(10) = 10a+50 = 10, using (ii)

Hence a = -4.

Thus us(x) = 50-4x (4)

Consequently,
ut(0, t) = u(0, t)- us(0)=50-50=0 (iv)
ut(10, t) = u(10, t)- us(10)=10-10=0 (V)
 and ut(x, 0) = u(x, 0)- us(x)= (2x+20)-(50-4x)
i.e, ut(x, 0) = 6x-30 (vi)

Now, ut(x, t) also satisfies (1) and (iv), (v), (vi).

Solving (1) and selecting a suitable solution

ut (x, t) = (Acosx+Bsinx) C𝑒−𝛼  𝑡

2 2
(5)
Using the boundary condition (iv) in (5),

ut (0, t) = A𝑒−𝛼  𝑡=0  t>0

2 2

A = 0

Using (v) in (5),

Bsin10𝑒−𝛼  𝑡=0,  t>0

2 2

Since B0, sin10=0.

Hence 10 = n


i.e. =𝑛, where n is any integer.
10

Therefore, (5) becomes,

−𝛼 2𝑛 2 2 𝑡
𝑛 𝑥
ut(x, t) = 𝐵𝑛𝑠𝑖𝑛 𝑒 100
10

The most general solution of (1) is,

𝑛𝑥 −𝛼 2𝑛 2 2 𝑡
ut(x, t) = ∑∞ 𝐵 𝑠𝑖𝑛 𝑒 100 (6)
𝑛=1 𝑛 10

Using the initial condition (vi) in (6),

𝑛𝑥
ut(x, 0) = ∑∞𝑛=1 𝐵𝑛 𝑠𝑖𝑛 = 6x-30, for 0<x<10.
10

Thus is itself Fourier half-range sine series for 6x-30 in 0<x<10 if

2 10 𝑛𝑥
Bn = ∫ (6𝑥 − 30) 𝑠𝑖𝑛 𝑑𝑥
10 0 10
= −60 [1 + (−1)𝑛] = −120, for n even
𝑛 𝑛

= 0, for n odd.

Substituting this value of Bn in (6).




  −𝛼 2𝑛 2 2 𝑡
−120 𝑛 𝑥
ut(x, t) = ∑∞𝑛=2,4,6⋯ 𝑠𝑖𝑛 𝑒 100
𝑛 10
−60 ∞ 1 𝑛𝑥 −𝛼 2𝑛 2 2 𝑡

= ∑𝑛=1 𝑛 𝑠𝑖𝑛 𝑒 25
𝜋 5
 u(x, t) = us(x)+ut(x, t)
−60 ∞ 1 𝑛𝑥 −𝛼 2𝑛 2 2 𝑡

= 50-4x- ∑𝑛=1 𝑛 𝑠𝑖𝑛 𝑒 25

𝜋 5

Problem 8. The ends A and B of a rod l cm long have the temperature 40 0C and
900C until steady state prevails. The temperature at A is suddenly raised to
900C and at the same time that at B is lowered to 40 0C. Find the temperature
distribution in the rod at time t. Also show that the temperature at the mid point of the
rod remains unaltered for all time, regardless, of the material of the rod.
(Apr’03)

solution
The P.D.E satisfied by u(x, t) is
u 2u
2 2
t x (1)
2𝑢
In steady-state, this equation reduces to 𝑑 = 0
𝑑𝑥2

Solving this, u = ax+b, where a and b are arbitrary constants.

u=40 at x=0 and u=90 at x= l.

 b = 40 and 90 = l a + b
Solving we get a = 50
𝑙

Thus the temperature function in steady-state is

u(x) = 50x+40 (2)

𝑙

When the temperatures at A and B are changed, the state is no longer steady. Then the
temperature function u(x, t) satisfies (1).

The boundary conditions in the second state are

(vii) u(0, t) =  t>0

 (viii) u(10, t) = 10  t>0
(ix) u(x, 0) = 2x+20 for 0<x<10

Consider, u(x, t) = us(x)+ut(x, t) (3)

where us(x) is a solution of (1) involving x only and satisfying the boundary conditions (i)
and (ii), ut(x, t) is a function defined by (3) and satisfying the equation (1).

Thus, us(x) is a steady-state solution of (1) and ut(x, t) may then be regarded as a transient
solution which decreases with increase of t; us(x) satisfies (1).
𝑑2𝑢𝑠
= 0, where u (0) = 50 and u (10) = 10.
s s
𝑑𝑥2
Solving, us(x) = ax+b.
us(0) = b = 50, using(i)
and us(10) = 10a+50 = 10, using (ii)
Hence a = -4.

Thus us(x) = 50-4x (4)

Consequently,

ut(0, t) = u(0, t)- us(0)=50-50=0 (iv)

ut(10, t) = u(10, t)- us(10)=10-10=0 (v)
and ut(x, 0) = u(x, 0)- us(x)= (2x+20)-(50-4x)
i.e, ut(x, 0) = 6x-30 (vi)
Now, ut(x, t) also satisfies (1) and (iv), (v), (vi).

Solving (1) and selecting a suitable solution

2𝑝 2 𝑡
ut(x, t) = (Acospx+Bsinpx) C𝑒−𝛼 (5)

Using the boundary condition (iv) in (5),

2𝑝2𝑡
ut(0, t) = A𝑒−𝛼 =0  t>0
A = 0
Using (v) in (5),
2𝑝2𝑡
Bsin10p𝑒−𝛼 =0,  t>0
Since B0, sin10p= 0
 Hence 10p=n
i.e. p=𝑛, where n is any integer.
10

Therefore, (5) becomes,

−𝛼 2𝑛 2 2 𝑡
𝑛 𝑥
ut(x, t) = 𝐵𝑛𝑠𝑖𝑛 𝑒 100
10

The most general solution of (1) is,

𝑛𝑥 −𝛼 2𝑛 2 2 𝑡
ut(x, t) = ∑∞ 𝐵 𝑠𝑖𝑛 𝑒 100 (6)
𝑛=1 𝑛 10

Using the initial condition (vi) in (6),

𝑛𝑥
ut(x, 0) = ∑∞𝑛=1 𝐵𝑛 𝑠𝑖𝑛 = 6x-30, for 0<x<10.
10

Thus is itself Fourier half-range sine series for 6x-30 in 0<x<10 if

2 10 𝑛𝑥
Bn = ∫ (6𝑥 − 30) 𝑠𝑖𝑛 𝑑𝑥
10 0 10
= −60 [1 + (−1)𝑛] = −120, for n even
𝑛 𝑛

= 0, for n odd.

Substituting this value of Bn in (6).


  −𝛼 2𝑛 2 2 𝑡
−120 𝑛 𝑥
ut(x, t) = ∑∞𝑛=2,4,6⋯ 𝑠𝑖𝑛 𝑒 100
𝑛 10
−60 ∞ 1 𝑛𝑥 −𝛼 2𝑛 2 2 𝑡

= ∑𝑛=1 𝑛 𝑠𝑖𝑛 𝑒 25
𝜋 5
 u(x, t) = us(x)+ut(x, t)
−60 ∞ 1 𝑛𝑥 −𝛼 2𝑛 2 2 𝑡
= 50-4x- ∑𝑛=1 𝑛 𝑠𝑖𝑛 𝑒 25
𝜋 5

Problem 9. A metal bar 10cm long with insulated sides has its ends A and B kept at
200C and 400C respectively, until steady state conditions prevail. The temperature at
A is then suddenly raised to 50 0C and at the same instant that at B is lowered to
100C. Find the subsequent temperature at any point at the bar at any time.
(Nov/Dec 2005)
Solution:

The P.D.E satisfied by u(x, t) is

u 2u
2 2
t x (1)
2𝑢
In steady-state, this equation reduces to 𝑑 = 0
𝑑𝑥2

Solving this, u = ax+b, where a and b are arbitrary constants.

u=20 at x=0 and u=40 at x=10.
 b=20 and 40=10a+b
Solving we get a=2
Thus the temperature function in steady-state is
u(x) = 2x+20 (2)
When the temperatures at A and B are changed, the state is no longer steady. Then
the temperature function u(x, t) satisfies (1).
The boundary conditions in the second state are
(x) u(0, t) = 50  t>0
(xi) u(10, t) = 10  t>0
(xii) u(x, 0) = 2x+20 for 0<x<10
Consider, u(x, t) = us(x)+ut(x, t) (3)
where us(x) is a solution of (1) involving x only and satisfying the boundary
conditions (i) and (ii), ut(x, t) is a function defined by (3) and satisfying the
equation (1).
Thus, us(x) is a steady-state solution of (1) and ut(x, t) may then be regarded as a
transient solution which decreases with increase of t; us(x) satisfies (1).
𝑑2𝑢𝑠
= 0, where u (0) = 50 and u (10) = 10.
s s
𝑑𝑥2
Solving, us(x) = ax+b.
us(0) = b = 50, using(i)
and us(10) = 10a+50 = 10, using (ii)
Hence a = -4.
Thus us(x) = 50-4x (4)
Consequently,
ut(0, t) = u(0, t)- us(0)=50-50=0 (iv)
ut(10, t) = u(10, t)- us(10)=10-10=0 (V)
and ut(x, 0) = u(x, 0)- us(x)= (2x+20)-(50-4x)
i.e, ut(x, 0) = 6x-30 (vi)
Now, ut(x, t) also satisfies (1) and (iv), (v), (vi).
Solving (1) and selecting a suitable solution
ut (x, t) = (Acosx+Bsinx) 𝑒−𝛼  𝑡
2 2
(5)
Using the boundary condition (iv) in (5),
ut (0, t) = A𝑒−𝛼  𝑡=0  t>0
2 2

A = 0
Using (v) in (5),

Bsin10𝑒−𝛼  𝑡=0,  t>0

2 2

Since B0, sin10=0.

Hence 10=n
i.e. =𝑛, where n is any integer.
10

Therefore, (5) becomes,

−𝛼 2𝑛 2 2 𝑡
𝑛 𝑥
ut(x, t) = 𝐵𝑛𝑠𝑖𝑛 𝑒 100
10

The most general solution of (1) is,

𝑛𝑥 −𝛼 2𝑛 2 2 𝑡
ut(x, t) = ∑∞ 𝐵 𝑠𝑖𝑛 𝑒 100 (6)
𝑛=1 𝑛 10

Using the initial condition (vi) in (6),

𝑛𝑥
ut(x, 0) = ∑∞𝑛=1 𝐵𝑛 𝑠𝑖𝑛 = 6x-30, for 0<x<10.
10

Thus is itself Fourier half-range sine series for 6x-30 in 0<x<10 if

2 10 𝑛𝑥
Bn = ∫ (6𝑥 − 30) 𝑠𝑖𝑛 𝑑𝑥
10 0 10
= −60 [1 + (−1)𝑛] = −120, for n even
𝑛 𝑛

= 0, for n odd.
Substituting this value of Bn in (6).
 −𝛼 2𝑛 2 2 𝑡
−120 𝑛 𝑥
ut(x, t) = ∑∞𝑛=2,4,6⋯ 𝑠𝑖𝑛 𝑒 100
𝑛 10
−60 ∞ 1 𝑛𝑥 −𝛼 2𝑛 2 2 𝑡

= ∑𝑛=1 𝑛 𝑠𝑖𝑛 𝑒 25
𝜋 5
 u(x, t) = us(x)+ut(x, t)
−60 ∞ 1 𝑛𝑥 −𝛼 2𝑛 2 2 𝑡

= 50-4x- ∑𝑛=1 𝑛 𝑠𝑖𝑛 𝑒 25

𝜋 5
 u 2u
Problem 10. Solve the heat flow equation 2 subject to the following
t x 2

boundary conditions

(i) u(0,t) = 0
(ii) u(l,t) = 0
(iii) u(x,0) = x
Solution:

The solution of one dimensional heat equation is given by

u = (Acospx+Bsinpx)Ce -  p t
2 2
(1)

Using boundary conditions (i) and (ii) in (1), we get,

A = 0, p = nπ/l

Substituting these values in (1) we get

The most general solution of (1) is

 2 p2 2t

nx
u = bn sin e l2
(2)
n1
l

Applying the condition (iii) in (2), we get


nx
x= b
n1
n
sin
l

The R.H.S is Half range sine series

2l nx 2l nx
bn  
l 0
f (x) sin
l
dx = 
l 0
x sin
l
dx

2
= l(1) n

n



   2 p2 2t

 2l nx
The solution is u =  (1)n sin e l2

n1 n l