L. J.

Institute of Engineering & Technology

Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Introduction
What is Process Control?
Process Control is an engineering
discipline that deals with architectures,
mechanisms & algorithms for
maintaining the output of specific
process within desires range.

Nimesh Prabhakar [email protected] 2

 Process Control
• In chemical plant there are number of units such
as,

To Constitute a chemical plant, we need to

assemble few of these units.
Nimesh Prabhakar [email protected] 3
 Process Control
What is the objective of Chemical plant?
The plants overall objective is to convert certain
raw materials into desired products using
available sources of energy, in the most
economical way.
Raw Product
Plant
material

Now to meet this objective, we need to meet

some requirements.

Nimesh Prabhakar [email protected] 4

 Requirements
• Safety
• Production specification
• Environmental regulation
• Operational Constraints
• Economics

Nimesh Prabhakar [email protected] 5

 Requirements
Safety
• Safety of personnel & equipment is of foremost
concern.
• Operating temperatures, pressures, flow rates etc
should not be allowed to exceed allowable limits.
100 psig

• We should have a control system that will

maintain the pressure below 100 psig value.
Nimesh Prabhakar [email protected] 6
 Requirements
Production specifications
• A plant should produce the desired quantity
without compromising with the desired
quality.

Nimesh Prabhakar [email protected] 7

 Requirements
Environmental Regulation
• The plant effluents (in the form
of solid, liquid, or gas) should not
exceed the limits decided by
federal & state laws.
• Effluents should not cause air,
water, or soil pollution.
• The amount of SO2 that a plant
inject to the atmosphere
• Quality of water returned to a
river or a lake.

Nimesh Prabhakar [email protected] 8

 Requirements
Operational Constraints
• The various type of equipments used in a chemical
plant have constraints inherent to their operation.
• Such Constraints should be satisfied throughout the
operation of plant.

Nimesh Prabhakar [email protected] 9

 Requirements
Economics
• Plant must operate at the optimum level.
Minimum operating cost
&
maximum profit
• It should be economical in utilization of
material, manpower energy & capital.

Nimesh Prabhakar [email protected] 10

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Classification of Variables

Nimesh Prabhakar [email protected] 2

What is Variable?
• A variable is something that can vary or
change.
Temperature
Pressure
Flow rate
Current
Nimesh Prabhakar [email protected] 3
 Classification of Variables
Variables

Input Output

Manipulated Disturbance Measured Unmeasured

Nimesh Prabhakar [email protected] 4

 Classification of Variables
Input Variables
• Input Variables denote the effect of the
surrounding on the chemical process
Output Variables
• Output variables denotes the effect of the
process on the surrounding.

Nimesh Prabhakar [email protected] 5

 Classification of Variables
Manipulated Variables
• Input variables are called manipulated variables if
their values can be adjusted freely by the human
operator or a control mechanism.
Disturbance variables
• Input variables are called disturbance variables if
their values are not the result of adjustment by
an operator or a control system.

Nimesh Prabhakar [email protected] 6

 Classification of Variables
Measured Variables
• Output variables are called measured variables
if their values are known by directly measuring
them.
Unmeasured Variables
• Output variables are called unmeasured
variables if they are not or cannot be measured
directly.

Nimesh Prabhakar [email protected] 7

 Classification of Variables
Control Variable
• Control variable is one that must be kept
constant for a fair test.
Reference or Setpoint
• The desired condition of the controlled
variable is called the reference or setpoint.

Nimesh Prabhakar [email protected] 8

 Example
Fi = Input steam flow rate
F0 = output steam flow rate
h = Height of liquid in tank
A = Cross sectional area of tank
Objective
To maintain the liquid height in the tank at it’s
desired value hd.
Objective : h = hd Manipulated Control Disturbance
Control variable = h Variables Variables variables

Input variable = Fi F &F h

i 0 F &F i 0

Output variable = F0
Nimesh Prabhakar [email protected] 9
 Continuous Stirred Tank Heater(CSTH)
FSt = Steam flow rate
Fi & Ti = Flow rate & temperature of feed
F0 & T0 = Flow rate & temperature of outlet stream
Input variable = Fi, Ti, FSt, Tst
Output Variable = F0 & T0
Objective
• T = Td
• h = hd
Manipulated Control Variables Disturbance
Variables variables
FSt T Fi , Ti, Tst
Fi & F0 h Fi & F0
Nimesh Prabhakar [email protected] 10
 L. J. Institute of Engineering & Technology
Department of Electronics & Communication

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Servo & Regulatory Control
• The objective of control system is one of the
following
• Elimination of disturbance
– Regulatory control
• Making the controlled variable follow the
changes in set point
– Servo control

Nimesh Prabhakar [email protected] 2

 Regulatory Control
• In this case deviation of the output from the
setpoint is minimized in the case of changing
circumstances by adjusting the inputs to the
system.
o Controlling the temperature in the room.
Setpoint = 250c
T = 270c
Deviation = 20c
T = 250c

Nimesh Prabhakar [email protected] 3

 Servo Control
• In this case, the aim is to get the control to follow a
desired trajectory specified by the input.
i. Tracking of missiles & aircrafts
ii. Guiding the direction of space ship.
shivranjani

Prhladnagar

YMC

LJIET

[email protected]
Nimesh Prabhakar 4
 Stability
• A system is defined as stable if the controlled
variables converge to a steady state after the
disturbance.
• Otherwise, the system is unstable.

[email protected]
Nimesh Prabhakar 5
 Stability
If closed-loop poles are located in left half then
system is stable.
G(s) = =
Poles: s =-2 & s = -4

[email protected]
Nimesh Prabhakar 6
 Stability
G(s) = =
Poles: s = 2 & s = 4

[email protected]
Nimesh Prabhakar 7
Mosam Pandya [email protected] 8
L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Mathematical Modelling

Nimesh Prabhakar [email protected] 2

State variable & State equation

Nimesh Prabhakar [email protected] 3

 Mathematical Modelling
Mathematical Model
• A mathematical model of a process is a description
of the characteristics of a number of phenomena
that take place in a process.
Experimental
Input Output
setup

Mathematics Output

• Model is a mathematical representation of a process,

intended to promote understanding of the real
system.
Nimesh Prabhakar [email protected] 4
 Types of Mathematical Model
• Theoretical model
• Empirical model
• Semi-empirical model

Nimesh Prabhakar [email protected] 5

 Theoretical model
• This model is also known as white box model and first
principle model.
• This model is based on principles of conservation law,
Mass conservation
Energy conservation
Momentum conservation
• The models are derived from the physics & chemistry
of the process.

Nimesh Prabhakar [email protected] 6

 Theoretical model
Advantages
• It is applicable for a wide range of condition
• It provide physical insight into process
behavior
Disadvantages
• It is time consuming to develop
• Some model parameters are not readily
available.
(reaction rate coefficient, overall heat transfer
coefficient these are not readily available)
Nimesh Prabhakar [email protected] 7
 Empirical Model
• This model also known as black box model.
• This model is obtained by fitting experimental
data.
Experimental
Input Output
setup

• If we have the experimental setup, and we have

different sets of input & output data.
• So by fitting those experimental data, we can
determine the co-efficient.
• By that way, we can construct a model and that
model is the empirical model.
Nimesh Prabhakar [email protected] 8
 Empirical Model
Advantages
• It is easy to develop & usually quite simple.
Disadvantages
• It require well-designed and accurate
experimental data.
• The behavior predicted by empirical model is
correct only around the experimental condition,
so extrapolation is usually invalid.
• It is applicable for a limited range of condition.

Nimesh Prabhakar [email protected] 9

 Semi-empirical model
• This model is also known as gray box model.
• This model is combined approach of theoretical as
well as empirical model.
• Model equations are derived from basic conservation
laws and certain model parameters.
• Reaction rate and heat transfer co-efficient evaluated
from physical experiments OR from process
operating data.

Nimesh Prabhakar [email protected] 10

 Use of mathematical modelling
• To improve understanding of the process
• To optimize process design and operating
conditions
• To design a control strategy for the process
• To train operating personnel

Nimesh Prabhakar [email protected] 11

 State variable
• State variable describe the natural state of the process.
• There are basically three fundamental quantities,
Mass
Energy
Momentum
• Now, these fundamental quantities are usually not
directly & conveniently measured.
• They are characterized by these variables
Temperature(T)
Pressure(P)
Flow Rate(F)
• These variables are called state variables.
Nimesh Prabhakar [email protected] 12
 State equations
• The equation, which are derived by the application of
conservation principle on the fundamental quantities
to relate the state variables with other variable are
called state equations.

Nimesh Prabhakar [email protected] 13

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Mass Balance
Energy Balance

Nimesh Prabhakar [email protected] 2

 Conservation Principle
Rate of rate of Rate of Rate of
= – +
accumulation input output generation
Mass
Rate of mass rate of Rate of
= –
accumulation input output

Energy
Rate of rate of Rate of energy Rate of
= – +
accumulation input output generation

Nimesh Prabhakar [email protected] 3

 Mass Conservation
Fi = Input water flow rate
F = output water flow rate
Fi = 10 Kg/s
Fi = 8 Kg/s

Rate of mass rate of Rate of

= –
accumulation input output
2 = 10-8
M= Q
Mass = (density of material)(Volume flow rate)
m3
= 3
m
=

Nimesh Prabhakar [email protected] 4

 Energy Conservation
Store in Boiler
3

9
Water In 10 Boiler Steam

Heat Supply
2
Rate of rate of Rate of energy Rate of
= – +
accumulation input output generation
3 = 10 – 9 + 2
Nimesh Prabhakar [email protected] 5
 Liquid Surge Tank
Fi = Input water flow rate
F = output water flow rate
h = Height of liquid in tank
A = Cross sectional area of tank
V = Volume of water in tank
Rate of mass rate of Rate of
= –
accumulation input output
( V) = Fi - F • Here State Variable is h.
V • Because it indicate the status of the system.
= Fi - F • If height is very close to the overflow condition then
the status of the tank is it is going to overflow.
A = Fi – F …(1) • If the height is very close to zero the status is it is
going to empty.
Nimesh Prabhakar [email protected] 6
 Stirred Tank Heater
Fi = Input water flow rate
Ti = inlet water temperature
F = output steam flow rate
T = outlet water temperature
h = Height of liquid in tank
A = Cross sectional area of tank F,T

V = Volume of water in tank

Rate of rate of Rate of energy Rate of
= – +
accumulation input output generation
[V cp(T – Tref)] = Ficp(Ti – Tref) - Fcp(T – Tref) + Q
Tref =Reference temperature for enthalpy calculation
Cpof=change
Rate Specific Heat Enthalpy coming Enthalpy of
of enthalpy Heat Supply
Content of the tank in input stream output stream
Nimesh Prabhakar [email protected] 7
 Stirred Tank Heater
[V cp(T – Tref)] = Ficp(Ti – Tref) - Fcp(T – Tref) + Q
• Consider Tref = 0 and divising by cp
Q
V = FiTi – FT +
cp
But V = Ah
Q
A = FiTi – FT +
cp
Q
Ah + AT = FiTi – FT +
cp
Q
Ah + T(Fi - F) = FiTi – FT +
cp
Q
Ah + TFi - TF = FiTi – FT +
A = Fi – F …(1) cp
Q
Ah = Fi(Ti – T) +
cp

Nimesh Prabhakar [email protected] 8

 Stirred Tank Heater
Q
Ah = Fi(Ti – T) +
cp
• This point we have not specify how heat entering to the
system.
Rate of Heat input
Q = UAt(Tst – T)
U = overall heat transfer co-efficient
At = total area of heat transfer
Tst = Temperature of the inlet steam
T = Temperature of tank
UAt(Tst– T)
Ah = Fi(Ti – T) +
cp

Nimesh Prabhakar [email protected] 9

 Stirred Tank Heater
Q
Ah = Fi(Ti – T) +
cp
State variable = h,T
Output Variable = h,T
Input variable
Disturbances – Ti,Fi
Manipulated variables = Q, F, Fi

Nimesh Prabhakar [email protected] 10

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Continuously Stirred Tank
Reactor (CSTR)

Nimesh Prabhakar [email protected] 2

Continuously Stirred Tank Reactor (CSTR)
• As name suggests, the tank
contents are continuously stirred.
Fi &Ti =flow rate & temperature of
inlet
F & T =flow rate & temperature of
outlet
CAi = Concentration of component A
CA = Concentration of Product
Fc & Tci = flow rate & temperature of
coolant inlet(flow around reactor)
Fc & Tco = flow rate & temperature of
coolant inlet(flow around reactor)

Nimesh Prabhakar [email protected] 3

 Continuously Stirred Tank Reactor (CSTR)

Rate of mass rate of Rate of

= –
accumulation input output
( V) = Fi - F
But V=Ah
V
= Fi – F …(1)

A = Fi – F …(2)
Nimesh Prabhakar [email protected] 4
Continuously Stirred Tank Reactor (CSTR)
Component A
Rate of mass rate of Rate of Rate of
accumulation = input – output + generation

(VCa) = FiCai – Fca – (-ra)V

V CaRate of disappearance of
Ca +V = FiCai – Fca – (-ra)V
A if we multiply with
C
Ca(Fi – F )+ V minus a
= Fisign
Cai –then,
Fca –we(-rget
a)V
the generation
Ca
CVaF=iF––CFaF…(1)
+ V = FiCai – Fca – (-ra)V
i
Ca
V = Fi (Cai – ca) – (-ra)V
Dividing both side by V
C a Fi
= (Cai – ca) – (-ra)
V
Nimesh Prabhakar [email protected] 5
Continuously Stirred Tank Reactor (CSTR)
Ca F i
= (Cai – ca) – (-ra)
V
• Here is the Rate if disappearance
• By Arrhenius equation we can write
-ra = K0 –E/RT ca
E = energy of active reaction
K0 = Frequency factor
R = Universal gas constant
T = Temperature
Ca Fi –E/RT
= (Cai – ca) – K0 ca
V

Nimesh Prabhakar [email protected] 6

Continuously Stirred Tank Reactor (CSTR)
Rate of rate of Rate of energy
= –
accumulation input output
Rate of Rate of
+ +

(V CpT) = Fi CpTi – F CpT – Q + (- H)(-ra)V

Dividing both the side by Cp
V Q (− H)(−ra)V
= FiTi – FT – +
Cp Cp
V Q (− H)(−ra)V
V + T = FiTi – FT – +
Cp Cp
V
Substituting the value of from eq (1)

Nimesh Prabhakar [email protected] 7

Continuously Stirred Tank Reactor (CSTR)
Q (− H)(−ra)V
V + T(Fi – F) = FiTi – FT – +
Cp Cp
Q (− H)(−ra)V
V + TFi – TF = FiTi – FT – +
Cp Cp
Q (− H)(−ra)V
V = Fi (Ti – T) – +
Cp Cp
• Dividing both the side by V
Fi(Ti – T) Q (− H)(−ra)
= – +
V V Cp Cp
Fi Q (− H)(K0 –E/RT ca)
= (Ti – T) – +
V V Cp Cp
Nimesh Prabhakar [email protected] 8
Continuously Stirred Tank Reactor (CSTR)
For CSTR we get three equation
• Total mass balance
V
= Fi – F
• Component mass balance
Ca Fi
= (Cai – ca) – K0 –E/RT ca
V
• Energy mass balance
Fi Q (−∆H)(K0 –E/RT ca)
= (Ti – T) – +
V V Cp Cp MV CV
State variables = V, ca, T
F V
Output variables = V, ca, T
Q T
Disturbances = Cai, Fi, Ti
Input variables = Cai, Fi, Ti, Q, (F)(when feedback control is used)
Manipulated variables = Q, F (Occasionally Fi, or Ti)

Nimesh Prabhakar [email protected] 9

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Continuously Stirred Tank
Heater (CSTH)

Nimesh Prabhakar [email protected] 2

Continuously Stirred Tank Heater (CSTH)

• As name suggests, the tank

contents are continuously
stirred.
Fi &Ti =flow rate & temperature of
inlet
F & T =flow rate & temperature of
outlet F, T
Cp = Specific heat of the liquid
V = volume of the liquid
Q = Rate of Heat(energy) supplied
by steam(mass/time)

Nimesh Prabhakar [email protected] 3

Continuously Stirred Tank Heater (CSTH)
• Content of the tank is well mixed
so that temperature of the liquid
in the tank and the outlet
stream remain the same.
• The physical properties , Cp do
not vary significantly with
temperature. F, T
• All heat given up by steam is
received by the liquid content of
the tank; i.e., no heat from the
steam is accumulated in the coil.
• Heat losses to the atmosphere
are negligible.

Nimesh Prabhakar [email protected] 4

 Continuously Stirred Tank Heater (CSTH)

Rate of mass rate of Rate of

= –
accumulation input output
( V) = Fi - F
But V=Ah
V
= Fi – F …(1) F, T

A = Fi – F …(2)
Nimesh Prabhakar [email protected] 5
Continuously Stirred Tank Heater (CSTH)

Rate of
Rate of rate of Rate of energy
= – +
accumulation input output

V Cp(T-Tref)] = Fi Cp(Ti-Tref) – F Cp(T-Tref) + Q

(V CpT) = Fi CpTi – F CpT + Q

Dividing both the side by Cp
V Q
= FiTi – FT +
Cp F, T
V Q
V + T = FiTi – FT –
Cp
V
Substituting the value of from equation (1)

Nimesh Prabhakar [email protected] 6

Continuously Stirred Tank Heater (CSTH)

Q
V + T(Fi – F) = FiTi – FT +
Cp
Q
V + TFi – TF = FiTi – FT +
Cp
Q
V = Fi (Ti – T) +
Cp
• Dividing both the side by V
Fi(Ti – T) Q
= –
V V Cp
Nimesh Prabhakar [email protected] 7
Continuously Stirred Tank Heater (CSTH)
For CSTH we get Two equation
• Total mass balance
V
= Fi – F F, T
• Energy balance
Fi(Ti – T) Q
= –
V V Cp MV CV
State variables = V, T F V
Output variables = V, T Q T

Input variables = Fi, Ti, Q, (F)(When feedback control use)

Manipulated variables = Q,F (Occasionally Fi, or Ti)

Nimesh Prabhakar [email protected] 8

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Degrees of Freedom

Nimesh Prabhakar [email protected] 2

 Degrees of Freedom

• Degrees of freedom is the difference between the

total number of independent process variables
and total number of independent equations.
F=V–E
F = Degrees of Freedom(DOF)
V = total number of independent process variable
E = total number of independent equation
To analyze this degree of freedom we consider three
case.

Nimesh Prabhakar [email protected] 3

 Degrees of Freedom
• Case – I
F=0
F=V–E
V=E
total number of
total number of
independent =
independent equation
process variable
• Then the system is exactly specified.
• No problem to find the solution of modeling equation.
2X + 3Y = 60
X + Y = 20
Y = 40 , X = -30

Nimesh Prabhakar [email protected] 4

 Degrees of Freedom
• Case – 2
F>0
F=V–E
V>E
total number of
total number of
independent
independent equation
process variable
• Then the system is Underspecified.
2X + 3Y + 4Z = 60
• How we can make this system exactly specified?
By Adding of F number of additional equation.
Nimesh Prabhakar [email protected] 5
 Degrees of Freedom
• Case – 3
F<0
F=V–E
V<E
total number of
total number of
independent <
independent equation
process variable
• Then the system is over specified.
2X + 3Y = 60
X = 2Y + 3
Y = 4X + 5
7X + 4Y = 45
• How we can make this system exactly specified?
By Removing F number of equation.

Nimesh Prabhakar [email protected] 6

 Degrees of Freedom

• Usually in practice Case – 2 is very common.

F>0
V>E
• How can make it exactly specified.
1. By specifying more number of disturbance
variables.(if we specified more number of
variable then unknown variable is reduce)
2. By adding more number of equation
Nimesh Prabhakar [email protected] 7
 Degrees of Freedom
For CSTH we get Two equation
• Total mass balance
V
= Fi – F
• Energy balance
F (Ti – T) Q
= i –
V V Cp
• How many Variable?
V, Fi, F, T, Ti, Q F, T
V=6
E=2
Degrees of freedom: F = V – E = 6 – 2
F=4
How to reduce Degree of freedom?

Nimesh Prabhakar [email protected] 8

 Degrees of Freedom
1. By specifying more number of
disturbance variables.
V
= Fi – F
F (Ti – T) Q
= i –
V V Cp

• What are the disturbance

variable in this equations? F, T
Fi & TI
• How we can specify these
variables?
By direct measurement
Degrees of freedom : F = 4 – 2 = 2

Nimesh Prabhakar [email protected] 9

 Degrees of Freedom
2. By adding more number of equation
V
= Fi – F
Fi(Ti – T) Q
= –
V V Cp
F = Fs + Kcf(hd - h)
F, T
Fs = Steady state value of F
Kcf = Tuning parameter
hd = Desired value
Q = Qs +KcQ(Td - T)
CV MV
Degrees of freedom: F = 2 – 2 = 0
h F
That means system is exactly specified.
T Q

Nimesh Prabhakar [email protected] 10

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Linearization of Systems
Deviation Variables
Dead Time
Self-regulating Process

Nimesh Prabhakar [email protected] 2

 Linearization of Systems

What is Linearization
Linearization is the process by which we
approximate nonlinear system with linear
once.

Nimesh Prabhakar [email protected] 3

 Linearization of Systems
Why we require linearization?
• Most physical system exhibit nonlinear
behavior, while characterization of dynamic
system by a transfer function can be done only
for linear systems characterized by linear
differential equations.
• It is important, then, that the nonlinear model
be linearized for control system analysis and
design purpose.

Nimesh Prabhakar [email protected] 4

 Linearization of Systems
• Consider the nonlinear differential equation,
𝑑𝑥
= f(𝑥) …(1)
𝑑𝑡
• Expand the nonlinear function f(𝑥) into Taylor series around
the point 𝑥 0
𝑑𝑓 𝑥 −𝑥0 𝑑𝑓2 (𝑥 −𝑥0)2 𝑑𝑓n (𝑥 −𝑥0)n
f(𝑥) = f(𝑥0)+ + 2 𝑥0 +…+ n 𝑥0
𝑑𝑥 𝑥0 1! 𝑑𝑥 2! 𝑑𝑥 𝑛!
…(2)
• If we neglect all terms of order two and higher, we take the
following approximation for the value of f(𝑥):
𝑑𝑓 𝑥 −𝑥0
f(𝑥) ≈ f(𝑥0)+ 𝑥0 …(3)
𝑑𝑥 1!

Nimesh Prabhakar [email protected] 5

 Linearization of Systems
𝑑𝑓 𝑥 −𝑥0
f(𝑥) ≈ f(𝑥0)+ …(3)
𝑥0 1!
𝑑𝑥
• It is well known that the error
introduced in the
approximation(3) is of the
same order of magnitude as
the term,
𝑑𝑓2 (𝑥 −𝑥0)2
I=
𝑑𝑥2 𝑥0 2!
• The linear approximation(3) is
satisfactory only when x is very
close to 𝑥0 ,where the value of
term I is very small.

Nimesh Prabhakar [email protected] 6

 Linearization of Systems
• In figure we can see the nonlinear
faction f(x) and its linear
approximation around 𝑥0
• The linear approximation depends
on the location of point 𝑥0 around
which we make the expansion
into a Taylor series.
• In eq(1), replace f(𝑥) by its linear
approximation given by eq(3)
𝑑𝑥 𝑑𝑓
= f(𝑥0) + (𝑥 − 𝑥0)
𝑑𝑡 𝑑𝑥 𝑥0
• This equation is the linearized
approximation of the equation(1).

Nimesh Prabhakar [email protected] 7

 Deviation Variables
• Consider the nonlinear differential equation,
𝑑𝑥
= f(𝑥) …(1)
𝑑𝑡
• Suppose that 𝑥s is the steady-state value of 𝑥.
𝑑𝑥s
= 0 = f(𝑥s) ..(2)
𝑑𝑡
𝑑𝑥 𝑑𝑓
= f(𝑥s) + −𝑥s) …(3)
(𝑥
𝑑𝑡 𝑑𝑥 𝑥s
• Now, subtracting eq(2) from eq(3) we get,
𝑑(𝑥−𝑥s) 𝑑𝑓
= −𝑥s) …(4)
𝑥s (𝑥
𝑑𝑡 𝑑𝑥
• If we define deviation variable X as
X = 𝑥 −𝑥s
Nimesh Prabhakar [email protected] 8
 Deviation Variables
• Then the equation (4) becomes
𝑑𝑋 𝑑𝑓
= X …(5)
𝑑𝑡 𝑑𝑥 𝑥s
• Equation (5) is the linearized approximation of the
nonlinear equation, expressed in terms of the
deviation variable X.

Nimesh Prabhakar [email protected] 9

 Dead Time
• Consider the flow of an
incompressible, nonreacting liquid
through a pipe.
• At t = 0, the temperature of the inlet
changes as shown by curve A in
figure b.
• It is clear that the temperature of the
outlet, Tout, will be remain the same
until the change reaches the end of
the pipe.
• Then we will observe the
temperature of the outlet changing,
as shown by curve B.

Nimesh Prabhakar [email protected] 10

 Dead Time
• We notice that the change of
outlet temperature follows the
same pattern as the change of the
inlet temperature with a delay of td
seconds.
• Td is the dead time and it is
expressed as,
𝑉𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑝𝑖𝑝𝑒 𝐴∙𝐿 𝐿
td = = =
𝑉𝑜𝑙𝑢𝑚𝑒𝑡𝑟𝑖𝑐 𝑓𝑙𝑜𝑤 𝑟𝑎𝑡𝑒 𝐴∙Uav
Uav
Where, Uav = average velocity of the
fluid
• Functionally, we can relate Tin and
Tout as follows:
Tout(t) = Tin(t – td)

Nimesh Prabhakar [email protected] 11

 Self-regulating Process
• A process is said to be self-regulated if a specific
value of the controlled variable is adopted for a
nominal load with no control action.
Example : Thermal System
• In a thermal system, there is a certain
temperature for a particular heat input.
• Due to disturbance, the system will acquire
another temperature, but it will not go out of
control.
Nimesh Prabhakar [email protected] 12
 Self-regulating Process
• A liquid level control system in which outflow is
through a pump is not a self-regulating process.
• If the inflow is equal to outflow, then the level
remain fixed.
• Since the outflow in this type of system does not
depend upon the head, the liquid is pumped out
at a constant rate.
• So, in case of an increased or decreased inflow,
the tank will over flow or go empty.
Nimesh Prabhakar [email protected] 13
L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Dynamics & Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 First-order System

Nimesh Prabhakar [email protected] 2

What is First-order System?
A First-order system is one whose output y(t) is
describe by the solution of First-order differential
equation.
𝑑𝑦
a1+ a0y = bf(t) …(1)
𝑑𝑡
Where f(t)=input(forcing function)
If a0≠0,
a1 𝑑𝑦 b
+ y = f(t)
a0 𝑑𝑡 a0
𝑑𝑦
𝜏p + y = kpf(t) …(2)
𝑑𝑡
a1
Where 𝜏p = = Time constant of the process
a0
kp = Steady-state gain or static gain or process gain
Nimesh Prabhakar [email protected] 3
 First-order System
𝑑𝑦
𝜏p + y = kpf(t) …(2)
𝑑𝑡
Taking laplace transform of eq(2),
𝜏psy(s) + y(s) = kpf(s)
y(s) 𝜏ps + 1 = kpf(s)
y(s) kp
=
f(s) 𝜏ps + 1
So we can write transfer function of first-order process as,
y(s) kp
G(s) = = …(3)
f(s) p 𝜏 s + 1
• A first-order process with a transfer function given in
eq(3) is also known as first-order lag, linear lag OR
exponential transfer lag.

Nimesh Prabhakar [email protected] 4

 First-order System
𝑑𝑦
a1 + a0y = bf(t) …(1)
𝑑𝑡
If a0=0
𝑑𝑦 b
= f(t)
𝑑𝑡 a1
𝑑𝑦
= k’pf(t)
𝑑𝑡
By taking laplace we get
sy(s) = k’pf(s)
y(s) k’p
=
f(s) s
y(s) k’p
G(s) = =
f(s) s
In this case the process is called purely capacitive OR pure
integrator.

Nimesh Prabhakar [email protected] 5

 Example of First-order System
A = uniform cross-sectional
area of tank
R = flow resistance
V = volume of tank = Ah
𝜌 = density of material
q0 = outlet flow rate
q = inlet flow rate
h = height of water level in tank
Relation between q0 & h can be written as,
ℎ
q0 = …(1)
𝑅
Aim: Determine the transfer function that relates height to flow.

Nimesh Prabhakar [email protected] 6

 Example of First-order System
Conservation equation
Rate of
Rate of mass rate of mass
– = accumulation
flow in flow out
of mass in tank
𝑑
𝜌q(t) - 𝜌q0(t) = (𝜌V)
𝑑𝑡
𝑑ℎ
q(t) - q0(t) = A …(2) h
𝑑𝑡 q0 = …(1)
ℎ 𝑑ℎ
q - = A …(3) Now we introduce R deviation
𝑅 𝑑𝑡 variable into the analysis
Initially, the process is operating
at steady state, which means that
𝑑ℎ
=0
𝑑𝑡
hs
qs - = 0…(3) Where subscript s indicate
𝑅 the steady state value of the variable
Subtracting eq(4) from eq(3)

Nimesh Prabhakar [email protected] 7

 h hs 𝑑h Taking laplace of eq(6) we get,
q – qs - + = A 1
R R 𝑑𝑡
Q(s) = H(s) + AsH(s)
h hs 𝑑h R
q – qs = - + A 1
R R 𝑑𝑡 Q(s) = H(s) + As
R
1 𝑑 H(s) 1
q – qs = (h-hs)+ A (h-hs) …(5) =
R 𝑑𝑡 Q(s) 1 + As
If we define deviation variable as R
H(s) 𝑅
Q = q – qs = …(8)
Q(s) (AR)s + 1
H = h-hs y(s) kp
G(s) = =
f(s) 𝜏ps + 1
1 𝑑H
Q= H+A …(6) 𝜏p = AR = Time constant
R 𝑑𝑡
Kp = R = Process gain

Nimesh Prabhakar [email protected] 8

• What is First-order System?
• Example of First-order System

Nimesh Prabhakar [email protected] 9

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Dynamics & Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Non-linear system

Nimesh Prabhakar [email protected] 2

 Steps for determining the transfer
function for a process
1) Write the appropriate balance equations (usually
mass OR energy balance)
2) Linearize terms if necessary.
3) Place balance equations in deviation variable form.
4) Laplace-transform the linear balance equations.
5) Solve the resulting transformed equations for the
transfer function (the output divided by input)

Nimesh Prabhakar [email protected] 3

What is Linear system?
• A system is said to be linear if all the function
in the equation are linear functions.
• A linear system should satisfy the
homogeneity and superposition properties.
Any system that
violates one of two
prosperities is said
to be nonlinear
system

Nimesh Prabhakar [email protected] 4

 First-order non Linear System
A = uniform cross-sectional
area of tank
R = flow resistance
V = volume of tank
𝜌 = density of material
q0 = outlet flow rate
q = inlet flow rate
h = height of water level in tank
Relation between q0 & h can be written as,
q0 = Ch1/2 …(0)(non-linear relation) q = ℎ
0 𝑅
Where C = Constant
Aim: Determine the transfer function that relates height to flow.
Nimesh Prabhakar [email protected] 5
 Conservation equation
Rate of
Rate of mass rate of mass
– = accumulation
flow in flow out
of mass in tank
𝑑
𝜌q(t) - 𝜌q0(t) = (𝜌V)
𝑑𝑡
𝑑h V = Ah
q(t) - q0(t) = A …(1)
𝑑𝑡
𝑑 h q0 = Ch1/2 …(0)
q - Ch1/2= A …(2)
𝑑𝑡
By Taylor series expansion, the function q0(h) may be expand around
the steady-state value hs , thus
q’’0(hs)(h−hs)2
q0 = q0(hs) + q’0(hs)(h-hs) + +…
2
Where q’0(hs) = First derivative of q0
q’’0(hs) = Second derivative of q0
If we keep only the linear term, the result is,
q0 ≈ q0(hs) + q’0(hs)(h-hs) …(3)

Nimesh Prabhakar [email protected] 6

 Taking the derivative of q0 , with At steady state the flow entering the
respect to h and evaluating the tank equals the flow leaving the tank
derivative at h = hs,
thus,
q0 = Ch1/2 …(0) qs = q0s …(6)
1
q’0(hs) = Chs−1/2 q0 ≈So
q0we(hs) +can
q’0(h s)(h-heq(5)
write s) …(3) as,
2 𝑑h h−hs
Putting this value in eq(3) we get A + = q –qs…(7)
1 𝑑𝑡 R1
q0 = q0s + R (h-hs)…(4) Introducing deviation variable
1
Where q0s = q0(hs) Q = q – qs
1 1 −1/2 H = h – hs in eq(7)
=
R1 2 s
Ch 𝑑H 𝐻
A + =Q
Substituting value of q0 from eq(4) 𝑑𝑡 𝑑h R1
into eq(1) we get q(t)
By- qtaking
0(t) = A laplace
…(1) we get,
𝑑𝑡
1 𝑑h H(s)
q - q0s - (h-hs) = A …(5) AsH(s) + = Q(s)
R1 𝑑𝑡 R1
1
H(s) 𝐴𝑆 + = Q(s)
R1
Nimesh Prabhakar [email protected] 7
H(s) 1
=
Q(s) 𝐴𝑆+ 1
R1
H(s) R1
=
Q(s) 𝐴R1𝑆+ 1
𝜏p = AR1
−1/2
2hs
Kp = R1 & R1 =
𝐶
• The resistance R1 is the reciprocal of the slope the
tangent line passing trough the point(q0s,hs)
• From the graph, it should be clear that the linear
approximation improve as the deviation in h
become smaller.

Nimesh Prabhakar [email protected] 8

First-order non Linear System

Nimesh Prabhakar [email protected] 9

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Dynamics & Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Second-order System

Nimesh Prabhakar [email protected] 2

What is Second-order System?
A Second-order system is one whose output y(t) is
describe by the solution of Second-order differential
equation.
d2𝑦 𝑑𝑦
a2 2 + a1 + a0y = bf(t) …(1)
𝑑t 𝑑𝑡
Where f(t)=input(forcing function)
If a0≠0,
a2 d2𝑦 a1 𝑑𝑦 b
2 + + y = f(t)
a0 𝑑t a0 𝑑𝑡 a0
2
d𝑦 𝑑𝑦
2
𝜏 2 +2ζ𝜏 + y = kpf(t) …(2)
𝑑t 𝑑𝑡
Where 𝜏 = natural period of oscillation of the system
ζ = damping factor
kp = Steady-state gain or static gain or process gain
Nimesh Prabhakar [email protected] 3
 Second-order System
d 2𝑦 𝑑𝑦
2
𝜏 2 +2ζ𝜏 + y = kpf(t) …(2)
𝑑t 𝑑𝑡
Taking laplace transform of eq(2),
𝜏2s2y(s) + 2ζ𝜏sy(s) + y(s) = kpf(s)
y(s) 𝜏2s2 + 2ζ𝜏s + 1 = kpf(s)
y(s) kp
= 22
f(s) 𝜏 s + 2ζ𝜏s + 1
So we can write transfer function of first-order
process as,
y(s) kp
G(s) = = …(3)
f(s) 𝜏2s2 + 2ζ𝜏s + 1

Nimesh Prabhakar [email protected] 4

 Example of Second-order System
U-tube manometer
L = length of the fluid column
in the manometer
D = Diameter of tube
The pressure on the both legs
of the manometer is initially
the same.
Aim: Determine the transfer
function between the applied
pressure difference Δ𝑃 and the
manometer reading h.

Nimesh Prabhakar [email protected] 5

 Example of Second-order System
Force balance/Newtown second law of motion
𝑠𝑢𝑚 𝑜𝑓 𝑓𝑜𝑟𝑐𝑒 𝑐𝑎𝑢𝑠𝑖𝑛𝑔 𝑅𝑎𝑡𝑒 𝑜𝑓 𝑐ℎ𝑎𝑛𝑔𝑒 𝑜𝑓
=
𝑓𝑙𝑢𝑖𝑑 𝑡𝑜 𝑚𝑜𝑣𝑒 𝑚𝑜𝑚𝑒𝑛𝑡𝑢𝑚 𝑜𝑓 𝑓𝑙𝑢𝑖𝑑
𝑠𝑢𝑚 𝑜𝑓 𝑓𝑜𝑟𝑐𝑒 𝑈𝑛𝑏𝑎𝑙𝑎𝑛𝑐𝑒𝑑 𝐹𝑟𝑖𝑐𝑡𝑖𝑜𝑛𝑎𝑙
𝑐𝑎𝑢𝑠𝑖𝑛𝑔 = 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 - 𝑓𝑜𝑟𝑐𝑒𝑠
𝑓𝑙𝑢𝑖𝑑 𝑡𝑜 𝑚𝑜𝑣𝑒 𝑓𝑜𝑟𝑐𝑒𝑠 𝑐𝑎𝑢𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛 𝑜𝑝𝑝𝑜𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛

Nimesh Prabhakar [email protected] 6

 𝑈𝑛𝑏𝑎𝑙𝑎𝑛𝑐𝑒𝑑
𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 = P1A - P2A - 𝜌ghA
𝑓𝑜𝑟𝑐𝑒𝑠 𝑐𝑎𝑢𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛
Net downward force
Net downward force
on Right-hand side
𝑈𝑛𝑏𝑎𝑙𝑎𝑛𝑐𝑒𝑑
on left-hand side 𝜋D 2 𝜋D 2
𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 =(P1 - P2) - 𝜌gh
𝑓𝑜𝑟𝑐𝑒𝑠 𝑐𝑎𝑢𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛 4 4
Weight of additional
mercury

Nimesh Prabhakar [email protected] 7

 𝐹𝑟𝑖𝑐𝑡𝑖𝑜𝑛𝑎𝑙
𝑆𝑘𝑖𝑛 𝑓𝑟𝑖𝑐𝑡𝑖𝑜𝑛
𝑓𝑜𝑟𝑐𝑒𝑠 =
𝑜𝑝𝑝𝑜𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛 𝑎𝑡 𝑤𝑎𝑙𝑙
𝑆ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 𝐴𝑟𝑒𝑎 𝑖𝑛 𝑐𝑜𝑛𝑡𝑎𝑐𝑡
=
𝑎𝑡 𝑤𝑎𝑙𝑙 𝑤𝑖𝑡ℎ 𝑤𝑎𝑙𝑙
𝑉 = average velocity of fluid in the =
8𝜇 Hagen-Posseuiller relationship
𝑉 (𝜋DL) for laminar flow
tube D
Which is also the velocity of the 8𝜇 1 dh
interface, = (𝜋DL)
1 dh D 2 dt
Which is equal to 2
dt
Nimesh Prabhakar [email protected] 8
𝑅𝑎𝑡𝑒 𝑜𝑓 𝑐ℎ𝑎𝑛𝑔𝑒 𝑜𝑓 𝑑
= (mass×velocity × momentum correction factor)
𝑚𝑜𝑚𝑒𝑛𝑡𝑢𝑚 𝑜𝑓 𝑓𝑙𝑢𝑖𝑑 𝑑𝑡
𝑑 𝜋D2
= 𝜌 L × 𝑉×𝛽
𝑑𝑡 4
𝑑 𝜋D2 1 dh
= 𝜌 ×L ×𝛽
𝑑𝑡 4 2 dt For laminar flow
𝜋D2 4 1 d2h 4
𝛽=
= 𝜌 L × × 3
4 3 2 𝑑t 2

Nimesh Prabhakar [email protected] 9

 𝑠𝑢𝑚 𝑜𝑓 𝑓𝑜𝑟𝑐𝑒 𝑐𝑎𝑢𝑠𝑖𝑛𝑔 𝑅𝑎𝑡𝑒 𝑜𝑓 𝑐ℎ𝑎𝑛𝑔𝑒 𝑜𝑓
=
𝑓𝑙𝑢𝑖𝑑 𝑡𝑜 𝑚𝑜𝑣𝑒 𝑚𝑜𝑚𝑒𝑛𝑡𝑢𝑚 𝑜𝑓 𝑓𝑙𝑢𝑖𝑑
𝑠𝑢𝑚 𝑜𝑓 𝑓𝑜𝑟𝑐𝑒 𝑈𝑛𝑏𝑎𝑙𝑎𝑛𝑐𝑒𝑑 𝐹𝑟𝑖𝑐𝑡𝑖𝑜𝑛𝑎𝑙
𝑐𝑎𝑢𝑠𝑖𝑛𝑔 = 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 - 𝑓𝑜𝑟𝑐𝑒𝑠
𝑓𝑙𝑢𝑖𝑑 𝑡𝑜 𝑚𝑜𝑣𝑒 𝑓𝑜𝑟𝑐𝑒𝑠 𝑐𝑎𝑢𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛 𝑜𝑝𝑝𝑜𝑠𝑖𝑛𝑔 𝑚𝑜𝑡𝑖𝑜𝑛
𝜋 D2 4 1 d2h 𝜋D2 𝜋D2 8𝜇 1 dh
𝜌 L × × =(P - P ) - 𝜌gh − (𝜋DL)
4 3 2 𝑑t 2 1 2 4 4 D 2 dt
𝜋 D2 4 1 d2h 8𝜇 1 dh 𝜋 D2 𝜋D2
𝜌 L × × + (𝜋DL) + 𝜌gh = (P1 - P2)
4 3 2 𝑑t 2 D 2 dt 4 4
𝜋D 2
Dividing both the side by 𝜌𝑔 , we get,
4
2L d2h 16𝜇 𝐿 dh (P1 − P2) ∆P
+ + h = =
3g 𝑑t2 𝜌D2g dt 𝜌g 𝜌g
Comparing with second order general equation
2 d2𝑦 𝑑𝑦
𝜏 2 +2ζ𝜏 + y = kpf(t)
𝑑t 𝑑𝑡
2L 2L
𝜏2 = so 𝜏 =
3g 3g
16𝜇 𝐿 8𝜇 3L
2ζ𝜏 = 2 so ζ = 2
𝜌D g 𝜌D 2g

Nimesh Prabhakar [email protected] 10

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Dynamic Behavior of
First Order System
Dynamic Behavior of
Second Order
System
Nimesh Prabhakar [email protected] 2
Nimesh Prabhakar [email protected] 3
Nimesh Prabhakar [email protected] 4
Nimesh Prabhakar [email protected] 5
Nimesh Prabhakar [email protected] 6
 Non-interacting system
Mass balance around tank 1:
𝑑ℎ
q – q1 = A1 1 …(1)
𝑑𝑡
Mass balance around tank 2:
𝑑ℎ
q1 - q2 = A2 2 …(2)
𝑑𝑡
The flow head relationship for the two
linear resistance are given by,
ℎ1
q1 = …(3)
𝑅1
ℎ2
q2 = …(4)
𝑅2
Now by introducing deviation variable Q1(s) = q1 – q1s
and steady state condition and taking
laplace transform we can write, H1(s) = h1 – h1s
H (s) Q2(s) = q2 – q2s
Q1(s) = 1 …(5)
𝑅1
H2(s) H2(s) = h2 – h2s
Q2(s) = …(6)
𝑅2

Nimesh Prabhakar [email protected] 7

 Non-interacting system
• At steady state equation(1) • By putting the value of H1(s)
become
from eq(5) we get
qs – q1s = 0 …(7)
Q(s) – Q1(s) =A1s Q1(s)R1
• Now subtracting eq(1) from
eq(7) we get, Q(s) = Q1(s)[A1R1s + 1]
Q1(s) 1
𝑑(ℎ −ℎ ) =1(s)
H
q - qs – (q1 - q1s) = A1 1 s Q(s)
Q1(s) = (A1R…(5)
1)s + 1
𝑑𝑡
𝑑ℎ1 𝑑H1 Q1(s) 𝑅1 1
q – q1 = A11 = A1…(1)
Q – Q …(8) = …(8)
𝑑𝑡 𝑑𝑡 Q(s) 𝜏1s + 1
• Same as for Tank-2 Where 𝜏1 = A1R1
𝑑H2 Same we can write for tank 2
Q 1 - Q 2 = A2 …(9)
𝑑𝑡 H2(s) R2
• By taking the laplace = …(9)
Q1(s) 𝜏2s + 1
transform of eq(8) we get, Where 𝜏2 = A2R2
Q(s) – Q1(s) =A1sH1(s) …(10)

Nimesh Prabhakar [email protected] 8

 Non-interacting system
• Now multiplying eq(8) and eq(9),
Q1(s) H2(s) 1 R2
× = ×
Q(s) Q1(s) (𝜏1s + 1) 𝜏2s + 1
H2(s) R2
= …(10)
Q(s) (𝜏1s + 1)(𝜏2s + 1)
• The overall transfer function of eq(10) is the
product of two first-order transfer function.

Nimesh Prabhakar [email protected] 9

 Interacting system
A1 &A2 = uniform cross-sectional
area of tank - 1&2
R1&R2 = flow resistance of
tank - 1&2
q = inlet flow rate of tank 1
q1 = outlet flow rate of tank 1
q1 = inlet flow rate of tank 1
q2 = outlet flow rate of tank 2
h1&h2 = height of water level in tank 1&2
• Here flow through R1 now depends on the difference between
h1 and h2 , hence variation in h2 affect variation in h1.
Aim: Determine the transfer function that relates h2 to q, that is,
H2(s)/Q(s).
Nimesh Prabhakar [email protected] 10
 Interacting system
Mass balance around tank 1:
𝑑ℎ
q – q1 = A1 1 …(1)
𝑑𝑡
Mass balance around tank 2:
𝑑ℎ
q1 - q2 = A2 2 …(2)
𝑑𝑡
The flow head relationship 𝑅1 is,
1
q1 = (ℎ - ℎ2) …(3)
𝑅1 1
The flow head relationship 𝑅2 is,
ℎ
q2 = 2 …(4)
𝑅2 Q1(s) = q1 – q1s
H1(s)−H2(s) H1(s) = h1 – h1s
Q1(s) = …(5)
𝑅1
H (s) Q2(s) = q2 – q2s
Q2(s) = 2 …(6) H2(s) = h2 – h2s
𝑅2

Nimesh Prabhakar [email protected] 11

 Interacting system
At steady state equation(1) become
qs – q1s = 0 …(7)
Now subtracting eq(1) from eq(7) we get,
𝑑(ℎ1−ℎs)
q - qs – (q1 - q1s) = A1
𝑑𝑡
𝑑H1
Q – Q1 = A1 …(8)
𝑑𝑡
Same as for tank -2
𝑑H2
Q1 - Q2 = A2 …(9)
𝑑𝑡
By taking the laplace transform of eq(8) and eq(9)we get,
Q(s) – Q1(s) =A1sH1(s) …(10)
Q1(s) – Q2(s) =A2sH2(s) …(11)

Nimesh Prabhakar [email protected] 12

 Interacting system
• Now we have four equation,
Tank 1 Q(s) – Q1(s) =A1sH1(s) …(10)
Tank 2 Q1(s) – Q2(s) =A2sH2(s) …(11)
Valve-1 𝑅1Q1(s) = H1(s) − H2(s) …(5)
Valve-2 𝑅2Q2(s) = H2(s) …(6)
• These equation may be combined to eliminate
Q1,Q2 and H1 to arrive the desired transfer
faction:
H2(s) R2
=
Q(s) 𝜏1𝜏2s2+(𝜏1 + 𝜏2 + A1R2)s + 1

Nimesh Prabhakar [email protected] 13

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Dynamic Behavior of
First Order System
Dynamic Behavior of
Second Order
System
Nimesh Prabhakar [email protected] 2
 Dynamic Behavior of First Order
System
𝑑𝑦
𝜏p + y = kpf(t) …(1)
𝑑𝑡
𝜏p = Time constant of the process
kp = Steady-state gain or static gain or process gain
• Transfer function of first-order process is,
y(s) kp
G(s) = = …(2)
f(s) 𝜏ps + 1
• Now we take Step signal as a Input.
f(t) = 0 t<0
A t≥0
A
f(s) =
s
• Now Substituting the value of f(s) in eq(1)
Akp
y(s) = …(3)
s(𝜏ps + 1)
• Now to take inverse Laplace transform of eq(3) we can write

Nimesh Prabhakar [email protected] 3

 Dynamic Behavior of First Order
System
𝑎 𝑏 • Now substituting value of s in
y(s) = Akp
+ … (4) eq(5) we get
𝜏ps + 1
𝑠
𝑎(𝜏ps + 1)+𝑏𝑠 b = - 𝜏p
y(s) = Akp • When we substitute s = 0 in
s(𝜏ps + 1)
• By Substituting the value of y(s) from eq(5) we get
eq(3) a =1
• Now substituting value of a & b
1 𝑎(𝜏ps + 1)+𝑏𝑠
Akp = Akp in eq(4) we get,
s(𝜏ps + 1) s(𝜏ps + 1) 1 𝜏p
𝑎(𝜏ps + 1) + 𝑏𝑠 = 1 …(5) y(s) = Ak p 𝑠 −
𝜏ps + 1
1 1
• Now to solve this equation(to get the y(s) = Akp − …(6)
𝑠 s + 1/𝜏p
value of 𝑎 & b) consider, • Now taking inverse laplace
1
S=- transform of eq(6) we get,
𝜏p
y(t) = Akp (1 - 𝑒 −t/𝜏p )

Nimesh Prabhakar [email protected] 4

 Dynamic Behavior of First Order
System
y(t) = Akp (1 - 𝑒 t/𝜏p )
When t = 0
y(t) = 0
When t = ∞
y(t) = Akp
When t =𝜏p
y(𝑡) = 0.632 Akp
• The transfer function y(t) reaches
63.2% of its ultimate value when t = 𝜏p. t y
• So time constant can be define as the 0 0
time require the response y(t) to reach
63.2% of its ultimate value. 𝜏p 0.632 Akp
• Speed of response is related of 𝜏p . 2𝜏p 0.865 Akp
Smaller the 𝜏p faster will be response. 3𝜏p 0.95 Akp
4𝜏p 0.98 Akp
5𝜏p 0.99 Akp
Nimesh Prabhakar [email protected] 5
 Dynamic Behavior of Second Order
System
d 2𝑦 𝑑𝑦
2
𝜏 2 +2ζ𝜏 + y = kpf(t) …(1)
𝑑t 𝑑𝑡
we can write transfer function of Second-order
process as,
y(s) kp
G(s) = = 22 …(2)
f(s) 𝜏 s + 2ζ𝜏s + 1
Where 𝜏 = natural period of oscillation of the system
ζ = damping factor
kp = Steady-state gain or static gain or process gain
Nimesh Prabhakar [email protected] 6
Dynamic Behavior of Second Order System
Case 1: ζ = 1 Critically damped
Case 2: ζ > 1 Overdamped
Case 3: ζ <1 Underdamped
• In this case, the output is a
damped sinusoid.
• The underdamped response
is initially faster than the
critically damped and
overdamped responses,
which are characterized as
sluggish.

Nimesh Prabhakar [email protected] 7

Dynamic Behavior of Second Order System
• Although the
underdamped response
is initially faster and
reaches its ultimate
value quickly, it does not
stay there; rather it
starts oscillating with
progressively decreasing
amplitude.
• Increase in ζ reduces
oscillations but also
slows down the
response.

Nimesh Prabhakar [email protected] 8

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Multicapacity
Process
Dynamic Systems
with dead time
Nimesh Prabhakar [email protected] 2
 Multicapacity Process

• N Capacities in series
• It can be classified in two parts
1. N non interacting capacities
2. N Interacting capacities

Nimesh Prabhakar [email protected] 3

 Multicapacity Process
N non interacting capacities

• Overall transfer function can be written as,

Yn(s)
= G1(s)G2(s)G3(s) …. … Gn(s)
f1(s)
Yn(s) k1 k2 k3 k𝑛
= × × ×…. ×
f1(s) 𝜏1s + 1 𝜏2s + 1 𝜏3s + 1 𝜏ns + 1
Yn(s) k1k2k3 ….k𝑛
=
f1(s) (𝜏1s + 1) (𝜏2s + 1) (𝜏3s + 1)… (𝜏ns + 1)
• For interacting capacities the overall transfer function is more
complex.
Nimesh Prabhakar [email protected] 4
 Multicapacity Process

N Noninteracting system
• The response has the characteristics of an overdamped system; that is,
it is S shaped and sluggish.
• Increasing the number of capacities in series increase the sluggishness
of the response.
N Interacting system
• Interaction increase the sluggishness of the overall response.

Nimesh Prabhakar [email protected] 5

 Dynamic Systems with dead time
• Whenever a change
took place in the input
variable, its effect was
not instantaneously
observed in the
behavior of the output
variable.
• Virtually all physical
will involve some time
delay between the
input and the output.

Nimesh Prabhakar [email protected] 6

Dynamic Systems with dead time

• Consider a first-order system with a dead time td between the

input f(t) and output y(t).
• For first order system we have following transfer function
L[y(t)] Y(s) kp
= =
L[f(t)] f(s) 𝜏ps + 1
• For the dead time we have following transfer function
L[y(t−td)]
= 𝑒 −tds
L[y(t)]
Nimesh Prabhakar [email protected] 7
Dynamic Systems with dead time

• So, the transfer function between the input f(t) and the
delayed output y(t−td) is given by,
L[y(t−td)] kp𝑒 −tds
= …(1)
L[f(t)] 𝜏p s + 1
• This is called First Order Plus Dead Time(FOPTD) system.

Nimesh Prabhakar [email protected] 8

 Dynamic Systems with dead time
• Similarly, the transfer
function for second –
order system with delay
is given by,
L[y(t−td)] kp𝑒 −tds
= 22
L[f(t)] 𝜏 s + 2ζ𝜏s + 1
• This is called Second
Order Plus Dead
Time(SOPTD) system.

Nimesh Prabhakar [email protected] 9

L. J. Institute of Engineering & Technology
Department of Instrumentation & Control

Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Inverse Response

Nimesh Prabhakar [email protected] 2

 Inverse Response
• When you push down on your car’s
accelerator, you expect the car to speed up,
right?
What if it slows down?
• Or even worse: You lift your foot off the
accelerator and your car speeds up.
• And the more you lift your foot, the more
the car speeds up.
• These are almost unthinkable and certainly
scary situations, yet they occur every day in
thousands of boilers and some other
processes around the world.
• The phenomenon is called an inverse
response.
• One of the most common occurrences of
inverse response is found in the control of
boiler drum level.

Nimesh Prabhakar [email protected] 3

 Inverse Response
• Consider the simple drum
boiler shown in figure.
• If the flow rate of the cold
feed water is increased by
a step, the total volume of
the boiling water and
consequently the liquid
level will be decreased for
short period and then it
will start increasing, as
shown by response in
figure.

Nimesh Prabhakar [email protected] 4

 Inverse Response
• Net result of two opposing
effects can be examined as
follows:
• The cold feedwater causes a
temperature drop which
decreases the volume of the
entrained vapor bubbles.
• This leads to decrease of the
liquid level of the boiling
water, following first-order
behavior(Curve 1), that is,
−k1
.
𝜏1s + 1

Nimesh Prabhakar [email protected] 5

 Inverse Response
• With constant heat supply,
the steam production
remains constant and
consequently the liquid level
of the bowling water will
start increasing in an integral
from, leading to a pure
k2
capacitive response,
s
• The overall transfer function
can be written as,
y(s) k2 k1
= -
f(s) s 𝜏1s + 1
y(s) (k2 𝜏1 − k1)s + k2
=
f(s) s(𝜏1s + 1)
Nimesh Prabhakar [email protected] 6
 Inverse Response
y(s) (k2 𝜏1 − k1)s + k2
=
f(s) s(𝜏1s + 1)
For
k2 𝜏1 < k1
• The second term –k1/ (𝜏1s + 1)
dominates initially and we take the
inverse response.
• If the condition above is not satisfied,
we do not have inverse response.
Note: when k2 𝜏1 < k1, then form eq(1)
we notice that the transfer function has
a positive zero, at the point
k2
S= >0
(k2 𝜏1 − k1)

Nimesh Prabhakar [email protected] 7

 Inverse Response of Two Opposing
First-Order System
• The figure shows two
opposing effects result
from two different first-
order process.
• The overall transfer
function can be written as,
y(s) k1 k2
= -
f(s) 𝜏1s + 1 𝜏2s + 1
y(s) (k1 𝜏2 − k2𝜏1)s + (k1−k2)
=
f(s) (𝜏1s + 1)(𝜏2s + 1)
Nimesh Prabhakar [email protected] 8
 Inverse Response of Two Opposing First-
Order System
• Initially(at t = 0+) process 2, which
reacts faster than process 1(k1 𝜏2 >
k2𝜏1), dominates the response of the
overall system, but
• Ultimately process 1 reaches a higher
steady-state value than process
2(K1>K2) and forces the response of
the overall system in the opposite
direction.
• Note: when (𝜏1/ 𝜏2) > (k1/ k2) > 1 the
process exhibits inverse response,
and we find that the system’s transfer
function has a positive zero.
k − k2
Z= 1 >0
k1𝜏2 + k2𝜏1

Nimesh Prabhakar [email protected] 9

 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Controller Modes
Process Lag
Error
Control Lag
 What is Controller
Modes
Controller modes refer to the
methods to generate different types
of control signals to final control
element to control the process
variable.
 Classifications of different
controller modes
Discontinuous Controller Modes
(a) Two-position (ON/OFF) Mode
(b) Multi-position Mode
(c) Floating Control Mode: Single Speed and Multiple Speed
Continuous Controller Modes
(a) Proportional Control Mode
(b) Integral Control Mode
(c) Derivative Control Mode
Composite Controller Modes
(a) Proportional-Integral Control (PI Mode)
(b) Proportional-Derivative Control (PD Mode)
(c) Proportional-Integral-Derivative Control (PID or Three
Mode Control)
 Classifications of different
controller modes
Based on the controller action on the control
element, there are two modes:
Direct Action
• If the controller output increases with increase in
controlled variable then it is called direct action.
Example
In level control system, if the level rises(increases),
the valve should be open(drive signal should be
increased)
 Classifications of different
controller modes
Reverse Action
• If the controller output decreases with increase in
controlled variable then it is called reverse action
Example
In the simple temperature control from a heater, if
the temperature increases, the drive to the heater
should be decreased.
 Controller Modes
• The choice operating mode for any given process
control system is complicated decision.
• It involves not only process characteristics but cost
analysis, product rate, and other industrial factors.
• The process control technologist should have good
understanding of the operational mechanism of each
mode and its advantages and disadvantages.
• In general, the controller operation for the error ep is
expressed as a relation:
p = F(ep)
• where F(ep) represents the relation by which the
appropriate controller output is determined.
 Process Lag
• A process load change or transient causes a
change in the controlled variable.
• The process-control loop responds to ensure that,
some finite time later, the variable return to the
set point value.
• Part of this time is consumed by the process itself
and is called the process lag.
 Process Lag
• Assume, the Inlet flow is
suddenly doubled
• A large process load
changes radically (reduces)
the liquid temperature.
• The control loop response
by opening the steam inlet
valve to allow more steam
and heat input to bring the
liquid temperature back to
the set point.
• The physical opening of
the control valve is slowest
part of the loop.
 Error
• Error is the difference between the measured
variable and the setpoint.
• Error can be either positive or negative.
• The objective of any control scheme is to minimize
or eliminate error.
• The deviation or error of dynamic variable from set
point is given by:
e=r–b
e = error
b = measured indication of variable
r = setpoint of variable(reference)
 Error
• To describe controller operation in a general
way, it is better to express the error as percent
of the measured variable range (i.e., the span).
• The measured value of a variable can be
expressed as percent of span over a range of
measurement by the equation,
c −cmin
cp = × 100
cmax−cmin
cp = measured value as percent of measurement
range
c = actual measured value
cmax = maximum of measured value
cmin = minimum of measured value
 Error
• We can express the error as percent of span,
r−𝑏
ep = × 100
bmax− bmin
ep = error expressed as percent of span
bmax = maximum value of measured indication of
variable
bmin = minimum value of measured indication of
variable
r = setpoint of variable(reference)
 Example
• The temperature in Figure
has a range of 300 to 440
K and a setpoint of 384 K.
Find the percent of span
error when the
temperature is 379 K.
• Here, r = 384, b = 379,
bmax = 440, bmin = 300
r−b
ep = × 100
bmax−bmin
384− 379
ep = × 100
440 − 300
ep = 3.6%
 Variable range
• Controlled variable has a range of values within
which control is required to be maintained at
set point.
• This range can be expressed as the minimum
and maximum values of the dynamic variable.
Example
• if a standard signal 4-20 mA transmission is
employed, then 4 mA represents the minimum
value of the variable and 20 mA the maximum
value.
 Control parameter (output)
range
• It is the possible range of values of final control element.
• The controller output is expressed as a percentage where
minimum controller output is 0% and maximum
controller is 100%.
• But 0% controller output does not mean zero output. For
example, it is necessary requirement of the system that a
steam flow corresponds to 1/4th opening of valve.
• The controller parameter output as a percent of full scale
when the output changes within the specified limits is
expressed as:
u −umin
p= × 100
umax −umin
p = controller outout as percent of full scale
u = value of output
umax= maximum value of controlling parameter
umin = minimum value of controlling parameter
 Example
• A controller outputs a 4- to 20-mA signal to control motor speed
from 140 to 600 rpm with a linear dependence. Calculate
(a) current corresponding to 310 rpm, and
(b) the value of (a) expressed as the percent of control output.
Solution
• We find the slope m and intersect S0 of the linear relation
between current I and speed S, where
Sp = mI + S0
• Knowing Sp and I at the two given positions, we write two
equations:
140 = 4m + S0
600 = 20m + S0
• Solving these simultaneous equations, we get,
m = 28.75 rpm/mA and S0 = 25 rpm
• Thus, at 310 rpm we have, 310 = 28.75I +25
• which gives I = 9.91 mA.
 Example
• Expressed as a percentage of the 4- to 20-mA
range, this controller output is,
u −umin
p= × 100
umax −umin
9.91 − 4
p= × 100
20 − 4
p = 36.9%
 Control Lag
• Control Lag refers to the
time for the process control
loop to make necessary
adjustments to the final
control element.
Example
• if a sudden change in liquid
temperature occurs, it
requires some finite time for
the control system to
physically actuate the steam
control valve.
Controller Modes
Process Lag
Error
Control Lag
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Two-Position Mode
 Two-Position Mode
• It is also known as ON/OFF Controller.
• The most general form can be given by:
p = 0% ep < 0
100% ep > 0
• This relation shows that when the measured
value is less than the setpoint, full controller
output results.
• When it is more than the setpoint, the
controller output is zero.
 Two-Position Mode
Example
Space Heater
• If the temperature drops below a setpoint, the
heater is turned ON.
• If the temperature rises above the setpoint, it
turns OFF.
 Two-Position Mode
Neutral Zone
• In practical implementation of the
two-position controller, there is an
overlap as ep increases through
zero or decreases through zero.
• In this span, no change in the
controller output occurs which is
illustrated in figure.
• It can be observed that, until an
increasing error changes by ∆ep
above zero, the controller output
will not change state.
• In decreasing, it must fall ∆ep
below zero before the controller
changes to 0%.
• The range 2∆ep is referred to as
neutral zone or differential gap.
 Applications
Generally the two-position control mode is best
adapted to:
• Large-scale systems with relatively slow process
rates
Examples
1. Room heating systems
2. Air-conditioning systems
• Systems in which large-scale changes are not
common
Examples
1. Liquid bath temperature control,
2. Level control in large-volume tanks.
 Example
• A liquid-level control system linearly converts
a displacement of 2 to 3 m into a 4 to 20 mA
control signal. A relay serves as the two-
position controller to open and close the inlet
valve. The relay closes at 12 mA and opens at
10 mA. Find
a) Relation between displacement level and
current
b) Neutral zone or displacement gap in meters.
Given data: Liquid- level range = 2 to 3 m
Hmin = 2m & Hmax = 3m
Control signal range = 4 to 20 mA
Imin = 4mA & Imax = 20mA
 Example
(a) The linear relationship between level and current
is given by
H = K I + H0
The simultaneous equations for the above range are:
For low range signal
2 m = K(4 mA) + H0
For higher range signal
3 m = K(20 mA) + H0
• Solving the above equations we get:
K = 0.0625 m/mA, & H0 = 1.75 m
Therefore, the relation between displacement level
(H) and current (I) is given by
H = 0.0625 I + 1.75
 Example
(b) The relay closes at 12 mA, which is high
level, HH
HH = 0.0625 x 12 + 1.75
HH = 2.5 m
• The relay opens at 10 mA, which is low level,
HL
HL = 0.0625 x 12 + 1.75
HL = 2.375 m
• Therefore, the neutral zone is,
NZ = (HH - HL) = (2.5 - 2.375) = 0.125 m
 Example
• The temperature of water in a tank is controlled by
two-position controller. When the heater is off the
temperature drops at 2 K per minute. When the
heater is on the temperature rises at 4 K per minute.
The setpoint is 323 K and the neutral zone is ±4% of
the setpoint. There is a 0.5 min lag at both the on
and off switch points. Find the period of the
oscillation and plot the water temperature versus
time.
Given data:
• Temperature drops = 2 K/min
• Temperature rises = 4 K/min
• Control Lag = 0.5 min
• Neutral zone = ± 4%
• Setpoint = 323 K
 Example
The natural zone in temperature
is
∆T = ± 4% of (323 K) = 13 K
• Therefore, the temperature
will vary from 310 to 336 K
(without considering the lag)
• Initially we start at setpoint
value. The temperature will
drop linearly, which can be
expressed by
T1(t) = T(ts) – 2 (t – ts)
where ts = Time at which we
start the observation
T(ts) = Temperature when we
start observation i.e. 323.
 Example
• The temperature will drop till
- 4% of setpoint (323K),
which is 310 K.
• Time taken by the system to
drop temperature value 310
K is
310 = 323 – 2 (t – 0)
t = 6.5 min
• Undershoot due to control
lag = (control lag) x (drop
rate)
= 0.5 min x 2 K/min = 1K
Due control lag temperature
will reach 309 instead of 310 K.
 Example
• From this point the temperature
will rise at 4 K/min linearly till
+4% of set point i.e.336K
which can be expressed by
T2(t) = T(th) + 2 (t – th)
where th = time at which heater
goes on
T(th) = temperature at which heater
goes on
336 = (310-1) + 4 [t – (6.5 +0.5)]
t = 13.75 min
• Overshoot due to control lag =
(control lag) x (rise rate) = 0.5
min x 4 K/min = 2K
Due control lag temperature will
reach 338 instead of 336 K.
 Example

• The oscillation period is

= 13.75 + 0.5 +0.5 + 6.5
= 21.25 ≈ 21.5 min
Two-Position Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Multiposition Control
Mode
 Multiposition Control Mode
What is Multiposition
Control Mode
• It is the logical extension of two-
position control mode to provide
several intermediate settings of the
controller output.
 Multiposition Control Mode
• This discontinuous control mode is used in an
attempt to reduce the cycling behavior and
overshoot and undershoot inherent in the two-
position mode.
• This control mode can be preferred whenever
the performance of two-position control mode
is not satisfactory.
• The general form of multiposition mode is
represented by
p = pi ep > ei i = 1, 2, ... , n
• As the error exceeds certain set limits ± ei , the
controller output is adjusted to present values
pi.
 Three-position Control Mode
• One of the best example for
multiposition control mode is three-
position control mode, which can be
expressed as,
p = 100% ep>-e1
50% -e1< ep >+e1
0% ep<-e1
• As long as the error is between +e1
and -e1 of the set point, the
controller stays at some nominal
setting indicated by a controller
output of 50%.
• If the error exceeds the set point by
+e1 or more, then the output is
increased to 100%.
• If it is less than the set point by -e1
or more, the controller output is
reduced to zero.
 Three-position Control Mode

Limitation
• The three-position control mode usually
requires a more complicated final control
element, because it must have more than
two settings.
 Three-position Control Mode

• Figure shows the relationship between the error and

controller output for a three-position control.
• The finite time required for final control element to change
from one position to another is also shown.
• The graph shows the overshoot and undershoots of error
around the upper and lower setpoints.
• This is due to both the process lag time and controller lag
time
Multiposition Control
Mode
Three-position Control
Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Floating Control Mode
 Floating Control Mode
• In floating control, the specific output of
the controller is not uniquely determined
by error.
• If the error is zero, the output does not
change but remains (floats) at whatever
setting it was when error went to zero.
• When error moves of zero, the controller
output again begins to change.
• Similar to two-position mode, there will be
a neutral zone around zero error where no
change in controller output occurs.
 Types of Floating Control Mode
1. Single Speed
2. Multiple Speed
 Single Speed
• In this mode, the output of the control element changes
at a fixed rate when the error exceeds the neutral zone.
• The equation for single speed floating mode is:
𝑑𝑝
= ±kF ep > ∆ep …(1)
𝑑𝑡
𝑑𝑝
= rate of change of controller output with time
𝑑𝑡
kF = rate constant
∆ep = half the neutral zone
• If the equation (1) is integrated for actual controller
output, we get
p = ±kF t + p(0) ep > ∆epk
where p(0) = controller output at t = 0
• The equation shows that the present output depends on
the time history of errors that have previously occurred.
 Single Speed
• The single- speed controller
action as output rate of
change to input error is
shown in figure a.
• The graph in figure-b shows a
reverse acting controller,
which means the controller
output decreases when error
exceeds neutral zone, which
corresponds to negative kF in
equation (1).
 Single Speed
• The graph shows that the
controller starts at some output
p(0).
• At time t1 , the error exceeds the
neutral zone, and the controller
output decreases at a constant
rate until t2 , when the error
again falls below the neutral
zone limit.
• At t3 , the error falls below the
lower limit of neutral zone,
causing controller output to
change until the error again
moves within the allowable
band..
 Multiple Speed
• In this mode several possible speeds (rate) are
changed by controller output.
• Usually, the rate increases as the deviation
exceeds certain limits.
• For speed change point epi error there will be
corresponding output rate change Ki.
• The expression can be given by,
𝑑𝑝
= ±kFi ep > ∆epi …(2)
𝑑𝑡
 Multiple Speed

• If the error exceeds epi, then the speed is KFi.

• If the error rises to exceed ep2, the speed is
increased to KF2, and so on.
• The graph of multiple-speed mode is shown in
figure.
 Floating Control Mode
Applications
• Primary applications are in single-speed
controllers with neutral zone
• This mode is well suited to self-regulation
processes with very small lag or dead time,
which implies small capacity processes. When
used for large capacity systems, cycling must
be considered.
Floating Control Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Proportional Control
Mode
 Limitation of two-position & multi-
step modes
• The two-position mode had the controller
output of either 100% or 0%, depending on the
error being greater or less than the neutral
zone.
• In multiple-step modes, more divisions of
controller outputs versus error are developed.
Solution
• The natural extension of this concept is the
proportional mode, where a smooth, linear
relationship exists between the controller
output and the error
 Proportional Control Mode
• In this mode a linear relationship exits between the
controller output and the error.
• For some range of errors about the setpoint, each
value of error has unique value of controller output
in one-to-one correspondence.
• The range of error to cover the 0% to 100%
controller output is called proportional band,
because the one-to-one correspondence exits only
for errors in this range.
• The analytical expression for this mode is given by:
p = kpep + p0 …(1)
kp = proportional gain between error and controller
output
p0 = controller output with no error (%)
 Proportional Control Mode
Direct and Reverse Action
• The equation (1) represents reverse action,
because the term Kpep will be subtracted from p0
whenever the measured value increases the above
setpoint which leads negative error.
• The equation for the direct action can be given by
putting the negative sign in front of correction term
i.e. - Kpep.
 Proportional Control Mode
• A plot of the proportional
mode output vs. error for
equation (1) is shown in
Figure.
• In the figure, p0 has been
set to 50% and two
different gains have been
used.
• It can be observed that
proportional band is
dependent on the gain.
 Proportional Control Mode
• A high gain (G1) leads to
large or fast response,
but narrow band of
errors.
• On the other side a low
gain (G2) leads to small
or slow response, but
wide band of errors.
• In general, the
proportional l band is
defined by the equation:
100
PB =
kp
 summary of proportional control
mode
1. If error is zero, output is constant and equal to
p0.
2. If there is error, for every 1% error, a correction
of Kp percent is added or subtracted from p0,
depending on sign of error.
3. There is a band of errors about zero magnitude
PB within which the output is not saturated at
0% or 100%.
 Offset error
What is Offset?
• Proportional control alone is unable to
eliminate the offset error, which is the
difference between the desired value and the
actual value, SP − PV error, as it requires an
error to generate an output.
• An important characteristic of the proportional
control mode is that it produces a permanent
residual error in the operating point of the
controlled variable when a load change occurs
and is referred to as offset.
• It can be minimized by larger constant Kp which
also reduces the proportional band.
 Proportional Control Mode
Advantages
• Easy to implement
• low cost
• Easy to tune the proportional constant
Disadvantages / Limits
• Responds only change in error.
• Error can not be reduced to zero (i.e. controlled
value cannot reach Set point)
• Fine controlling is not possible.
 Applications
• Whenever there is one-to-one correspondence of
controller output is required with respect to error
change proportional mode will be ideal choice.
• The offset error limits the use of proportional mode,
but it can be used effectively wherever it is possible to
eliminate the offset by resetting the operating point.
• Proportional control is generally used in processes
where large load changes are unlikely or with moderate
to small process lag times.
• If the process lag time is small, the PB can be made
very small with large Kp, which reduces offset error.
• If Kp is made very large, the PB becomes very small,
and proportional controller is going to work as an
ON/OFF mode, i.e. high gain in proportional mode
causes oscillations of the error.
 Example
• Consider the proportional-mode level-
control system of figure. Value A is
linear, with a flow scale factor of 10
m3/h per percent controller output.
The controller output is nominally 50%
with a constant kp = 10% per %. A load
change occurs when flow through
valve B changes from 500 m3/h to 600
m3/h. Calculate the new controller
output and offset error.
 Certainly, valve A must move to a new
position of 600 m3/h flow or the tank
will empty.
 This can be accomplished by a 60%
new controller output because
10 m3/h
QA = (60%) = 600 m3/h
%
 Example
• This is a proportional controller, we have
p = kpep + p0
with the nominal condition p0=50%. Thus
p−p0 60 −50
ep = = %
kp 10
ep = 1%
• so a 1% offset error occurred because of the
load change.
 Example
• For a proportional controller, the controlled variable
is a process temperature with a range of 50 to
1300C and a setpoint of 73.50C. Under nominal
conditions, the setpoint is maintained with an
output of 50%. Find the proportional offset
resulting from a load change that requires a 55%
output if the proportional gain is (a) 0.1 (b) 0.7 (c)
2.0 and (d) 5.0.
• Given data:
Temperature Range = 50 to 1300C
Setpoint (Sp) = 73.50C
P0 = 50%
P = 55%
ep = ?
Offset error = ? for Kp =0.1, 0.7, 2.0 & 5.0
 Example
• For proportional controller:
P = Kpep + P0
p− p0 55 – 50 5
ep = = = %
Kp Kp Kp
(a) when Kp = 0.1 Offset error, ep = 5/0.1 = 50%
(b) when Kp = 0.7 Offset error, ep = 5/0.7 = 7.1%
(c) when Kp = 2.0 Offset error, ep = 5/2.0 = 2.5%
(d) when Kp = 5.0 Offset error, ep = 5/5.0 = 1%
[It can be observed from the results that as
proportional gain Kp increases the offset error
decreases.]
Proportional Control
Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Integral Control Mode
 Integral Control Mode
Advantage
• The integral control eliminates the offset
error problem by allowing the controller
to adapt to changing external conditions
by changing the zero-error output.
 Integral Control Mode
Working
• Integral action is provided by summing the error over
time, multiplying that sum by a gain, and adding the
result to the present controller output.
• If the error makes random excursions above and below
zero, the net sum will be zero, so the integral action will
not contribute.
• But if the error becomes positive or negative for an
extended period of time, the integral action will begin to
accumulate and make changes to the controller output.
• The analytical expression for integral mode is given by
the equation
𝑡
p(t) = KI 0 ep 𝑑𝑡 + p(0) …(1)
p(0) = controller output when the integral action starts
(%)
KI = Integral gain
 Integral Control Mode
• Another way of expressing the integral action is by
taking derivative of equation (1), which gives the
relation for the rate of change of controller output
with error.
𝑑𝑝
= KI ep …(2)
𝑑𝑡
• The equation (2) shows that when an error occurs,
the controller begins to increase (or decrease) its
output at a rate that depends upon the size of the
error and the gain.
• If the error is zero, controller output is not changed.
If there is positive error, the controller output begins
to ramp up at a rate determined by Equation (2).
 Integral Control Mode
• Fig a show two different
values of gain.
• It can be observed that the
rate of change of controller
output depends upon the
value of error and the size of
the gain.
• Figure b shows how controller
output will vary for a constant
error & gain.
• It can be observed that the
controller output begins to
ramp up at a rate determined
by the gain.
• In case of gain K1, output
finally saturates at 100%, and
no further action can occur.
 summary of integral control mode
• If the error is zero, the output stays fixed at a
value equal to what it was when the error went
to zero (i.e. p(0))
• If the error is not zero, the output will begin to
increase or decrease at a rate of KI %/sec for
every 1% of error.
 Area Accumulation
• From calculus we learn that an integral
determines the area of the function being
integrated.
𝑡
p(t) = KI 0 ep 𝑑𝑡 + p(0) …(1)
• Thus, Equation (1) can be interpreted as
providing a controller output equal to the net
area under the error-time curve multiplied by KI.
• We often say that the integral term accumulates
error as a function of time.
• Thus, for every 1% of accumulated error-time
area, the output will be KI percent.
 Example
• An integral controller is used for speed control with
a setpoint of 12 rpm within a range of 10 to 15 rpm.
The controller output is 22% initially. The constant KI
= -0.15% controller output per second per
percentage error. If the speed jumps to 13.5 rpm,
calculate the controller output after 2s for a
constant .
• We find ep from Equation
r−𝑏
ep = × 100
bmax−bmin

12 − 13.5
ep = × 100
15 − 10
ep = -30%
 Example
• The rate of controller output change is then given by
Equation,
𝑡
p(t) = KI 0 ep 𝑑𝑡 + p(0)
• but ep is constant
p(t) = KI ept + p(0)
• After 2s, we have
p = (0.15)(30%)(2) + 22
p = 31%
 Application
• In general, integral control mode is not used
alone, but can be used for stems with small
process lags and correspondingly small
capacities.
Integral Control Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Derivative Control Mode
 Derivative Control Mode
• Suppose you were in charge of controlling some
variable, and at some time, t0 ,
• Your helper yelled out,
“The error is zero”
What action do you want to take?
Well, it would seem perfectly rational to answer
“None”
because, after all, the error was zero.
 Derivative Control Mode
• But suppose you have a screen that shows the
variation of error in time and that it looks like
Figure.
• You can clearly see that even though the error at t
is zero, it is changing in time and will certainly not
be zero in the following time.
• Therefore, some action should be taken even
though the error is zero! This scenario describes
the nature and need for derivative action.
 Derivative Control Mode
• Derivative controller action responds to the rate
at which the error is changing- that is, derivative
of the error.
• The analytical expression for derivative control
mode is given by;
𝑑ep
p(t) = KD …(1)
𝑑𝑡
where KD = Derivative gain (s)
• Derivative action is not used alone because it
provides no output when the error is constant.
• Derivative controller action is also called rate
action and anticipatory control.
 Derivative Control Mode
• Figure illustrates how derivative
action changes the controller
output for various rates of change
of error.
• For this example, it is assumed
that the controller output with no
error or rate of change of error is
50%.
• When the error changes very
rapidly with a positive slope, the
output jumps to a large value, and
when the error is not changing,
the output returns to 50%.
• Finally, when error is decreasing -
that is negative slope - the output
discontinuously changes to a
lower value.
 Derivative Control Mode
• The derivative mode must be used with great
care and usually with a small gain, because a
rapid rate of change of error can cause very
large, sudden changes of controller output and
lead to instability.
 Summary of Derivative Control Mode
• If the error is zero, the mode provides no
output.
• If the error is constant in time, the mode
provides no output
• If the error is changing with time, the mode
contributes an output of KD percent for every
1% per second rate of change of error.
• For direct action, positive rate of change of
error produces a positive derivative mode
output
Derivative Control Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Composite Control
Modes
 Composite Control Modes
• It is found from the discontinuous and
continuous controller modes, that each mode
has its own advantages and disadvantages.
• In complex industrial processes most of these
control modes do not fit the control
requirements.
• It is both possible and expedient to combine
several basic modes, thereby gaining the
advantages of each mode.
• In some cases, an added advantage is that the
modes tend to eliminate some limitations they
individually posses.
 Type of Composite Control Modes
• Proportional-Integral (PI) control
• Proportional- Derivative (PD) control
• Proportional- Integral - Derivative (PID) control
 Proportional-Integral (PI) control
• This control mode results from combination
of proportional and integral mode.
• The analytical expression for the PI mode is
given by:
𝑡
p = KPeP + KPeI 0 eP 𝑑𝑡 + pI(0)…(1)
pI(0) = integral term value at t=0 (initial value)
 Proportional-Integral (PI) control
Advantage of this composite control mode
• The main advantage of this composite
control mode is that one-to-one
correspondence of the proportional control
mode is available and integral mode
eliminates the inherent offset.
 Proportional-Integral (PI) control
Working
𝑡
p = KPeP + KIeI 0 eP 𝑑𝑡 + pI(0)…(1)
• It can be observed from the equation (1) that the
proportional gain also changes the net integration mode
gain, but the integration gain, through KI, can be
independently adjusted.
• The proportional mode when used alone produces
offset error whenever load change occurs and nominal
controller output will not provide zero error.
• But in PI mode, integral function provides the required
new controller output, thereby allowing the error to be
zero after a load change.
• The integral feature effectively provides a reset of the
zero error output after a load change occurs.
 Proportional-Integral (PI) control
• Figure shows the PI mode
response for changing error.
• At time t1 , a load change
occurs that produces the
error shown.
• Accommodation of the new
load condition requires a
new controller output.
• It can be observed that the
controller output is provided
through a sum of
proportional plus integral
action that finally brings the
error back to zero value.
 Summary of PI mode
• When the error is zero, the controller output is
fixed at the value that the integral term had
when the error went to zero, i.e. output will be
pI(0) when eP=0 at t = 0.
• If the error is not zero, the proportional term
contributes a correction, and the integral term
begins to increase or decrease the accumulated
value [i.e. initial value pI(0)], depending on the
sign or the error and direct or reverse action.
 Application

• This composite proportional-integral mode

eliminates the offset problem of proportional
controllers.
• It follows that the mode can be used in systems
with frequent or large load changes.
 Disadvantage

• Because of the integration time, however, the

process must have relatively slow changes in
load to prevent oscillations induced by the
integral overshoot.
• Another disadvantage of this system is that
during start-up of a batch process, the integral
action causes a considerable overshoot of the
error and output before settling to the operation
point.
 Disadvantage
• This is shown in Figure, where we see
the proportional band as a dashed band.
• The effect of the integral action can be
viewed as a shifting of the whole
proportional band. The proportional
band is defined as that positive and
negative error for which the output will
be driven to 0% and 100%.
• Therefore, the presence of an integral
accumulation changes the amount of
error that will bring about such
saturation by the proportional term.
• In Figure, the output saturates whenever
the error exceeds the PB limits. The PB is
constant, but its location is shifted as the
integral term changes
Derivative Control Mode
Composite Control
Modes
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Proportional-Derivative
Control Mode (PD Mode)
Proportional-Integral-
Derivative Control Mode (PID
or Three Mode)
 Proportional-Derivative Control Mode
• The PD mode involves the serial (cascaded) use
of proportional and derivative modes and this
mode has many industrial applications.
• The analytical expression for PD mode is given
by
𝑑eP
p = KPeP + KPKD + p0
𝑑𝑡
• This system will not eliminate the offset of
proportional controller, however, handle fast
process load changes as long as the load change
offset error is acceptable.
 Proportional-Derivative Control Mode
• Figure shows a typical PD
response for load changes.
• It can be observed that the
derivative action moves the
controller output in relation
to the error rate change.
• Figure show PD control
mode response, showing
offset error from
proportional mode and
derivative action for
changing load, for reverse
acting system.
Proportional-Integral-Derivative Control
Mode (PID or Three Mode)
• One of the most powerful but complex
controller mode operations combines the
proportional, integral, and derivative modes.
• This PID mode can be used for virtually any
process condition.
• The analytical expression is given by
𝑡 𝑑eP
p = KPeP + KPKI 0 eP 𝑑𝑡 + KPKD + pI(0)…(1)
𝑑𝑡
pI(0) = integral term value at t=0 (initial value)
Proportional-Integral-Derivative Control
Mode (PID or Three Mode)
• This mode eliminates
the offset of the
proportional mode and
still provides fast
response for changing
loads.
• A typical PID response
is shown in Figure.
• In figure, the response
of the three-mode
system to an error is
shown.
 Example
• Suppose the error, Figure, is
applied to a proportional-
derivative controller with KP = 5,
KD = 0.5s, and p0 = 20%. Draw a
graph of the resulting controller
output.
• In this case, we evaluate
𝑑eP
p = KPeP + KPKD + p0
𝑑𝑡
over the two spans of the error.
• In the time of 0 to 1 s where ,
eP = at we have
p1 = Kpat + KPKDa + p0
Here a = 1%/sec
p1 = 5t + 2.5 + 20
 Example
• Note the instantaneous
change of 2.5% produced
by this error. In the span
from 1 to 3 s, we have
p2 = 5 + 20 = 25
• The span from 3 to 5 s has
an error of , eP = -0.5t + 2.5
so that we get for 3 to 5s
p3 = -2.5t + 12.5 - 12.5 + 20
OR
p3 = -2.5t + 31.25
 Example
• Let us combine everything and
see how the error of Figure
produces an output in the
three-mode controller with KP
= 5, KI = 0.7s-1 ,KD=0.5 s, and
pI(0) = 20%. Draw a plot of the
controller output.
• From Figure, the error can be
expressed as follows:
 Example
𝑡 𝑑eP
p = KPeP + KPKI 0 P
e 𝑑𝑡 + K P D 𝑑𝑡 + pI(0)
K
𝑡 𝑑eP
p = 5eP + 3.5 0 P
e 𝑑𝑡 + 2.5 + 20
𝑑𝑡
From 0 to 1 s, we have
𝑡
p1 = 5t + 3.5 e 𝑑𝑡
0 P
+ 2.5 + 20
p1 = 5t + 1.75t2 + 22.5
• This is plotted in Figure in the span
of 0 to 1s.
• At the end of 1 s, the integral term
has accumulated to pI(1) = 21.75%.
Now, from 1 to 3 s, we have
𝑡
p2 = 5 + 3.5 0 (1) 𝑑𝑡 + 21.75
p2 = 3.5(t - 1) + 26.75
 Example
• This controller variation is shown
in Figure from 1 to 3 s.
• At the end of 3s, the integral term
has accumulated to a value of
pI(3)= 21.75%.
• Finally, from 3 to 5s, we have

p3 = -0.875t2 + 6.25t + 21.625

• After 5 s, the error is zero.
Therefore, the output will simply
be the accumulated integral
response providing a constant
output of pI = 32.25%.
 Proportional-Derivative
Control Mode (PD Mode)
Proportional-Integral-
Derivative Control Mode (PID
or Three Mode)
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 ANALOG CONTROLLERS
Two - position Controller
Proportional Mode
Integral Mode
 ANALOG CONTROLLER
• An analog controller is a device that implements
the controller modes described previously,
using analog signals to represent the loop
parameters.
• The analog signal may be in the form of an
electric current or a pneumatic air pressure.
 Two - position Controller

• A method using op amp implementation of ON/OFF

control with adjustable neutral zone is given in Figure.
• For this circuit, we assume that if the controller input
voltage, V, reaches a value VH, then the comparator
output should go to the ON state, which is defined as
some voltage,V0.
• When the input voltage falls below a value VL, the
comparator output should switch to the OFF state,
which is defined as 0 V.
 Two - position Controller

• The comparator output switches states when the voltage

on its input, V1, is equal to the setpoint value, Vsp.
• Analysis of this circuit shows that the high (ON) switch
voltage is
VH = Vsp …(1)
And the low(OFF) switching voltage is
R1
VL = Vsp – V0 …(2)
R2
 Two - position Controller
• Figure shows the typical
two-position
relationship between
input and output voltage
for this circuit.
• The width of the neutral
zone between VL and VH
can be adjusted by
variation of R2.
• The relative location of
the neutral zone is
calculated from the
difference between
Equations (1) and (2).
 Proportional Mode

• The analytical expression for this mode is given by:

p = kpep + p0 …(1)
kp = proportional gain between error and controller output
p0 = controller output with no error (%)
eP = error in percent of variable range
• If we consider both the controller output and error to be
expressed in terms of voltage, we see that Equation (1) is
simply a summing amplifier.
• The op amp circuit in Figure shows such an electronic
proportional controller.
 Proportional Mode

• In this case, the analog electronic equation for the output

voltage is
Vout = GPVe + V0 …(2)
Vout = output voltage
R
GP = 2 = gain
R1
Ve = error voltage
V0 = output with zero error
 Proportional Mode

• The design of a proportional controller calls for

specification of the proportional gain described by KP
in Equation (1) that expresses the percent of output
for an error of 1% of the measurement range.
• Alternatively it could be described as the
proportional band,
PB = 100/KP
• This must now be expressed in terms of the voltage
gain, GP in Equation (2).
 Proportional Mode

• The relationship between GP and K is given by,

∆Vout
Gp = K P
∆Vm
∆Vout = range of output voltage
∆Vm = range of measurement voltage
 Integral Mode

• The analytical expression for integral mode is given by

the equation
𝑡
p(t) =
KI 0 ep 𝑑𝑡 + p(0) …(1)
p(0) = controller output when the integral action starts (%)
KI = Integral gain
• This function is easy to implement when op amps are
used as the building blocks. A diagram of an integral
controller is shown in Figure.
 Integral Mode

• The corresponding equation relating input to output

is
 Integral Mode

• The values of R and C can be adjusted to obtain the

desired integration time.
• The initial controller output is the integrator output at t
= 0.
• As we noted earlier, the integration time constant
determines the rate at which controller output
increases when the error is constant.
• If KI is made too large, the output rises so fast that
overshoots of the optimum setting occur and cycling is
produced.
 ANALOG CONTROLLERS
Two - position Controller
Proportional Mode
Integral Mode
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Derivative Mode
Composite Controller
Modes
 Derivative Mode

• The derivative mode is never used alone because it cannot

provide a controller output when the error is zero.
• The analytical expression for derivative control mode is given by;
𝑑ep
p(t) = KD …(1)
𝑑𝑡
where KD = Derivative gain (s)
p = controller output in percent of full output
eP = error in percent of full-scale range
• In principle, this mode could be implemented by the op amp
circuit presented in figure.
 Derivative Mode

• Indeed, the theoretical transfer function for this

circuit given in Equation (2) looks just like Equation
(1) with appropriate identifications in terms of
circuit elements:

• where the input voltage has been set equal to the

controller error voltage
 Derivative Mode

• From a practical perspective, this circuit cannot be

used because it tends to be unstable; that is, it may
begin to exhibit spontaneous oscillations in the
output voltage.
• The reason for this instability is the very large gain
at high frequencies where the derivative is very
large.
 Composite Controller Modes
• The combination of several controller modes
was found to combine the advantages of
each mode and, in some cases, eliminate
disadvantages.
• Composite modes are implemented easily
using op amp techniques.
• Basically, this consists of simply combining
the mode circuits introduced in the previous
section.
 Proportional-Integral Mode

• A simple combination of the proportional

and integral circuits provides the
proportional-integral mode of controller
action. The resulting circuit is shown in
Figure.
 Proportional-Integral Mode

• For this case, the relation between input and

output is most easily found by applying op amp
circuit analysis. We get (including the inverter)

• The definition of the proportional-integral

controller mode includes the proportional gain in
the integral term, so we write
 Proportional-Derivative Mode

• A powerful combination of controller modes is the

proportional and derivative modes.
• This combination is implemented using a circuit
similar to that shown in Figure.
• Analysis shows that this circuit responds according
to the equation
 PID (Three-Mode)
• The ultimate process controller is the one that
exhibits proportional, integral, and derivative
response to the process-error input.
• The analytical expression is given by
𝑡 𝑑eP
p = KPeP + KPKI 0 eP 𝑑𝑡 + KPKD + pI(0)…(1)
𝑑𝑡
pI(0) = integral term value at t=0 (initial value)
p = controller output in percent of full scale
eP = process error in percent of the maximum
KP = proportional gain
KI = integral gain
KD = derivative gain
 PID (Three-Mode)
• The zero-error term of the
proportional mode is not
necessary because the
integral automatically
accommodates for offset and
nominal setting.
• This mode can be provided by
a straight application of op
amp circuits, resulting in the
circuit of Figure.
• It must be noted, however,
that it is possible to reduce
the complexity of the circuitry
of Figure and still realize the
three-mode action, but in
these cases an interaction
results between derivative
and integral gains.
 PID (Three-Mode)
• We will use the circuit of
Figure because it is easy
to follow in illustrating
the principles of
implementing this mode.
• Analysis of the circuit
shows that the output is
 Derivative Mode
Composite Controller
Modes
 Classifications of different
controller modes
Discontinuous Controller Modes
(a) Two-position (ON/OFF) Mode
(b) Multi-position Mode
(c) Floating Control Mode: Single Speed and Multiple Speed
Continuous Controller Modes
(a) Proportional Control Mode
(b) Integral Control Mode
(c) Derivative Control Mode
Composite Controller Modes
(a) Proportional-Integral Control (PI Mode)
(b) Proportional-Derivative Control (PD Mode)
(c) Proportional-Integral-Derivative Control (PID or Three
Mode Control)
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Velocity & Position
Algorithm of PID
Control
Bumpless Transfer
 Velocity & Position Algorithm of PID
Control
• In the digital control mode, the
error signal is first sampled
and the controller output is
computed numerically through
a digital processor.
• Now Controller output for a
continuous-type P-I-D
controller:

…(1)
• The above equation can be
discretised at small sampling
interval T0 as shown in Figure.
Velocity & Position Algorithm of PID Control
• Taking the first order
derivative,

• and using rectangular

integration, we can
approximate as:

• Now replacing the

derivative and integral
terms in eqn. (1), one can
obtain
Velocity & Position Algorithm of PID Control

…(2)
• The above algorithm is known as Position
algorithm.
• But the major problem here is that the error
values at all the time instants are to be stored
(or at least the second term of the r.h.s of Eqn.
(2) at each instant have to be stored).
• An alternative approach known as velocity
algorithm can be obtained as follows.
Velocity & Position Algorithm of PID Control

…(2)
• From (2), one can write the error signal at the
(k-1)th instant as:

..(3)
• Subtracting eqn. (3) from (2), we can have:
Velocity & Position Algorithm of PID Control

• The above algorithm is known as Velocity algorithm.

• The major advantage of this algorithm is that it is of
recursive type.
• It calculates the incremental output at each sample instant.
As a result, it requires only to store three previous values:
e(k), e(k-1) and e(k-2).
 Bumpless Transfer
What is Bumpless Transfer?
• Bumpless Transfer is the method by which a
controller can be transitioned from Manual
mode to Automatic mode without disrupting
the process.
• Its aim is to keep the controllers output the same
when switching auto/manual, that is if the
controller is at 50% output in auto it should retain
that 50% output as you switch it to manual. If you
switch from manual to auto the same should apply.
• Most modern PID controllers have bumpless
transfer built in, including PLC and DCS PID
controllers.
 Example

• At 10:00 a.m. the operator places a

temperature control loop in Manual. The SP
= 280°F at the time of this transition.
• The operator decides to manually decrease
the control valve position from 40% to 30%.
 Example

• The PV, after several minutes, settles out at 220°F. At

11:00 a.m. the operator decides to switch the loop
back in Automatic mode. At that time the SP = 280°F
and the PV = 220°F.
• As soon as the operator flips the switch, the control
valve goes to 100% open (from 30%).
• Alarms start going off, levels and pressures start
swinging and, in general, everyone associated with
operating the plant gets very upset.
 Bumpless Transfer
• The remedy here is quite straightforward.
• The control logic should be designed to
1) Inform the PID controller when the loop has
been placed in Manual mode, and
2) Have the PID controller track the value of
the control valve position that is being
established by the plant operator.
• By having the PID controller track the actual
valve position, we will prevent it from
winding up to either the completely Closed
or Open position.
 Bumpless Transfer
Proportional Bump at the Time of Transfer
from Manual to Auto
• Assuming that we did our job and properly
designed a PID tracking system that
prevented the PID controller from “winding
up” when the control loop was in Manual.
• So let’s take a look at what could happen in
our previous example:
 Bumpless Transfer

• At 10:00 a.m. the operator places a temperature

control loop in Manual. As before, the SP = 280°F at
the time of this transition.
• The operator decides to manually decrease the
control valve position from 40% to 30%.
• The PV, after several minutes, settles out at 220°F. At
11:00 a.m. the operator decides to switch the loop
back in Automatic mode.
 Bumpless Transfer

• At that time the SP = 280°F and the PV = 220°F.

• As soon as the operator flips the switch the
control valve goes to 50% open (from 30%).
• Alarms may still trigger and disturbances to the
process may still occur. Maybe not quite as bad
as without tracking, but still very disruptive.
What happened?
 Bumpless Transfer
• Note in the above example that at the time that
the switch from Manual to Automatic occurred
there was an Error present (SP-PV; 280°F -220°F
= 60 °F).
• So even though we had the PID Controller
tracking the position of the Control Valve the
first time it calculated a control valve position
the Proportional Term of the PID controller
would “bump” due to the presence of the Error.
• The positive aspect is that the “bump” was less
than that which occurred without PID Tracking;
but there was a substantial “bump”
nonetheless.
• So now what should we do? Well there are two
options:
 Bumpless Transfer
Design the SP to track the PV when the control
loop is placed in Manual
• Most modern control systems today will give
you the option to have the SP track the PV
when the control loop is placed in Manual.
 Bumpless Transfer
Advantage
• Since the SP tracks the PV when in Manual the value
of the Error at the time the operator “flips the
switch” will be zero.
• As such there will be no “proportional bump” at the
time of the switch.
Disadvantage
• Operators will have to “remember” what the normal
SP should be for the control loop when it is returned
to Automatic mode.
• By having the SP track the PV the operator no longer
has the SP pegged at its normal value.
• There could still be a bump due to P action if large
change in SP is made after transferring to Auto mode.
 Bumpless Transfer
Suppress the Proportional Action at the instant
that the switch from Manual to Auto is made
• Some modern control systems provide the
option of “suppressing” the Proportional action
at the time the operator “flips” the switch from
Manual to Automatic.
 Bumpless Transfer
Advantage
• The SP remains at its nominal value during the
period of time that the control loop is in Manual
thus preventing the operator from having to
remember where the SP should be during
normal operations. Eliminates the Proportional
Bump.
Disadvantage
• If there is an Error present at the time the
operator “flips the switch” it may take a
considerable amount of time to “drive the PV to
the SP” since the proportional action of the PID
controller has been suppressed.
Velocity & Position
Algorithm of PID
Control
Bumpless Transfer
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Integral Wind Up
Controller Bias
 Integral Wind Up
What is Wind Up ?
• When the controller reaches the actuator limit,
then the actuator becomes saturated and the
system effectively operates in open loop
• The integral term and the controller output may
become really large = large overshoot
• The controller signal remains saturated even if
the error begins to increase; hence, very bad
transients
• Example: When a car is on a steep hill, the
throttle saturates when the cruise control
attempts to maintain speed.
 Integral Wind Up
The PI Algorithm

Where:
CO = controller output signal (the wire out)
CObias = controller bias or null value
e(t) = current controller error, defined as SP – PV
SP = set point
PV = measured process variable (the wire in)
KC = proportional gain, a tuning parameter
KI = integral gain, a tuning parameter
 Integral Wind Up

• The integral sum starts accumulating when the

controller is first put in automatic and continues to
change as long as controller error exists.
• If an error is large enough and/or persists long
enough, it is mathematically possible for the integral
term to grow very large (either positive or negative):
 Integral Wind Up

• This large integral, when combined with the other terms

in the equation, can produce a CO value that causes the
final control element (FCE) to saturate. That is, the CO
drives the FCE (e.g. valve, pump, compressor) to its
physical limit of fully open/on/maximum or fully
closed/off/minimum.
• And if this extreme value is still not sufficient to eliminate
the error, the simple mathematics of the controller
algorithm, if not jacketed with protective logic, permits
the integral term to continue growing.
• If the integral term grows unchecked, the equation above
can command the valve, pump or compressor to move to
110%, then 120% and more. Clearly, however, when FCE
reaches its full 100% value, these last commands have no
physical meaning and consequently, no impact on the
process.
 Integral Wind Up

• To better visualize the problem of windup and the

benefit of anti-windup protection, consider the
plot from heat exchanger process below.
• To the left is the performance of a PI controller
with no windup protection. To the right is the
performance of the same controller protected by
an anti-windup strategy.
 Integral Wind Up

• For both controllers, the set point is stepped from 200 °C

up to 215 °C and back again. As shown in the lower trace on
the plot, the controller moves the CO to 0%, closing the
valve completely, yet this is not sufficient to move the PV
up to the new set point.
• To the left in the plot, the impact of windup is a degraded
controller performance. When the set point is stepped
back to its original value of 200 °C , the windup condition
causes a delay in the CO action. This in turn causes a delay
in the PV response.
• To the right in the plot, anti-windup protection permits the
CO, and thus PV, to respond promptly to the command to
return to the original SP value of 200 °C.
 Avoiding windup (1)

• There are many ways to avoid windup.

• One method is illustrated in figure, the system has an
extra feedback path that is generated by measuring
the actual actuator output, or the output of a
mathematical model of the saturating actuator, and
forming an error signal(es) as the difference between
the output of the controller(v) and actuator
output(u).
 Avoiding windup (1)

• The signal es is fed to the input of the integrator

through gain kt.
• The es is zero when there is no saturation and extra
feedback loop has no effect on the system.
• When the actuator saturates, the signal es is feedback
to the integrator in such a way that es goes toward
zero.
• This implies that controller output is kept close to the
saturation limit.
 Avoiding windup (1)

• The controller output will then change as soon as the

error changes sign and integral windup is avoided.
• The rate at which the controller output is reset is
governed by the feedback gain, kt, a large value of kt
give a short reset time.
• The parameter kt, can, however, not be too large
because measurement error can then cause an
undesirable reset.
• A reasonable compromise is to choose Kt = 1/Ti for PI
control and control and as kt = 1/ TiTd for PID control.
 Avoiding windup (2)
• Back-calculation: Use feedback of the difference
between the desired (v) and the actual
control(u) as input for the integral term.
• If v is not saturated, then set u = v
• If v is at the saturation value, then set u =
saturation value -> the process is in open loop.
• The error signal es = u–v. It is non-zero only
when the actuator is saturated => no effect on
normal operation.
• The normal feedback path around process is
broken and a new feedback path around the
integrator is formed. The integrator input
becomes: es/kt + e*k/ki.
 Controller Bias
• The P-Only controller computes a Controller output
action every loop sample time T as:
CO = CObias + Kc∙e(t)
Where:
CO = Controller output
CObias = controller bias or null value
Kc = controller gain, a tuning parameter
e(t) = controller error = SP – PV
SP = set point
PV = process variable
 Controller Bias

• Let’s suppose the P-Only control algorithm shown

above is used for speed control in an automobile and
CO is the throttle signal adjusting the flow of fuel to
the engine.
• Let’s also suppose that the speed SP is 70 and the
measured PV is also 70 (units can be mph or kph ).
Since PV = SP, then e(t) = 0 and the algorithm reduces
to:
CO = CObias + Kc∙(0) = CObias
 Controller Bias

• If CObias is zero, then when set point equals

measurement, the above equation says
that the throttle signal, CO, is also zero. This
makes no sense. Clearly if the car is
traveling 70 kph, then some baseline flow
of fuel is going to the engine.
 Controller Bias

• This baseline value of the CO is called the bias or

null value. In this example, CObias is the flow of fuel
that, in manual mode, causes the car to travel the
set speed of 70 kph.
• A P-Only controller bias (sometimes called null
value) is assigned a value as part of the controller
design and remains fixed once the controller is put
in automatic.
Integral Wind Up
Controller Bias
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Tuning of Controller
Design Problems
Performance Criteria
Time-Integral Performance
Criteria
 Why study Controller Tuning?
• Controller tuning refers to the
adjustments in the tuning parameters
(proportional gain, integral gain and
derivative gain) in order to ensure the
best response of the controller.
• In any control process, when a controller
is properly tuned, there is reduced
process variability, maximum efficiency,
minimum energy costs and high
production rates.
• Hence it is important to study controller
tuning.
 Design Problems
• Consider the block diagram of general closed-
loop system shown in figure.
• When the load or the set point change, the
response of the process deviates and the
controller tries to brings the output close to the
desired set point.
 Design Problems
• Figure shows the response of the controlled
process to a unit step change in the load when
different type of controllers have been used.
• We notice that different controller have different
effects on the response of the controlled process.
 Design Problems
• Thus the few design question arises:
1. What type of feedback controller should be used
to control a given process?
2. How do we select the best values for the
adjustable parameters of a feedback controller?
3. What performance criterion should we use for the
selection and the tuning of the controller?

• Let us now study the questions above in more

detail and provide the initial guideline for the
design of a feedback controller.
 Performance Criteria
• Consider two different feedback control
systems producing the two closed-loop
responses shown in figure.
• Response A has reached the desired level of
operation faster than response B.
 Performance Criteria
• If our criterion for the design of the controller
had been
Return to the desired level of operation as soon as
possible
• Then, we would select the controller which
gives the closed-loop response of type A.
 Performance Criteria
• But, if our criterion is
To keep the maximum deviation as small as possible
OR
Return to the desired level of operation and stay close
to it in the shortest time
• We would have select the other controller, yielding
the closed-loop response of type B.
 Performance Criteria
• For every process control application, we can
distinguish
Steady-state performance criteria
Dynamic response performance criteria
• The principle steady-state performance criterion
usually is zero error at steady state.
• The evaluation of dynamic performance of closed-
loop system is based on two type of commonly used
criteria:
1. Criteria that use only a few point of response. They
are simple but only approximate.
2. Criteria that use the entire closed-loop response
from time t=0 until t= very large. These are more
precise but also more cumbersome to use.
 Performance Criteria
• The performance criteria are based on some
characteristic features such as,
• Overshoot
• Rise time ( time need for the response to
reach the desired value for first time)
• Settling time ( Time need for the response to
settle within ±5% of desired value)
• Decay ratio (Decay ratio is a reasonable
trade-off between a fast rise time and a
reasonable settling time)
• Frequency of oscillation of the transient
 Time-Integral Performance Criteria
• The shape of the complete closed-loop
response, from time t=0 until steady state has
been reached, could be used for the
formulation of a dynamic performance
criterion.
• Unlike the simple criteria that use only isolated
characteristics of the dynamic response, the
criteria of this category are based on the entire
response of the process.
 Time-Integral Performance Criteria
• Integral of the square error(ISE)
∞ 2
ISE = 0
𝜖 (t)𝑑𝑡
• Integral of the absolute value of the
error(IAE)
∞
IAE = 0
𝜖(t) 𝑑𝑡
• Integral of the time-weighted absolute
error(ITAE)
∞
ITAE = 0
𝑡 𝜖(t) 𝑑𝑡
 Time-Integral Performance Criteria
• If we want to strongly suppress large errors, ISE
is better than IAE because the errors are
squared and thus contribute more to the value
of the integral.
• For the suppression of small errors, IAE is better
than ISE because when we square small number
they become even smaller.
• To suppress error that persist for long time, the
ITAE criterion will tune the controllers better
because the presence of large t amplifies the
effect of even small errors in the value of the
integral.
 Tuning of Controller
Design Problems
Performance Criteria
Time-Integral Performance
Criteria
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Controller Tuning
Process reaction curve method
ZN tuning
 Controller Tuning
• When the type of feedback controller has been
selected, we still have the problem of deciding
what value to use for its adjusted parameters.
• This is known as the controller tuning problem.
• There are three general approaches we can use for
tuning a controller:
1. Use simple criteria such as
The one-quarter decay ratio,
minimum settle time,
minimum largest error and so on
• This approach is simple and easy to implement
 Controller Tuning
2. Use time integral performance criteria
such as
ISE
IAE
ITAE
• This approach is rather cumbersome and
relies heavily on the mathematical model of
process.
3. Use the semi empirical rules which have
been proven in practice.
 Controller Tuning
• Process reaction curve method
• ZN tuning
 Process reaction curve method
• This method is developed by Cohen and Coon.
• This is basically an open loop technique of tuning.
• Here the process is assumed to be a stable first
order system with time delay.
• The closed loop system is broken as shown in Fig; a
step input is applies at m’ , output is measured at b.
 Process reaction curve method
• In fact, a bias input may be necessary so that
the plant output initially becomes close to
the nominal value.
• The step input is superimposed on this bias
value. The input and the output response
are plotted by suitable means as shown in
Figure.
 Process reaction curve method
• M,L and K are measured.
• Let us define the following terms
corresponding to Figure.
Slope = N,
Time Constant T=K/N
Lag Ratio R=L/T
 Process reaction curve method
• Then, the recommended optimum settings,
for P, P-I and P-I-D controller are as follows.
 ZN Tuning Technique
• Unlike the process reaction curve method which uses
data from the open-loop response of a system, the
Ziegler- Nichols tuning technique is a closed loop
procedure.
• It goes through the following steps:
1. Bring the system to the desired operation level
2. Using propotiona; control only and with the
feedback loop closed, introduce a set point change
and vary proportional gain until the system oscillates
continuously.
3. The frequency of continuous oscillation is the
crossover frequency 𝜔CO. Let M be the amplitude
ratio of the system response at the crossover
frequency
 ZN Tuning Technique
4. Compute the following two quantities
1
Ultimate gain = KU =
𝑀
2𝜋
Ultimate period of sustained cycling = PU =
𝜔CO
min/cycle
• Using value of KU and PU, set gain according to
given in table.
Controller KC 𝝉l 𝝉D
P Ku/2
P-I Ku/2.2 PU/1.2
P-I-D Ku/1.7 PU/2 PU/8
 ZN Tuning Technique
Advantages
• Does not need require expert knowledge.
• Does not require model of system
• Often produce good initial Kp, KD, KI.
Disadvantages
• Not mathematically rigorous
• It require a stable system
• If you do not have model real experiment could be
costly.
• Require a system that can be driven a ustable with
increasing promotional gain
 Controller Tuning
Process reaction curve method
ZN tuning
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 Feedback control
Feedforward Control
 Feedback control

• The idea of the basic controller is that the output of

the controlled process, the "measured variable" or
"process variable' is compared with the "desired
value" or "setpoint".
• The difference between the two is the "system
error" or simply "error" which is also called the
"deviation" because it is not actually erroneous or a
mistake, it is a fundamental aspect of the loop.
 Feedback control

• The deviation is then multiplied by a control

function. The control function may be a 3-term or
PID controller or it may be another type of
controller of which there are many to choose from.
• The output of the controller is then used to drive
the process by feeding it to an "actuating device"
sometimes called a "final control element".
 Feedback control
Advantages
• It is stabilising. It balances or regulates the
measured variable.
• It improves the accuracy of control.
• It partially linearises any non-linearities in the
process.
• Within the limits of the system headroom or
available power in the process, it speeds up the
process response.
• It counteracts process disturbances or loads on the
process, such as opening oven doors etc.
 Feedback control
Advantages
• It allows the operator to alter the setpoint and
therefore the balance point of the control
• With a regulator, the control loop holds the
process output (measured variable) steady
• With a tracking device or "servomechanism" it
accurately tracks the changing setpoint.
 Feedback control
Disadvantages
• It waits until the effect of disturbance hits the
process, so it is unsatisfactory for the processes
with a large dead time.
• It may also create instability in closed-loop
response.
• It is expensive and complex to implement.
 Feedforward Control
Why feedforward Control require?
• Conventional feedback control loops can never
achieve perfect control.
• It is difficult for the conventional loops to keep the
process output continuously at the desired set point
value in the change of load or set point changes.
• This is because a feedback controller reacts only after
it has detected a deviation in the value of the output
from the set point.
 Feedforward Control

• In feedforward control strategy, information

concerning one or more conditions that might
disturb the control variable is converted into
corrective action to minimize deviation of
controlled variable.
• The signals which have the potential to upset
the process are transmitted to the controller.
 Feedforward Control

• The controller makes appropriate

computation on these signals and calculates
new values for the manipulated signals and
send those to the final control element.
• Therefore, the control variable remains
unaffected in spite of load changes.
 Feedforward Control

Heat Exchanger
• Figure depict the use of feedforward control in
heat exchanger.
• The objective is to keep the exit temperature of
the liquid constant by manipulating the steam
flow rate.
 Feedforward Control

• There are two principal disturbances that are

measured for implementing feedforward control
1. Liquid flow rate
2. Liquid inlet temperature
• Measurement these disturbance menipulations the
steam flow.
 Feedforward Control
Drum Boiler
• Here the objective is to keep
the liquid level in the drum
constant.
• The two disturbances are
the steam flow from the
boiler, which is dictated by
varying demand elsewhere
in the plant and the flow of
the feed water.
• The flow rate of feed is also
the principal manipulated
variable.
Comparison
 Feedback control
Feedforward Control
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Cascade Control
Ratio Control
 Cascade Control
Why cascade control Require?
• It should be noted that a simple closed loop control
system is not a panacea. It can not perfectly solve
all problems.
• One of the problems is that non-control engineers
often expect a simple loop to operate perfectly.
Happily there are many design changes we can
make to improve the degree of control.
• Cascade control allows us to share the work
between two or more controllers. Ratio control
allows us to control two or more parameters from a
couple of related setpoints.
 Cascade Control
• Cascade control is the technique of nesting one
control loop inside another. It allows the
workload to be shared between the two loops.
• The name derives from the fact that the output
of the outer controller is cascaded into the
setpoint of the inner controller.
 Cascade Control
• By sharing the workload between the two
controllers the advantages of the closed loop
control (above) are increased.
• Accuracy, speed of response, settling time
after a disturbance, and any control errors
are all improved.
• The cost is that you need a second
measuring device which can be expensive,
especially when installation costs are
considered.
 Cascade Control
• The following is a cascade control system
with level control cascaded into flow control.
The level controller output is fed into the
flow control setpoint.
 Cascade Control

• The inner loop of the cascade pair is there to assist

the outer loop. It would only be a control loop in
isolation if the cascade were to be taken off and the
control room operator were to run the flow controller
by hand (manually).
• Usually this would not be the case and so the job of
the inner loop is to reduce the effects of disturbances
within the inner loop.
 Cascade Control

• As a result, removing steady state error is not a

necessary function. The outer loop can remove
the steady state error in the level and the flows
will balance anyway.
 Cascade Control
Advantages
• Better control of primary variable
• Primary variable is less affected by
disturbances
• Faster recovery from disturbance
• Increases the natural frequency of the
system
• Reduces the effective magnitude of a time
lag
• Improves dynamic performance
• Provides limits on the secondary variable
 Ratio Control
• Ratio control is used to ensure that two or more
flows are kept at a constant ratio even if the
flows are changing.
• Ratio control is a special type of feedforward
control where two disturbances are measured
and held at constant ratio with each other.
 Ratio Control
Application
• To keep a constant ratio between the feed
flow-rate and the steam in the reboiler of a
distillation column.
• To hold the reflux ratio constant in a
distillation column.
• To control the ratio of two reactants entering
a reactor at a desired ratio.
• To maintain correct air and fuel mixture for
optimal combustion.
 Ratio Control
Configuration 1
• In this control configuration, we measure both
the flow rates, and take their ratio.
• This ratio is compared with the desired ratio(the
set point, as shown in the figure), and the
deviation between the two is used to generate
the actuating signal for the ratio controller.
 Ratio Control
Configuration 2
• In this configuration, flow measurement of
the wild stream, A, provides set point for the
controllable flow stream, B.
 Ratio Control
Configuration 3
• This configuration is applicable
when both the stream are
controllable.
• It is implemented with a double
feedback loop as shown in figure.
• Each flow is maintained by
individual feedback loop.
• Here, the error signal for stream A
is generated with a given set point
signal SP ; whereas for stream B,
the set point is generated using
the signal from A multiplied by the
desired ratio.
Cascade Control
Ratio Control
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
 SWELL AND
SHRINK REACTION
Inverse Response
 SWELL AND SHRINK REACTION
• A sudden increase or decrease in
steam flow changes the pressure
in the steam drum and boiler
circuit.
• The change in pressure will cause
a change in both the boiling point
and density of the water and
steam.
• These combined reactions will
cause the level in the steam drum
to increase or decrease rapidly.
• The increase and decrease in the
water level caused by the pressure
change are commonly referred to
as the swell and shrink reactions.
 SWELL REACTION
• A sudden steam load
increase will naturally
produce a drop in the
pressure in the steam
drum, because, initially
at least, the firing rate
cannot increase fast
enough to match the
steam production rate at
the new demand level.
• When the pressure in the
drum drops, it has a
dramatic effect on the
natural convection within
the boiler.
 SWELL REACTION
• The drop in pressure
causes a small fraction of
the saturated water in the
boiler to immediately
vaporize, producing a large
amount of boil-up from
most of the tubes in the
boiler.
• During the transient, most
of the tubes temporarily
become risers. The result is
that the level in the steam
drum above the
combustion chamber rises.
 SHRINK REACTION
• If the steam output
from the boiler is
reduced suddenly, the
pressure will increase
in the boiler circuit.
• This increase of
pressure will cause the
drum level to shrink
initially and then
increase due to the
higher inflow than
outflow.
 Inverse Response
• When you push down on your car’s
accelerator, you expect the car to speed
up, right?
What if it slows down?
• Or even worse: You lift your foot off the
accelerator and your car speeds up.
• And the more you lift your foot, the
more the car speeds up.
• These are almost unthinkable and
certainly scary situations, yet they occur
every day in thousands of boilers and
some other processes around the
world.
• The phenomenon is called an inverse
response.
 Inverse Response
• The temperature of the
feedwater flowing into
the drum is normally
below boiling point.
• When we add more of
this colder water to the
boiler, some of the
steam bubbles in the
boiler condense.
• This causes the drum
level to decrease and
the effect is called
inverse response.
 Inverse Response
• However, the effect is only
temporary.
• After a while, the higher
rate of feedwater flow
overcomes the lost volume
and the drum level rises .
• The opposite is also true:
when we decrease the flow
rate of the colder feedwater,
steam production increases,
and the additional steam
bubbles cause the drum
level to rise. But after a
while, the drum level begins
to fall, as expected.
 SWELL AND
SHRINK REACTION
Inverse Response
 L. J. Institute of Engineering & Technology
Department of Instrumentation & Control
Process Control

INSTRUCTOR:
Mr. Nimesh Prabhakar
Mob:9537723297
[email protected]
Feed water &
Drum-Level
Control
 Feed water & Drum-Level Control

• The main purpose of the water level controller

is to bring the drum up to level at boiler start-up
and maintain the level at all time.
• A sudden decrease in this level may uncover
boiler tubes, causing them to become
overheated and damaged.
• Boiler high water level may interfere with the
process of separating moisture from steam(i.e.
carryover of saturated steam) within the drum,
thus reducing boiler efficiency and carrying
moisture into the turbine.
Types of Boiler water level controller
• Single element drum level control
• Double element drum level control
• Three element drum level control
 Single Element

• This consists of a proportional signal(element) coming

from the drum level transmitter.
• This signal is compared to an input setpoint of water
level and the difference between these two is a
deviation value.
• This signal is an input valve for the controller which
will generate corrective action in the form of a
proportional output.
• The output is then passed to the boiler feedwater
valve, which then adjusts the level of feedwater flows
into the boiler drum.
 Single Element

Limitations
• Single element water level control system is
unable to control the level in big boilers.
• Possible inadequate control option because
of the swelling effect of Boiler
 Two Element

• In two element controlling the swelling and

shrinking effect of the boiler drum is taken care.
Drum Level Element
• signal coming from the drum level transmitter is
compared to a setpoint of water level and the
resultant is a deviation value.
• On the basis of this signal, the controller will
generate corrective action.
 Two Element

Steam Flow Element

• A mass flow rate signal from flow sensor on
steam outlet line (corrected for density) is
used to control the feedwater flow, giving
immediate corrections to feedwater demand
in response to fluctuation in the amount of
steam consumption(load change).
 Three Element

• This above-defined control system is not suited for a

steam system where consumption is very high and
fluctuating.
• The three-element drum level control is ideally suited
where a boiler plant consists of multiple boilers (as in
case of tankers) and multiple feedwater pumps or
where the feedwater has variations in pressure or
flow(if connected with turbines etc. sudden rise and
fall of the load are very frequent).
 Three Element

Drum Level Element

• A proportional signal coming from the drum level
transmitter.
Steam Flow Element
• A mass flow rate signal (corrected for density) is used
to control the feedwater flow, giving immediate
corrections to feedwater demand in response to load
changes.
Feedwater Flow Element
• A mass flow rate signal on the feed water pipe,
responds rapidly to variations in feedwater demand.
 Three Element

• Any difference between steam mass flow and

feedwater mass flow into the drum is corrected
by the level controller.
How difference between these flow rate can
arise?
• Blowdown of boiler
• Leaks in the steam or sudden increase in steam
consumption
• Feedwater pressure or flow fluctuations
 Three Element

Salient features
• The three-element system provides tighter
control for drum level with fluctuating steam
load.
• This system is ideal for fluctuating feedwater
pressure or flow.
• More sophisticated level of control.
Feed water &
Drum-Level
Control