The Laplace Transform


Definition

F ( s )  L[ f (t )]   f (t )e  st dt
0

» Time (t) is replaced by a new independent variable (s)

» We call s the Laplace transform variable

The Laplace domain
» Often more convenient to work in Laplace domain than time domain
» Time domain  ordinary differential equations in t
» Laplace domain  algebraic equations in s

General solution approach
» Formulate model in time domain
» Convert model to Laplace domain
» Solve problem in Laplace domain
» Invert solution back to time domain
 Laplace Transform of Selected Functions

Constant function: f(t) = a

 a  st  a a
F ( s )  L(a )    st
ae dt   e  0   
0 s 0  s s

Exponential function: f(t) = e-bt

0

F ( s )  L(e )   e e dt   e
bt bt  st


0
(b  s )t
dt  
1
bs

e (b  s )t  

0

1
sb

Derivatives and integrals
 df   df  st 
 st 
L    e dt   f (t )e sdt  f (t )e
 st
 sF ( s)  f (0)
 dt  0 dt 0 0

 dn f  n
L n   s F ( s )  s n 1 f (0)  s n  2 f (1) (0)    sf ( n  2) (0)  f ( n 1) (0)
 dt 
 t
 
 t
 1
L  f (t*)dt *    f (t*)dt * e dt  F ( s)  st
 0  0  0  s
 Properties of Laplace Transforms

Superposition
L[af (t )  bg (t )]  aF ( s )  bG( s )


Final value theorem
If lim y (t ) exists  lim y (t )  lim[ sY ( s )]
t  t  s 0


Initial value theorem
lim y (t )  lim[ sY ( s )]
t 0 s 


Time delay

f d (t )  f (t  t0 ) S (t  t0 ) 
L
Fd ( s)  e  st0 F ( s)
L1
Fd ( s)  e  st0 F ( s)  f d (t )  f (t  t0 ) S (t  t0 )
 Linear ODE Example

ODE
dy
5  4y  2 y (0)  1
dt

Laplace transform
2
5[ sY ( s )  y (0)]  4Y ( s ) 
s

Substitute y(0) & rearrange
5s  2 s  0.4
Y ( s)  
s (5s  4) s ( s  0.8)


Inverse Laplace transform
1  s  0.4 
1
y (t )  L [Y ( s )]  L  
 s ( s  0.8) 
 Linear ODE Example cont.

Table 3.1

1 s  b3  b3  b1 b1t b3  b2 b2t
L   e  e
 ( s  b1 )( s  b2 )  b2  b1 b1  b2


Our problem
 s  0.4 
1
y (t )  L    b1  0 b2  0.8 b3  0.4
 s ( s  0.8) 


Substitute and simplify

1 s  0.4  0.4  0 0t 0.4  0.8 0.8t  0.8t
L  
 0.8  0 e  e  0. 5  0. 5e
 ( s  0 )( s  0. 8)  0  0.8
General ODE Solution Procedure

Procedure
» Transform to Laplace
domain
» Solve resulting algebraic
equations
» Transform solution back
to time domain

Partial fraction
expansion
» Necessary when inverse
Laplace transform not
tabularized
» Break complex functions
into simpler tabularized
functions
 Partial Fraction Example

Partial fraction expansion

s5 s5 1 2
Y (s)  2   
s  5s  4 ( s  1)( s  4) s  1 s  4


Determination of coefficients

s5 4 s5 1
1   2  
s  4 s  1 3 s  1 s  4 3


Inverse Laplace transform
1 1  4 / 3 1 / 3  4 t 1  4t
y (t )  L [Y ( s )]  L     e  e
 s 1 s  4 3 3
 Repeated Factor Example

s 1 1 2 3
Y ( s)    
s ( s  2) 2
s  2 ( s  2) 2
s

s 1 1 s 1 1 1
2   3   1  
s s  2 2 ( s  2) 2 s 0
4 4

s 1 1 2 3 1/ 4 1/ 2 1/ 4
Y (s)       
s ( s  2) 2
s  2 ( s  2) 2
s s  2 ( s  2) 2
s

1 1  2t 1  2t 1
y (t )  L [Y ( s )]   e  te  S (t )
4 2 4
 Quadratic Factor Example
s 1 1  2  3 s   4
Y (s)  2 2   2  2
s ( s  4s  5) s s s  4s  5

s  1  1s ( s 2  4s  5)   2 ( s 2  4 s  5)  ( 3 s   4 ) s 2

(1   3 ) s 3  (41   2   4 ) s 2  (51  4 2  1) s  (5 2  1)  0

s 1 1  2  3 s   4 0.04 0.2  0.04 s  0.36

Y (s)  2 2   2  2   2  2
s ( s  4s  5) s s s  4s  5 s s s  4s  5

 0.04 s  0.36  0.04( s  2)  0.28

 
s  4s  5
2
( s  2)  1 ( s  2) 2  1
2

y (t )  L1[Y ( s )]  0.04 S (t )  0.2t  0.04e  2t cos t  0.28e  2t sin t

 Transient Response Example


Component balance
dc1 dc1
V  q (ci  c1 )  4  2c1  2ci c1 (0)  1
dt dt

Pulse input
1 5

ci (t )  S (t )  5[1  S (t  0.25)]  Ci ( s )   1  e 0.25s
s s

 Transient Response Example cont.

Laplace transform
4 2
4[ sC1 ( s )  1]  2C1 ( s)  2Ci ( s )  C1 ( s )   Ci ( s )
4s  2 4s  2


Substitute input
4 2  1 5(1  e 0.25s )  2 6 5e 0.25s
C1 ( s)       
4s  2 4s  2  s s  2 s  1 s(2s  1) s (2 s  1)


Inverse Laplace transform

 
c1 (t )  e t / 2  61  e  t / 2   5 1  e  (t 0.25) / 2 S (t  0.25)
Transient Response Example cont.
 Transfer Functions

The transfer function
» Represent relation between input U(s) & output Y(s) in the
Laplace domain
U(s) Y(s)
» Usually denoted as G(s) G(s)
Input Output
» Y(s) = G(s)U(s)
Transfer
» Only applicable to linear models! function

Deviation variables
» Defined as difference between variable and its steady-state
value
u ' (t )  u (t )  u U ' ( s)  L[u ' (t )] y ' (t )  y (t )  y Y ' ( s)  L[ y ' (t )]

» Transfer functions always specified in terms of deviation

variables
» Y’(s) = G(s)U’(s)
» Usually often omit primes for notational simplicity
 Transfer Function Example

Stirred tank heater
dT
VC p  wC p (Ti  T )  Q
dt

Steady-state equation: 0  wC p (Ti  T )  Q


Initial conditions: T (0)  T Ti (0)  Ti Q(0)  Q


Subtract steady-state equation
 0  wC p  (Ti  Ti )  (T  T )   Q  Q 
dT
VC p
dt
VC p
dT '
dt
 
 wC p Ti '  T '  Q'  Deviation model
 Transfer Function Example cont.

Laplace transform
C pV [ sT ' ( s )  T ' (0)]  wC p [Ti ' ( s)  T ' ( s )]  Q' ( s )


Rearrange noting that T’(0) = 0
1 1 / wC p
T ' (s)  Ti ( s)  '
Q' ( s)
V V
s 1 s 1
w w

Definitions
pV
Time constant : 
w
1
Steady - state gain : K 
wC p

Transfer functions – 1st-order system
1 K
T ' (s)  Ti ' ( s )  Q' ( s )  G1 ( s )Ti ' ( s )  G2 ( s )Q ' ( s )
s  1 s  1
 Properties of Transfer Functions


Additive property
» Y(s) = G1(s)U1(s)+ G2(s)U2(s)


Multiplicative property
» Y2(s) = G1(s)G2(s)U(s)


ODE equivalence
Y ( s) b1s  b0
Transfer function  G ( s) 
U (s) a2 s 2  a1s  a0
d2y dy du
Equivalent ODE a2 2  a1  a0 y  b1  b0u
dt dt dt