Laplace Transformation

1 Def of the Laplace Transform

Transforms: transform a function into another function.

Differentiation and integration are transforms.

Laplace Transform: special type of integral transform.

We’re working with functions of two variables?

e.g f (x , y )

when we take a definite integral (a single integral), we get a function of only one of
the variables (the one we held constant).

e.g.

[ ]
2 2
x2 y2 4y
2 2
y 3y
2

∫ x y
2
dx=¿
2 1
¿ ¿
2
−
2
=
2
1

So in general, a definite integral

∫ K ( s , t ) f ( t ) dt
b

Transforms a function f of the variable t into a function F(s) of the variable s.

Interested in an integral transform (maps a function to another function under the

process of integration) – where the interval of integration is an unbounded interval ¿

If f (t) is defined for t ≥ 0 then the improper integral ∫ K ( s , t ) f ( t ) dt is defined as a limit:

0

∞ b

∫ K ( s , t ) f ( t ) dt=lim
b→∞
∫ K ( s , t ) f ( t ) dt ,(1)
0 0
If the limit exists, call the integral convergent. If not, it is divergent.

In general, the integral will only exist for some s.

Def: Kernel

The function K ( s , t) is called the kernel of the transform. (why?)

The choice K ( s ,t )=e−st gives us the Laplace transform.

Def 7.1.1 Laplace Transform

Let f be a function defined for t ≥ 0 then the integral

∞
L { f ( t ) }=∫ e
− st
f ( t ) dt ,(2)
0

Is said to be the Laplace transform of f, provided that the integral converges.

If the integral converges – result is a function of s.

Use lowercase to denote the function being transformed – uppercase to denote the
function that results.

Remember, we use the notation L{f(t)}, to emphasize that the Laplace transform is
an operator that acts on a function – rather than a simple function itself.

Operator – transforms a function into another function.

Typically - transforms a time-domain function, into a frequency domain

representation.

Example 1:

Evaluate L {1 }

∞ b
L { 1 }=∫ e dt¿ lim ∫ e
−st −st
dt
0 b→∞ 0
First compute,

1 −st 1 u 1 −ts
−st
∫e dt let u=−st Then du=dt⇒ ∫ e dt = ∫ e du¿ e
−s −s −s

Compute new bounds:

At t = 0, u=−0

And t=b , u=−s ⋅ b

So we have:

[ ] ( )
b − sb 2

1 −ts 1 −(−sb ) s 1 1 s b 1 1−e s b

∫e
2
−st
dt = e ¿ e − ¿ e + ¿
0 −s 0 −s −s −s s s

Now evaluate:

[ ]
2 2 2 2
s b
1−e s b 1−e s ⋅ eb 1 e ⋅e 1 es
lim =lim ¿ lim − ¿ − ⋅ lim eb
b→∞ s b→∞ s b→∞ s s s s b→∞

Their solution:

Evaluate L { 1 }

∞
L { f ( t ) }=∫ e
− st
f ( t ) dt
0

∞ b
¿ L {1 }=∫ e dt =lim ∫ e
−st − st
dt
0 b→∞ 0

If we use a change of variables:

du 1
u=−st =−s du=dt
dt −s

If t = b

U = -sb
 | ( ) ( ) ( )
b −st b −sb 0 −sb
1 1 1
1 u −e −e 1 e
¿−∫ e du ¿ lim −e ¿ lim − ¿ lim − ¿ ⋅ lim 1− sb ¿ , s >0
0 −s b→∞ s 0 b→∞ s s b→∞ s s s b→∞ e s

Example 2:

Evaluate L {t }

∞ ∞
L { t } =∫ e ⋅t ⋅dt¿ lim ∫ e ⋅t ⋅ dt Need to evaluate
−st −st

0 b→∞ 0

−st
∫ e ⋅ t ⋅dt

d d d
( h ⋅ g )= ( h ) ⋅ g+ ( g ) ⋅ h
dt dt dt

∫
d
dt [
( h ⋅ g ) dt = ∫
d
dt
d
]
( h ) ⋅ g+ ( g ) ⋅h dth ⋅ g= ∫
dt ([ dhdt ) ⋅g ] dt+ ∫ [( dgdt ) ⋅h] dt
Our possibilities are:

∫
[( ) ]
dh
dt
⋅ g dt=h⋅ g−∫
[( ) ]
dg
dt
⋅h dt ,∨¿ ∫
dg
dt [( ) ]
⋅ h dt =h ⋅ g− ∫
dh
dt
⋅ g dt
[( ) ]
Possibilities:

 We have g and the derivative of h – and need to recover h (by finding its
primitive)
 We have h and the derivative of g – and need to recover g (by finding its
primitive)

Let g=t

And assume h' t=e−st

To recover h,

−st
h=∫ e dt
 du 1
Let , u=−st =−s du=dt
dt −s

Then

−st 1 u −1 u −1 −st
h=∫ e dt=− ∫ e du= e= e
s s s

Also,

dg
=1
dt

And we get

∫
[( ) ]
dh
dt
⋅ g dt=h⋅ g−∫
[( ) ]
dg
dt
⋅h dt ¿ ∫ e−st ⋅t ⋅dt=
−1 −st
s
1 −st 1 −st 1 −st
e ⋅t+ ∫ e dt¿− e ⋅t+ ∫ e dt
s s s

|
b
1 −st 1 1 −st 1 1 t 1 b 1 b
¿− ⋅ e ⋅t+ ⋅ L { 1 }¿ ⋅e ⋅t+ ⋅ but , lim ⋅ st ¿ lim = ⋅ lim sb =0
s s s s s b→∞ s e
0
b→∞ s ⋅ e
sb
s b→∞ e

So,

1
L {t }= 2
s

Now

)| ¿ lim (e ⋅( 1s − s1 )−( 1s − s1 ))¿ lim (( 1s − s1 ) ⋅ (e )

∞ b

¿ lim ∫ e
b→∞ 0
−st −st
b →∞ (
1 1
⋅t ⋅ dt=lim e ⋅ − 2
s s 0
b→0
− sb
2 2
b→0
2
−sb
−1 )

Example 4:

Evaluate L { sin 2t }

∞
L { sin 2t }=∫ e
− st
⋅sin ( 2 t ) ⋅ dt
0

Use integration by parts:

 −st
−e d
∫e
−st
dt= , afte using a substitutuion. sin ( 2t )=2 ⋅cos ⁡(2 t)
s dt

Then integration by parts gives:

∞ −st ∞ −st
∫ e ⋅ sin ( 2t ) ⋅ dt=−−es
−st
⋅sin ( 2 t )−∫ 2 cos ( 2 t ) ⋅
−e
s
dt
0 0

| ( |)
−st ∞ ∞ −st b ∞
−e 2 −e 2
⋅sin ( 2 t ) + ∫ e ⋅cos ( 2 t ) dt¿ lim ⋅sin ( 2 t ) + ∫ e ⋅cos (2 t ) dt
−st −st
¿
s 0 s 0 b→∞ s 0 s 0

∞ ∞
¿ lim
(−1
b → ∞ s ⋅e
sb
2 −st
) 2 2
⋅sin ( 2 b ) + ∫ e ⋅cos ( 2 t ) dt ¿ 0+ ∫ e− st ⋅cos ( 2t ) dt ¿ ⋅ L { cos ( 2 t ) } , s>0
s 0 s 0 s

Evaluating L{cos(2t)}, using integration by parts:

|
−st ∞ − st −st ∞ ∞
−e −e
L { cos ( 2 t ) }= ⋅ cos ( 2 t )−∫ ⋅−2sin ( 2 t ) dt ¿ −e 2
⋅cos ( 2t ) − ⋅∫ e ⋅sin ( 2t ) dt
−st
s 0 s s 0 s 0

|
∞
−e
−st
2 1 2
¿ ⋅cos ( 2t ) − ⋅ L {sin ( 2 t ) }¿ − ⋅ L { sin ( 2 t ) }
s 0 s s s

Substituting into the first equation, we get:

L { sin ( 2t ) } =
2 1 2
( ) 2 4
− ⋅ L { sin ( 2 t ) } L { sin ( 2t ) } = − ⋅ L { sin ( 2 t ) }
s s s s s
2
L { sin ( 2t ) } = ⋅ ( 1−2 ⋅ L{sin ( 2t ) })
s

Use algebra to find:

2
L { sin ( 2t ) } = 2
, s> 0
s +4

Linear Transform:

For a linear combination of functions, we have:

∞ ∞ ∞

∫ e ⋅ ( αf ( t ) + βg ( t ) ) dt=α∫ e
−st −st
f ( t ) dt+ β ∫ e
−st
g ( t ) dt
0 0 0
Whenever both integrals converge for s> c.

Hence

L { αf ( t ) + βg ( t ) }=α L { f (t ) } + β { g ( t ) } ,(3)

L is said to be a liner transform.

Theorem 7.1.1 Transforms of some basic functions

1 n! 1 d . L { sin kt }= k e . L { cos kt }= s
a . L {1 }= b . L { t } = n+1 , n=1 ,2 , 3 , …c . L { e } =
n at
2 2 2 2
s s s−a s +k s +k
k s
f . L { sinh kt }= 2 2
g . L { cosh kt }= 2 2
s −k s −k

Def: Sufficient conditions for existence of L{f(t)}

The integral that defines the Laplace transform does not have to converge?

The sufficient conditions that guarantee the existence of L { f ( t ) } are that:

1. F be piece-wise continuous on [a, ∞); and

2. That f be of exponential order for t > T

Def: Exponential Order:

A function f is said to be of exponential order if there exist constants

1. c
2. M > 0
3. T > 0

s.t |f ( t )|≤ M e ct for all t > T

i.e.

if ∃c, M, T s.t. ∀t > T, |f ( t )|≤ M e

ct
in other words, if f is an increasing function, the condition ¿ f ( t )∨≤ M ect , just means
that the function f does not grow faster than the exponential function M ect , where c is
a positive constant.

Theorem 7.1.2. Sufficient conditions for existence

If f is piecewise continuous on [0, ∞) and of exponential order, then ℒ{f(t)} exists, for
some c > 0.

Theorem 7.1.3

If f is

1. piecewise continuous on [ 0 , ∞ ) and

2. of exponential order
3. and F ( s )=L{f ( t ) }

Then slim F ( s )=0

→∞