Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
246 views

1 Laplace Transform of Periodic Function

1. The document discusses Laplace transforms of periodic functions. It provides an expression for the Laplace transform of a periodic function in terms of its period. Examples are provided to illustrate this formula. 2. Convolution is introduced, which relates the inverse Laplace transform of the product of two Laplace transforms to the convolution of the original functions. The convolution theorem is stated. 3. Several examples of applications to differential equations and integral equations are worked out using Laplace transforms and convolution. Periodic functions, non-homogeneous terms, and variable coefficients are handled.

Uploaded by

asbadg
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
246 views

1 Laplace Transform of Periodic Function

1. The document discusses Laplace transforms of periodic functions. It provides an expression for the Laplace transform of a periodic function in terms of its period. Examples are provided to illustrate this formula. 2. Convolution is introduced, which relates the inverse Laplace transform of the product of two Laplace transforms to the convolution of the original functions. The convolution theorem is stated. 3. Several examples of applications to differential equations and integral equations are worked out using Laplace transforms and convolution. Periodic functions, non-homogeneous terms, and variable coefficients are handled.

Uploaded by

asbadg
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

S.

Ghorai

Lecture XIX
Laplace Transform of Periodic Functions, Convolution, Applications

Laplace transform of periodic function

Theorem 1. Suppose that f : [0, ) R is a periodic function of period T > 0, i.e.


f (t + T ) = f (t) for all t 0. If the Laplace transform of f exists, then
T

f (t)est dt
0

F (s) =

1 esT

(1)

Proof: We have

f (t)est dt

F (s) =
0

(n+1)T

f (t)est dt

=
nT

n=0

f (u + nT )esusnT du

=
n=0

esnT

u = t nT

f (u)esu du
0

n=0

T
su

f (u)e

esnT

du

n=0

f (u)esu du
=

1 esT

The last line follows from the fact that

esnT
n=0

is a geometric series with common ration esT < 1 for s > 0.


Example 1. Consider f (t) = sin(t), which is a periodic function of period 2/.
Solution: Using (1), we find
F (s) =

1
1 e2s/

2/

est sin(t) dt =
0

1 e2s/

= 2
2
2
2s/
s + 1e
s + 2

Example 2. Consider a saw-tooth function (see Figure 1)


f (t) =

t,
0 t < 1,
f (t 1),
t 1.

S. Ghorai

Figure 1: A saw-tooth function.


Solution: Here period T = 1. Using (1), we find
1
F (s) =
1 es

test dt =
0

1 es (1 + s)
1
es
=

s2 (1 es
s2 s(1 es )

Example 3. Consider the following function (see Figure 2)

0 t < 1,
t,
2 t,
1 t < 2,
f (t) =

f (t 2),
t 2.

f(t)

1
o

Figure 2: A saw-tooth function.


Solution: Here f (t) is a periodic function of period T = 2. Hence, using (1), we find
1
F (s) =
1 e2s

2
st

f (t)e
0

1
dt =
1 e2s

te

st

(2 t)est dt

dt +
1

Simplifying the RHS, we find


F (s) =
Aliter: Note that

(1 es )2
1 es
1
=
= 2 tanh(s/2)
2
2s
2
s
s (1 e )
s (1 + e )
s

1,
0 < t < 1,
1,
1 < t < 2,
f (t) =

f (t 2),
t > 2.

S. Ghorai

Since f is piecewise continuous and is of exponential order, its Laplace transform exist.
Also, f is periodic with period T = 2. Hence,
1
=
1 e2s

L f

f (t)e

st

1
dt =
1 e2s

2
st

est dt

dt +

(1 es )2
.
s(1 e2s )

Hence,
L f

1
1
1
tanh(s/2) = sF (s) f (0) = tanh(s/2) = F (s) = 2 tanh(s/2)
s
s
s

Comment: Is it possible to do similar calculations (like in aliter) in Example 2? If


not, why not?

Convolution

Suppose we know that a Laplace transform H(s) can be written as H(s) = F (s)G(s),
where L f (t) = F (s) and L g(t) = G(s). We need to know the relation of h(t) =
L1 H(s) to f (t) and g(t).
Definition 1. (Convolution) Let f and g be two functions defined in [0, ). Then
the convolution of f and g, denoted by f g, is defined by
t

(f g)(t) =

f ( )g(t ) d

(2)

Note: It can be shown (easily) that f g = g f . Hence,


t

(f g)(t) =

g( )f (t ) d

(3)

We use either (2) or (3) depending on which is easier to evaluate.


Theorem 2. (Convolution theorem) The convolution f g has the Laplace transform property
L (f g)(t) = F (s)G(s).
(4)
OR conversely
L1 F (s)G(s) = (f g)(t)
Proof: Using definition, we find

L (f g)(t)

(f g)(t)est dt

=
0

f ( )g(t ) d

=
0

est dt

The region of integration is the area in the first quadrant bounded by the t-axis and

S. Ghorai

=t

0110
1 1
0
10
1010
2
00000000
11111111
000000001010
11111111
0000000010
11111111

Figure 3: Effects of unit step function on a function f (t). Here b > a.


the line = t. The variable limit of integration is applied on which varies from = 0
to = t.
Let us change the order of integration, thus apply variable limit on t. Then t would
vary from t = to t = and would vary from = 0 to = . Hence, we have

L (f g)(t)

est g(t ) dt f ( ) d

=
0

esu g(u) du f ( )es d,

=
0

t =u

es f ( ) d

esu g(u) du

= F (s)G(s)
Example 4. Consider the same problem as given in Example 4 of Lecture Note 18,
i.e. find inverse Laplace transform of 1/s(s + 1)2 .
Solution: We write H(s) = F (s)G(s), where F (s) = 1/s and G(s) = 1/(s + 1)2 . Thus
f (t) = 1 and g(t) = tet . Hence, using convolution theorem, we find
t

e d = 1 (t + 1)et .

f (t )g( ) d =

h(t) =
0

Note: We have used f (t )g( ) in the convolution formula since f (t) = 1. This helps
a little bit in the evaluation of the integration.
Example 5. Find inverse Laplace transform of 1/(s2 + 2 )2 .
Solution: Let H(s) = F (s)G(s), where F (s) = 1/(s2 + 2 ) and G(s) = 1/(s2 + 2 ).

S. Ghorai

Thus, f (t) = sin(t)/ = g(t). Hence,


t
1
sin( ) sin((t ))d
2 0
1
sin(t) t cos(t) .
=
2 3

h(t) =

Applications

Example 6. (Differential equation) Solve the IVP


y + y = t,

y(0) = 0, y (0) = 2

Solution: Take Laplace transform on both sides. This gives


s2 Y 2 + Y =

1
2
1
=
Y
=
+
s2
s2 (s2 + 1) s2 + 1

Using partial fraction, we find


Y =

1
1
+
= y(t) = t + sin t
s2 s2 + 1

Aliter: In the method above, we evaluated Laplace transform of the nonhomogeneous


term in the right hand side. Now here we dont evaluate it. Let g(t) be nonhomogeneous
term (in this case g(t) = t). Let G(s) be the Laplace transform of g. Now Take Laplace
transform on both sides. This gives
s2 Y 2 + Y = G(s) = Y =

2
G(s)
+ 2
2
s +1 s +1

Taking inverse transform and convolution, we find


t

(t ) sin( ) dt + 2 sin t

g(t ) sin( ) dt + 2 sin t = y(t) =

y(t) =
0

OR (using integration by parts)


y(t) = t + sin t
Example 7. (Differential equation) Solve the IVP
y + 9y =

8 sin t,
0,

0 < t < ,
t > ,

y(0) = 0, y (0) = 4.

Solution: Consider g(t) = 8 u0 (t) u (t) sin t. Then Laplace transform of the
nonhomogeneous term is the same as that of g(t). Now we write g(t) as
g(t) = 8u0 (t) sin t + +8u (t) sin(t ).

S. Ghorai

Now taking Laplace transform of the ODE, we get


s2 Y 4+9Y =

1
8
es
4
1
s
+8
=
Y
=
+8
+8e
s2 + 1 s2 + 1
s2 + 9 (s2 + 1)(s2 + 9)
(s2 + 1)(s2 + 9)

Using partial fraction, we get


Y =

s2

4
+
+9

s2

1
1
2
+1 s +9

Now

1
sin 3t
3

L sin t

+ es

s2

1
1
2
+1 s +9

1
1

.
s2 + 1 s2 + 9

Hence, using shifting theorem and inverse transform, we find


y(t) =

4 sin 3t
1
1
+ sin t sin 3t + u (t) sin(t ) sin 3(t )
3
3
3

Further, this can be break up as

sin 3t + sin t,
y(t) =
4 sin 3t,
3

0 t ,
t ,

Example 8. (Differential equation) (Variable coefficient) Solve the IVP


y 2xy + 4y = 0,

y(0) = 1, y (0) = 0

Solution: Take Laplace Transform on both sides, we find


s2 Y sy(0) y (0) + 2

d
L (y ) + 4Y = 0,
ds

OR
s2 Y s+2

d
s 3
sY y(0) +4Y = 0 = 2sY +(s2 +6)Y = s = Y +
+
ds
2 s

This is linear equation. Hence,


2 /4

Y s3 es
OR

1
2

s3 es

2 /4

ds + C
2

s2 4
es /4
Y =
+
C
s3
s3

OR

1
4
es /4
Y = 3 +C 3 .
s s
s
Now it can be shown by Bromwich integral method (not in the syllabus) that
L

x2 1

2
4

es /4
s3

Y =

1
2

S. Ghorai

Hence, we find
y(t) = (1 2x2 ) + C

x2 1

2
4

OR
y(t) = (1 C/4) + (C/2 2)x2
Now y(0) = 1 = C = 4. Hence
y(x) = (1 2x2 )
Comment: If we expand es

2 /4

/s3 then we find

es /4
1
1
+ non-negative power of s.
=

s3
s3 4s
If we assume L1 (sk ) = 0, k = 0, 1, 2, , then we find
L

x2 1

2
4

es /4
=
s3

Example 9. (Integral equation) Solve


t

y(t )e2 d = 1,

y +

y(0) = 1.

Solution: Take Laplace Transform on both sides, we find


sY y(0) +

Y
1
s+2
2
1
=
= Y =
= Y =
s+2
s
s(s + 1)
s s+1

Hence,
y(t) = 2 et

You might also like