12 - Laplace Transforms and Their Applications PDF

12 LAPLACE TRANSFORMS AND
T HEIR APPLICATIONS
aaaaa
12.1 INTRODUCTION
This subject was enunciated first by English Engineer Oliver Heaviside (1850–1925) from
operational methods while studying some electrical engineering problem. However,
Heaviside`s treatment was not very systematic and lacked rigour which later on attended to
and recapitulated by Bromwich and Carson.
Laplace transform constitutes an important tool in solving linear ordinary and partial
differential equations with constant coefficients under suitable initial and boundary conditions
with first finding the general solution and then evaluating from it the arbitrary constants.
Laplace transforms when applied to any single or a system of linear ordinary differential
equations, converts it into mere algebraic manipulations. In case of partial differential
equations involving two independent variables, laplace transform is applied to one of the
variables and the resulting differential equation in the second variable is then solved by the
usual method of ordinary differential equations. Thereafter, inverse Laplace transform of
the resulting equation gives the solution of the given p.d.e.
Another important application of Laplace Transform is in finding the solution of
Mathematical Model of physical problem where in the right hand of the differential equation
involves driving force which is either discontinuous or acts for short time only.
∞
Definition: Let f(t) be a function defined for all t ≥ 0. Then the integral, L{ f (t)} = ∫ e−st f (t)dt
0
if exists, is called Laplace Transform of f(t). ‘s’ is a parameter, may be real or complex number.
Clearly L{f(t)} being a function of s is briefly written as f (s) i.e. L{ f (t)} = f (s) . Here the symbol
L which transforms f(t) into f (s) is called Laplace Transform Operator. Then f(t) is called
inverse Laplace transform of f (s) or simply inverse transform of f (s) i.e. L−1{ f (s)} .
Note: There are two types of laplace transforms. The above form of integral is known as one sided or unilateral transform.
∞
However, if the transform is defined as L{ f (t)} = ∫ e−st f (t)dt , where s is complex variable, called two sided or bilateral
0
laplace of f(t), provided the integral exists. If in the first type transform, variable s is a complex number with a result that
755
756 Engineering Mathematics through Applications
laplace transform is defined over a portion of complex plane. If L{f(t)} exists for s real and then L{f(t)} exists in half of the
complex plane in which Re s>a (Fig.12.1). The transform f (s) is an analytic function with properties:
(i) Lt f (s) = 0 viz. the necessary condition for f (s) to be a transform.

Re. s→∞
(ii) Lt. s f (s) = A , if the original function has a limit Lt. f (t) = A
s→∞ t →∞
Imgn.s
Fig. 12.1: Real axis
12.2 EXISTENCE CONDITIONS

The laplace transform does not exist for all functions. If it exists, it is uniquely determined.
For existence of laplace, the given function has to be continuous on every finite interval and
of exponential order i.e. if there exist positive constants M and ‘a’ such that f (t) ≤ Meat for
all t ≥ 0. The function f(t) is some times termed as object function defined for all t ≥ 0 and f (s)
is termed as the resultant image function. Here the parameter should be sufficiently large to
make the integral convergent. In the above discussion, the condition for existence of f (s) is
sufficient but not necessary, which precisely means that if the above conditions are satisfied,
the laplace transform of f(t) must exist.
But if these conditions are not satisfied, the laplace transform may or may not exist. For
1 1
e.g. in case of f ( t ) = , f (t ) → ∞ as t → 0 , precisely means f ( t ) = is not piecewise
t t
continuous on every finite interval in the range t ≥ 0. However, f(t) is integrable from 0 to
any positive value, say t0 . Also f (t) ≤ Meat for all t > 1 with M = 1 and a = 0. Thus,
π
Lf (t) = , s > 0 exists even if 1 is not piecewise continuous in the range t ≥ 0.
s t
12.3 EXISTENCE THEOREM ON LAPLACE TRANSFORM

If f(t) is a function which is piecewise* continuous on every finite interval in the range t ≥ 0
and satisfy f (t) ≤ Meat for all t ≥ 0 and for some positive constants ‘a’ and M means, f(t) is of
∞
− st
exponential order ‘a’, then the laplace transform of f(t) i.e. ∫ e f (t) dt exists.
0
Laplace Transforms and their Application 757
∞ t0 ∞
Proof: We have L { f ( t )} = ∫ e− st f ( t ) dt = ∫ e− st f (t ) dt + ∫ e− st f ( t ) dt … (1)
0 0 t0
t0
Here ∫ e−st f ( t ) dt exists since f(t) is piecewise continuous on every finite interval 0 ≥ t ≥ t0.
0
∞ ∞ ∞
Now ∫ e−st f (t) dt ≤ ∫ e−st f (t) dt ≤ ∫ e−st Me atdt, since f ( t ) ≤ Meat
t0 t0 t0
∞
e−( s −a)t0 ;
= ∫ e−( s −a)t M dt = M s>a
t0 ( s − a) … (2)
−( s − a)t0
But M e can be made as small as we please by making t0 sufficiently large. Thus,
( s − a)
from (1), we conclude that L{f(t)} exists for all s > a.
*Note: A function is said to be piecewise (sectionally) continuous on a closed interval [a, b], if this closed interval can
be divided into a finite number of subintervals in each of which f(t) is continuous and has finite left hand and right hand
limits.
A function is said to be of exponential order ‘a’ (a > 0) as t → ∞, if there exist finite positive constants t0 and M such
that f (t ) ≤ Meat or e− at f (t) ≤ M for all t ≥ t0.
12.4 TRANSFORMS OF ELEMENTARY FUNCTIONS

By direct use of definition, we find the laplace transform of some of the simple functions:
1
1. L(1) = , (s > 0)
s
1
2. L(eat ) = , (s > a)
s−a
n ! Γ (n + 1)
3. L(t ) = n+1 , where n = 0,1, 2, 3, … otherwise
n
s sn +1
a
4. L(sin at) = , (s > 0)
s2 + a2
s
5. L(cos at) = 2 , (s > 0)
s + a2
a
6. L(sinh at) = 2 , (s > a )
s − a2
s
7. L(cosh at) = 2 , (s > a )
s − a2
Proofs:
∞ ∞
 e−st  1
1 . L (1) = ∫ e dt = 
− st
 = , s>0
0  −s 0 s
∞
∞ ∞  e− ( s − a ) t 
2. L ( e )= ∫e − st at −( s − a ) t 1
at
e dt = ∫ e dt =   = , s>a
0 0  − (s − a)  ο ( s − a)
Remarks: Here the condition s > a is necessary to ensure the convergence of the integral since otherwise
divergent for s ≤ a.
∞ ∞ n
− p  p  dp  dp 
3. L ( t ) = ∫ e t dt = ∫ e  
− st n
 on taking st = p, dt =
n

0 0 s s s 
∞ ∞
1 1 Γ(n + 1)
= n+ 1 ∫ e− ppndp = n+ 1 ∫ e− pp( n+1) −1dp = , if n > – 1 and s > 0
s 0 s 0 sn+ 1
∞ ∞
 e−st 
2 (
−s sin at − a cos at ) = 2
a
4. L (sin at ) = ∫ e sin at dt =  2
− st , s>0
0 s + a  ο s + a2
∞ ∞
 e−st
5. L (cos at ) = ∫ e−st cos at dt =  2 ( −s cos at − a sin at ) = 2 s 2 , s > 0
0  s + a2
ο s + a
∞ ∞
 eat − e−at  1 ∞  − ( s − a )t
6. L (sinh at ) = ∫ e− st sinh at dt = ∫ e− st   dt = ∫ e − e−( s + a)t  dt
0 0  2 2 0
1 1 1  a
= − + = s> a.
2  s − a s + a  s2 − a2
∞ ∞
 eat + e−at  1 ∞ −( s − a)t
7. L (cosh at ) = ∫ e−st cosh at dt = ∫ e−st   dt = ∫ e + e−( s + a)t  dt
0 0  2 2 0
= 
1 1 1  s
+ = , s> a.
2  s − a s + a  s2 − a2
12.5 PROPERTIES OF LAPLACE TRANSFORMS
Sr. No. Property f(t), function f (s) , laplace transform
1. Linearity a f(t) + bg (t) – c h(t) a f (s) + b g(s) − c h(s)
f   , a > 0
1 s
2. Change of scale f(at)  a
a
3. Initial value Lt Lt
t→0 f (t) s→∞ sf (s)
4. Final value Lt
f (t) Lt
t →∞ s→0 sf (s)
5. First shifting eat f(t) f ( s − a)
6. Second shifting f(t – a)u(t – a) e−as f (s)
d
7. Derivatives
dt
f (t) s f (s) − f (0), s > 0
d2
f (t) s2 f (s) − s f (0) − f '(0), s > 0
dt2
dn
f (t) sn f (s) − sn −1 f (0) − sn − 2 f ´(0) … − f n − 1(0), s > 0
dtn
t
f (s)
8. Integral ∫ f (u) du , s>0
0 s
t t
∫ ∫ f ( du)
2 1
f (s)
0 0 s2
t t t
∫ ∫ …∫ f ( du)
n 1
f (s) , s > 0 for any positive integer n.
0 0 0 sn
d
9. Multiplication by t t f(t) − f (s)
ds
d2
t 2f(t) ( − 1)2 f (S)
ds2
dn
t nf(t) ( −1)n f (s)
dsn
∞
f (t)
10. Division by t ∫ f (s) ds , provided the integral exists.
t s
f (t) ∞ ∞
t2 ∫ ∫ f (s)(ds)2 , provided the entegral exists.

s s
∞ ∞ ∞
f (t)
∫ ∫ …∫ f (s)(ds)n , provided the integral exists
tn s s s
11. Convolution f(t)*g(t), f (s)g(s)

Theorem
 t

 where f ( t)* g( t ) = ∫ f (u) g ( t − u) du,
0 
T
∫ e− st f (t) dt
12. Periodic Function f(t + T) = f(t) 0
1 − e− sT
Proofs of Some of the properties are taken in subsequent discussions.

P 1 Linearity Property:
Let f(t), g(t), h(t) be any functions whose laplace transforms exist, then for any constants a,
b, c and k, we have
(i) L[af (t) + b g(t) – hf(t)] = a L[f(t)] + b L[g(t)] – cL[h(t)]
(ii) L[k f(t)] = k L[f(t)].
Proofs:
∞
(i) LHS = ∫ e−st  a f (t) + b g(t) − c h(t) dt.
0
∞ ∞ ∞
a ∫ e−st f (t) dt + b ∫ e−st g(t) dt − c ∫ e−st h(t) dt = aL  f (t) + bL  g(t) − cL [ h(t)]
0 0 0
This result can easily be generalized.

∞ ∞
(ii) LHS = L  kf (t) = ∫ e−st kf (t) dt = k ∫ e−st f (t) dt = kL  f (t)
0 0
In view of above discussion, Laplace operator is a linear in nature.
P 2 First Shift Property: [KUK, 2010]

r
If L  f (t) = f (s) then L  eat f (t) = f ( s − a)
∞ ∞ ∞
L  eat f (t) = ∫ e− st  eat f (t) dt = ∫ e−( s − a)t f (t) dt = ∫ e−pt f (t) dt = f (p), where p = ( s − a)
0 0 0
Thus, if we know the transform of f(t), we can write the transform of eat f(t) simply replacing
s by (s – a) in f (s)
f (s)
f (s) Translation along s axis
f (s – a)
s
0
Fig. 12.2
Application of this property give arise to following simple results:
1. L ( eat ) =
1 1
, as L(1) =
s−a s
2. L ( e t ) =
n! , as L ( tn ) = n !
at n
( s − a)n+1 sn +1
3. L ( e sin bt ) =
b b
, as L sin(bt) =
at
( s − a )2 + b2 s + b2
2
s−a
4. L ( e cos t ) = , as L ( cos bt ) = s
at
( s − a )2 + b2 s2 + b2
5. L ( eat sinh bt ) =
b , as L (sinh bt ) = b
( s − a)2 − b2 s2 − b2
s−a
6. L ( e cosh bt ) = , as L ( cosh bt ) = s
at
( s − a)2 − b2 s − b2
2
where in each case s > a.
P 3 Change of Scale:
1  s
If Lf (t) = f (s) then L  f (at) = f 
a  a
∞ ∞ −s p
dp  dp 
L  f (at)  = ∫ e−st f (at) dt = ∫ e a f (p)  on putting at = p, dt = 
0 0 a a
1 ∞ −( s a ) p
f (p) dp = f   .
1 s
= ∫e
a0 a  a
Note: Changing a to   , the result changes to L  f  t   = a f (as)

1
 a   a  
P 4 Change of Scale with Shifting:

1  s − b
If L  f (t) = f (s), then L ( ebt f (at)) = f 
a  a 
∞ ∞
 dp 
L  ebt f (at) = ∫ e− st ebt f (at) dt = ∫ e−( s − b )t f (at) dt  on taking at = p, dt = 
0 0 a
∞  s−b
1 − 1  s − b
∫
p
= a  f (p) dp = f
e 
a a  a 
0
Example 1: Find the laplace transform of the followings:

(i) sin2t cos3t (ii) sin3 2t (iii) sin t
(iv) e–tsin2t (iv) cosh at sin bt (v) sin at sinbt
Solution:
(i) L (sin 2t cos 3t ) = 1 L (sin 5t − sin t ) = 1 L (sin 5t) − 1 L sin t
2 2 2
1 5 1 
=  − 2 , s>0
2 s + 5
2 2
s + 1 
2s2 − 10 2 ( s2 − 5)
= =
( s2 + 1)( s2 + 25 ) ( s2 + 1)( s2 + 25)
(ii) L(sin3 2t) = 1 L ( 3 sin 2t − sin 6t ) , Using sin3 θ = 3 sin θ − sin 3θ
4 4
L (sin 2t ) − L (sin 6t ) =  2
3 1 3 2  1 6 
= −  
4 4 4  s + 22  4  s2 + 62 
3 1 1  48
= − = , s>0
2  s2 + 4 s2 + 36  ( s2 + 4)( s2 + 36 )
(

)
(iii) L sin t = L  t −
( t )3 + ( t )5 − …
 3! 5! 
Γ  Γ  Γ 
3 5 7
 2 1  2 1  2
= 3 − 5 + 7 − ……
3! 5!
s2 s2 s2
1 3 1 5 3 1
π ⋅ π ⋅ ⋅ π
1 1
= 2 3 − 2 25 + 2 2 2
7 ……
6 120
s2 s2 s2
π   1 1 1 
2 3
π − 4s
1
1 1
= 1 −   +   −   + … = ⋅ e
3  4s  2!  4s  3!  4s  3
2s 2   2s 2
(iv) L ( e−t sin2 t ) = L  e−t 1 − cos 2t  which comparable to L ( eat f (t)) = f ( s − a)

 2 
 1 − cos 2t  1 1 1 1 s
where L = L(1) − L (cos 2t ) = −
 2  2 2 2s 2 s2 + 4
1 ( s + 1) 1  ( s + 2s + 5 ) − ( s + 1)( s + 1) 
2
L ( e−t sin2 t ) =
1
∴ − =  
2 ( s + 1) 2 ( s + 1)2 + 4 2  ( s + 1) ( s2 + 2s + 5 ) 
2
=
( s + 1) ( s2 + 2s + 5)
−at
(v) L (cosh at sin at ) = L  e + e sin at  = 1  L ( eat sin at ) + 1 L ( e−at sin at)
at
 2  2 2 
a
But L (sin at ) = , s > 0 ; then by 1st shift property,
s2 + a2
1 
L (cosh at sin at ) = 
a a
+ 
2  ( s − a ) + a
2 2
( s + a) + a 
2 2
a 1 1 
=  2 + 2 
2  ( s + 2 a ) − 2 as ( s + 2 a ) + 2 as 
2 2
=
{ 2 2 2
} {
a  ( s + 2a ) + 2as + ( s + 2a ) − 2as
2
} 
{ }{
2  ( s2 + 2a2 ) − 2as ( s2 + 2a2 ) + 2as
 } 
=
a ( 2s2 + 4a2 ) =
a ( s2 + 2a2 )
2 ( s2 + 2a2 )2 − 4a2 s2  s4 + 4a4
 
Alternately: Finding the imaginary part of L {eiat cosh at} when L ( cosh at ) = s
, s > a.
s − a2
2
−iat
(vi) L (sin at sin bt ) = L  e − e sin bt = 1  L ( eiat sin bt ) − L ( e−iat sin bt) 
iat
 2i  2i  
But L ( sin bt ) = b
, s > 0 , then by 1st shift property,
s + b2
2
1 
L ( sin at sin bt ) =
b b
 − 
2i  ( s − ia ) + b2 ( s + ia ) + b2 
2 2
b  1 1 
=  − 2
2i  s − a − 2ias + b
2 2 2
s − a + 2ias + b2 
2
b  1 1 

= −
{ }
2i  s2 + ( b2 − a2 ) − 2ias
 {
s2 + (b2 − a2 ) + 2ias 
 }
=
 2 ({ 2 2
} 2
) ({
b  s + ( b − a ) + 2ias − s + ( b − a ) − 2ias
2 2
} ) 
2i  { }
s2 + ( b2 − a2 ) − 4i2 a2 s2 
2
 
 
b  4ias 
=
{ }
2i  s4 + ( b2 − a2 )2 + 2s2 ( b2 − a2 ) + 4a2 s2 
 
 
2abs
= 
 s4 + ( b2 − a2 ) + 2s2 ( b2 + a2 ) 
2
 
2abs
=
(s
2
+ (b + a)
2
)(s 2
+ (b − a )
2
)
Alternately: L ( sin at sin bt ) = 1  L cos ( a − b ) t − L cos ( a + b ) t  , using 2sinA sinB = cos(A – B)
2
– cos(A + B)
Example 2: Find the Laplace transform of
 et , 0 < t < 1
(i) f (t) =  0 , t > 1

(ii) f (t) = { sin t ,
0,
0<t< π
t>π [Madras, 2005]
 t , when 0 < t < T  cos ( t − 2π 3 ) , t > 2π 3

(iii)* f (t) =  T (iv) f (t) = 
t < 2π 3
 1, when t > T  0,
*[Kerala Tech. 2005, NIT Kurukshetra, 2010]
(v) f (t) = t − 1 + t + 1 , t ≥ o [NIT Kurukshetra, 2003]
Solution:
(i) For function f(t) defined for all t ≥ 0, laplace transform is defined as:
∞
L  f (t) = f (s) = ∫ e−st f (t) dt, Provided that this integral exists.
0
1
∞
e(1− s)t
(e( )
1 1
1
− st t
= ∫ e e dt + ∫ e− st&o dt = ∫ e(1− s)t dt = = 1− s )
−1
0 1 0 (1 − s) 0 (1 − s)
s is any parameter, may have real or complex value.
(ii) f (t) = { sin t,

0,
0<t<π
t>π
∞
By definition, L ( f (t)) = ∫ e−st f (t) dt = I ( say )
0
π ∞ π
⇒ I = ∫ e−st sin t dt + ∫ e−st 0 dt = ∫ e−st sin t dt
0 π 0
 π 
= ( e−st . − cos t )0 − ∫ ( −se−st ) ( − cos t ) dt 
π
 0 
  π 
= ( e−sπ + 1) − s ( e−st sin t ) − ∫ ( e−st. − s) sin t dt 
π
ο
  0 
 π 
I = ( e−sπ + 1) − s2 ∫ e−st sin t dt 
 0 
(Since value of the expression e–st sint is zero at both the limits)
I = (e–sπ + 1) – s2I or I(1 + s2) = (1 + e–sπ)
 1 + e−πs 
I= = L  f (t)
 1 + s2  Hence the result.
 t , when 0 < t < T

(iii) Here f (t) =  T
1, when t > T
∞ T ∞ T ∞
t
Lf (t) = ∫ e−st f (t)dt = ∫ e−st f (t)dt + ∫ e−st f (t)dt = ∫ e−st dt + ∫ e−st 1dt
0 0 T 0 T T
1  e−st  
T ∞
e−st  e−st 
∫ 1 −s dt +  −s 
T
=  t  − (Integration by parts)
T  −s  o ο T
1  Te−sT  1  e−st   1
T
=  − 0 +    − ( 0 − e−sT )
T  −s  s  −s  0  s

1  Te−sT  e−sT
= 
T  −s
− (
1 −sT
e − 1)  + = 2 (1 − e−sT )
1
Hence the result
s2
 s Ts
  2π  2π
cos  t − 3  , t>
3
(iv) Here f (t) = 
0, 2π
t<
 3
2π
∞ ∞ ∞
2π 
L ( f (t)) = ∫ e f (t) dt = ∫ e−st f (t) dt + ∫ e−st f (t) dt = ∫ e−st cos  t −
3
− st
∴   dt
0 0 2 π 2 π 3 
3 3
Now integrate by parts taking cos (t – 2π/3) as 2nd function and e–st as 1st function.
∞
  2π   ∞
 2π   ∞
 2π 
I =  e− st sin  t −   − ∫ ( −se− st ) sin  t −  dt  = s ∫ e− st sin  t −  dt
   3   2π 2π  3   2π  3
 3 3  3
 2π 
(Since the value of the expression e–stsin  t −  is zero at both the limits).
3
∞
  2π  ∞
  2π   
I = s  e−st ⋅ − cos  t −  − ∫ ( −se−st )  − cos  t −   dt
 3  2π 2π   3  

 3 3 
 s2π ∞

 −  2π  
⇒ I = s 0+e 3 cos 0 − s ∫ e cos  t −  dt
2 − st
 2π  3 
 3 
2π s 2π s
− −
⇒ I = se 3 − s2 I or I(1 + s2 ) = se 3
2 πs
−
se 3
I = Lf (t) =
(1 + s2 ) Hence the result.
(v) Here f (t) = t − 1 + t + 1 , t ≥ 0

Clearly f(t) = 2 for 0 ≤ t ≤ 1
= 2t for t > 1
Say, if t = 0, |t – 1| + |t + 1| = |0 – 1| + |0 + 1| = 2
1 1 1 1 3
if t = , t −1 + t +1 = −1 + +1 = + = 2
2 2 2 2 2
if t = 1, t − 1 + t + 1 = 1 − 1 + 1 + 1 = 0 + 2 = 2
i.e. for all value of t varying from 0 to 1, value of t − 1 + t + 1 is always 2. Likewise

for t > 1, it is 2.
∞ 1 ∞
− st − st − st
Hence, L  f (t) = ∫ e f (t)dt = ∫ e 2dt + ∫ e 2t dt
0 0 1
1  −st ∞ ∞ 
e−st e−st
∫
e
=2 + 2 t − 1 ⋅ dt
−s 0  −s 1 −s 
1
 1 1∞ 
=− ( e − e ) + 2 − (0 − 1 ⋅ e−s ) + ∫ e−st dt
2 −s 0
s  s s1 
∞
2 e−st
=−
s
(
2 −s
e − 1) + e−s +
2
s s −s 1
2 2e− s 2  e− s 
− 2 ( e − e−s ) = + 2 =  1 +
2 2 −∞
=  Hence the result.
s s s s s s 
P 5. Transforms of Functions Multiplied by tn
If f(t) is a function of class A and if L  f (t) = f (s), then
L (t f (t)) = − f (s) and L (tn f (t)) = (−1)n n f (s)

d dn
ds ds
∞
− st
By definition of laplace of f(t), we have ∫ e f (t) dt = f (s) … (1)
0
d ∞ −st d
Differentiating (1) with respect to s, ∫ e f (s)dt = f (s) … (2)
ds 0 ds
By Leibnitz rule of differentiation under integral sign, (2) reduces to
∞
∂ −st
∫
∂s
( e ) f (t) dt = dsd f (s)
0
∞ ∞
∫ ( −te−st ) f (t) dt = f (s) ∫ e−st (tf (t) ) dt = −
d d
or f (s) …(3)
0 ds 0 ds
This proves the theorem for n = 1 (i.e. first result)
Now assume the theorem is true for n = m (say), so that (3) gives
∞
∫ e−st (tm f (t)) dt = (−1)m
dm
f (s) … (4)
0 dsm
d ∞ −st m dm +1
then ∫ e (t f (t)) dt = (−1)m m+1 f (s) … (5)
ds 0 ds
Again by Leibnitz rule, (5) reduces to
∞ ∞ m +1
∫ ( −te−st )( tm f (t)) dt = − ∫ e−st (tm+1 f (t)) dt = (−1)m m+1 f (s)
d
0 0 ds
∞
dm + 1
or ∫ e−st (tm + 1 f (t)) dt = (−1)m + 1 f (s) … (6)
0 dsm + 1
This clearly shows that if the theorem is true for n = m, it is true for n = m + 1 i.e. for n =
1 + 1 = 2, n = 2 + 1 = 3, so on. Hence the theorem is true for all positive integer value n.
Note: A function which is piecewise or sectionally continuous on every finite interval in the range t ≥ 0 is of exponential
( )
order s i.e. Lt. e− st f (t) = 0 is known as a function of class A.
t →∞
Example 3: Find Laplace of

(i)* t (sin2 t) (ii) t2 cos at (iii) ** sinh 3t cos2 t
[* KUK, 2000; ** NIT Kurukshetra, 2005]
Solution:
(i) L ( t sin2 t ) = L  (1 − cos 2t ) = L(t) − L (t ⋅ cos 2t )
t 1 1
 2  2 2
L ( tn ) = n +1 and L ( t ⋅ f (t)) = −
As n ! d
f (s)
s ds
∴ L ( t sin2 t ) =
1 1 1
⋅ 2 −
2 s
d
− ⋅ 2
s
{
2 ds s + 22
1
= 2 −
2s
1 s2 − 4
2 ( s2 + 4 )2 =}2 ( 3s2 + 4 )
s2 ( s2 + 4 )
2.
(ii) L ( t2 cos at) = ( − 1)2 d f (s), where

2
s
f (s) = L (cos at ) = ,s>0
ds2 s2 + a2
d  ( s + a ) − s.2s  d  a2 − s2 
 2 2   
d2  s 
∴ L ( t2 cos at ) = 2   = =
ds  s + a  ds  ( s2 + a2 )
2 2 2
 ds  ( s2 + a2 ) 
2
   
=
( s2 + a2 ) ( −2s) − ( a2 − s2 ) ⋅ 2 ( s2 + a2 ) ⋅ 2s 2s ( s2 − 3a2 )
2
= .
( s2 + a2 )4 ( s2 + a2 )3
− 3t
(iii) L ( sinh 3t cos2 t ) = L  e − e ⋅  1 + cos 2t   = 1 ( e3t − e−3t ) + ( e3t − e−3t ) cos 2t 
3t
 2  2   4  
L ( e − e ) + L ( e3t cos 2t − e−3t cos 2t )

1  3t − 3t
=
4
1  1 1   ( s − 3) ( s + 3 )  
=  − + −
4  s − 3 s + 3   ( s − 3 ) + 22 ( s + 3 )2 + 22  
2
(using first shift property, replace s by s – a)
=

1 6
+
{ }
( s − 3 ) ( s + 3)2 + 22 − ( s + 3 ) ( s − 3 )2 + 22{ } 
4  s2 − 9
 { }{
( s − 3)2 + 22 ( s + 3)2 + 22 } 

3 1 s2 − 13 
= + 4
2  s − 9 s − 10s2 + 169 
 2
P 6 Transforms of Functions Divided By t

 f (t)  ∞
If L  f (t) = f (s) then L = ∫ f (s) ds , provided the integral exists.
 t  s
∞
− st
Proof: We have f (s) = L  f (t) = ∫ e f (t) dt
0
Integrating the above equation with respect to s between the limits s and ∞, we get
∞ ∞ 
∞ ∞
∫ f (s) ds = ∫ ∫o e−st f (t) dt  ds = ∫ f (t)  ∫ e−st ds dt
∞
s s s 0
(On assumption that the change of order of integration exists).
∞
∞
e−st ∞
f (t)   e−st  
= ∫ f (t) dt = ∫ e−st dt  Since lim   = 0, s > o
0 −t s s t x →∞  t  
∞
 f (t) 
∴ ∫ f (s) ds = L 
s  t 
Remarks: Since L  f (t)  corresponds to the integration of laplace transform of f(t) with respect to s between
 t 
∞∞ ∞
 f (t) 
the limits s and ∞, therefore the repeated application of the result gives L  n  =
 t 
∫ ∫ …∫ f (s)(ds)n provided for
1442443
s s s
n times
 f (t) 
positive integer n, lim  n  exists.
x →∞  t 
1 − cos t
Example 4: Find the laplace transform of . [Osmania, 2003]
t2
∞
Solution: Here, L  1 − cos t  =  1 − s  ds , since L(1) = 1 , L cos t = s
 t  ∫
 s s2 + 1 
s
s s2 + 1
∞ ∞ ∞
 s2   1 
= log s − log ( s2 + 1) = log 2
1
 = log
 2 
s  s + 1 s 1 
 1+ 2 
 s s
  As 1 → 1 = 1, when s → ∞
s2 
= log 1 − log 2  1 1 
 s + 1  ,  1+ 2
s
1+
∞

 s2 + 1 
= log 
 s 
 1 − cos t  =  f (t)  = ∞ log s2 + 1 = ∞ log  s2 + 1  ⋅ 1
Again L  L ∫ ds ∫  ds
 t2   t  s s s s 
  ∞ ∞ 
s2 + 1   d  s2 + 1 
∫
s
=   log s − s ds  , integration by parts
  s  s 
s2 + 1 ds  s2   
 s 
1 2s
∞ ∞ s⋅ − 1 ⋅ s2 + 1
s2 + 1 2 s2 + 1
∫
s
= −s log − × s ⋅ ds
s2 s s +1
2 s2
s
∞
∞ ∞
∫ + ( tan−1 s )s
1 ∞
= −s log 1 + s−2 + ds = −s log 1 + s−2
s s +1
2 s
s
π
= − s log (1 + s−2 ) +  − tan −1 s = cot−1 s − s log (1 + s−2 )
1 1
2 2  2
Example 5: Prove that L 

sin t 
= tan− 1 and hence find L  sin at  . Does the laplace
1
 t  s  t 
transform of cosat exist?

t
f (t) sin t 1
Solution: For f (t) = sin t, lt = lt =1 and L (sin t ) = = f (s)(say)
t →o t t →o t s2 + 1
sin t  ∞ ∞
π
L  ds = ( tan−1 s)s = − tan−1 s = cot−1 s = tan−1
1 ∞ 1
  = ∫ f (s) ds = ∫ 2
t s s s +1 2 s
L 
sin at   sin at  = a  1 tan−1 1  = tan−1  a 
Now
  = a L    
t  at  a  s  s
 
 a 
(Using change of scale property, L ( f (at)) =

1  s
f  , a>0)
a  a
s
Again, L (cos at ) = = f (s),(say), then
s + a2
2
L 

cos at  ∞ s
t
 = ∫ 2
s s + a2
2
(
ds = log ( s2 + a2 ) =
1 ∞
s
1
2 s →∞
) (
lt. log ( s2 + a2 ) − log ( s2 + a2 ) )
lt log (s2 + a2) is infinite. Hence L  cos at  does not exist.
Which does not exist as s→∞  
t 
∞
π
∫
sin t
Example 6: Using Laplace Transform, show that dt =
t 2
0
a
Solution: We know that L[sin(at)] = , s>0
s + a2
2
∞ ∞
 sin at  a  s π  s  s
L  =∫ 2 ds =  tan−1  = − tan−1   = cot−1   … (1)
 t  s s + a2  a s 2  a   a
∞
 sin at  = −st  sin at 
However by definition, L 
 t 
e 
 t 
dt ∫
0
… (2)
∞
 sin at  = cot−1  s 
From (1) and (2),
∫e
0
− st

t 
 dt  
a
∞
sin t π
On taking a = 1 and s = 0 in the above relation, we get ∫ dt = cot−1 0 =
0 t 2
P 7 Laplace Transform of Integral of f(t)
t  1
L ∫ f (u) du  = f (s) where Lf (t) = f (s)
0  s
t
Let φ ( t ) = ∫ f (u) du and φ(0) = 0 then φ '(t) = f (t)
0
We know that, L [ φ '(t)] = sL ( φ(t)) − φ(0)
i.e. L [ φ′(t)] = sL (φ(t)) , sin ce φ(0) = 0
L [ φ(t)] =
L ( φ′(t))
1
i.e.
s
On putting the values of φ(t) and φ’(t), we get
1 1
L ∫o f (u) du  = L  f (t) = f (s)
t
  s s
Remarks: If the above preposition holds, then the inverse transforms
 t  1 
t
1 
i.e. L−1  L∫ f (u) du = L−1  f (s) or ∫ f (u) du = L−1  s f (s) also holds.
 0  s  0
P 8 Laplace Transforms of Derivatives

Let f(t) be a continuous function for all t ≥ 0 and be of exponential order as t → ∞, and f ’(t)
if also be piece wise continuous for all t ≥ 0 and be of exponential order as t → ∞, then
L[( f ’(t)] = s L( f(t)) – f(0).
∞  ∞ 
L ( f ′ ( t )) = ∫ e−st f ′ (t ) dt =  e−st f (t) o − ∫ (−s) e−st . f (t) dt  = lim e−st f (t) − f (0) + s f (s)
∞
0  0  t →∞
Now, assuming f(t) to be such that Lt e−st f (t) = 0 (When this condition is satisfied, f(t) is
t →∞
said to be of exponential order s).
Therefore, we conclude that L(f ’(t)) exists and L ( f ´(t)) = sL ( f (t)) − f (0)
Remarks: By successive application of the above result, we obtain the laplace transform of f n(t) as follows:
L f n(t) = sn f (s) − sn −1 f (0) − sn −2 f ′(0) − …… − f n −1(0) , if f (t), f ′(t), …… , f n −1(t) are continuous, f n(t) is piecewise
continuous for all t ≥ 0 and if f (t), f ′(t), … , f n(t) all are of exponential order, when t → ∞.
Example 7: Find L(cosat) and deduce from it L(sinat) [KUK, 2002]

∞ −iat + e−iat
Solution: By def n., L (cos at ) = ∫ e−st  e + e  dt , as cos at = e
iat iat
0  2  2
1  ∞ −( s − ia)t ∞ 
=  ∫e dt + ∫ e−( s+ ia)t dt 
2 0 0 
 ∞ ∞

1  e−( s − ia)t e−( s + ia)t 
= +
2  − ( s − ia) o − ( s + ia) o 
 
1  1   1 
=  0 −  + 0 − 
2  − ( s − ia)   − ( s + ia )  
1  2s  s
=  2 2
= 2 , i2 = −1 … (1)
2  s − i a  s + a2
2
t
Now to deduce L (sin at) using L (cos at), write, sin at = a∫ cos atdt … (2)
0
t 
Take laplace on both sides, L (sin at ) = aL  ∫ cosatdt … (3)
0 
t  f (s)
, where f (s) = L f (t) =  2
s 
By formula, L ∫ f (t) dt  = 
0  s  s + a2 
 s 
 2 2
L (sin at ) = a s + a =  2
a 
  … (4)
s s + a2 
−1 d
Alternately: Write, sin at = (cos at ) = − 1 f ’(t)
a dt a
Take laplace on both sides, L (sin at ) = − 1 Lf '(t) … (5)
a
Now by the formula,
s
L  f ’(t) =  s f ( s) − f (0) , where f (s) = and f (0) = (cos at)t =0 = 1
s + a2
2
Using above, (5) becomes
1  s − (s + a ) 
2 2 2
L (sin at ) = −  s. 2 
1 s a
∴ − 1 = −  = 2
 ( s + a )
a s + a 2 
 a  s + a
2 2 2 … (6)
Hence the result.
Example 8: Find sin t and hence deduce cos t . [KUK, 2010]

t
d cos t
Solution: Let sin t = f (t), then f ’(t) = (sin t ) =
dt 2 t
π
1
( )
−
But Lf ’(t) = s f (s) − f (0), where f (s) = L sin t = 3 e 4s
2 s2
 cos t  π − 4s
1
L = −0 f (0) = sin 0 = 0
 2 t 
∴ s. 3 .e as
2s 2
 cos t  π − 4s
1
or L
 = e
t  s
Laplace Transforms and Their Applications 773
 t sin t 
Example 9: Evaluate L  ∫ et dt 
0 t 
∞
 sin t  = 1
f (s) = L (sin t) =
 t  ∫s
Solution: L f (s) ds, where
s +12
∞
1 ∞ π
=∫ ds = tan −1 s s = − tan−1 s = cot−1 s
s s +1
2
2
∴  sin t  = cot−1( − 1), by first shift property

L et s
 t 
 t sin t  f (s) cot−1(s − 1)
Now L ∫ et dt  = =
0 t  s s
P 9 Evaluation of Integrals by Laplace Transforms
∞ ∞
− 3t 3 −t
Example 10: Evaluate (i) ∫ t e sin t dt (ii) ∫ t e sin t dt
0 0
OR
Find laplace of (i) t sin t at s = 3 (ii) t3e–tsint at s = 1
[MDU, 2000; KUK, 2002, 2004, 2005, 2006; VTU, 2007]
Solution:
∞ ∞
− st
(i) The Integral ∫ t e−3t sin t dt is comparable to ∫ e f (t) dt with s = 3 and f(t) = t sin t.
0 0
1
Now L(sin t) = = f (s), s > 0
s2 + 1
f (s) = −  2
d d 1  2s
∴ L (t sin t ) = − =
ds ds  s + 1  ( s2 + 1)2
Hence the integral
∞
2s 2.3 3
∫ e−3tt sin t dt = L(t sin t) = = = , when s = 3.
0 (s 2
+ 1)
2
(3
2
+ 1)
2
50
∞ ∞
− st
(ii) The integral ∫ t3e−t sin t dt is comparable to ∫ e f (t) dt
0 0
Hence, f(t) = t3sin t
Now L(sin t) = 1
= f (s), (s > 0)
s2 + 1
 
d3  1  d2 2s d  1 − 3s2  24s(s2 − 1)
L(t3 sin t) = (−1)3   =   = 2 =
∴ ds3  s2 + 1  ds2  ( s2 + 1)2  ds  (s2 + 1)3  (s2 + 1)4
24 s (s2 − 1)
Hence the integral ∫ t3 e−t sin t dt = L (t3 sin t ) = =0 when s=1
(s2 + 1)4
∞
e−t − e−3t
Example 11: Evaluate ∫ dt [KUK, 2005]
0 t
∞ −t ∞ ∞
e − e −3 t  1 − e −2 t  − st
( ) , when s = 1
Solution: Write ∫ dt = ∫ e−t 
  dt comparable to ∫0 e f t dt
0 t 0 t
∞ ∞
 e−2t 
= ∫ e− st   dt − ∫ e− st 
1
 t 
dt
0
 t 0
 f (t )   g (t ) 
= L  − L
 t 
, where f(t) = 1, g(t) = e–2t
 t 
∞ ∞
1 1 ∞   f (t)  ∞ 
ds = log s s − log ( s + 2 ) s  using L 
∞
= ∫ ds − ∫  = ∫ f (s) ds
s s s s+2   t  s 
∞
= log 
s 
= 0 − log 
1 
 s + 2  s  1 + 2  = – log1 + log3 = (log3), when s = 1
 
s
Alternately:
∞ −t ∞ −t ∞ − 3t
e − e−3t e e
∫ dt = ∫ dt − ∫ dt
0 t 0 t 0 t
∞ ∞
f (t)  f (t) 
= ∫ e−st dt − ∫ e−st  n fn
 d t (def . By de . when f(t) = 1)
0 t 0  t
for s = 1 for s = 3
∞ ∞ ∞ ∞
1 1 ∞ ∞ 1 1
= ∫ ds − ∫ ds = log s s − log s s = − log + log (On reformatting)
s s s s for s = 1 for s = 3 ss ss
for s = 1 for s = 3 for s = 1 for s = 3
= ( − 0 + log 1) +  0 − log  = log 3  as s → ∞, → 0

1 1
 
3 s 
ASSIGNMENT 1
Find Laplace Transforms of followings:
1.
(i) cosh at cos at (ii) *cosh at sin at
(iii) sin h at cos at (iv) sinh at sin at

(v) sin at cosh at + cos at sinh at (vi) sin at cosh at – cos at sinh at
*[NIT Kurukshetra; Delhi 2002, 2003]
2.
at − sin at
(i) sin at cos at (ii)
a3
3.
t2 , 0<t<2   2π  2π
 sin  t − 3  , t>
3
(i) f (t) = t2 − 1, 2<t<3 (ii) f (t) = 
7, t>3 0, 2π
t<
 3
4.
(i) t sin at [Raipur 2005] (ii) t sinh at
(iii) t cos at (iv) t cosh at
5.
(i) sin at + at cos at (ii) sinh at + at cosh at
(iii) sin at – at cos at (iv) sinh at – at coshat
6.
1 1
(i) cos at − at sin at (ii) cosh at + at sinh at
2 2
7.
cos at − cos bt 1 − cos 2t
*(i) (ii)
t t
*[VTU 2006; INTU 2005; UPTU 2005]
sin at e−t sin t
(iii) [WBTU, 2005] **(iv) [KUK,*2003-04, **2006]
t t
8. Find the laplace transform sin at and hence deduce L(cos at). using transforms of
differentiation.
9. Find the laplace transform of cosh at and hence deduce L(sinh at). using the transforms
of derivatives.
10. Find the laplace transform of the following:
t
−t
t
cos at − cos bt
*(i) ∫ e cos t dt (ii) ∫ dt
0 0 t
*[Punjabi Univ., 2003]
11. Show that
∞ ∞ −t
− 2t 3 e sin2 t 1
(i) ∫ te cos t dt = , * (ii) ∫ dt = log 5,
0 25 0 t 4
∞
 cos at − cos bt  dt = 1 log b 2
(iii) ∫   *[KUK, 2006, 2010]
0 t 2 a2
12.6 INVERSE LAPLACE TRANSFORMS

To find the inverse laplace transform of given functions, we try to recognize the given
function either in the given form comparable to some standard expression whose inverse
function is a known standard function or split the given function of s into number of expressions
in s (with the help of partial fractions) comparable again to some standard functions of s of
which inverse functions are known.
Let us list first inverse laplace transform of some standard elementary functions and
associated functions.
L−1   = 1  1  = at
1
L−1 
 s − a 
e
 s
 1 tn −1  1  eat t n −1
L−1  n  = L−1  =
 s  (n − 1)!  (s − a)n  n − 1!
L−1  2
1  1  s  = cos
= sin at L−1  2
 s + a2  a  s + a2 
at
L−1  2
1  1  s  = cosh
= sinh at L−1  2
 s − a2  a  s − a2 
at
 1  1 at  s−a 
L−1  2
= e sin bt L−1  2
= eat cos bt
 (s − a) + b  b
2
 (s − a) +b 
2
 s  1  s  1
L−1  2 2 2
= t sin at L−1  2 2 2
= t sinh at
 ( s + a )  2 a  ( s − a )  2a
 s2 − a2   s2 + a2 
L−1  2 = t cos at L−1  2 = t cosh at
 (s + a2 )2   (s − a2 )2 
 1  1  1  1
L−1  2 2 2
= (sin at − at cos at) L−1  2 2 2
= − 3 (sinh at − at cosh at)
 (s + a )  2a3  (s − a )  2a
 s2  1  s2  1
L−1  2 2 2
= (sin at + at cos at) L−1  2 2 2
= (sinh at + at cosh at)
 (s + a )  2a  (s − a )  2 a
 s3  1  s3  1
L−1  2 2 2
= (cos at − at sin at) L−1  2 2 2
= (cosh at + at sinh at)
 (s + a )  2  (s − a )  2
 s3 
L−1  4
s  1
= L−1  4 = cosh at.cos at
 s + 4a4  2a2  s + 4a4 
sinh at.sin at
 s2 + 2a2  1  s2 − 2a2  1
L−1  4 = cosh at.sin at L−1  4 = sinh at.cos at
 s + 4a4  a  s + 4a4  a
L−1  4
1  1
= (cosh at sin at − sinh at cos at)
 s + 4a4  4a3
 s2  1
L−1  4 = (cosh at sin at + sinh at cos at)
 s + 4a4  2a
 1   1 
L−1  2 = J0(at) L−1  2 = I0(at)
 s + a 
2  s − a 
2
   s 
s
L−1  3  = tJ0 (at) L−1  ( s2 − a2 )32  = t I0(at)
 ( s2 + a2 )2   
Inverse transforms tabulated above will be frequently used as ready reckner, particularly
for solving differential equations using laplace transforms. Without going into details of
derivation of these, we may discuss some of them as under:
s3 s(s2 + a2 − a2 ) s s
= = 2 − a2 2
(s + a )
2 2 2
(s + a )
2 2 2
(s + a )
2
(s + a2 )2
∴
 s3   s   s 
L−1  2 2 2
= L−1  2 2 
− a2 L−1  2 2 2
 (s + a )   (s + a )   (s + a ) 
t sin at 1
= cos at − a2 = cos at − at sin at
2a 2
 s3   s  2 −1  s 
Likewise, L−1  2 2 2
= L−1  2 2 
+a L  2 2 2
 (s − a )   (s − a )   (s − a ) 
t sinh at 1
= cosh at + a2 = cosh at + at sinh at
2a 2
As already stated, in deducing these results, we need to make use of partial fractions.
Some of the properties of inverse laplace transforms and Convolution Theorem are taken up
in subsequent discussions.
a0 sm + a1sm−1 + … + am
Tips for Partial Fractions: In any general expression
b0 sn + b1sn −1 + … + bn
A
Corresponding to a non-repeated linear factor (s – a), write
(s − a)
A1 A2 Ar
Corresponding to a non-repeated linear factor (s – a), write + +…+
(s − a) (s − a)2
(s − a)r
As + B
Corresponding to a non–repeated quadratic factor (s2 + as + b), write
(s + as + b)
2
Corresponding to repeated quadratic factors (s2 + as + b), write

A1s + B1 A2 s + B2 A s + Br
+ + …+ 2 r
(s2 + as + b) (s2 + as + b)2 (s + as + b)r
Then determine unknown constants A, B, C, … A1, A2, A3, … by equating the coefficients
of equal powers of s on both sides.
Example 12: Find the inverse Laplace transform of

1 , 1 ,
(i) (ii)
( s + a )n ( As + B )n
where A, B and ‘a’ are constants and n is a positive integer.
Solution:
 1  tn −1 , using first shift property.
(i) L−1  −at −1  1 
n = e L  n  = e−at
s 
 ( s + a )  n − 1!
 
 1  1 −1  1  1 − AB t −1  1  1 − AB t tn−1
(ii) L−1  n = L   = e L  n  = e
 ( As + B)  A n − 1!
n
n
 B   An s An
  s + A  
As + B As + B
Example 13: Find the inverse laplace transform of and .
Cs + Ds + E ( Cs + Ds + E )
2 2 2
Solution:
As + B A s B 1
(i) = + ⋅
Cs + Ds + E C s2 + s +
2 D E C s2 + s + E
D
C C C C
A s B 1 , where F = D and G = E
= +
C s2 + Fs + G C s2 + Fs + G C C
A s B 1
= ⋅ 2 + ⋅ 2
C   F   S + F  +  G − F2 
S +  + G − 
F 2 C
  
2  4  2 4
(By completing squares) … (1)
2
Case I: When G > F , then the expression (s2 + Fs + G) takes the form (s + a)2 + b2, where
4
F  F2 
a= and b2 =  G −  , then
2  4
As + B  A −1   B −1  
L−1  2
s 1
 = L  + L  
 Cs + Ds + E  C  ( s + a ) + b  C  ( s + a ) + b 
2 2 2 2
A −1  ( s + a ) − a  B −1  1 
= L   + L  
 ( s + a ) + b  C  ( s + a ) + b 
2 2 2 2
C
A  − 1  ( s + a)   a   B −1  1 
= L   − L−1    + L  
C   ( s + a) + b2 
2
 ( s + a ) + b2   C
2
 ( s + a) + b2 
2
A −at  −1  s  −1  1   B −at −1  1 
= e L  2  − a L  2  + e L  2
C   s + b2  s + b2   C  s + b2 
A −at  a  B sin bt
= e cos bt − sin bt  + e−at .
C  b  C b
B − aA
= e−at  cos bt + 
A
sin bt  … (2)
C bC 
2
Case II: when G < F , then
4
As + B  A −1   B  
L−1  2
s 1
 = L   + L−1  
 Cs + Ds + E  C  ( s + a ) − b  C  ( s + a ) − b 
2 2 2 2
A −1  ( s + a ) − a  B −1  1 
= L   + L  
 ( s + a ) − b  C  ( s + a ) − b 
2 2 2 2
C
A  −1  ( s + a )   1   B −1  1 
= L   − a L−1   + L  
C   ( s + a ) − b 
2 2
 ( s + a ) − b   C
2 2
 ( s + a ) − b 
2 2
A −at  −1  1  −1  1   B −at −1  1 
= e L  2  − a L  2  + e L  2
C   s − b2  s − b2   C  s − b2 
A  −at a  B sinh bt
=  e cosh bt − sinh bt  + e−at .
C b  C b
+ −
∴ L−1  2As B  = e−at  A cosh bt + B aA sinh bt  … (3)
 Cs + Ds + E   C bC 
2
Case III: when G = F , then
4
 As + B  = A −1  s  + B 1 
L−1  2 L
 Cs + Ds + E  C  F
2
 F2   C   F
2
 F2  
  s +  −  G +     s +  −  G −  
2 4 2 4
A −1  s  B −1  1 
= L  2 + L  2
C  ( s + a )  C  ( s + a ) 
A  − 1  ( s + a) − a   B − 1  1 
= L + L  2
C   ( s + a )2   C  ( s + a ) 
A  −1  1   1   B −1  1 
= L   − a L−1  2  + L  2
C   s + a   ( s + a )   C  ( s + a ) 
A −at  −1  1 1 B  1
= e L   − a L−1 2  + e−at L−1  2  .
C   s  s  C s 
A −at B
= e (1 − at ) + e−at t
C C
−1  As + B 
= e−at  (1 − at ) + t 
A B
∴ L  2 
 Cs + Ds + E  C C 
2s + 1
Example 14: Find inverse laplace of .
( s + 2 )2 ( s − 1 )2
2s + 1 A B C D
Solution: Write 2 = + 2 + +
( s + 2) ( s − 1) s + 2 ( s + 2) ( s − 1) ( s − 1)2
2
Implying
(2s + 1) = A(s + 2)(s – 1)2 + B(s – 1)2 + C(s + 2)2(s – 1) + D(s + 2)2 … (1)
= A(s + 2)(s2 – 2s + 1) + B(s2 – 2s + 1) + C(s – 1)(s2 + 4s + 4) + D(s2 + 4s + 4)
= A(s3 – 3s + 2) + B(s2 – 2s + 1) + C(s3 + 3s2 – 4) + D(s2 + 4s + 4)
(2s + 1) = s3(A + C) + s2(B + 3C + D) + s(– 3A – 2B + 4D) + (2A + B – 4C + 4D) … (2)
From (1), if s = 1; then (2 + 1) = D(1 + 2)2 1

implying D= … (3)
3
if s = 2; then (–4 + 1) = B(–2 – 1)2 implying B = − 1 … (4)

3
Further on equating coefficients of s3 and constant terms on both sides of equation (2), we
get,
A+C=0 and 2A + B – 4C + 4D = 1 … (5)
Thus on putting D = 1 and B=−

1 ; equations (5) results in A = 0 = C
3 3
 2s + 1   1 1  − 1 − 2t 1 t
∴ L−1  −1
2  = L − 2 + 2 = − e + e
 ( s + 2 ) ( s − 1)   3 ( s + 2 ) 3 ( s − 1) 
2
3 3
s
Example 15: Find L−1
s + 2s2 + 1
4
Solution: Write s4 + s2 + 1 = (s4 + 2s2 + 1) –s2 = (s2 + 1)2 – s2 = (s2 + 1 + s)(s2 + 1 – s)

s s As + B Cs + D
= = +
s4 + s2 + 1 ( s2 + 1 + s)( s2 + 1 − s) ( s2 + 1 − s ) ( s2 + 1 + s )
Thus,
or s = (As + B)(s2 + 1 + s) + (Cs + D)(s2 + 1 – s)

= s3(A + C) + s2(A + B – C + D) + s(A + B + C – D) + (B + D)
On comparing coefficients of equals powers of s on both sides,
For s3 , A+C = 0 …(1) 

For s2 , A+B−C+D =0 …( 2 )
For s, A+B+C−D =1 …( 3 )
Constant, B+D= 0 …( 4)
All these equation together give B = 1 , D = − 1

2 2
L−1  4
s  L−1  1 1 1 1 
∴ =  − 
 s + s + 1 
2
 2 ( s + 1 − s) 2 s + 1 + s 
2 2
 1   1 
1 −1   2 1  3  1 −1   
= L  s + − s +  − L  s2 + 1 + s + 3 
2  4  4 2  4  4
   
   
 1   1 
= L−1   s − 1  +  3   − L−1    3 
1 2 2 1 2 2
1
     s+ + 
2  2  
2 2   2
  2
   
   
1  1  −1  1 
e 2 L−1 
1 t 2 1 2 t −1  2
=  3  − e L  3  By Ist shift property
 s2 +   2  s2 +  
 2    2  
2
 
   
 3  3
sin 
 2  1 sin 
 2 
t t t t
−
1 1
t −
1 2 t  2  e2 − e 2 3
= ⋅ − e =
 3 
e2 sin t
2  3 2  3 2 2
   
2  2 
2 t 3
= sinh sin t
3 2 2
s3
Example 16: Find inverse Laplace transform of [NIT Jalandhar, 2007]
s − a4
4
s3 s3 As + B Cs + D
Solution: write = 2 = +
s −a
4 4
( s − a )(s2 + a2 ) (s2 − a2 ) ( s2 + a2 )
2
Implying s3 = (As + B)(s2 + a2) + (Cs + D)(s2 – a2)

= s3(A + C) + s2(B + D) + s(a2A – a2C) + a2(B – D)
On comparing coefficients of equal powers of s3, s2, s and constant on both sides we get,
A+C =1 
B+D= 0  1
A= = C, B = 0 = D.
a2 ( A − C ) = 0
All together implying
 2
a2 (B − D ) = 0 
 s3 
= L−1  2
1 s 1 s  1
∴ L−1  4 + = [ cosh at + cos at ]
 s − a4  2 s − a2
2 s2 + a2  2
Example 17: Find inverse transforms of
1 , s , s2 ,
(i) 4 (ii) (iii)*
s + 4a4
s + 4a4
4
s4 + 4a4
s3 , a(s2 − 2a2 )
(iv) (v)** [*KUK, 2001, 2002; **NIT Kurukshetra, 2010]
s + 4a4
4
s4 + 4a4
Solution:
(i) Write, s4 + 4a4 = s4 + 4a4 + 4a2s2 – 4a2s2 = (s2 + 2a2)2 – (2as)2
= (s2 + 2a2 – 2as)(s2 + 2a2 + 2as)
1 As + B Cs + D 1  s + 2a s − 2a 
= + = −
s4 + 4a4 s2 + 2a2 + 2as s2 + 2a2 − 2as 8a3  s2 + 2as + 2a2 s2 − 2as + 2a2 
so that
 By partial fractions, A = 1 = − C, B = 1 = D
 
8a3 4a2
1 −1  ( s + a ) + a ( s − a) − a 
Now L−1  4 
1
 s + 4a4  = 8a3 L  ( s + a )2 + a2 − ( s − a)2 + a2 
 
1  1 s+a 1 s−a 1 
=  L− + a L−1 − L−1 + a L−1 
 ( s + a ) + a ( s + a ) + a2 ( s − a) + a2 ( s − a) + a2 
2 2 2 2
8a3 2
=
1  − at −1 s + − at −1 a − at −1 s + at −1 a 
8a3 e L s2 + a2 e L s2 + a2 e L s2 + a2 e L s2 + a2 
(By first shift property)

1 −at
= e cos at + e−at sin at − eat cos at + aeat sin at 
8a3 
=
1  (eat + e−at ) + cos at (e−at − eat ) 
 sin at 
4a3  2 2 
1
= sin at cosh at − cos at sinh at 
4a3 
(ii) L−1  s  1 −1  1 1 
 s4 + 4a4  = 4a L  s2 − 2as + 2a2 − s2 + 2as + 2a2 
By partial fraction
1  −1  1   1 
= L   − L−1  
4a   ( s − a ) + a2 
2
 ( s + a ) + a2  
2
1  at −1 1 1 
=  e L 2 − e−at L−1 2
4a  s +a2
s + a2 
=
1  at sin at −at sin at 
e − e =
1
sin at
(eat − e−at ) = 1 sin at sinh at
4a  a a  2a2 2 2a2
 s2  1 −1  s s  , (By partial fractions)

L−1  4 = −
 s + 4a4  4a
(iii) L  2
 s − 2as + 2a2 s2 + 2as + 2a2 
1 −1  s s 
= L
 − 2
4a  ( s − a )2
+ a2
( s + a )2
+ a 
1 −1  ( s − a ) + a ( s + a) − a 
= L  − 
 ( s − a ) + a ( s + a)2 + a2 
2 2
4a
1  −1 ( s − a ) a (s + a ) + L−1 a 
= L + L−1 − L−1 
4a  ( s − a ) + a2
2
(s − a ) + a2
2
(s + a ) + a2
2
(s + a ) + a2 
2
1  −1 ( s − a ) a ( s + a) + L−1 a 
= L + L−1 − L−1 
4a  ( s − a ) + a2
2
( s − a) + a2
2
( s + a) + a2
2
( s + a) + a2 
2
1  at −1 s a s a 
= e L 2 + eat L−1 2 − e−at L−1 2 + e−at L−1 2
4a  s + a2 s + a2 s + a2 s + a2 
1 at
= e cos at + eat sin at − e−at cos at + e−at sin at 
4a 
1   eat − e−at   eat + e−at  

=  cos at   + sin at   
2a   2   2 
1
= cos at sinh at + sin at cosh at 
2a 
−1  s3  1 −1  s−a s+a 
(iv) L  s4 + 4a4  = 2 L  ( s − a )2 + a2 + ( s − a )2 + a2  (By partial fractions)
 
1 at 1
= e cos at + e−at cos at = cosh at cos at
2 2
a ( s2 − 2a2 ) As + B Cs + D
(v) Here = + 2
( s4 − 4a4 ) ( s2
+ 2 a2
− 2as ) s + 2a2 + 2as
1 −a
On resolving partial fractions we get A = = C and B = =D
2 2
 a ( s2 − a2 )  1  ( s − a) (s + a) 
∴ L−1  4  = L−1
 − 
 ( s + 4a )  2  ( s − a ) + a ( s + a )2 + a2 
4 2 2
1  at −1 s s 
= e L 2 − e−at L−1 2
2  s + a2 s + a2 
1 at 1
= e cos at − e−at cos at.
2 2
eat − e− at
= cos at = sinh at cos at.
2
12.7 SOME IMPORTANT PROPERTIES ON INVERSE LAPLACE TRANSFORMS:
( )
If L−1 f (s) = f (t), then
( )
(i) L−1 f (s − a) = eat f (t), (ii) ( )
L−1 s f (s) =
d
dt
f (t), provided f(0) = 0
 f (s)  t  d ( ) = ( ),
(iii) L−1   = ∫ f (u) du, (iv) L−1  − f s tf t
 s  0  ds 
∞  f (t)
(v) L−1  ∫ f (s) ds =
s  t
 f (t − a ) ,
( )
(vi) L−1 e− as f (s) = F(t), where F(t) = 
0
t>a
t<a
Example 18: Find the inverse laplace of
( s + 2)  1 cos 1  1
(i) , (ii)   (iii)
( s + 4s + 5) s s3 ( s2 + 1 )
2
2 s
Solution:
( s + 2) =−
1 d  1  1 d
 2 =− f (s)
2 ds  ( s + 4s + 5) 
(i) Write …(1)
( s2 + 4s + 5) 2
2 ds
Now on using property; if L−1 f (s) = f (t) then  d ( ) = ( )

L−1  − f s tf t
 ds 
1  1   1 
Here f (t) = L−1 = L−1  2  = L−1   = e−2t sin t
( s + 4s + 5 )
2
 ( s + 4 s + 4 ) + 1   ( s + 2 )2
+ 12

∴ L−1  s+2  = 1 t f (t) = 1 te−2t sin t

 s2 + 4s + 5 2  2
( ) 2
(ii) We know that
1 1 1  x2 x 4 
cos = 1 − 2 + 4 + ……  using, cos x = 1 − + …
s s 2! s 4!  2! 4! 
1 
L−1  cos  = L−1   1 − − … 
1 1 1 1
∴ +
s s s  2! s2
4! s2  
= L−1  − − …
1 1 1
+
 s 2! s3 4! s5 
t2 t4
= 1− 2
+ −…
(2!) (4!)2
t
(iii) As L−1  2
1 
L−1  2  = sin u du = (1 − cos t)
1
 ( s + 1)  = ∫
sin t, therefore
 s ( s + 1)  0

 
t
implying L−1  2 2  = (1 − cos u) du = ( t − sin t )
1
 s ( s + 1)  0∫
 
 = (u − sin u) du =  t2 + cos t − 1
t
and L−1  3 2
1
 ∫  
 s ( s + 1)  0 2
s−1
Example 19: Find inverse laplace transform of s.log
s+ 1
[NIT Kurukshetra 2003]
s − 1 d
f (s) = −  s log  = ( s log ( s + 1) − s log(s − 1))
d d
Solution: Here, −
ds ds  s + 1  ds
 s    s  
= log ( s + 1) +    − log ( s − 1) +  
  s + 1   s − 1  
 s 
= {log ( s + 1) − log ( s − 1)} + 
s
−
  s + 1 s − 1  
, where φ(s) = {log(s + 1) − log(s − 1)}

2s
= φ(s) − …(1)
s2 − 1
Taking inverses laplace on both sides,
t. f (t) = L−1 φ(s) − 2 cosh t …(2)
t φ(t) = L−1 − φ(s) = L−1 log ( s − 1) − log ( s + 1)

d d
Again,
 ds  ds
= L−1 
1 1  2
− = L−1 2 = 2 sinh t
 s − 1 s + 1  s −1
2 sinh t
or φ(t) = …(3)
t
Using result (3) in (2), we get
sinh t − t cosh t 
− 2 cosh t or f (t) = 2 
2 sinh t
t f (t) = 
t  t2
Example 20: Find inverse laplace transform of

1 s s2
(i)
( s2 + a )2 2 , (ii)
( s2 + a )
2 2
(SVTU 2007) (iii)*
( s2 + a2 )
2
[Madrass 2000; *KUK, June 2004]
1 1 s f (s)
Solution: = 2 is comparable to … (1)
(s 2
+a )
2 2 s (s + a )
2 2 s
s 1 d  1  d
=− is comparable to − φ(s)
Now
(s 2
+a 2 2
) 2 ds  ( s + a2 ) 
 2
ds
L−1 
s  = 1 t φ(t), where φ(t) = L−1  1  sin at
∴  s2 + a2  = a ( )
 (
 s2 + a2 2  2
 )  
Implying L−1 
s  = 1 t sin at , which proves result (ii) ...(2)
 (
 s2 + a2 2  2 a )
For (i),
L−1 
1  = L−1  1 s  = L−1  f(s) 
 s 
 (
 s2 + a2 2 
 )  s s2 + a2 2 
  ( )
t t
= ∫ f (u) du = ∫
2
1
u sin au du, (using (2))
0 0 a
t
1  − cos au  t
 cos au 
=  u ⋅  − ∫ 1  −  du
2a a 0 0 a 
1  t cos at sin at  1
=  − + 2 
= 3 (sin at − at cos at)
2a  a a  2a
∞
 f (t) 
For (ii), Alternately ∫ f (s) ds = L 
s  t 
1 1 
∞
 f (t)  ∞ s 1∞ 2s 1 1 
or L  = ∫ ds = ∫ ds = − =  2 
 t  s ( s2 + a2 )2
2 s ( s2 + a2 )2  
2  s + a s
2 2
2  ( s + a2 ) 
f (t) 1 sin at t sin at

Taking inverses, = or f (t) =
t 2 a 2a
s2 s
For (iii) = s⋅ comparable to s f (s)
(s2
+a 2 2
) (s 2
+ a2 )
2
∴ L−1
 s2 
 ( s2 + a2 )2 
= L−1 s f (s) =
d
dt
( )
f (t), where f (t) = L−1 f (s) and f (0) = 0 ( )
 
Here L−1 f (s) = L−1  s  = t sin at , result (ii)

 ( s2 + a2 )2  2a
Implying L−1   d  t sin at 

 ( s2 + a2 )2  dt  2a  2a (
sin at + at cos at )
s2 1
= =
 
12.8 CONVOLUTION THEOREM

Convolution: Let f(t) and g(t) be two functions of class ‘A’, then the convolution of the two
functions f(t) and g(t) denoted by f ⊗ g is defined as:
t
f ⊗ g = ∫0 f (u) g(t − u) du, where f ⊗ g is also known as falting of f(t) and g(t).
The convolution f ⊗ g is
(i) Commutative i.e. f ⊗ g = g ⊗ f
(ii) Associative i.e. (f ⊗ g) ⊗ h = f ⊗(g ⊗ h)
(iii) Distributive with respect to addition i.e. f ⊗ (g + h) = f ⊗ g + f ⊗ h
( )
t
Statement: If L−1 f (s) = f (t) and L−1 g(s) = g(t), then L−1 f (s)g(s) = ∫ f (u)g(t − u) du
0
∞ ∞ t 
Proof: L( f ⊗ g) = ∫ e− st ( f ⊗ g) dt = ∫ e− st ∫ f (u)g(t − u) du dt … (1)
0 0 0 
∞t
= ∫ ∫ e−st f (u) g(t − u) dudt … (2)
00
(Expression (1) clearly shows that the integration is Ist taken along the vertical strip PQ
from P to Q, P resting on the curve u = 0 and Q resting on the curve u = t and finally the strip
PQ slides between t = 0 to ∞ to cover the full dotted region)
On changing the order of integration, we get u
∞∞ 
L( f ⊗ g) = ∫  ∫ ( e−st f (u) g(t − u) dt ) du … (3)
u=t
0u  Q
t=u t →∞
∞∞ P

= ∫  ∫ (e−s(t −u + u) f (u) g(t − u) dt ) du 0 u=0 t
0u  Fig: 12.3
∞ ∞ 
= ∫ e−su f (u)  ∫ ( e−s(t −u) g(t − u) dt ) du
0 u 
∞ ∞ 
= ∫ e−su f (u)  ∫ e−sp g(p) dp du, (on putting t – u = p so that dt = dp)
0 0 
∞ ∞ 
0
( )
= ∫ e−su f (u) g(s) du =  ∫ e−su f (u) du g(s) = f (s) g(s) (Using defn. of laplace)
0 
Whence the desired result.
Example 21: Using Convolution theorem, find
−1  s   s2 
(i) L  2 L− 1  4
( ) (s + a ) 
2  (ii)* 4 
s + a
*[KUK, 2002, 03, NIT Kurukshetra, 2005]
Solution:
 s  = L−1  s 1 ,
(i) Write, L−1  2 where in L−1  s  = cos at and
 ( s + a )
2 


 ( s2
+ a2
) ( s2
+ a )
2 
 (
 s2
+ a )
2 
L−1  2  = sin at
s
 ( s + a2 )  a
 
∞
−1  s   1  sin a(t − u) 
∴ L  2 2  2 2  = (cos au) 
∫  du

 s −a  
s +a   a
0
1t
= ∫ cos au ⋅ (sin at cos au − cos at sin au) du
a0
sin at t cos at t
= ∫ cos2 au du − ∫ sin 2au du
a 0 2a 0
sin at t cos at t
= ∫ (1 + cos 2au) du − ∫ sin 2au du
2a 0 2a 0
t
sin at  sin 2at  cos at  cos 2au 
= t+ − −
2a  2a  2a  2a 0
t sin at ( 2 sin2 at ) ( cos at ) cos at 1

= + + ( cos 2at − 1)
2a 2a 2a 2a 2a
t sin at
=
2a
−1  s2  −1  s s  −1  
(ii) L  ( s4 − a4 )  = L  (s2 − a2 ) (s2 + a2 )  = L  f (s) g(s) = f (t) ⊗ g(t)
   
 s  = cosh −1  s 
at and g(t) = L g(s) = L  2  = cos at
−1
f (t) = L−1 f (s) = L−1  2
 s − a2 
where,
s + a2 
 s2  t
∴ L−1  4 4 
= ∫ cosh au cos a(t − u) du = I ,
 ( s − a )  0
Integrate by parts, taking coshau as Ist function and cos a(t – u) as second function,

{ } sin a(t − u) 
t
sin a(t − u) t
I =  cosh au. − ∫ a sinh at du 
 −a 0 0 −a 
 t
sin a(t − u  t 
=  cos au  + ∫ sinh au sin a(t − u) du
 −a 0 0 
 
 1  t
− cos(t − u) t − cos a(t − u)  
= − (0 − sin at) +  sinh au. −∫ a cosh au. du 
 a  −a 0 0
−a  
 sin at  sinh au cos a(t − u)  t 

I= +  − I
 0
 a a 
2I = 
sin at sinh at  1
implying +  or I= (sin at + sinh at)
 a a  2a
Hence L−1  s  = 1 [sin at + sinh at]

2
 s4 + a4  2a
eau + e− au
Alternately, may take cosh au = in integral I.
2
1
Example 22: Using convolution theorem, find L−1 . [ KUK, 2004]
s ( s2 + 4 )
1 1 1
Solution: Write = comparable to f (s) g(s) with
s ( s2 + 4 ) s s2 + 22
f (s) =
1
s
( )
and L−1 f (s) = 1; g(s) = 2
1
s + 22
and L−1  2
1  sin 2t
 s + 22 
=
2
 1  1t 1  − cos 2(t − u) 
t
χ L−1  2  = ∫ 1 ⋅ sin 2(t − u) du =

 s ( s + 4)  2 0 2 −2 0
t
1 1
= cos(2t − 2u) = (1 − cos 2t)
4 0 4
1 1 s
Alternately: = 2 2 ; f (s) = 1 and g(s) = cos 2t
s ( s + 4 ) s s + 22
2
s2 2
 1  1t 1
so that L−1  2 2  = ∫ u cos 2(t − u) du = (1 − cos 2t)
 s (s + 4)  2 0 4
ASSIGNMENT 2
Find the inverse laplace transform of
s s 1
, , ,
(s − 1) ( s − 3) (s − 4)
1. (i) 2 (ii) (iii)
2 2 s − a3
3
s2 + s − 2 s
( s + 1)2 ( s2 + 1)
(iv) (v)* *[NIT Kurukshetra, 2005]
s(s + 3)(s − 2)
As + B
2. where A, B, C, D, are constants.
(Cs 2
+ Ds + E)
2
2s − 3 3s + 7 3s
3. (i) , (ii) , (iii)
s2 + 4s + 13 s2 − 2s − 3 s2 + 2s − 8
s+2 s+3 s
4. (i)* , (ii) , (iii)**
(s2
+ 4s + 5)
2
(s2
+ 6s + 13)
2
(s2
− 1)
2
[*Madrass, 2003; PTU, 2005; **KUK, 2005]
s2 + 1
cot−1   or tan−1   (iii)** tan−1  2 
log s 2 2
5. (i)* (ii)
s ( s + 1)  2  s s 
[*VTU, 2004; **Bombay, 2005]
(s + a)  (s + 1) 
(vi)* log  1 − a 
2
(iv)* log (v) log 
(s + b)  (s + 2)(s + 3)   s2 
[*Anna 2003, UP Tech 2003]
 ( s2 + 1)  1  s2 + b2 
(viii) cot− 1(s + 1) log  2
 s + a2 
(vii)* log  2
(ix)**
 ( s − 1)  2
[*Madras, 2005; **VTU, 2007]
s s s
6. (i)* (ii) (iii)**
a s + b2
2 2 ( s + a)2 ( s + a)3
[*Madras, 2005; **KUK, 2004]
1 1 1
s ( s + 2) s (s + 2) s2 ( s2 + a2 )
7. (i) 3 (ii) 2 (iii)*
[*NIT Kurukshetra, 2010]
1 1 t3 t5
8. Show that L−1  sin  = t − + − …… ∞
s s (3!) (5!)2
3
9. Use Convolution theorem to find inverse of the following:
1 1 1 1
(i) (ii) (iii)∆ (iv)
(s + a)(s + b) (s2
+a )
2 2 s ( s + 1)
2 2
s ( s + a2 )
2 2 2
1 s s2
(v)* (vi)** (vi)***
( s + 1)( s + 9)2 (s2
+ 1)( s + 4 )
2
( s2 + a2 )( s2 + b2 )
[KUK, *2004, 2005-06; **2004-05 *** JNTU, 05, SVTU, 07; ∆KUK, 2010]
Note: Most of the above problems can be handled either by partial fractions or convolution theorem. However, for easy
( )
handling of problems 4 & 5, apply property L−1  − d f (s) = tf (t); in problems (6), apply property L s f (s) =
−1 d
f (t);
 ds  dt
 f (s)  t
in problems 7 & 8, apply property L−1   = ∫ f (u)du.
 s  0
12.9 APPLICATION TO DIFFERENTIAL EQUATIONS
Example 23: Using laplace transforms, solve (D4 – K4)y = 0

where y(0) = 1, y’(0) = y”(0) = y”’(0) = 0 [NIT Kurukshetra, 2004]
Solution: Taking laplace transform of each individual term of the given equation,
L(D4y) – K4L(y) = 0
Implying s4y(s) − s3y(0) − s2y′(0) − sy′′(0) − y′′′(0) − K 4y(s) = 0
s3
i.e. ( s4 − K 4 ) y(s) = s3 implying y(s) = ( s − K )( s + K ) ( s2 + K2 )
… (1)
Then by partial fractions,

s3 A B Cs + D
= + +
(s − K )( s + K ) ( s2 + K2 ) ( s − K ) ( s + K ) s2 + K2 … (2)
implying s3 = A(s + K)(s2 + K2) + B(s – K)(s2 + K2) + (Cs + D)(s2 – K2)
On comparing coefficients of equal powers of s on both sides
1 = (A + B + C) … (i)
0 = (A – B) K + D … (ii)
0 = (A + B – C)K2 … (iii)
0 = (AK – BK – D)K2 … (iv)
Subtracting (iii) from (i), we get C = 1

2
Subtracting (iv) from (ii), we get D = 0
With values of C and D, equations (i) and (iv) becomes
1
A+B= 1 1
2 implying A = and B =
A−B=0 4 4
Putting the values of A, B, C and D in equation (2) or more precisely in (1), we get
1 1 1 s
y(s) = + +
4 ( s − K ) 4 ( s + K ) 2 ( s2 + K 2 ) … (3)
Taking inverse transforms of each individual terms on both sides,

1 Kt 1 −Kt 1 1  eKt + e−Kt  1 1

y(t) = e + e + cos Kt =   + cos Kt = (cosh Kt + cos Kt )
4 4 2 2 2  2 2
d3y d2 y dy dy d2 y
Example 24: Solve + 2 − − 2y = 0 when y = 1, = =2 at x=0
dx3 dx2 dx dx dx2
Solution: Taking laplace transforms on both sides of the given equation

 s3 y(s) − s2 y(0) − sy ′(0) − y ′′(0) + 2  s2 y(s) − sy(0) − y ′(0) − sy(s) − y(0) − 2y(s) = 0 … (1)
Making use of the given conditions,
 s3 y − s2 ⋅ 1 − s ⋅ 2 − 2  + 2  s2 y − s ⋅ 1 − 2  − sy − 1 − 2 y = 0
i.e. ( s3 + 2s2 − s − 2) y(s) = (s2 + 4s + 5) … (2)
y(s) =
(s2 + 4s + 5) =
(s2 + 4s + 5)
( s3 + 2s2 − s − 2) ( s − 1)( s + 1)( s + 2) … (3)
5 1   1  1 1 
y(s) =
  − +   (By partial fractions) … (4)
3  s − 1  s + 1 3  s + 2 
Taking inverse laplace on both sides of equation (4), we get
5 x 1
y(x) =
e − e−x + e−2x … (5)
3 3
which is the desired solution.
Example 25: Solve the equation (D2 + 1)x = t cos 2t; x = 0 = Dx at t = 0 [KUK, 2010]
Solution: Taking laplace on both sides,

d  s 
 s x − sx(0) − x′(0) + x(s) = − ds  s2 + 4 
2
s2 − 4
Using the given conditions (s2 + 1) x(s) =
( s2 + 4)2
x(s) =
( s2 − 4 )
i.e.
( s2 + 4)2 ( s2 + 1) … (1)
By partial fractions,
( s2 − 4) = As + B + Cs + D + Es + F
( s2 + 4 )2 ( s2 + 1) ( s2 + 1) ( s2 + 4 ) ( s2 + 4)2
or (s2 – 4) = (As + B)(s2 + 4)2 + (Cs + D)(s2 + 1)(s2 + 4) + (Es + F)(s2 + 1)
On comparing coefficients of equal powers of ‘s’ on both sides, we get
s5 , A+C = 0 ...(i)  A = 0 = C = E
s4 , B+D= 0 ...(ii)  5 
 B=− 
s3, 8A + 5C + E = 0 ...(iii) 9 
⇒
...(iv)  
5 … (2)
s2 , F + 8B + 5D = 1 D=
9 
s1 , 16A + 4C + E = 0 ...(v)  8 
s0 , 16B + 4D + F = − 4 ...(vi) F= 
3 
5 1 5 1 8 1
x(s) = − + +
9 ( s + 1) 9 ( s + 4 ) 3 ( s + 4 )2
Thus, 2 2 2
… (3)
Taking inverse laplace transform on both sides,

5 5 8 1
x(t) = − sin t + sin 2t + (sin 2t − 2t cos 2t )
9 18 3 2.23
 
 using L−1  1  = 1 (sin at − at cos at ) where a = 2 in this case 
  ( s + a )  2a 
2 2 2 3
 
5 5 1
x(t) = − sin t + sin 2t + (sin 2t − 2t cos 2t )
9 18 6
5 4 1
x(t) = − sin t + sin 2t − t cos 2t. Hence the result.
9 9 3
d2 x
Example 26: Solve π/2) = – 1. [UPTech 2002; Raipur 2004]
+ 9x = cos 2t if x (0) =1 and x(π
dt2
Solution: Taking laplace on both sides of the given equation,

s
 s2 ⋅ x(s) − sx(0) − x′(0) + 9 x(s) = 2
s +4
s
i.e. s2 x(s) − s.1 − a + 9 x(s) = , taking x ′(0) = a, say
s2 + 4
( s2 + 9) x = s2 s+ 4 + s + a
s s a
x(s) = + 2 + 2
or ( s + )( s + ) s + s + 9
2
4 2
9 9
1 s s  s a
x(s) = − + +
5  s2 + 4 s2 + 9  s2 + 9 s2 + 9
(By partial fractions) … (1)
Taking inverse laplace transforms on both sides

1 a
x(t) = [cos 2t − cos 3t ] + cos 3t + sin 3t … (2)
5 3
π 1 a 12
At t = , − 1 = [ − 1 − 0] + 0 + ( − 1) or a = … (3)
2 5 3 5
1
∴ x = (cos 2t + 4 cos 3t + 4 sin 3t )
5
t
dy
Example 27: Solve + 2 y + ∫ ydt = sin t , y ( 0 = 1 ) .
dt 0
Solution: Taking laplace on both sides,
 t  f (s)
( s y(s) − y(0)) + 2y(s) +
y(s) 1
= 2
s s +1
, using  L∫ f (t) dt  = s
 0 
On using the given condition,
 s + 2 + 1  y(s) = s2 + 2 s ( s2 + 2 )
  or y(s) =
s ( s2 + 1) ( s + 1)2 ( s2 + 1)
s ( s2 + 2)
A B Cs + D
+ = 2 +
(s + 1) ( s + 1) ( s + 1) ( s + 1) ( s2 + 1)
By partial fractions, 2 2
i.e. s(s2 + 2) = A(s + 1)(s2 + 1) + B(s2 + 1) + (Cs + D)(s + 1)2

On comparing the coefficients of equal powers of ‘s’ and solving for A, B, C, D.
3 1
A = 1, B=− , C = 0, D=
2 2
1 3 1 1 1
∴ y(s) = − 2 +
( s + 1) 2 ( s + 1) 2 s2 + 1
3 −t 1
Taking inverses, y(t) = e−t + te + sin t
2 2
t
Example 28: Solve y′(t) = t + ∫ y(t − u)cos u du if y(0) = 4.
0
Solution: Taking laplace of each individual term on both sides of the given equation,
t 
L ( y′(t)) = L(t) + L  ∫ y(t − u)cos u du 
0 
t
or ( s y(s) − y(0)) = s12 + L y ⊗ cos t; using def n. f ⊗ g = ∫ f (u)g(t − u) du
0
; L ( f ⊗ g ) = f (s)g(s)
1 s
s y(s) − 4 = + y(s) 2
s2 s +1
implying  s − 2
s 
=
1
+ =
( 4s2 + 1)( s2 + 1) = 4 + 5 + 1
  y (s) 4 or y (s)
s + 1 s2 s2 s3 s s3 s5
5 1 4
Taking inverses, y(t) = 4 + t2 + t
2 24
t
Example 29: Show that the solution of the equation y(t) = φ(t) + ∫ y(u) g(t − u) du can be
0
 φ(s) 
represented as y = L− 1  , where φ(s) = L ( φ(t) ) and g(s) = L ( g(t) ) . Hence,
 1 − g(s) 
t2 t
solve y(t) = − ∫ u y(t − u) du.
2 0
Solution: Laplace transform of the given integral equation results in

t 
y(s) = φ(s) + L  ∫ y(u) g(t − u) du = φ(s) + L ( y ⊗ g ) = φ(s) + y(s)g(s)
0 
 φ(s) 
On simplification, (1 − g(s)) y(s) = φ(s) or y(s) =  
 1 − g(s) 
 φ(s) 
Implying, y(t) = L−1  
 1 − g(s) 
 1   1 
or y(t) = L−1  φ(s)  = φ(t) ⊗ ψ(t) , where L−1   = ψ(t)
 1 − g(s)   1 − g(s) 
t
= ∫ φ(u) ψ(t − u) du
0
Now, for hence part
y(s) = 3 − L (t ⊗ y ) = 3 − 2 y(s)
1 1 1
s s s
 1 + 1  y(s) = 1 or y(s) = 1 1
or y(s) = − 2
s
 
or s2  s3 s ( s2 + 1) s s +1
Taking inverses, y = (1 – cost).
2
Example 30: Solve the equation t d x + dx + t x = 0, x(0) = 1, x ′(0) = 0.
dt2 dt
[NIT Kurukshetra, 2006]
Solution: Taking laplace of each individual term on both sides of the given equation,
 d2 x 
L  t 2  + L   + L ( t x) = 0
dx
 dt   dt  … (1)
⇒ −
d 2
ds
( s x(s) − s x(0) − x '(0)) + ( s x(s) − x(0)) − ( x(s)) = 0
d
ds
 using formula, L t f (t) = − d f (s)
 ( ) ds 
⇒
ds
( s x − s − 0 ) − ( sx − 1) + dsd x = 0 , (Taking x(0) = 1, and x’(0) = a(say))
d 2
 s2 dx + 2sx − 1 − sx − 1 + dx = 0
⇒   ( )
ds ds
dx 1  2s 
( s2 + 1) dx
ds
+ sx = 0 ⇒
x
+  2  ds = 0
2  s + 1
On integrating, log x + log ( s2 + 1)
1
C
= log C ⇒ x = 2
2
… (2)
s +1
Taking inverse laplace transforms,
x(t) = C J0(t) (It is a standard laplace of Bessel Function) …(3)
On using given condition, x(0) = 1; 1 = CJ0(0) or C = 1 as J0(0) = 1.
Hence x(t) = J0(t)
d2 y dy
Example 31: Solve the equation x + + xy = 0; y(0) = 2, y '(0) = 0.
dx2 dx
[NIT Kurukshetra, 2010]
Solution: Taking inverse laplace on both sides of the given equation

d2 y dy
x + + xy = 0, … (1)
dx2 dx
 s y − s y(0) − y´(0) + ( s y − y(0)) − ( y ) = 0

d 2 d
We get, −
ds  ds
− ( s2 y − 2s) + ( s y − 2 ) −
d dy
= 0 , using the conditions y(0) = 2, y’(0) = 0
ds ds
dy  s  dy 1  2s 
+ 2 y = 0 or +  ds = 0
 2  2 ( s2 + 1) 
implying
ds  s + 1 y
On integration, we get
C
log y + log ( s2 + 1) 2 = log C y=
1
i.e. … (2)
s +1
2
Taking Inverse laplace on both sides of equation (2), we get y(x) = C J0(x),
 1 
(Since L−1  2
 s + 1 
is a standard transform of Bessel Function J0(x)
Whence the result.

d2 x dx
Example 32: Solve the equation 2
−t + x = 1, x(0) = 1, x ′(0) = 2
dt dt
Solution: Taking Laplace Transform of an each individual terms of the given equation,
 d2 x 
L  2  − L  t  + L(x) = L(1)
dx
 … (1)
 dt  dt 
d 1
⇒  s2 x(s) − sx(0) − x’(0) +  s x(s) − x(0) + x(s) =
ds s
d 1
⇒  s2 x(s) − s − 2 +  s x(s) − 1 + x(s) =
ds s
dx 1
⇒ s2 x(s) − s − 2 + s +x+x =
ds s
+ ( s2 + 2 ) x =  + s + 2
dx 1
⇒ s
s 
ds
dx  s2 + 2  1 2
⇒ +  x =  2 + 1 +  … (2)
ds  s  s s
Now this equation can be treated as Leibnitz linear differential equation in x(s) and s.
 s2 + 2 
, Q =  1 + + 2 
2 1
P=
 s 
Here,
 s s 
 s2 + 2  s2 s2 s2
∴ ∫   ds + 2log s
I.F. = e∫ P ds = e s  = e2 = e2e log s2
= e 2 ⋅ s2 … (3)
∫
Now for solution, x(s).I.F. = (I.F.) Q ds + c … (4)
 
2
s
= ∫ e  2 1
  1 + s + s2  ds + c
2 s2
 
 s 
2 2 2
s s
∫
=  e 2 s2 ds + e 2
 ∫ ∫
⋅ 2s ds + e 2 ds + c

 s2   s2  s2
 ∫
=  se 2  sds + 2
 ∫ 
2

2
∫
 se  ds + e ds + c
 s   s 
2 2 2
s
∫ ∫
= d  e 2  sds + 2  de 2  ds + e 2 ds + c
    ∫
 s2 s2  s2 s2
∫
=  se 2 − e 2 1ds + 2e 2 + e 2 ds + c
 
∫
s2
s2 s2
1 2 ce − 2
= ( s + 2 ) + c or x = + 2 + 2
x(s) ⋅ e 2 s2 e2 … (5)
s s s
Taking Inverse Laplace Transforms on both sides, we get
x(t) = 1 + 2t + c.F(t), where F(t) is some function of t. … (6)
On using boundary conditions,
When x(0) = 1, x(0)= 1 + 2.0 + c.F(0) i.e. c.F(0) = 0 … (7)

when x’(0) = 2, c.F’(0) = 0 … (8)
(7) & (8) together imply c = 0
Whence x(t) = 1 + 2t.
Example 33: t = y” + 2y’ + ty = sint, y(0) = 1 y'(0) = 0

[Punjab Univ., 2003; NIT Kurukshetra, 2005]
Solution: L(t y”) + 2 L(y’) + L(t y) = L(sin t),
−
ds
(
d 2
) d
s y(s) − sy(0) − y′(0) + 2s y( s) − 2y(0) − y(s) = 2
ds
1
s +1
d d 1
−2s y(s) − s2 y(s) + y(0) + 2s y(s) − 2y(0) − y( s) = 2
ds ds s +1
− ( s2 + 1)
d 1
y(s) − y(0) = 2 ,
ds s +1
− ( s2 + 1)
d 1
y(s) = 2 + 1, y(0) = 0
ds s +1
d 1 1
−
y(s) = 2 + … (1)
ds s + 1 (s2 + 1)2
Taking Inverse Laplace on both sides of (1),
1 1 1
ty(t) = sin t +
(sin t − t cos t) , using L−1 = (sin at − at cos at) … (2)
2 (s2 + 1)2 2a3
3 sin t cos t
y(t) = −
2 t 2
∫ ∫ ∫
1 1 1
Alternately: y(s) = − 2 ds − ds = − tan −1 s − ds … (3)
s +1 (s2 + 1)2 (s2 + 1)2
∫
1
2 ds = I , take s = tan θ so that ds = sec θ
Now for 2
(s )
2
+ 1
sec2 θ 1 + cos 2θ θ sin 2θ
Implying I=
∫
(1 + tan θ)
2 2
dθ = cos2 θ dθ =
∫ ∫2
dθ = +
2 4
tan−1 s 2s 2 tan θ
I= + , as sin 2θ = … (4)
2 4 ( s2 + 1) (1 + tan2 θ)
1 s
Putting (4) into (3), y(s) = − tan−1 s − tan −1 s − 2
2 2(s + 1)
3 1 s
y(s) = − tan−1 s −
2 2 s2 + 1
y(t) = − L−1 ( tan−1 s ) − L−1 2

3 1 s
⇒
2 2 ( s + 1)
3 1
⇒ y(t) = − L−1(tan−1 s) − cos t … (5)
2 2
let g(t) = L–1(tan–1s) so that L ( t g(t)) = −

d 1
Now tan−1 s = − …(6)
ds 1 + s2
Taking Inverse Laplace on both sides,
sin t
t g(t) = – sin t or g(t) = − … (7)
t
−3  − sin t  cos t
Putting (7) in (5), f (t) = −
2  t  2
3 sin t cos t
f (t) = −
2 t 2
Example 34: A voltage Ee–at is applied at t = 0 to a circuit of inductance, L and Resistance
E  −at
−R 
t
R. Show by the transformation method that current at any time t is e − e L  .
R − aL  
[VTU, 2000]
Solution: Equation governing the current flow in L-R circuit is

di di R E
L + Ri = Ee−at or + i = e−at
dt dt L L
Taking laplace on both sides.
 si(s) − i(0) + R i(s) = E 1
  L L (s + a)
 L L
−
L L 
E 1 E
i (s) = =  R − aL (s + a) RaL  R 
or
L (s + a)  s + R  L  s + 
 L 
 L
E  −at − t
R
Taking inverse laplace transform on both sides, i(t) =  e −e L 
R − aL  
Example 35: Use laplace transform method to obtain the charge at any instant of a capacitor
which is discharged in R – C – L circuit, after the switch is closed if R = 2.25 ohms, self
inductance L = 1 Henry, capacitance, C = 2 farads, and the capacitor has initially a charge
of 100 coulombs. Initially the switch is open and therefore, no current is flowing.
Solution: The desired equation in L – C – R circuit is

d2 q dq q
L 2
+R + = E. … (1)
dt dt C
 dq 
Here given, L = 1, R = 9/4, C = 2, q(0) = 100, q′(0) =   = 0, E = 0
dt t = 0
With above values, equation (1) becomes
d2 q 9 dq q
+ + =0 … (2)
dt2 4 dt 2
Taking laplace on both sides of the above equation
9 1
 s q (s) − sq(0) − q´(0) + 4  s q(s) − q(0) + 2 q (s) = 0
2
 s2 + 9 s + 1  q(s) − 100s − 9 × 100 = 0

i.e.  4 2  4
100 ( 4s + 9) 100  32 1 
q(s) = = −
i.e.
( 4s + 9s + 2) 7  ( 4s + 1) ( s + 2) 
2  (By partial fractions)
100  1 1 
q(s) = 8 ⋅ − 
7  ( s + 4 ) s + 2 
1 … (3)
100  −4t 
Taking inverse laplace transforms on both sides, q(t) =  8e − e − 2 t 
7  
Example 36: Alternating voltage 200 sin(100t) is applied at t = 0 to a circuit with an
inductance 50 mH (millihenry), Capacitance 2000µf (micro farad) and resistance 10 Ω (ohms).
Find the current i at any time t seconds if the initial current i and charge q are zero.
Solution: Equation of current flow in R – L – C circuit is

d2q dq q dq
L 2
+R + = E, where =i … (1)
dt dt c dt
Here we are given R = 10 ohms
L = 50 × 10–3 Henrys
C = 2000 × 10–6 Farads
E = 200 sin(100t)
With above values, equation (1) becomes
d2q dq q
50 × 10−3 2 + 10 + = 200 sin (100t )
dt dt 2000 × 10−6
d2 q dq
or 2
+ 200 + 104 q = 4000 sin (100t ) …(2)
dt dt
Taking Laplace on both sides,
100
 s2 q (s) − sq(0) − q′(0) + 200  sq(s) − q(0) + 104 q(s) = 4000 2
s + 1002
1
 s2 + 200 s + 104  q( s) = 400000 2
( 1002 )
s +
1  1 1 
or q(s) = 4 × 105 = (2 × 103 )  2 − 2 … (3)
(s + 100 )(s + 100)
2 22
 (s + 100 ) (s + 100) 
2
Taking Inverse Laplace Transform,

q(t) = (2 × 103 ) 100 sin(100t) − t e−100 t  … (4)
dq
Also = i = (2 × 103 ) cos(100t) − e−100 t(1 − 100t) … (5)
dt
Example 37: Obtain the equation for the forced oscillations of mass m attached to the
lower end of an elastic spring whose upper end is fixed and whose stiffness is k, when
the driving force is F0 sinat. Solve this equation (Using the Laplace Transforms) when
a2 ≠ k/m, given that initially velocity and displacement (from equilibrium position) are zero.
Solution: The problem lies under forced oscillations without damping.
(If the point of support of the spring is also vibrating with some external periodic force,
then the resulting motion is called forced oscillatory motion, otherwise the motion is termed
as forced oscillations without damping)
Taking the external periodic force to be F0 sin at, the equation of motion is
d2 x
m = mg − k ( e + x ) + F0 sin at … (1)
dt2
where,
x is the length of the stretched portion of the spring (displacement) after time t,
e is the elongation produced in the spring by the mass m,
k is the restoring force per unit stretch of the spring due to elasticity;
a is any arbitrary constant and p is any scalar.
A
e
B
x
P
F0 sin at
k(e + x)
mg
Fig. 12.4
In this particular problem, tension mg = ke, therefore equation (1), becomes
d2 x d2 x k F
m = −kx + F0 sin at or + x = 0 sin at
dt2 dt2 m m
d2 x F
2
+ n2 x = 0 sin at … (2)
dt m
Taking laplace on both sides of (2), we get
F a
 s2 x(s) − sx(0) − x´(0) + n2 x(s) = 0 2
m s + a2
 s2 + k  x(s) = F0 a , using x(0) = 0, x’(0) = 0

 
m m s2 + a2
aF 1 aF 1 k
x(s) = 0 ⋅ = 0 2 ,µ= … (3)
m   k 
2
m (s + a )(s2 + n2 )
2
m
( s + a )  s +  m  
2 2 2
 
1 As + B Cs + D
Now by partial fractions, = +
(s2 + a2 )(s2 + n2 ) ( s2 + a2 ) (s2 + n2 )
1 = ( As + B ) ( s2 + n2 ) + (Cs + D ) ( s2 + a2 )
⇒ 1 = s3 ( A + C ) + s2 ( B + D ) + s ( n2 A + a2C ) + (n2 B + a2 D )
1 1
On solving for A, B, C, D; we get A = 0, B = , C = 0, D=−
n − a2
2
(n − a2 )
2
aF0  1 1 
x(s) = 2  2
− 2 2 
m (n − a )  ( s + a ) ( s + n ) 
Hence 2 2
Taking Laplace Inverse on both sides,
x(t) =
aF0  1 sin at − n sin nt  ,
m (n − a )  a
2 2  
F0 k k
= n sin at − a sin nt  , where n = and ≠ a2
mn (n2 − a2 )  m m
ASSIGNMENT 3
Solve the following Linear Differential equations:
dx
1. + x = sin wt, x(0) = 2 [KUK, 2001]
dt
d2 y  dy 
2. + w2 y = 0, where y(0) = A,   = B
dx2 dx x = 0
3. y” – 3y’ + 2y = 4t + e3t, when y(0) = 1 and y’(0) = – 1
[Andhra, 2000; NIT Kurukshetra, 2008]
4. (D + n )x = a sin(nt + α), x = Dx = 0 at t = 0
2 2
5. y”’ + 2y” – y’ – 2y = 0, given y(0) = y’(0) = 0 and y”(0) = 6

6. (D3 – 3D2 + 3D – 1)y = t2et, given y(0) = 1, y’(0) = 0, y”(0) = –2
d2 x dx
7. −2 + x = et with x = 2, x = − 1 at t = 0 [KUK, 2004]
dt2 dt
8. yir(t) + 2y” + y = sint, when y(0) = y’(0) = y”(0) = y”’(0) = 0
d2 x dy
9. 2
+2 + 5y = et sin t, where y(0) = 0, y′(0) = 1
dt dx
[KUK, 2002, 2004; Bombay, 2005; PTU, 2005]
10. t y” + 2y’ + t y = cost, given that y(0) = 1
11. t y” + (1 – 2t)y’ – 2y = 0, when y(0) = 1, y’(0) = 2 [Punjabi Univ. 2003]
t
−u
12. f (t) = cos t + ∫ e f (t − u) du
0
12.10: APPLICATIONS TO SIMULTANEOUS LINEAR EQUATIONS WITH CONSTANT

COEFFICIENTS
Example 38: Solve the simultaneous equations
dx dy
− y = et , + x = sin t , given x(0) = 1, y(0) = 0 [Delhi, 2002]
dt dt
Solution: Taking Laplace Transforms of the given simultaneous equations,
[ sx(s) − x(0)] − y(s) = s −1 1 or sx(s) − 1 − y(s) =

1
or s x − y =
s
… (1)
s−1 s−1
1 1
and sy(s) − y(0) + x( s) = 2 or x + sy = … (2)
s +1 s2 + 1
On solving (1) and (2) for x(s) and y(s) ,
s2 1
x(s) = +
( s − 1) ( s2 + 1) ( s2 + 1)2
1 1 1 s  1
=  + + +
2  s − 1 s2 + 1 s2 + 1  ( s2 + 1)2 … (3)
s s
y(s) = −
and
(s2
+ 1)
2
( s − 1) ( s2 + 1)
s 1 1 1 s 
= =  + 2 − 2 
(s 2
+ 1)
2
2 s − 1 s + 1 s + 1 … (4)
Taking Inverse Laplace Transforms on both sides of equations (3) and (4), we get
 
1 −1  1 1 s  1
x= L  + 2 + 2  + L−1  
2  s − 1 s + 1 s + 1  ( s2 + 1)2 
=
2
(
1 t
e + sin t + cos t ) + (sin t − t cos t )
1
2
=
2
( e + 2 sin t + cos t − t cos t)
1 t
… (5)
 
y = L−1 
s 1  1 + 1 − s 
 − L−1 
and  ( s + 1)  2
2 2  s − 1 s2 + 1 s2 + 1
t sin t − ( et + sin t − cos t )

1 1
=
2 2
= ( t sin t − sin t − et − cos t )
1
… (6)
2
x=
2
(e + 2 sin t + cos t − t cos t )
1 t
Hence, 
y = (t sin t − sin t − et − cos t ) 
1
2 
Example 39: The coordinates (x, y) of a particle moving along a plane curve at any time t
dy dx
are given by + 2 x = sin 2t , − 2y = cos t , (t > 0) . If at t = 0, x = 1 and y = 0, show by
dt dt
using transforms that the particle moves along the curve 4x2 + 4xy + 5y2 = 4.
[UPTech, 2003]
Solution: On taking Laplace on both sides of the given simultaneous equations,

2 2
sy = y(0) + 2x = 2 or 2x + sy = 2 , using (y(0) = 0) … (1)
s +4 s +4
Similarly [ sx = x(0)] + 2y = s2 2 4 or sx + 2y = s2 s+ 4 + 1 , using (x(0) = 1) … (2)
+
For solving x and y , multiply throughout (1) by s, (2) by 2 and take the difference of the
two,
2
y=− 2 or y(t) = − sin 2t … (3)
s +4
1 2 s  1
Further x=  2 + 2⋅ 2  or x(t) = (sin 2t + cos 2t ) … (4)
2 s +4 s +4 2
Equation (4) implies, 2x = sin 2t + 2 cos 2t i.e. 2x + y = 2 cos 2t, using (3)
(2x + y) = 4(1 – sin 2t) or 4x + y + 4xy = 4 – 4y2, using (3)
2 2 2 2
4x2 + 4xy + 5y2 = 4

Hence the desired particle (x, y) moves along the curve 4x2 + 4xy +5y2 = 4
d2 x dy dx d2 y
Example 40: Solve + + sin t = 0, − + cos t = 0 with the initial conditions
dt2 dt dt dt2
dx dy
x = 0, = 1; y = 1, =0 when t = 0.
dt dt
Solution: On applying Laplace transform on each individual terms of the two given equations
 d2 x   dy 
L  2  + L   + L sin(t) = 0 ⇒ ( s2 x − 1) + ( sy − 1) + 1
=0 … (1)
 dt   dt  ( s2 + 1)
 d2 y  sx − ( s2 y − s ) +
s
L   − L  2  + L (cos t ) = 0
dx =0
and ⇒ … (2)
 dt   dt  s2 + 1
On simplification of (1) and (2),
2s2 + 1
s2 x + sy = … (3)
s2 + 1
s2 + 2
and x − sy = − … (4)
s2 + 1
Adding (3) and (4), we get
s2 − 1
( s2 + 1) x = 2ss2 ++11 − ss2 ++ 21 = ss2 −+ 11
2 2 2
or x= … (5)
( s2 + 1)2
Using (5) in (4), we get
s2 − 1 s2 + 2 s2 + 1 − 2 s2 + 1 + 1
sy = + 2 = +
( s2 + 1) (s + 1) ( s2 + 1)2 (s2 + 1)
2
1 2 1
= = 2 +
( s2
+ 1) ( s + 1) ( + 1)
2 s2
2 2 2s2
= − + 1 = +1
( s2 + 1) ( s2 + 1)2
( s2 + 1)
1 2s
or y=+ 2 … (6)
s (s + 1)2
Taking inverse of (5) and (6),
x = t cos t and y = (1 + t sin t).
ASSIGNMENT 4
Solve the following simultaneous equations
dx dy
1. + y = sin t, + x = cos t given x = 2, y = 0 when t = 0.
dt dx
[UPTech, 2004; Kerala, 2005]
dy dy
2. 3 dx + + 2x = 1,
dx
+4 + 3y = 0; given x = 3, y = 0 when t = 0.
dt dt dt dt
dx dy
3. + 5x − 2y = t, + 2x + y = 0; being given that x = y = 0 at t = 0
dt dt
4. The currents i1 and i2 in a mesh are given by the differential equations:
di1 di2
− wi2 = a cos pt, + wi1 = a sin pt . Find the currents i1 and i2 by Laplace transforms
dt dt
if i1 = i2 = 0 at t = 0.
5. Small Oscillations of a certain system with two degrees of freedom are given by the
equations
D2 x + 3x − 2y = 0 

D x + D y − 3x + 5y = 0
2 2 where x = y = 0 and
Dx = 3
Dy = 2
when t = 0. }
Find x and y when t = 1 .
2
12.11 LAPLACE TRANSFORMS OF SOME SPECIAL FUNCTIONS
Sr.No. Function Name f(t) f (s)
1. Heaviside's Unit Step Function u(t − a) = {0,

1
t<a
t≥a
e−as
s
, s>0
 1 , 0 ≤ t ≤∈
2. Dirac Delta function δ∈ =  ∈ 1
s∈
(1 − e−s∈)
ο t >∈
T
− st
∫ e f (t) dt
3. Periodic Function f (t) = f (t + T ) 0
1 − e−sT
t
sin u 1 1
4. Sine Integral Si (t ) = ∫ du tan−1
0 u s s
∞
log ( s2 + 1)
cos u 1
5. Cosine Integral Ci(t) = ∫ du
t u 2s
6. Error Function erf ( t) = 2 t −u2

∫ e du
π 0
1
s s+1
1
7. Complementary Error Fun. erfc ( t) = 1− 2 t −u2
∫ e du
π 0 {( s + 1) + 1 } s+1
∞ −u
e 1
8. Exponential Fun. of Order Zero eL (t) = ∫ du log ( s + 1)
t u s
t2 t4 t6 1
9. Bessel Function of Order Zero J0 (t) = 1 − + 2 2 − 2 2 2…
22
24 246 1 + s2
A few of the above functions are taken up in the subsequent discussions.
1. Unit Step Function (Heaviside’s Unit Function): In engineering, many times we come
across such fractions of which inverse laplace is either very difficult or can not be
obtained by the formulae discussed so far. To overcome such problem, Unit step
function (Heaviside’s Unit Function) has been introduced.
The unit step function is defined as follows:
u(t – a)
0 for t < a
Ua(t) = u(t − a) = 
1 for t ≥ a,
where a is always positive.
From the graph, it is apparent that function is zero for all values
of t up to ‘a’ and after which it is unity. In physical problems, t is a
time variable. t
t=a
u(t) = 0, t < 0 Fig. 12.5
As a particular case, = 1, t ≥ 0
Observations: Generally the unit step function in mechanical engineering comes into picture as a force suddenly
applied to a machine or a machine component, where as in electrical engineering it manifests as an electromotive
force of a battery in circuit.
Transform of Unit Step Function:

∞ ∞ ∞
a
e−st e−as
L {u (t − a )} = ∫ e u ( t − a ) dt = ∫ e 0 dt + ∫ e−st 1 dt =
− st − st
=
0 0 a −s a s
L {u(t − a)} =
and in particular, when a = 0, 1
s
Second Shifting Theorem (t-shifting):
f (t) u ( t − a ) = 0, when t < a
On the basis of the u(t – a), we had the functions, = f (t), when t ≥ a,
and the function f(t – a) represents the graph of f(t) displaced through a distance ‘a’ to the
right i.e. ua(t) has a jump type of discontinuity at t = a.
If L ( f (t)) = f ( s), then L { f (t − a) u(t − a)} = e−as f (s)
∞
L { f (t − a) u(t − a)} = ∫ e−st { f (t − a)u(t − a)} dt
a)
f(t)
–
(t
0
)u
–a
∞ ∞
= ∫ e−st odt + ∫ e−st f (t − a) ⋅ 1 dt f(t)
f(t
0 0
∞
= ∫ e−s( a + x ) f (x) dx (on taking t – a = x) 0 t0 a t0 + a t
0 Fig. 12.6
∞
= e−as ∫ e−sx f (x) dx = e−as f (s)
0
The second shift theorem is also known as t–shifting i.e. t is replaced by t – a in f(t).
Example 41: Find Laplace Transform of

(i) f(t) = (t – 1)2u(t – 1) (ii) f(t) = e–t {1 – u(t – 2)}
(iii) f(t) = sint·u(t – a) (iv) f (t) = {

t−1
3−t
for
for
1<t<2
2<t<3
Solution:
(i) Given f(t) = (t – 1)2u(t – 1) … (1)
On comparing the given function with f(t – a)u(t – a),
We see that in this case a = 1, f(t) = t2 and f (s) = L ( f (t)) = L ( t2 ) = 2 … (2)

s3
Consequently by second shift theorem,
2 2 e−s
L { f (t − a) u(t − a)} = e−as f (s) = e−s = 3
s3 s
1
(ii) Here in this case taking f (t) = e−t , f (s) = and then by second shift theorem
s+1
( )
L e−t {1 − u ( t − 2)} = L ( e−t − e−t u ( t − 2 )) f(t)
 e−(t − 2) 
= L  e−t  − L  2 u ( t − 2)
 e 
o 1 2 3 t
1 1 1 Fig. 12.7
= − 2 e−2s
s+1 e s+1
Example 42: Find the inverse laplace transforms of the following:

se−as se− s / 2 + π e− s e−as
(i)* 2 , a > 0 (ii)** (iii)*** , a>0
s − w2 s2 + π 2 s ( s + b)
2
e− s e− s
(iv) (v)
(s − 1)(s − 2) ( s + 1 )3
[*KUK, 2002; **NIT Kurukshetra, ***KUK 2005; VTU 2000, 2003]
Solution:
s
(i) Let f (s) = , then f (t) = cosh wt
s − w2
2
{ }
Now by second shifting theorem, L−1 e−as f (s) = f (t − a) u(t − a)
⇒ {
L−1 e−as
s
s − w2
2 }
= cosh w(t − a) u(t − a)
s
(ii) For f (s) = , f (t) = cos π
s2 + π2
π
f (s) = , f (t) = sin πt
s + π2
2
{
Also by second shifting theorem L−1 e−as f (s) = f (t − a) u(t − a) , }
{ }
 −s 
−1  se+ πe−s 
2  − 2s s  −1 − s π
L  2  = L−1
 e +L e 2
 s + π  s + π2  s + π2
2 2

= cos π  t −  u  t −
1 1
+ sin π(t − 1) u(t − 1)
 2  2
= cos π  t −  u  t −  + sin ( −π + πt ) u(t − 1)

1 1
 2  2
= cos π  t −  u  t −  − (sin πt) u(t − 1)

1 1
 2   2
e−as
= e−as  2
1 1 1 1 1 1
(iii) Given F(s) = − + ⋅ , (By partial fractions)
s ( s + b)
2
 b s + b b2 s b s2 
=  2 e−as
1 1 1 1 1 −as 1 
− 2 e−as + e
b s+b b s b s2 
Further L−1 
1   1  1
= e−bt , L−1   = 1, L−1  2  = t,
 s + b   s s 
(
Also by second shifting theorem, L−1 e−as f (s) = f (t − a) u(t − a), )
 e−as  −1  1 −as 1  −1  1 −as 1 −1  1 − as 1 
L−1  2  = L  b2 e s + b  + L  − b2 e s  + L  b e s2 
 s ( s + b ) 
1 −b ( t − a ) 1 1
= e u(t − a) − 2 1 ⋅ u(t − a) + (t − a) u(t − a)
b2 b b
=
b2
e {
1 − b (t − a )
}
+ b(t − a) − 1 u(t − a)
e− s
= e−s 
1 1  1 1
(iv) F(s) = − = e− s − e−s
( s − 1)(s − 2 )  s − 2 s − 1  s−2 s−1
 e−s  −1  −s 1  −1  −s 1 
⇒ L−1   = L  e s − 2  − L  e s − 1
 ( s + 1)( s − 2 ) 
[By second shifting theorem]
=e 2(t −1)
u(t − 1) − e
(t −1)
u(t − 1)
= e2(t −1) − e(t −1)  ⋅ u(t − 1)
e−s − as ( ) 1
(v) Here F(s) = 3 = e f s , where f (s) =
( s + 1) ( s + 1)3
 e −s  1 − t −1
L−1  3 = e ( )(t − 1)2 u(t − 1)
 ( s + 1)  2
1
Example 43: Find the inverse laplace transform of .
s ( 1 − e− s )
  1 −1  1 
= L−1  (1 − e−s )  = L−1  (1 + e−s + e−2s + e−3s + …)
Solution: L−1  1
−s 
 s ( 1 − e )   s   s 
= L−1   + L−1  e−s  + L−1  e−2 s  + L−1  e−3 s  + …

1 1 1 1
 s  s  s  s
Comparable to L−1 e−as f (s) ,( ) where f (s) =

1
s
in each case .
f (t )
1 −1 
⇒ L−1  (1 − e− s )  = 1 + u(t − 1) + u(t − 2) + u(t − 3) + …
s 
Thus, the function can be written as 3
1, 0<t≤1 2
2, 1< t≤ 2
 1
f (t) = 3, 2<t≤3
… ………… t
… …………
 O 1 2 3
Fig. 12.8
Sometimes it is called as staircase function.
Example 44: In an electrical circuit with e.m.f. E(t), resistance R and inductance L, the
current i builds up at the rate given by L di + Ri = E(t) . If the switch is connected at t = 0

dt
and disconnected at t = a, find the current i at any instant.
E, for 0 < t < a

Solution: Given i = 0 at t = 0 E(t) = 
and
 o, for t > a
On taking laplace of each term of the given equation, we get
∞
( )
L s i (s) − i(0) + R i (s) = ∫ e−st E(t) dt (L and R are constants)
0
… (1)
a ∞
( Ls + R ) i (s) = ∫ e−stE dt + ∫ e−st 0 dt
0 a
a
e−s t
= (1 − e−as )
E
(Ls + R) i (s) = E
−s o s
 1 e−as 
⇒ i (s) = E  − 
 s ( Ls + R ) s ( Ls + R ) 
… (2)
Now taking inverse laplace on both sides, we get
 1   1 
i(t) = EL−1   − EL−1  e−as
 s ( Ls + R )   s ( Ls + R )  … (3)
1 11 L 1 
By partial Fractions, s ( Ls + R ) =  R s − R Ls + R  … (4)
∴ Equation (3) becomes,
i(t) = E L−1 { 11 L
−
1
R s R Ls + R


}
− E L−1 e −as 
11 L
−
1 
 R s R Ls + R  
E  −1  1  1   E  −1  e− as  1 
i(t) = L   − L−1  − − L−1  e−as
 s + R   R   s 
L
  s  R 
R
      s +  
 L   L 
i(t) =
E ( − t E
1 − e L − L−1 F(s)
R
) ( ) … (5)
R R
( )
Now L−1 F(s) = L−1 e−as f (s) = f (t − a) u(t − a) , (By second shifting property)
 1
L−1  e−as  − L−1  e−as

1 

− ( t − a)
= 1.u(t − a) − e L
R
(
− (t − a)
u(t − a) = 1 − e L
R
u(t − a))
∴  s   s +
R
 … (6)
L
On putting (6) into (5), we get the value of the current
i(t) =
E ( − t
1− e L −
E
1− e L
R
) (
− (t − a )
u(t − a)
R
) … (7)
R R
Now, for 0 < t < a, i(t) =

E( − t
1−e L
R
) Q u(t − a) = 0,
 = 1,
0 < t < a
t ≥ a 
R
E − t E − (t − a )  E −Lt  La 
R R R R
for t > a, i(t) =  1 − e L
 −  1 − e L
 = e  e − 1
R  R  R  
II Unit Impulse Function (Dirac Delta Funtion)

Impulse is considered as a force of very high magnitude applied
1
for just an instant and the function representing the impulse is called
∈
Dirac–Delta Function. δ∈(t – a)
Mathematically, it is the limiting form of the function
 1 , a ≤ t ≤ a+ ∈
δ∈(t − a) =  ∈
0, otherwise, as ∈→ 0 t
O aa+∈
It is apparent from the figure 12.9, that as ∈ → 0, the height of Fig. 12.9
the strip increases indefinitely and the width decreases in such a
way that the area of the strip is always unity.
The above fact can notionally be stated as δ(t − a) = ∞, for t = a such that
o, for t ≠ a
∞
∫ δ(t − a) dt = 1 for a ≥ 0
0
Observation: In mechanics, the impulse of a force f(t) over a time interval, say a ≤ t ≤ (a + ∈) is defined to be the
integral of f(t) from a to (a + ∈). The analog of an electric circuit is the integral of e.m.f. applied to the circuit,
integrated from a to (a + ∈). Of particular practical interest is the case of very short ε with limit ε → 0 i.e. impulse
of a force acting for an instant.
Transform Of Unit Impulse Function:

∞ a a+∈ ∞
L [ δ(t − a)] = ∫ e−st δ(t − a) dt = ∫ e−st odt + ∫ e−st
1
dt + ∫ e−st o dt
0 0 a ∈ a+∈
a +∈ a+∈
1 1 e  − st
1  −s( a+∈) −as  e−as  1 − e−s∈ 
=
∫ e−stdt =   = −e +e  =
∈ a ∈  −s a s∈  s  ∈ 
Taking Limit as ∈ → o,
e−as  1 − e−s∈  1 − e−s∈
L [ δ(t − a)] = Lt   = e−as Lt ⋅ = e−as
∈→ o s  ∈  ∈→ 0 s∈
In particular, if a = o, L[δ(t)] = 1.
Filtering of unit Impulse Function:
∞
If f(t) is continuous and integrable for all t ≥ 0, then ∫ f (t) δ(t − a) dt = f (a)
0
Relation Bewteen u’(t – a) and δ(t – a)

e−as
L [ u′(t − a)] = sL [ u(t − a)] − u(− a) = s − o = e− as = L [ δ(t − a)]
s
Example 45: The differential equation governing the flow of a current i(t) in an LR circuit
is given by
di
Ri + L = E δ(t)
dt
where E δ(t) is the impulse voltage and i(0) = 0. Find the current i(t), t > 0.
Solution: On taking laplace of each individual term in the given equation,

R i (s) + L  s i ( s) − i(o) = E.1
E
implying (R + Ls) i (s) = E or i (s) = L
R
s+
L
R
E − t
On taking inverse, laplace, i(t) = e L is the required expression for the current i.
L
Example 46: An impulsion voltage E δ(t) is applied to an L C R circuit with zero initial
condition. If i(t) be the current at any subsequent time t, find the limit of i(t) as t → 0.
Solution: The governing equation for an L C R circuit is

di 1t
+ Ri + ∫ i dt = E δ(t) , where i(t) = 0 when t = 0
L
dt Co
Taking Laplace of each individual term on both sides of the equation,
 s i (s) − i(0) + R i (s) + 1 i (s) = E , where i(0) = 0

  L CL s L
 + R + 1 1 E  2 + R + 1  ( )= E
Implying,  s L CL s  i (s) = L or  s
L
s i s
CL  L
s
E s E s
i (s) = =
L s2 + 2as + ( a2 + b2 )
R 1
L s2 + R s + 1 where = 2a and = a2 + b2
L CL L CL
E  (s + a) − a  E  ( s + a ) 1 
i (s) =  2  −a ,
L  (s + a) + b  L  ( s + a ) + b
2 2 2
( s + a) + b 
2 2
On inversion, i(t) = E  e−at cos bt − a e−at sin bt 

L  b 
Taking limits as t → 0, i → E . Though, initially current is applied, yet a large current will
L
E
develop instantaneously due to impulse applied at t = 0 and it is .
L
ASSIGNMENT 5
1. Sketch the graph of the following function and express them is terms of unit step
function. Hence find their laplace transforms:
2t, for 0<t<π
(i) f (t) 1, for t>π

0 for 0 < t < a

(ii) f (t) = t2 , for t > a [Hint: f(t) = t2ua(t)]

2. Find the laplace transform of the followings:
0, t>T
(i) f (t) = cos ( ω t + φ) , 0<t<T

k, for a < t < b, where a > 0
(ii) f (t) = 0 for o < t < a and t > b

 1 , 0<t<h
= 1
(iii) f (t) h [Hint: Expessible as f(t) = [1 – uh(t)] ]
0, t>h h
 e3t , 0<t<1
(iv) f (t) = 0 t>1 [Hint: f(t) = [1 – u1(t)]e3t]

{
0, 0<t<π
(v) f (t) = sin t, t > π [Hint: f(t) = sin t uπ(t)]
(vi) f (t) = 0,{

sin t , 0 < t ≤ π
t>π
Hint : f (t) = sin t [1 − uπ (t)] + t uπ(t) = sin t + [ t − sin t ] uπ (t)
3. Find the inverse transform of

e−πs e−as se−as 1 − e−πs
(i)* 2 (ii) 2 (iii) 2 (iv) *[KUK, 2004]
s +1 s s + b2 s2 + 4
f (t) = 1, 0 < t < a
4. Solve y” + 4y’ + 3y = f(t), where = 0, t > a  ; given that y(0) = y’(0) = 0
Hint: f (t) = 1 − u (t), y(s) =
 a
1
s( s + 1)(s + 3)
(1 − e−as ) , y(t) = f (t − a) ua(t)

5. A beam has its ends clamped at x = 0 and x = l. A concentrated load W acts vertically
downwards at the point x = l . Find the resulting defection.

3
 d4 y ω  l
Hint: The differential equation and boundary conditions are dx4 = EI δ  x − 3  
 
 and y(0) = y′(0) = y(l) = y′(l) = 0 
6. A cantilever beam is clamped at the end x = 0 and is free at the end x = l. It carries a
uniform load per unit length from x = 0 to x = l/2. Calculate the deflection y at any
point.
 d4 y ω(x) 
Hint: The differential equation and boundary condition dx4 = EI , 
 
  l 
 w0, 0 < x <
(0 < x < l) ; ω(x) =  2 and y(0) = y′(0) = 0 = y′′(0) = y′′′(0) 
 0, x>l 2 
 
III Laplace Transform of Periodic Functions

If f(t) be a periodic function of period, T(> 0) i.e. f(t + T) = f(t), then
T
∫ e−st f (t) dt
Lf (t) = 0
(1 − e−sT )
Proof: By definition,
∞ t 2t 3t
Lf (t) = ∫ e−st f (t) dt = ∫ e−st f (t) dt + ∫ e−st f (t) dt + ∫ e−st f (t) dt + ……
0 0 t 2t
Substituting t = u, t = u + T, t = u + 2T, in the successive integrals, then
T T T
Lf (t) = ∫ e− su f (u) du + ∫ e− s(u+ T) f (u + T ) du + ∫ e− s(u+ 2T) f (u + 2T) du + ……
0 0 0
Since f(u) = f(u + T) = f(u + 2T) = …,

T T T
∴ L[ f (t) = ∫ e− su f (u) du + e− sT ∫ e− su f (u) du + e−2 sT ∫ e− su f (u) du + ……
0 0 0
T
= (1 + e−sT + e−2sT + …) ∫ e−su f (u) du
0
T
1
= e−su f (u) du (On changing the dummy variable u to t)
(1 − e−sT ) ∫0
T
∫ e−st f (t) dt
L ( f (t)) = 0
(1 − e−sT )
Observation: Hence the laplace transform of a periodic function f(t) with period T is expressible in the form of
T
integral ∫ e−st f (t) dt over one period of f(t) instead of over the semi-infinite interval 0 ≤ t ≤ ∞.
0
Example 47: Find the laplace transform of the triangular wave of period 2a given by
f (t) = t ,
= 2a − t ,
0<t<a
a < t < 2a }
[Madras, 2000; UP Tech, 2002; VTU, 2003; KUK, 2004; NIT Jalandhar, 2006]
Or
Find the laplace transform of the triangular wave function f(t) whose graph is shown
below.
Solution: Clearly the graph of f(t) is a straight line from the origin with slop 1 in the interval
o≤t≤a
∴ f(t) = t, o ≤ t ≤ a
However in the interval a ≤ t ≤ 2a, the graph is a straight line whose slope is -1.
∴ f(t) = (– 1)(t – 2a) = 2a – t, a ≤ t ≤ 2a

Thus, the function is a periodic one, with period 2a
f(t)
t
0 a 2a 3a 4a
Fig. 12.10
2a
1
∴ L[ f (t)] = ∫ e−st f (t) dt
1 − e−2as 0
1  a −st 2a 
= −2 as  ∫
e f (t) dt + ∫ e−st (2a − t) dt 
1−e o a 
1  e−st e−st a  e−st e−st  
2a
=  t − +
  (2 a − t ) +  
1 − e−2 as  −s s2 o  −s s2 a 
1   e−as e−as 1   e−2as e−as e−as 

= − 2 as  
a − 2 + 2+ 2 +a − 2 
1− e   −s s s   s s s 
 1 −2as − 2 −as + 1 
=
1
1 − e−2as  s2
(e e )
− 2 as (
1 − e−as )
1 2
=
s (1 − e )
2
1  e−as/2 ( e−as/2 − e−as/2 )

2
= 2 − as
se (e − e )
as − as  
=
s (e
1
(eas/2 − e−as/2 )2
2 as/2
−e − as/2
)(e as/2
+e − as/2
)
tanh  
1 as
=
s2  2
ωt, o < t < π/ω

Example 48: Find the laplace transform of the full wave rectifier f(t) = E sinω ω;
having period π/ω.
ω.
π/ w
1
Solution: Lf (t) = − sπ ∫ e−st E sin ωt dt
1− eω 0
π/ w
 e−st 
 s2 + ω2 ( −s sin ω t − ω cos ωt )
E
= − sπ
1− e ω 0
E  e−sπ/ ω  π 1 
= −s π  s2 + ω2  − sin π − ω cos ω ω  − s2 + ω2 (0 − ω )
1− eω  
E  e−sπ/ ω ω 
= − sπ  s2 + ω2 ω + s2 + ω2 
1− eω
E ⋅ω
− sπ/ ω (
= e−sπ/ ω + 1)
( s2
+ ω2
)(1 − e )
Eω 1 + e−sπ/ ω
=
(s2 + ω2 ) 1 − e−sπ/ ω
Eω sπ 
ex + e− x 1 + e−2 x 
= coth Q= coth x =

(s + ω2 )
2
2ω ex − e− x 1 − e−2 x 
The graph for the function f(t) which is a full-sine wave rectifier can be drawn as below:
Y
y=a
X
– π/ω 0 π/ω 2π/ω 3π/ω
Fig. 12.11
 π
Since sin w  t +  = sin ω t for all t.
 ω
Example 49: Find the laplace transform of the square wave (or meander) function of a
period a defined as: f(t) = 1 when 0 < t < a/2 

= –1, when a/2 < t < a  [UP Tech, 2004]
1 a −st
Solution: L [ f (t)] = ∫ e f (t) dt
1 − e−as o
1  a/2 a 
=  ∫ 1 ⋅ e− st
dt + ∫ (− 1) e− st
dt 
1 − e−as  o a/2 
1  e−st 
a /2 a
e−st
=  − 
1 − e− as  − s o −s 
a /2 
1  1 −sa/2
− 1 +  e−as − e−as/2  
1
= e
1 − e−as  −s  s 
1
= 1 − 2 e−sa/2 + e−as 
s (1 − e−as ) 
− as (
1 − e−sa/2 )
1 2
=
s (1 − e )
1 (1 − e ) = 1 (esa/4 − e−sa/4 ) = 1 tanh as

− sa /2
=
s (1 + e−sa/2 ) s (esa/4 + e−sa/4 ) s 4
Note: It may be defined like, f(t) = k, when 0 < t < a/2

= – k, when a/2 < t < a
Y
k 



X 
–a o a/2 a 3a 2a 5 X
— — —a
2 2 2
–k
Fig. 12.12
Example 50: Find the Laplace Transform of the periodic function (saw tooth wave),
t
f(t + T) = f(t); f (t) = k for 0 < t < T
T
1 T −st 1 T −st t
Solution: L[ f (t)] = ∫ e f (t) dt = ∫ e k dt
1 − e−sT 0 1 − e−sT 0 T
1 k T −st
= ∫ e t dt
1 − e−sT T 0
T
 e−st e−st 
∫
k
= t − 1 ⋅ dt
T (1 − e−sT )  −s −s  o (By parts)
T
k  t e−st e−st 
= − 2 
T (1 − e−sT )  −s s o
k  Te−sT e −sT 1
= − 2 + 2
T (1 − e )  −s
− sT 
s s 
 Te−sT 
+ 2 (1 − e−sT )
k 1
=
T (1 − e )  −s
− sT 
s 
k ke− sT
= −
s T s (1 − e−sT )
2
Note: It may result to a particular case: when k = 1, f (T ) = t

T
f( t)
t
– 2T –T 0 T 2T
Fig. 12.13
Example 51: Find the Laplace transform of the function (Half wave rectifier)
 sin ω t , for π
0<t<
 ω
f (t) = 
π 2π [Madras, 2003; KUK, 2005]
 0, for <t<
 ω ω
Solution: L[ f (t)] = 1 T −st

∫ e f (t) dt
1 − e−sT 0
π/ ω
1 Q 2π 
= − 2π s ∫ e−st f (t) dt f (t) has a period, T =
1− e ω 0  ω 
1  π/ ω − st 2π / ω
− st

= − 2π s  ∫ e sin ω t dt + ∫ e × o × dt 
1− e ω  0 π/ ω 
π/ ω
1
= − 2π s ∫ e− st sin ωt dt,
1− e ω o
Using ∫ eax sin bx dx =

eax
a + b2
2 ( a sin bx − b cos bx ) ,
 e−st ( −s sin ω t − ω cos ω t ) 

π /ω
1
= − 2π s  
   s2 + ω2  0
 1 − e
ω

 −π s 
=
1  ωe ω + ω 
(1 − e )
− 2π s
ω
 s2 + ω2 
πs
 +ω 
ω 1 + e w 
= −2 πs
 
( s2 + ω2 ) 1 − e ω 
ω
= −2 πs
 
( s2 + ω2 ) 1 − e ω 
0 π/ω 2π/ω 3π 4π 5π
ω ω ω
Fig. 12.14
Note: Clearly the function f(t) = sinωt represents the half sine wave rectifier with period 2π . The graph of f(t) in the
ω
π π π 2π
internal 0 < t < is sine function which varnishes at t = 0 and t = and f(t) = 0 for the rest half from to .
ω ω ω ω
Example 52: A periodic square wave function f(t), in terms of unit step functions, is written
as f(t) = k [u0(t) – 2ua(t) + 2u2a(t) – 2u3a(t) + …]. Show that the Laplace transform of f(t) is
tanh   .
given by L [ f (t)] =
k as
s  2
Solution: f(t) = k[u0(t) – 2ua(t) + 2u2a(t) – 2u3a(t) + ……]
∴ L f(t) = k0[L u0(t) – 2 L ua(t) + 2L u2a(t) – 2 L u3a(t) + ……]
1 e−as e−2as e−3as 
= k −2 +2 −2 + ……
 s s s s 
k
= 1 − 2e−as + 2e−2as − 2e−3 as + ……
s
= 1 − 2 ( e−as − e−2as + e−3as − ……)
k
s
k e−as 
= 1 − 2
s  1 + e−as 
k  1 + e−as − 2e−as 
=
s  1 + e−as 
 as −as

k  1 − e−as  k  e 2 − e 2 
=  =
s  1 + e−as  s  as2 −as

 e + e 2

tanh  
k as
=
s  2
f(t)
–T T 2T
t
0
T
Fig. 12.15
ASSIGNMENT 6
Find the Laplace transform of the following periodic functions:
π
1. f (t) = sin ωt , t ≥ 0 when f (t) = f (t + T ); T =
ω
[Hint: f(t) represents full sine wave rectifier. Ex. 48]
2at
2. Saw tooth wave function of period T given by f (t) = − a + , 0≤t<T
T
 a, 0 ≤ t < b
3. Non–Symmetric square wave of period T, given by f (t) = 
−a, b ≤ t < T
4. Staircase function defined by f(t) = kn, nT < t <(n + 1)T; where n = 0, 1, 2,…
ANSWERS
Assignment 1
s3 a ( s2 + 2a2 ) a ( s2 − 2a2 )
1. (i) , (ii) , (iii) ,
s4 + 4a4 s4 + 4a4 s4 + 4a4
2a2 s 2as2 4a3
(iv) , (v) , (vi) ,
s4 + 4a4 s4 + 4a4 s4 + 4a4
a 1
2. (i) , (ii) s2 ( s2 + a2 )
s + 4a2
2
2π s
−
2 e− 2 s e− 3 s
3. (i) 3 − 3 ( 2 + 3s + 3s ) + 2 ( 5s − 1) (ii) e
2 3
, s>0
s s s s +1
2
2as s2 − a2 2as
,s> a s2 + a2
4. (i) , (ii) , (iii) (iv) ,
( s2 + a ) 2 2
( s2 + a )
2 2 (s2
−a )
2 2
( s2 − a2 )2
2as2 2a3
5. (i) , (ii) ,
( s2 + a2 )2 ( s2 + a2 )2
2as2 − 2a3
(iii) , s> a (iv) , s> a
( s2 − a ) 2 2
( s2 − a2 )2
s3 s3
6. (i) , (ii) ,
( s2 + a2 )2 ( s2 − a2 )2
 s2 + 4 
1/2
7. (i) log  s + b 
2 2 1
(ii) log  2 
 s + a2 
2 2  s 
(iii) cot−1  s  (iv) cot–1(s + 1)

 a
a s a s
8. , 2 9. , 2
s +a
2 2
s + a2 s −a
2 2
s − a2
1 s + 1  1  s2 + b2 
  log  2
10. (i)
s  s2 + 2s + 2 
(ii)
2s  s + a2 
Assignment 2
1. (i)
t sinh t
, (ii)
1
13
(3 e3t − 3 cos 2t + 2 sin 2t )
2
2  
(iii) a  eat − eat /2 cos 3 at + 3 sin 3 at  ,
3   2 2 
(iv) 1 + 4 e−3t + 2 e2t (v) 1 sin t − 1 te−t

3 15 5 2 2
e−at 
2. ( aA − B) bt ⋅ cos bt + (B − aA + Ab2t ) sin bt 
2b3C 
3. (i) 1 e−2t ( 6 cos 3t − 7 sin 3t ) , (ii) 4 e3t − e−t , (iii) e2t + 2 e−4t
3
4. (i) 1 te−2t sin t, (ii) 3t e– 3tsin t, (iii) 1 t sinh t

2 2
2 sinh t sin t
5. (i) 1 − 2 cos t + e
−t
(ii) sin 2t , (iii) ,
t t t
(iv) e
− bt
− e− at
,
− 3t − 2t
(v) e + e − e ,
−t
(vi)
2 (1 − cosh at ) ,
t t t
2 ( et − cos t )
(ix) cos at − cos bt
−t
(vii) , (viii) e sin t ,
t t t
6. (i)
1
a2
cos  t
b
a 
(ii) (1 – at)e–at, (iii)
2
( a t − 4at + 2) e− at
1 22
1 1 2 1  −2t
7. (i) − t +t+ e (ii) 1 ( e−2t + 2t − 1) (iii) 1 ( at − sin at )
8 4 2 4 a3
e−bt − e−at
9. (i) , (ii)
1
(sin at − at cos at ) , (iii) t ( e−t + 1) + 2 ( e−t − 1)
a−b 2a3
1 e−t 1
(iv) (at − sin at ) , (v) 1 − e−8t (1 + 8t ) , (vi) (cot t − cos 2t )
a3 64  3
(a sin at − b sin bt )
(vii)
( a2 − b2 )
Assignment 3
1. x =  2 + 2
ω  −t sin ωt − ωt cos ωt 2. y = A cos ωx +
B
sin ω x
 e +
ω + 1 ω2 + 1 ω
1 3t t a
3. y = (2t + 3) + (e − e ) − 2 e2t 4. x = [sin nt cos α − nt cos(nt + α)]
2 2n2
 1 1 5 t
5. y = et − 3 e−t + 2 e−2t 6. y =  1 − t − t2 + t e
2 60 
 1  1
7. x =  2 − 3t + t2  et 8. y = [(3 − t2 )sin t − 3t cos t]
2 8
11 1 2
9. y = (sin t + sin 2t) e
−t
10. y =  1 +  sin t
3 2 t
11. y = e2t 12. f(t) = cos t + sin t
Assignment 4
1  −6
t
x=  5 − 2e −t
− 3 e 11

x = et + e−t 10 
1. 2.
y = et − e−t + sin t 1 −6
t
y =  e−t − e 11 
5 
1
x=− (1 + 6t ) e−3t + 1 (1 + 3t ) , i1 =
a
(sin ωt + sin pt)
27 27 p+ω
3.
y=−
2
( 2 + 3t ) e−3t + 2 ( 2 − 3t )
4.
i2 =
a
p+ω
(cos ωt − cos pt )
27 27
1
sin 3t , x   = 1.4015
1 1
x= 11sin t +
4 3   2
1 
y   = 1.069
5. 1 1
y =  11sin t − sin 3t ,
4 3   2
Assignment 5
2  1 − 2π 2  −as − as  a 2a 2 
1 (i) (1 − 2t) u(t − π) + 2t u(t), + − 2e (ii) t ua(t), e  + 2 + 3 
2
s2  s s  s s s
2. (i) [u(t) − u(t − T )cos(ω t + φ) (ii)

s
(e − e )
k − as −bs
(iii)
1
sh
[1 − e−sh ]
1  (3 − s) −e−πs π 1
+  + +
e − 1 1 1  −πs
( 3 − s) 
(iv) (v) , (vi) e
s2 + 1 s2 + 1  s s2 s2 + 1 
sin 2t (1 − uπ (t))
1
3. (i) – sint uπ(t) (ii) (t – a) ua(t) (iii) cosb(t – a) ua(t) (iv)
2
1 1 −t 1 −3t
4. y(t) = f (t) − f (t − a) ua(t), where f (t) = − e + e
3 2 6
 2ω x2 ( 3l − 5x ) l
 81 EI , 0<x<
 3
5. y(x) = 
 2ω x ( 3l − 5x ) + ω  x − l  ,
2 3
l
<x<l
 81 EI 6EI  3 3
ω l2 2 ωl 3 ω 4 ω
( ) ( )
4
6. y(x) = x − x + x − x−l 2 u x−l 2
16EI 12EI 24EI 24EI
Assignmen 6
 
 sT 
− cot h   
Eω sπ a 1
1. cot h 2.   2 
s2 + w2 2ω s   sT  
  2  
a ( e−sT − 2e−sb + 1) k e− sT
s (1 − e− sT ) s (1 − e−sT )
3. 4.

12 - Laplace Transforms and Their Applications PDF

Uploaded by

Copyright:

Available Formats

12 - Laplace Transforms and Their Applications PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

12 - Laplace Transforms and Their Applications PDF

Uploaded by

Copyright:

Available Formats

12 LAPLACE TRANSFORMS AND

(i) Lt f (s) = 0 viz. the necessary condition for f (s) to be a transform.

Fig. 12.1: Real axis

12.2 EXISTENCE CONDITIONS

12.3 EXISTENCE THEOREM ON LAPLACE TRANSFORM

12.4 TRANSFORMS OF ELEMENTARY FUNCTIONS

12.5 PROPERTIES OF LAPLACE TRANSFORMS

Sr. No. Property f(t), function f (s) , laplace transform

1. Linearity a f(t) + bg (t) – c h(t) a f (s) + b g(s) − c h(s)

t2 ∫ ∫ f (s)(ds)2 , provided the entegral exists.

11. Convolution f(t)*g(t), f (s)g(s)

Proofs of Some of the properties are taken in subsequent discussions.

This result can easily be generalized.

In view of above discussion, Laplace operator is a linear in nature.

P 2 First Shift Property: [KUK, 2010]

f (s) Translation along s axis

Application of this property give arise to following simple results:

where in each case s > a.

Note: Changing a to   , the result changes to L  f  t   = a f (as)

P 4 Change of Scale with Shifting:

Example 1: Find the laplace transform of the followings:

(iv) L ( e−t sin2 t ) = L  e−t 1 − cos 2t  which comparable to L ( eat f (t)) = f ( s − a)

Example 2: Find the Laplace transform of

 t , when 0 < t < T  cos ( t − 2π 3 ) , t > 2π 3

(ii) f (t) = { sin t,

 t , when 0 < t < T

(v) Here f (t) = t − 1 + t + 1 , t ≥ 0

i.e. for all value of t varying from 0 to 1, value of t − 1 + t + 1 is always 2. Likewise

L (t f (t)) = − f (s) and L (tn f (t)) = (−1)n n f (s)

Example 3: Find Laplace of

(ii) L ( t2 cos at) = ( − 1)2 d f (s), where

L ( e − e ) + L ( e3t cos 2t − e−3t cos 2t )

(using first shift property, replace s by s – a)

P 6 Transforms of Functions Divided By t

Example 5: Prove that L 

transform of cosat exist?

(Using change of scale property, L ( f (at)) =

We know that, L [ φ '(t)] = sL ( φ(t)) − φ(0)

i.e. L [ φ′(t)] = sL (φ(t)) , sin ce φ(0) = 0

P 8 Laplace Transforms of Derivatives

Example 7: Find L(cosat) and deduce from it L(sinat) [KUK, 2002]

Using above, (5) becomes

Hence the result.

Example 8: Find sin t and hence deduce cos t . [KUK, 2010]

∴  sin t  = cot−1( − 1), by first shift property

Hence, f(t) = t3sin t

= ( − 0 + log 1) +  0 − log  = log 3  as s → ∞, → 0

(iii) sin h at cos at (iv) sinh at sin at

12.6 INVERSE LAPLACE TRANSFORMS

Corresponding to repeated quadratic factors (s2 + as + b), write

Example 12: Find the inverse Laplace transform of

From (1), if s = 1; then (2 + 1) = D(1 + 2)2 1

if s = 2; then (–4 + 1) = B(–2 – 1)2 implying B = − 1 … (4)

Thus on putting D = 1 and B=−

Solution: Write s4 + s2 + 1 = (s4 + 2s2 + 1) –s2 = (s2 + 1)2 – s2 = (s2 + 1 + s)(s2 + 1 – s)

or s = (As + B)(s2 + 1 + s) + (Cs + D)(s2 + 1 – s)

For s3 , A+C = 0 …(1) 

All these equation together give B = 1 , D = − 1

Implying s3 = (As + B)(s2 + a2) + (Cs + D)(s2 – a2)