Transfer Functions

That knowledge of
the process is
contained in a model
of the process
Outputs Inputs
G(s)
Input-Output Models: The Transfer Function
Introduction
Modeling activity often leads to a set of nonlinear
ordinary differential equations.
Such models are difficult to work with and may not yield
analytical solutions.
Input-output models require simple expressions that
can represent process dynamics at least locally.
Local behavior can be obtained by linearizing the
nonlinear ODEs around a desired operating point.
A linear ODE is the starting point for input-output
models expressed in the Laplace domain.
Transfer Functions
Process Dynamics Mixing Tank
Two process streams are mixed to produce one of the
feeds for a chemical reactor.
After mixing, the blended stream is fed to a heating
vessel before being sent to the reactor.

Process Dynamics Mixing Tank
The process is running along at steady state.
At 3:00 P.M. the shift changes at the plant.
The new operator misreads the flow meters for the
process and switches the flow rates of the two streams.
Stream 1 is switched to 20 L/min, and stream 2 is
switched to 10 L/min.
At 3:30 P.M. the shift supervisor hurries to the control
room to determine the source of the problem now being
experiencing with the reactor.
Use your knowledge of chemical engineering to
determine what has happened to the exit concentration
from the heating vessel over the first half-hour of the
shift.
Mixing Tank Balance on A
After the change, the new concentration to the heating
tank is
Mixing Tank Balance on A
around heating vessel
Mixing Tank - heating vessel
Mixing Tank Energy balance
Mixing Tank Energy balance
T
3
= 45
Mixing Tank Energy balance
around heating vessel
= 1000 g/L, Cp = 1cal / g.
o
C

Amount of heat required to raise the temperature of tank
outlet to 80
oC
, Q = 73.2 kW
Mixing Tank Energy balance for
new conditions
T
3
= 35
o
C









The outlet temperature drops by 10
o
C from 80 to 70
o
C
Mixing Tank Energy balance for new
conditions around heating vessel
Mixing Tank Energy balance for new
conditions around heating vessel
Example - Tank Level
F
in
F
V=A h
A : Cross sectional area
h : Height of the liquid level
V : Volume of the tank
R
R: Resistance
The fundamental quantity that provides the
information about the dynamics of the tank is:

The total mass of the liquid in the tank


: density of the liquid
V: volume of the liquid
A: cross sectional area of the tank
h: height of the liquid level
Ah V Mass Total = =
Example - Tank Level
State Variable: h
Constant parameters: and A

Note: It is assumed that the density is
constant and independent of temperature.
Conservation principle on the fundamental
quantity: the total mass
| | | | | |
time time time
mass total of Output mass total of Input mass total of on Accumulati
=
Example - Tank Level
( )
F F
dt
Ah d
i

=

F
i
: inlet volumetric
flow rate (m
3
/min)

F : outlet volumetric
flow rate (m
3
/min)
F F
dt
dh
A
i
=
: constant
Additional Equation:
h
R
h
F o = =
h : State Variable and Output
Example - Tank Level
Steady-State Condition
A very important concept associated with the model for
control is the notion or condition of steady-state of the
system.

Steady-State is the state at which the state variables no
longer change with time. In this case, the accumulation term is
equal to zero.

The application of the steady-state condition leads to a
set of algebraic equations

The solution of this set of equations gives us the values of the
state variables at steady-state, also called the equilibrium
state.
Example - Tank Level
Steady-State Equation
F F
dt
dh
A
in
= = 0
R
h
F
s
s in
=
,
0
Define Deviation Variables
s in in in s
F F F h h h
,
, = =
h
RA
in
F
A dt
h d
1
1
=
The linear process model represents the relationship
between the state (output) variable (h) and the input
variable (F
in
).
Example Tank Process
Desired operating point
around which the
linearization is performed.
Laplace Transform allows us to obtain a simpler
representation of the system dynamics..



Associated with the use of Laplace Transforms, we
can define a fundamental concept
..in terms of algebraic equations, instead of
differential equations
Transfer Function
Transfer Function
Definition: The TF of a linear dynamic system is
given as the ratio of the Laplace transform of the
output to the Laplace transform of the input.
u(t) y(t)
Process
( )
) (
) (
Input the of Transform Laplace
Output the of Transform Laplace
Function Transfer
s u
s y
s g = = =
Transfer Function (TF)
u(s) y(s)
g(s)
) (
0 1
t u y a
dt
dx
a = +
Consider
with y(0)=0 ; y(t): deviation variable
) ( = ) ( + ) (
0 1
s u s y a s y s a
0 1
1
) (
) (
) (
a s a s u
s y
s g
+
= =
Example
Defining,
0 0
1
1
;
a
k
a
a
= = t
1
) (
+
=
s
k
s g
t
Transfer function
of a first-order
process
k: Steady-State Gain t : Time constant
Example
The transfer function is represented schematically as,
Cause and effect relationship !!!
Example
u(s) y(s)
1
) (
+
=
s
k
s g
t
Level of liquid in a tank
R k = AR = t
Example - Tank Process
1 ) (
) (
+
=
s
k
s F
s h
in
t
F
in
V=Ah
F
out
Consider
u b u b h b y a y a y a
m
m m
n
n n
+ + + = + + +

... ...
) 1 (
1
) (
0
) 1 (
1
) (
0
) ( ... ) ( ) ( ) ( ... ) ( ) (
1
1 0
1
1 0
s u b s s u b s s u b s y a s s y a s s y a
m
m m
n
n n
+ + + = + + +

Applying Laplace
Transform..
( ) ( ) ) ( ... ) ( ...
0 )
1
1 0
s u b s b s y a s a s a
m
m
n
n n
+ + = + + +

Collecting Terms..
TF for a General System
Finally
n
n
m
m
a s a
b s b
s u
s y
s g
+ +
+ +
= =
....
...
) (
) (
) (
0
0
Using the Transfer Function concept, we
can represent the system dynamics by
algebraic equations in s.
TF for a General System

The transfer function allows the development of a
simpler input-output model than one would obtain in
the time domain.

It describes completely the dynamic expression of the
output variable when the corresponding input variable
changes are specified.

The transfer function is a property of the system and
is independent of the functional form of the input
variable.
Characteristics of the TF
We cannot form the TF with nonzero initial
conditions..

TF is only defined for deviation variables..

To find the transfer function for a nonlinear system, we
must :

First linearize around a steady-state operating point,
and
Express the differential equation in terms of deviation
variables
Characteristics of the TF
Example Tank Process
in
F h
dt
dh
A = +o
F
in
V=Ah
F
out
) (
) (
,
,
,
,
,
,
s in in
F h
in
in
s
F h
in
s in s
in
F F
A
F
A
h
F
h h
A
F
A
h
h A
F
A
h
A
F
A
h
dt
dh
s in s
s in s

+
c
c
+

+
c
c
+
(
(

+ =
+ =
o
o
o
o
Taylor Expansion
Define Deviation Variables
h b F a
dt
h d
in
=
s
h h
h A
dh
h d
A
b
A
a
s
2
) ( 1
,
1 o o
=
(

= =
=
Example Tank Process
? = k ? = t
1 ) (
) (
+
=
s
k
s F
s h
in
t
Develop the transfer function by considering the unsteady-state behavior of
an ordinary mercury-in-glass thermometer.
Consider the thermometer to be located in a flowing stream of fluid for
which the temperature x varies with time
Calculate the response or the time variation of the thermometer reading y
for a particular change in x.

Example Mercury Thermometer
Assumptions:

1. All the resistance to heat transfer resides in the film surrounding
the bulb (i.e., the resistance offered by the glass and mercury is
neglected).

2. All the thermal capacity is in the mercury. Furthermore, at any
instant the mercury assumes a uniform temperature throughout.

3. The glass wall containing the mercury does not expand or contract
during the transient response.
Example Mercury Thermometer
It is assumed that the thermometer is initially at steady state.
This means that, before time 0, there is no change in temperature
with time.
At time 0, the thermometer will be subjected to some change in the
surrounding temperature x(t).
Example Energy balance
Example Energy balance
General procedure for development of
transfer functions
Step 1. Write the appropriate balance equations (usually mass or
energy balances for a chemical process).

Step 2. Linearize terms using Taylors series, if necessary.

Step 3. Place balance equations in deviation variable form.

Step 4. Laplace-transform the linear balance equations.

Step 5. Solve the resulting transformed equations for the transfer
function, the output divided by the input.
Standard form of a first order process
where y is the output variable and x ( t ) is the input forcing function
The initial condition is
y(0) = y
s
Introducing deviation variables gives
Standard form of a first order process
where y is the output variable and x ( t ) is the input forcing function
The initial condition is
y(0) = y
s
Introducing deviation variables gives
Response of a first order process for
different forcing functions
1. Step input
2. Impulse input
3. Ramp input
4. Sinusoidal input
Example
A thermometer having a time constant of 0.1 min is at a steady-state
temperature of 90F. At time t = 0, the thermometer is placed in a
temperature bath maintained at 100F. Determine the time needed for
the thermometer to read 98F.
Example
The ultimate thermometer reading will be 100F, and the ultimate value
of the deviation variable Y(t) is 10F. When the thermometer reads 98F,
Y(t) is 8F.

Derive the expression for Y(t) for a step change in the input
Time Delay (Deadtime)
rate flow volumetric
pipe of volume
velocity fluid
pipe of length
=
=
D
t
Whenever material or energy is physically transported in a process
or plant, there is a time delay associated with it.
The transportation time between points 1 and 2 is given by:

>
<
=
0 ) (
0 0
) (
t t t x
t
t y
D
1
2
x(t) y(t)
Time
x(t)
y(t)
t
D
The output y(t) is simply the same input function
shifted back in time by the amount of delay.
s t
e s g
s x
s y
D

= = ) (
) (
) (
Transfer Function
Time Delay (Deadtime)
Example
L
T
1
(t)
T
i
(t)
T(t)
The following are assumed:
The pipe between the tank and
point 1 is well insulated
The flow of the liquid through the
pipe is ideal plug flow
The response of T
1
(t) to inlet disturbances in T
i
will be
the same as T(t) except that it will be delayed by
some amount of time, equal to
F
AL
t
D
=
1 ) (
) (
+
=
s
k
s T
s T
i
t
s t
e
s T
s T
D

=
) (
) (
1
s t
e
s
k
s T
s T
D
i

+
=
1 ) (
) (
1
t
Overall Transfer Function
Example