Laplace Transforms

Part-II

Pradeep Boggarapu

December 1, 2022

Pradeep Boggarapu Laplace Transforms December 1, 2022 1 / 24

 Application to Differential Equations:
Laplace transforms of derivatives: Suppose we wish to find the particular
solution of the differetial equation
y 00 + ay 0 + by = f (x) (0.1)
that satisfies the initial conditions y (0) = y0 and y 0 (0) = y00 .
The use of Laplace transforms provides an alternative way of attacking
this problem that has several advantanges. We will prove the following
propostions
Proposition. If f (x) is a continuously differentiable function of
exponential order defined on [0, ∞), then
L[f 0 (x)] = pL[f (x)] − f (0). (0.2)
Proof. An integration by parts gives
Z ∞  ∞ Z ∞
0 −px 0 −px
L[f (x)] = e f (x) dx = e f (x) +p e −px f (x) dx
0 0 0
= −f (0) + pL[f (x)],

Pradeep Boggarapu Laplace Transforms December 1, 2022 2 / 24

here in obtaining the above, we used that

lim f (x)e −px = 0

x→∞

which is true for all large values of p, as the function f (x) is of exponential
order. Thus
L[f 0 (x)] = pL[f (x)] − f (0).
Now we will see that how the Laplace transform will be used to solve the
initial value problem (0.1), let us apply Laplace transform on both sides of
(0.1):
L[y 00 ] + aL[y 0 ] + bL[y ] = L[f (x)]. (0.3)
Here we assume that the function f (x) is piecewise continuous and of
exponential order. Once this assumption is made, then it can be shown
(we omit the proof) that y , y 0 and y 00 necessarly have the same
properties, so they have Laplace transforms.

Pradeep Boggarapu Laplace Transforms December 1, 2022 3 / 24

Now we use the formula (0.2) including the initial conditions to write the
Laplace transforms L[y 0 ] and L[y 00 ] in terms of L[y ],
L[y 0 ] = pL[y ] − y0 (0.4)
00 2
L[y ] = p L[y ] − py0 − y00 . (0.5)
If we now substitute these expressions in (0.3), we get
p 2 L[y ] − py0 − y000 + apL[y ] − apL[y ] − ay0 + bL[y ] = L[f (x)];
and solving for L[y ] yields
L[f (x)] + (p + a)y0 + y00
L[y ] = . (0.6)
p 2 + ap + b
The function f (x) is known, so its Laplace transform L[f (x)] is a specific
function of p; and since a, b, y0 and y00 are known constants, L[y ] is
completely known as functions of p.
If we can now find which function y (x) has the right side of the above
equation as its Laplace transform, then this function will be the solution of
our initial value problem. For this, we need inverse Laplace transforms.
Pradeep Boggarapu Laplace Transforms December 1, 2022 4 / 24
Example-1: Find the Laplace tranform of y that satifies the initial value
problem
y 00 + y 0 = 3x 2 , y (0) = 0 and y 0 (0) = 1.
Sol. Compare the given equation with
y 00 + ay 0 + by = 0; y (0) = y0 y 0 (0) = y00 , then a = 1, b = 0, f (x) = 3x 2
y0 = 0 and y00 = 1. We substitute these values in (0.6), then we get

L[3x 2 ] + (p + 1) · 0 + 1 3(2!/p 3 ) + 1 6 + p3
L[y ] = = = .
p2 + 1 · p + 0 p2 + p p 4 (1 + p)

Inverse Laplace transform: If F (p) is the Laplace tranform of f (x) then

we can call f (x) as the inverse Laplace transform of F (p). We denote the
inverse Laplace transform by L−1 and we write it as:

L−1 [F (p)] = f (x) if F (p) = L[f (x)].

Note that inverse Laplace transform of given F (p) may or may not exists,
if it exists, it may not be unique but continuous inverse Laplace transform
is unique.
Pradeep Boggarapu Laplace Transforms December 1, 2022 5 / 24
Example-2: Find the inverse Laplace transforms of
p 2
(a) p 2 +4
(b) p 2 +4

1 6+p 3
(c) pn (d) p 4 (1+p)
.

Sol.  
−1 p p
(a). L 2
= cos 2x; since L[cos 2x] = 2 .
p +4 p +4
 
2 2
(b). L−1 = sin 2x; since L[sin 2x] = 2 .
p2 + 4 p +4

x n−1
 
−1 1 n!
(c). L n
= ; since L[x n ] = n+1 .
p (n − 1)! p

Pradeep Boggarapu Laplace Transforms December 1, 2022 6 / 24

 6 + p3
(d). We first decompose the rational function into partial
p 4 (1 + p)
fractions:
6 + p3 6 6 6 5 5
4
= 4− 3+ 2− + .
p (1 + p) p p p p p+1
Now if we apply inverse Laplace transform on both sides, then we get

6 + p3
   
−1 −1 6 6 6 5 5
L =L − + − +
p 4 (1 + p) p4 p3 p2 p p + 1
         
−1 6 −1 6 −1 6 −1 5 −1 5
=L −L +L −L +L
p4 p3 p2 p p+1
= x 3 − 3x 2 + 6x − 5 + 5e −x .

Pradeep Boggarapu Laplace Transforms December 1, 2022 7 / 24

We shall considere a number of general properties of Laplace transforms
that are greatly useful. The first of these is the shifting formula:
Shifting formula: If F (p) is the Laplace transform of f (x) then for a ∈ R

L[e ax f (x)] = F (p − a) and L−1 [F (p + a)] = e −ax L−1 [F (p)].

Since
Z ∞
ax
L[e f (x)] = e −px e ax f (x) dx
Z0 ∞
= e −(p−a)x f (x) dx = F (p − a).
0

The formula can be used to find the transforms of product of the form
e ax f (x) when F (p) = L[f (x)] is known and also to find the inverse
transforms of functions of the form F (p + a) when L−1 [F (p)] = f (x) is
known.

Pradeep Boggarapu Laplace Transforms December 1, 2022 8 / 24

Example-3: Find the Laplace transforms of
(a) x 5 e −2x (b) (1 − x 2 )e −x
(c) e 3x cos 2x (d) e ax sin bx.
Sol. We first recall the shifting formula: L[e ax f (x)](p) = L[f (x)](p − a)
5!
(a). L[x 5 e −2x ](p) = L[x 5 ](p + 2) = .
(p + 2)6
1 2
(b). L[(1 − x 2 )e −x ](p) = L[(1 − x 2 )](p + 1) = − .
p + 1 (p + 1)3
p−3
(c). L[e 3x cos 2x] = L[cos 2x](p − 3) = .
(p − 3)2 + 4
2
(d). L[e ax sin bx](p) = L[sin bx](p − a) = .
(p − a)2 + b 2

Pradeep Boggarapu Laplace Transforms December 1, 2022 9 / 24

Example-4: Find the inverse Laplace transforms of
6 12 p+3
(a) (b) (c) .
(p + 2)2 + 9 (p + 3)4 p2 + 2p + 5
Sol. We first recall the shifting formula for inverse Laplace transform
L−1 [F (p + a)] = e −ax L−1 [F (p)].

   
−1 6 −2x −1 6
(a). L =e L = e −2x sin 3x.
(p + 2)2 + 9 p2 + 9
   
−1 12 −3x −1 12
(b). L =e L = 2x 3 e −3x .
(p + 3)4 p4
(c).
     
p+3 (p + 1) + 2 −x −1 p + 2
L−1 2 = L−1 = e L
p + 2p + 5 (p + 1)2 + 4 p2 + 4
  
p 2
= e −x L−1 2 + 2 = e −x (cos 2x + sin 2x).
p +4 p +4

Pradeep Boggarapu Laplace Transforms December 1, 2022 10 / 24

Laplace transforms of Integrations: If f (x) is an integrable function on
[0, ∞) and F (p) is its Laplace transform, then show that
Z x 
F (p)
L f (t) dt = ,
0 p
 
−1 1
and verify this formula by finding L in two ways.
p(p + 1)
Z x
Sol. Let y (x) = f (t) dt, then we can verify that y 0 (x) = f (x) and
0
y (0) = 0. Now we use the formula L[y 0 (x)] = pL[y (x)] − y (0),
Z x 
1 1 F (p)
L f (t) dt = L[y (x)] = (L[y 0 (x)] + y (0)) = L[f (x)] = .
0 p p p
   
−1 1 −1 1 1
L =L − = x − e −x or
p(p + 1) p p+1
  Z x
−1 1/(p + 1)
=L = e −t dt = x − e −x
p 0

Pradeep Boggarapu Laplace Transforms December 1, 2022 11 / 24

Derivatieves of Laplace transforms: If f (x) is such that x k f (x) has Laplace
transforms for all k = 0, 1, 2, . . . n and F (p) is its Laplace tranform then
L[(−1)n x n f (x)] = F (n) (p) (0.7)
Proof. Consider the general Laplace transform formula
Z ∞
F (p) = e −px f (x) dx.
0
The differentiation of this with respect to p under the integral sig can be
justified, and yields
Z ∞
F 0 (p) = e −px (−x)f (x) dx (0.8)
0
or
L[−xf (x)] = F 0 (p)
By differentiating (0.8), we find that L[x 2 f (x)] = F 00 (p) and more
generally, that
L[(−1)n x n f (x)] = F (n) (p)

Pradeep Boggarapu Laplace Transforms December 1, 2022 12 / 24

Example-1: Since L[sin ax] = a/(p 2 + a2 ), we have
 
d d a 2ap
L[x sin ax] = − L[sin ax] = − 2 2
= 2 .
dp dp p + a (p + a2 )2

d2
 2
a − 3p 2
  
2 d 2ap
L[x sin ax] = 2 L[sin ax] = − = −2a
dp dp (p 2 + a2 )2 (p 2 + a2 )3
Exampl-2: We have shown that L[x −1/2 ] = π/p, so
p

r
1/2 −1/2 d p
1 π
L[x ] = L[x(x )] = − π/p = .
dp 2p p

d2 p
r
3/2 2 −1/2

3 π
L[x ] = L[x (x )] = 2 π/p = 2 .
dp 4p p

Pradeep Boggarapu Laplace Transforms December 1, 2022 13 / 24

Integration of Laplace transforms: If F (p) is the Laplace transform of f (x)
then   Z ∞
f (x)
L = F (q) dq. (0.9)
x p

And hence Z ∞ Z ∞
f (x)
dx = F (p) dp (0.10)
0 x 0
which is valid whenever the integral on the left exists.
 
f (x)
To prove this, we let G (p) = L , as F 0 (p) = L[−xf (x)] we get that
x
 
dG f (x)
=L −x = L[−f (x)] = −F (p)
dp x

If we integrate the above from p to a then we will have that

Z a
G (a) − G (p) = − F (q) dq
p

Pradeep Boggarapu Laplace Transforms December 1, 2022 14 / 24

 Z ∞
Let a → ∞, then G (a) → 0 and hence −G (p) = − F (q) dq which
  Z ∞ p
f (x)
implies (0.9): L = F (q) dq. If we write the this as
x
Z ∞ Z ∞ p
f (x)
e −px dx = F (q) dq and let p → 0, we obtain
x
Z0 ∞ Z ∞ p
f (x)
dx = F (p) dp which proves (0.10).
0 x 0

Example-3: Since L[sin x] = 1/(p 2 + 1), (0.10) gives

Z ∞ Z ∞ ∞
sin x dp −1 
 π
dx = 2+1
= tan p  = 2.
0 x 0 p 0

Example-4: Since L[e ax ] = 1/(p − a), (0.10) gives

∞
e −ax − e −bx ∞
p + a ∞
Z Z 
1 1
b
dx = − dp = log = log .
0 x 0 p+a p+b p+b 0  a

Pradeep Boggarapu Laplace Transforms December 1, 2022 15 / 24

Product of Laplace transforms and convolutions of functions:
Theorem. (Convolution theorem) If L[f (x)] = F (p) and L[g (x)] = G (p)
then Z x 
F (p)G (p) = L f (x − t)g (t) dt . (0.11)
0
The integral on the right is a function of x and is called the convolution of
the functions f (x) and g (x). Thus, the inverse Laplace transform of the
product F (p)G (p) of Laplace transforms of f (x) and g (x) is the
convolution of the functions of f (x) and g (x) i.e.
Z x
−1
 
L F (p)G (p) = f (x − t)g (t) dt. (0.12)
0

Proof. To prove this, we use dummy variables s and t in the integrals

defining the transforms and write
Z ∞  Z ∞ 
−ps −pt
F (p)G (p) = e f (s)ds e g (t)dt
0 0

Pradeep Boggarapu Laplace Transforms December 1, 2022 16 / 24

 Z ∞Z ∞
F (p)G (p) = e −p(s+t) f (s)g (t)dsdt
Z0 ∞ 0Z ∞ 
−p(s+t)
= e f (s)ds g (t)dt,
0 0

where the double integration is taken over the first quadrant (s ≥ 0,

t ≥ 0) in the st-plane. We now introduce a new variable x in the inner
integral of the last expression by putting s + t = x, so that s = x − t and
(t being fixed during this integration) ds = dx. This enables us to write
Z ∞Z ∞ 
−px
F (p)G (p) = e f (x − t)dx g (t)dt
0 t
Z ∞Z ∞
= e −px f (x − t)g (t)dxdt.
0 t
Now by reversing the order of integration, we get what we need:
Z ∞ Z x  Z x 
−px
F (p)G (p) = e f (x − t)g (t)dt dx = L f (x − t)g (t)dt .
0 0 0

Pradeep Boggarapu Laplace Transforms December 1, 2022 17 / 24

Example-5: Since L[x] = 1/p 2 and L[sin ax]1/(p 2 + 1), we have
    
1 1 1
L−1 2 2 = L−1 2
p (p + 1) p p2 + 1
Z x
= (x − t) sin tdt = x − sin x,
0

which can also be obtained by partial fractions.

Problems
Find L−1 1/(p 2 + a2 )2 by using convolution theorem.
 
1

2 Show that
p 2 − a2
L[x cos ax] = ,
(p 2 + a2 )2
use theis result to fine L−1 1/(p 2 + a2 )2 .
 

Show that the differential equation 00 2

Z x y + a y = f (x),
3

1
y (0) = y 0 (0) = 0 has y (x) = f (t) sin a(x − t) dt as its solution.
a 0
Pradeep Boggarapu Laplace Transforms December 1, 2022 18 / 24
Now we will see that how we use the Laplace transforms to solve the
second order linear differential equations of the form:

(Ax + B)y 00 + (Cx + D)y 0 + (Gx + K )y = f (x), y (0) = y0 , y 0 (0) = y00 .

(0.13)
For this we derive the some useful formulas. We let Y = L[y ] and note
that

L[y 0 ] = pY (p) − y (0); (0.14)

00 2
L[y ] = p Y (p) − py0 − y00 ; (0.15)

In view of (0.7), we also have

dY
L[xy ] = − ; (0.16)
dp
d  0 dY
L[xy 0 ] = − L[y ] = −Y (p) − p ; (0.17)
dp dp
d  00  dY
L[xy 00 ] = L[y ] = −p 2 − 2pY − y0 . (0.18)
dp dp

Pradeep Boggarapu Laplace Transforms December 1, 2022 19 / 24

Example-5: Solve the differential equation,

xy 00 + (3x − 1)y 0 − (4x + 9)y = 0, y (0) = 0;

Sol. We apply Laplace transform on both sides of the differential equation

to get
L[xy 00 ] + 3L[xy 0 ] − L[y 0 ] − 4L[xy ] − 9L[y ] = 0.
If we use the formulas (0.18), (0.17), (0.14) and (0.16) respectively, then

2 dY dY dY
−p −2pY −y0 +3 −Y −p −[pY (p)−y (0)]+4 −9Y = 0.
dp dp dp

dY dY 3Y
Or [−p 2 − 3p + 4] − [3p + 12]Y = 0 which implies = . Now
dp Z Z dp 1 −p
dY dp
we separate the variables and integrate, = then we get
Y 1−p
C
log Y = −3 log (p − 1) + c or L[y ] = Y = (p−1) 3 . Apply inverse Laplace
 
2
transform y (x) = L−1 (p−1)
C
3 = C x2 e x .
Pradeep Boggarapu Laplace Transforms December 1, 2022 20 / 24
Problems
1 Solve each of the following differential equations:

(a) xy 00 + (3x − 1)y 0 − (4x + 9)y = 0, y (0) = 0;

(b) xy 00 + (2x + 3)y 0 + (x + 3)y = 3e −x , y (0) = 0.
2 If J0 is the Bessle function which satisfies the Bessel’s equation of
order zero, xy 00 + y 0 + xy = 0 with initial condition y (0) = 1, then
1
(a) show that L[J0 ] = ,
+ p 2 )1/2
(1 Z
∞
(b) and hence show that J0 (x) dx = 1.
Z0 π
1
(c) Show that J0 (x) = cos(x cos t)dt.
π 0
3 If f (x) is periodic with period a, so that f (x + a) = f (x), show that
Z ∞
1
F (p) = e −px f (x) dx.
1 − e −ap 0

4 Find F (p), if f (x) = 1 in the interval from 0 to 1, 2 to 3, 4 to 5, etc.,

and f (x) = 0 in the remaining intervals.
Pradeep Boggarapu Laplace Transforms December 1, 2022 21 / 24
A more interesting class of applications for Laplace transforms with the
above convolution property arises as follows:
Integral Equations: If f (x) and k(x) are given functions, then the equation
Z x
f (x) = y (x) + k(x − t)y (t) dt, (0.19)
0
in which the unknown function y (x) appears under the interal sign, is
called an integral equation.
Because of its special form, in which the integral is the convolution of the
two function k(x) and y (x), this equation lends itself to solution by means
of Laplace transforms. Indeed if we apply Laplace transform to both sides
of equation (0.19), we get

L[f (x)] = L[y (x)] + L[k(x)]L[Y (x)],

or
L[f (x)]
L[y (x)] =
1 + L[k(x)]

Pradeep Boggarapu Laplace Transforms December 1, 2022 22 / 24

The right side of the last equation in the above is known as a function p;
and if this function is recognizable, then we have our solution y (x).
Problems: Solve each of the following integral equations:
Z x
1 y (x) = 1 − (x − t)y (t) dt;
0
 Z x 
x −t
2 y (x) = e +− e y (t) dt ;
0
Z x
−x
3 e = y (x) + 2 cos x − ty (t) dt;
0
Z x
4 3 sin 2x = y (x) + (x − t)y (t) dt.
0

Pradeep Boggarapu Laplace Transforms December 1, 2022 23 / 24

 Thank you for your attention

Pradeep Boggarapu Laplace Transforms December 1, 2022 24 / 24