INDUSTRIAL INSTRUMENTATION AND PROCESS CONTROL

LINEARIZATION & DEVIATION VARIABLES

RONALD F. RINGOR, CHE
LINEARIZATION
•The models discussed were easily solved because they involved
linear equations, which were a natural result of the conservation
balances and constitutive relationships for the specific physical
systems.

•However, the conservation and constitutive equations are non-

linear for most systems, and general methods for developing
analytical solutions for nonlinear models are not available.

•Fortunately, methods exist for obtaining approximate linearized

solutions to nonlinear systems, and experience over decades has
demonstrated that linearized methods of control systems analysis
provide very useful results for many (but not all) realistic
processes.
LINEARIZATION
Linearity:
An operator F is linear if it satisfies the properties of additivity and
proportionality, which are included in the following superposition,
where xi are variables and a and b are constants:

F(ax1 + bx2) = aF(x1) + bF(x2)

A linearized model can be developed by approximating each non

linear term with its linear approximation. A non linear term can be
approximated by a Taylor Series expansion to the nth order about a
point if derivatives up to the nth order exist at that point; (refer to
Table 3.3 pp76, Marlin)
LINEARIZATION
Examples:
1 1/ 2
F ( x)  x F ( x )  xs  xs ( x  x s )
1/ 2 1/ 2

x xs 1
F ( x)  F ( x)   ( x  xs )
1  ax 1  axs (1  axs ) 2
DEVIATION VARIABLE

A deviation variable simply translates the total variable by a

constant value, and the total value of the variables is easily
recovered by adding the initial steady state value xs to its
deviation value x’.

The use of deviation variables is not necessary and provides

no advantage in analysis at this point. However, expressing a
model in deviation variables will be useful in future
discussions.

Deviation variable: (x-xs) = x’

EXAMPLE 1. ISOTHERMAL CSTR
Goal: Determine the transient response of the tank concentration in
response to a step in the inlet concentration for the nonlinear and
linearized models.

Information: The process equipment and flow are the same as in the
first example of CSTR. The important variable is the reactant
concentration in the reactor.

Assumptions. The same as first example

Data: The same as in first example except that the chemical reaction
rate is second-order, with rA = -kCA2 and k=0.5[ (mole/m3)min]-1
EXAMPLE 1. ISOTHERMAL CSTR
dC A
Formulation: V  F (C A0  C A )  VkCA
2

dt

C A2  C AS
2
 2C AS (C A  C AS )

dC A
V  F (C A0  C A )  [VkCAS2
 2VkCAS (C A  C AS )]
dt
dC A
V  F (C A0 S  C AS )  [VkCAS
2
 2VkCAS (C AS  C AS )]
dt
EXAMPLE 1. ISOTHERMAL CSTR

dC A '
V  F (C ' A0 C ' A )  2VkCAS C ' A
dt

dC A ' 1 F
 C'A  C ' A0
dt  V
V
 
F  2VkC AS
EXAMPLE 1. ISOTHERMAL CSTR
Solution: by application of the integrating factor,
the following relationship is derived
 
1  e t /   C A0 Kp 1  e t / 
F
C ' A  C ' A0 
 F  2VkCAS

F
Kp 
F  2VkCAS

C ' A  0.9250.1461  e t / 3.62


EXAMPLE 1. ISOTHERMAL CSTR

Results Analysis: The linearized solution is plotted

in comparison to the original nonlinear
differential equation. The linear solution can
give a good semi quantitative description of the
true process response. (refer to Figure 3.12
pp79, Marlin, also refer to table 3.4)
EXERCISE: TANK LEVEL CONTROL

Consider a liquid surge tank (liquid level control).

If the outlet flowrate, q is related to the tank
level, h by q=βh0.5 , find the linearized
approximate model for the system and compare
it with the nonlinear solution through a graphical
representation.
EXAMPLE 2. TANK DRAINING
Goal: Determine a model for this system. Evaluate the
accuracies of the linearized solution for small (10m3/hr)
and large (60m3/hr) step changes in the inlet flowrate.

Information: the system is the liquid in the tank, and the

important variables are level and flow out.

Assumptions:
1.the density is constant.
2.The cross-sectional area of the tank, A, does not
change with height.
EXAMPLE 2. TANK DRAINING

 Data: 1.The initial steady-state conditions are:

 Flows = Fo = F1 = 100m3/h
 Level = L = 7m

2.The cross sectional area is 7m2

 Formulation: By TMB:
dL
A  F0  F1
dt
EXAMPLE 2. TANK DRAINING
 This single balance does not provide enough
information, because there are two unknowns, F1 and
L. Thus, the number of degrees of freedom (1)
indicates that another equation is required.

 The following relationship relating the flow out to the

liquid level is estbalished:

F1  k F L 0.5

CONTINUE…
INDUSTRIAL INSTRUMENTATION AND PROCESS CONTROL

LAPLACE DOMAIN DYNAMICS

RONALD F, RINGOR, CHE
 THE LAPLACE TRANSFORMS
 In the past, we developed a number of mathematical
models that describe the dynamic operation of selected
processes. Solving such models – that is, finding the output
variables as functions of time for some change in input
variable(s) – requires either analytical or numerical
integration of the differential equations.

 The use of Laplace Transforms offers a very simple and

elegant method of solving linear or linearized differential
equations which result from the mathematical modeling of
the chemical process.
 DEFINITION
 The Laplace Transform of a function f (t ) is defined as:

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

 where F(s) is the symbol for the Laplace Transform, s is a

complex independent variable, f(t) is some function of
time to be transformed, and L is an operator, defined by
the integral.
 THE LAPLACE TRANSFORMS

 The Laplace transformation of a function of time,

f (t ) consists of “operating on” the function by
 st
multiplying it by e and integrating with respect
to time t from 0 to infinity.

 Thus, Laplace transformation converts functions

from the time domain (where t is the
independent variable) to the Laplace domain
(where s is the independent variable).
 LINEAR PROPERTY

 One of the most important properties of the

Laplace transformation is that it is linear. A linear
operator satisfies the superposition principle:

L[ax(t )  by(t )]  aX (s)  bY (s)

 Therefore the Laplace Transform of a sum of
functions x(t) and y(t) is the sum of the individual
Laplace Transforms X(s) and Y(s); in addition
multiplicative constants can be factored out of the
operator.
LAPLACE TRANSFORMS OF IMPORTANT
FUNCTIONS

 Constant Function
 Step Function
 Derivatives
 Exponential Functions
 Ramp
 Others (Trigonometric Functions, Rectangular
Pulse, Impulse…)

 For more transforms, refer to Table 4.1 pp110

SOLUTION OF LINEAR DIFFERENTIAL EQUATIONS
USING LAPLACE TRANSFORMS
 INVERSION (PARTIAL FRACTION EXPANSION)

 As mentioned, the primary use of the Laplace

Transforms is to solve linear differential equations or
systems of linear (or linearized nonlinear) differential
equations with constant coefficients.

 The procedure was developed by the English

engineer Oliver Heaviside and it enables us to solve
many problems without going through the trouble of
finding the complementary and particular solutions
of differential equations.
 INVERSION (PARTIAL FRACTION EXPANSION)
 The procedure used to solve a differential
equation is quite simple:
 First,‘Laplace transform’ both sides of the differential
equation, substituting values for the initial conditions
in the derivative transforms;

 Rearrange the resulting algebraic equation, and solve

for the transform of the dependent (output) variable;

 Finally find the inverse of the transformed output

variable.
 INVERSION (PARTIAL FRACTION EXPANSION)

Example #1. Solve the differential equation,

dy
5  4y  2
dt
y (0)  1

*the critical point in finding the solution to a

differential equation using Laplace transforms is
the inversion of the Laplace transform.
Ans : y(t )  0.5  0.5e0.8t
 INVERSION (PARTIAL FRACTION EXPANSION)
Solve the differential equation: d 3 y d2y dy
 6 2  11  6 y  1
d y 3
3
dt 3
dt dt
L 3   s Y ( s ) y (0)  y ' (0)  y ' ' (0)  0
 dt 
 d2y 
 
L 6 2   6s Y ( s)
2

 dt 
 dy  Y ( s)( s  6s  11s  6) 
3 2 1
L11   11sY ( s )
 dt  s
1
1 Y (s) 
L (1)  s( s 3  6s 2  11s  6)
s
 INVERSION (PARTIAL FRACTION EXPANSION)
 In general, a transform equation may not exactly
match any of the entries in “tables”. This problem
always arises for higher-order differential equations
because the order of the denominator polynomial of
the transform is equal to the order of the original
differential equation.

 The factors of the characteristic polynomial

correspond to the roots of the polynomial set equal to
zero. Once the factors are obtained, the Laplace
transform is then expanded into partial fractions.
 INVERSION (PARTIAL FRACTION EXPANSION)

 The fastest and most popular method is called the

Heaviside expansion. In this method multiply both
sides of the equation by one of the denominator
terms (s+bi) and then set s= -bi, which causes all
terms except one to be multiplied by zero.
s5
 Consider: Y ( s ) 
s  5s  4
2
 INVERSION (PARTIAL FRACTION EXPANSION)

 We now state a
general procedure
to solve ODE using
Laplace Transforms
SEATWORK
Transform the following Laplace Transform
expressions to the time domain expressions:
a) 1
Y ( s) 
s( s 3  6s 2  11s  6)
b) s 1
Y (s) 
s( s 2  4s  4)
c)Solve the model of the isothermal CSTR
example through Laplace Transforms
dC A ' 1 1
 C A '  C AO ' C ' A(t )  CA0(1  e t / )
dt  
 TRANSFER FUNCTIONS
AND INPUT-OUTPUT MODELS
 TRANSFER FUNCTIONS
 The use of Laplace Transforms allows us to form a very
simple, convenient, and meaningful representation of
chemical process dynamics.

 simple because it uses only algebraic equations not

differential ones;

 convenient because it allows a quick analysis of process

dynamics;

 meaningful because it provides directly the relationships

between inputs (disturbances, manipulated variables)
and the outputs (controlled variables) of a process.
 TRANSFER FUNCTIONS

 A very common manner for presenting input-

output models, which finds considerable
application in process control, is the transfer
function.

 The transfer function is a model based on

Laplace transforms with special assumptions as
follows.
 TRANSFER FUNCTIONS

 The transfer function of a system is defined as the

Laplace transform of the output variable, y(t),
divided by the Laplace transform of the input
variable, x(t), with all initial conditions equal to
zero.
Y ( s)
TransferFunction  G( s) 
X ( s)
 TRANSFER FUNCTIONS

 The assumptions of Y(0) = 0 and X(0) = 0 are

easily achieved by expressing the variables in the
transfer functions as deviations from the initial
conditions.

 Thus, all transfer functions involve variables that

are expressed as deviations from an initial steady
state. All derivatives are zero if the initial
conditions are at steady state.
 TRANSFER FUNCTIONS
 Note that the transfer function relates one output
to one input variable. IF more than one input or
output exists, an individual transfer function is
defined for each input-output relationship. The
effects of several inputs can be summed to
determine the net effect on the output.

 The overall behavior of a complex system can be

modeled considering only the input and output
variables by the use of transfer functions and
block diagrams.
 TRANSFER FUNCTIONS (CSTR – MIXING)
Recall the model for the CSTR mixing system. In deviation
variables: dC ' A
V  F (C ' A0  C ' A)
dt
Taking the Laplace Transform of each term:
V [ sC ' A( s )  C ' A(t ) t  0]  F [C ' A0( s )  C ' A( s)]
 CA0 
V [ sC ' A( s )]  F   C ' A( s ) 
 s 
CA0 1 CA( s) 1
TF : 
C ' A( s )  CA0( s) s  1
s s  1
t / 
C ' A(t )  CA0(1  e )
Properties shown by the Laplace transform:
 Linearity
 Real differentiation theorem
 Real integration theorem
 Real translation theorem
 Final value theorem
 Complex differential theorem
 Complex translation theorem
 Initial value theorem
PROPERTIES SHOWN BY TRANSFER FUNCTIONS

 ORDER
 POLE
 ZERO
 ORDER OF NUMERATOR AND DENOMINATOR
 CAUSALITY
 STEADY STATE GAIN

(page 125, Marlin)

 PROPERTIES SHOWN BY TRANSFER FUNCTIONS

 Order: The order of the system is the highest

derivative of the output variable in defining
differential equation, when expressed as a
combination of all individual equations. For
transfer functions of physical systems, the order
can be easily determined to be the highest power
of s in the denominator.
 PROPERTIES SHOWN BY TRANSFER FUNCTIONS
 Pole: A pole is defined as a root of the denominator
of the transfer function, thus, it is the same as a root
of the characteristic polynomial. Important
information on the dynamic behavior of the system
can be obtained by analyzing the poles, such as: a)
the stability of the system b) the potential for
periodic transients.
 The analysis of poles is an important topic on feedback
systems since feedback control affects the poles.
 Zero: A zero is a root of the numerator of the
transfer function.
 PROPERTIES SHOWN BY TRANSFER FUNCTIONS

 Order of numerator and denominator: Physical

systems conform to specific limitations between
the orders of the numerator and denominator,
that is, the order of the denominator must be
larger- than the order of the numerator.
 This limitation results from the observation that
real physical systems do not contain pure
differentiation as would be required for a system
with numerator order greater than denominator
order.
 PROPERTIES SHOWN BY TRANSFER FUNCTIONS
 Causality: The direction of the cause-effect relationship is
essential to control system design. This direction is
presented in the transfer function by identifying the variable
in the denominator as an input (cause) and an output (effect).
 In designing feedback control strategies, the variable chosen
to be adjusted must be an input, and the measured controlled
variable used for determining the adjustment must be an
output.
 When the physical system is causal, the order of the
denominator is greater than that of the numerator and the
value of the transfer function as s  ∞ is equal to 0 (zero).
Such a transfer function is referred to as Strictly Proper.
 PROPERTIES SHOWN BY TRANSFER FUNCTIONS

 Steady-State Gain: The steady-state gain is the

steady state value of Y/X for all systems
whose outputs attain steady state after an input
perturbation X. The steady-state gain is normally
represented by K, often with the subscript and
can be evaluated by setting s = 0 in the (stable)
transfer function. This is exact for linear systems
and gives the linearized approximation for non-
linear systems.
BLOCK DIAGRAMS
 CONSIDER: TWO SURGE TANKS IN SERIES
 Suppose that two liquid surge tanks are placed in series so that
the outflow from the first tank is the inflow to the second tank,
as shown in the figure. If the outlet flow rate from each tank is
proportional to the height of the liquid (head) in that tank find
the transfer function relating changes in flow rate from the
second tank, Q’2(s), to changes in flow rate into the first tank,
Q’i(s).
 Show how this transfer function is related to the individual
transfer functions
 H’1(s)/Q’i(s),
 Q’1(s)/H’1(s),
 H’2(s)/Q’1(s),
 Q’2(s)/H’2(s)
 Assume that the two tanks have different cross-sectional areas
A1 and A2, and that the valve resistances are R1 and R2.
EXAMPLE
 BLOCK DIAGRAMS – FUNDAMENTALS
 In modeling, the initial steps involve balances based on
fundamental principles, which can be applied to each process of
a complex plant. Then the transfer function of each system is
determined by taking the laplace transform of the linearized
model.

 The overall transfer function can be used to determine some

important properties of the system without solving the defining
differential equations. These properties include: (1) The final
value of the output variable. (2). The stability of the response
and (3). The response of the output to sine input

 The block diagram can be constructed to present the

interactions among the individual transfer functions and the
overall transfer function for the integrated system can be
derived through block diagram manipulation.
 BLOCK DIAGRAMS – FUNDAMENTALS

A block diagram is a shorthand, pictorial

representation of the cause-and-effect
relationship between the input and output of
a physical system. It provides a convenient
and useful method for characterizing the
‘functional relationships among the various
components of a control system.
 BLOCK DIAGRAMS – FUNDAMENTALS
 The simplest form of the block diagram is the single
block, with one input and one output, as shown

INPUT OUTPUT
BLOCK

 The interior of the rectangle representing the block

usually contains a description of or the name of the
element, or the symbol for the mathematical
operation to be performed on the input to yield the
output. The arrows represent the direction of
information or signal flow.
BLOCK DIAGRAMS – FUNDAMENTALS
 BLOCK DIAGRAMS – FUNDAMENTALS
 In general, a block diagram consists of a specific
configuration of four types of elements: blocks,
summing points, takeoff points, and arrows
representing unidirectional signal flow:
EXAMPLE: CSTR MIXING

 Recall the Transfer Function of the continuous

stirred mixing tank:
CA( s) 1
TF : G( s)  
CA0( s) s  1

 Block Diagram:
__1__
CAO (s) CAO (s)
τs + 1
 EXAMPLE: TWO SURGE TANKS IN SERIES
Getting the required TFs:

H1' ( s) K1

Qi' ( s)  1S  1
Q1' ( s) 1
'

H1 ( s) K1
H 2' ( s) K2

Q1' ( s)  2 S  1
Q2' ( s) 1

H 2' ( s) K 2
Obtaining the Input – Output Diagram:
 EXAMPLE: TWO SURGE TANKS IN SERIES

 The desired transfer function relating the outflow

of Tank 2 to the inflow to Tank 1 can be derived by
forming the product of the individual transfer
functions ' ' ' ' '
Q2 ( s) Q2 ( s) H 2 ( s) Q1 ( s) H1 ( s)
 '
Qi ( s) H 2 ( s) Q1' ( s) H1' ( s) Qi' ( s)
'

or
Q2' ( s) 1 K 2 1 K1
 '
Qi ( s) K 2  2 s  1 K1  1s  1
'
EXAMPLE: TWO SURGE TANKS IN SERIES
'
Q ( s) 1
2

Q ( s) ( 2 s  1)( 1s  1)
i
'

OVERALL TRANSFER FUNCTION:

Q i’ 1 Q2’
( 2 s  1)( 1s  1)
 BLOCK DIAGRAM ALGEBRA
 Block diagrams of complicated control systems may be
simplified using easily derivable transformations.
BLOCK DIAGRAM ALGEBRA
BLOCK DIAGRAM ALGEBRA
BLOCK DIAGRAM EXERCISES

 Given the following block diagrams, derive the

overall input-output transfer functions:
+
X0(s) G1 G2 G3 G4 X1(s)

+ e + e
X0(s) G1 G2 G3 G4 X1(s)

- -
 Gd d(s)

Xsp(s) + e(s) Gc X1(s) G X2(s) Gp + Y(s)

f

- Ym(s)
Gm