C&I 504 Process Control

Course Objectives: This course aims to provide in-

depth understanding of modelling, design and control
strategies in process industries.
Course Outcomes:
After completion of this course, students will be able to:
CO1: Understand the basic principles of process
modelling & importance of process control in industrial
processes.
CO2: Understand the use of process mathematical
models for the design of control strategies.
CO3: Design and tune process (PID) controllers.
CO4: Understand sequential control using
 C&I 504 Process Control
• Process Dynamics:
• Introduction to process control- objective of modeling – models of industrial process – hydraulic tanks – fluid flow systems –
mixing process – chemical reactions – thermal systems-heat exchangers and distillation column.
• Control Action and Controller Tuning:
• Basic control action-on/off, P, P+I, P+I+D, floating control – pneumatic and electronic controllers – controller tuning – time
response and frequency response methods – non-linear controllers, Inverse time response of system, effect of pole and
Zero on right hand side of S- plane.
• Complex Control Techniques:
• Feed forward, ratio – cascade – split range – inferential – predictive – adaptive and multivariable control.
• Programmable Logic Controllers:
• Evolution of PLC – Sequential and Programmable controllers – Architecture – Programming of PLC – Relay logic and Ladder
logic – Functional blocks – Communication Networks for PLC, field bus such as profi-bus, mod-bus etc.
• Distributed Control System:
• Evolution of DCS – Architecture – Local control unit – Operator interface – Engineering interface.
• Suggested Readings:
• John.P. Bently , “Principles of Measurement Systems “, Pearson Education
• B.Wayne Bequette, “Process Control-Modeling, Design and Simulation, Prentice Hall of India Pvt. Ltd.
• George Stephanopolus, “Chemical Process control”, Prentice Hall India?
• Harriot P., “Process control”, Tata McGraw-Hill. New Delhi,.
• Norman A Anderson, “Instrumentation for process measurement and control” CRC Press LLC, Florida, .
• Dale E. Seborg, Thomas F Edgar, Duncan A Mellichamp, “Process dynamics and control”, Wiley John and Sons, .
• Marlin T.E., “Process control”, second edition McGraw hill, New York.
• Balchan J.G. and Mumme G., “Process control structures and applications”, Van Nostrand Renhold Co., New York, .
• Lucas M.P, “Distributed Control System”, Van Nostrand Reinhold Co. NY .
• Pertrezeulla, “Programmable Controllers”, McGraw-Hill, .
 Continuous Evaluation Scheme
Course
Course Name Cr L T P CWS PRS MTE ETE PRE
Code

C&I 504 Process Control 4 3 0 2 15 25 20 40 -

CWS 15% Assignments/Project work and

Attendance
PRS 25% Laboratory Work
MTE 20% Project Work based on Process
Control
ETE 40%
 Introduction
• Process-The conversion of feed materials to
products using chemical and physical
operations. In practice the term process tends
to be used for both the processing operation
and processing equipments.
• Process automation is automating the various
activities involved in a process.
 Need of Process automation
• Any process involves few or more variables which
need to be controlled to get desired output.
• If number of variable which are to be controlled are
less then they can be controlled manually.
• But as the complexity of process increases, number
of variable need to be controlled increases.
• Process automation is important for operation of
large scale and more complex process as they
involve large number of variable which needed to be
controlled.
 Modern Industrial Automation

Needed for the following reasons

• Safe Operation
• Compliance with environment regulation
• Effective use of Raw Material & Energy
• Higher Productivity & Effective Production
• Flexibility in changing the Product & Process
requirements
Steps involved in developing a control strategy

• Control objective : first step is to find out the

control objective.
• Input variables : input variables can be
classified as
1) Manipulated – They variables which can
be controlled/adjusted.
2) Disturbance-the variables which cant be
controlled/adjusted but they affect the
process output.
• Output variables-output variables can be
classified as
1. Measured-which we can measure, Value of these
variables are used to formulated the control
strategy.
2. Unmeasured-Which we can not measure.
• Constraints-Every process has certain
limitations in the control strategy.
They can be classified as
1. Hard-which cant be violated. Example-Minimum
or maximum flow rate in a process.
2. Soft-which can be violated up to some extent.
example-product composition.
• Operating characteristics-Any process can be
classified into
1. Continuous process-Long operating time and
operating conditions remains constant.
 Batch process-Operate for a short operating
time and operating conditions may vary.
Semi Continuous process-it may have setup but
operating conditions may vary during the process.
• Safety, Environment and economic considerations
An unsafe process will ultimately cost more.
effect of process output and by products on
environment should not be harmful and if they are
found to be harmful, proper corrective measures
should be taken.
 Type of valves
• Two types of valves are used
1. Fail open(Air to close) : Fail open valve would be fully
open in case of signal failure (will allow maximum flow
rate)
2. Fail close(Air to open) : Fail close valve will be fully closed
in case of signal failure (will allow minimum (zero) flow
rate)
For example :
If we have a process which is designed to operate
between 3 and 15 psig. A fail close valve will be fully
closed at 3 psig and fully open at 15 psig.
Similarly a fail open valve of same size would function
oppositely.
 SURGE TANK
From
process
one (F1)

h=height of
liquid Two
process
two(F2)
Tank Level Protection
• The surge tank shown in last slide is used as
intermediate storage for fluid streams being
transferred between two processes

• Applying Control strategy to the surge tank

system :
1. Control Objective – To maintain the height of
water up.
2. Input Variables-Flow rate F1 and flow rate F2 are
two input variables.
3. Output variables-Liquid level (h)
• Constraints: No of constraints for above
system are
1. Maximum liquid level-tank should not overflow
2. Minimum and maximum flow rate.
• Operating Characteristics –It’s a continuous
process.
• Safety and Environment and economic
considerations-it depends on fluid
characteristics.
• Control structure-It can be feed back or feed
forward
• Control Structure : Type of controllers
1. Feedback –It measures the output variable and
compared the output with a reference input and
produces the error.
controller action is such that it minimize the
error.
2. Feed forward-It measures the disturbance
variables and sent this value to a controller which
adjusts the manipulated variable before output is
affected.
Feedback control schematic (Block Diagram)

F1 is manipulated and F2 is a disturbance

 FEEDBACK CONTROL STRATERGY 1

The level is measured and the inlet flow rate (Valve

position) is manipulated
 FEEDBACK CONTROL STRATERGY 2.

Outlet Flow rate is manipulated

FEEDBACK CONTROL SCHEMATIC (Block Diagram)
Scenario2 : F2 is manipulated and F1 is Disturbance
 FEED FORWARD CONTROL STRATERGY FOR LIQUID LEVEL
CONTROL

Inlet Flow rate is measured and outlet flow rate is manipulated

• A control strategy can be made combining
both feed forward as well as feedback control
strategy which is depicted in the next slide.
FEED-FORWARD/FEEDBACK CONTROL SCHEMATIC
BLOCK DIAGRAM
 FEED-FORWARD/FEEDBACK CONTROL STRATERGY.

The inlet flow rate is the measured disturbance, tank height is

the measured output, and outlet flow rate is manipulated.
 Taking a Shower : A common multivariable control
Problem

•We analyze this process step by step:

1. Control Objectives:

a) To become clean
b) Convenience (Suitable temperature and water flow through shower)
c) To become fresh

2. Input Variables: Hot Water and Cold Water valve positions, Body position under
shower.
3. Output Variables: Temperature and flow of water from shower.
4. Constraints: Maximum and Minimum Valve positions for temperature control
and flow of water from shower. Max temperature of hot water and min
temperature of cold water falls under – Hard Constraints. Moreover, Mixed
Stream water temperature falls under – Soft Constraints.
HOT COLD
WATER WATER

3. Operating Characteristics: It can be Continuous or

semi continuous.
4. Safety, Environment and Economic consideration:
Temperature control, treatment of waste water for
further use and duration of operation (consumption of
energy for heating water) should be taken care for
above respectively.
3. Control Structure: More than one variable to control,
hence it is a multivariable Control Problem, therefore
appropriate strategy should be planned.
• Also type of control i.e. Feedback or Feed-forward
control should be appropriately selected.
• 8. It is important to note that individual taking the
shower served as the controller.
 INSTRUMENTATION INVOLVED IN A CONTROL
PROBLEM

•A Sensor : Measurement Device

Common Sensors used are mainly for Chemical Processes

such as temperature control, flow control, level control,
composition pH etc.

•Actuator : Manipulated Input Device

The most common manipulated input is the Valve Actuator

Signal (which is usually Pneumatic)
• Controller

In control loop, it operates to receive or supply

signal from another device.

• Signals may be : Continuous or Discrete.

• Continuous Signals : Such as electrical voltage or

current, are called Analog Signals.

• Discrete Signals : Signals communicated at different or

discrete intervals of time, are called Digital Signals.
 Understanding : Analog type of signals used

•Provides basis for control theory and design and analysis.

•Current range for analog signals : 4 to 20 mili A.

•Voltage range for analog signals : 0 to 5 V.

•Since most usual controllers are of Pneumatic analog Controllers

(developed in 1930, still find application even now) type, use
instrument air, bellows and spring arrangement to calculate
controller output.

•Pneumatic range for analog signal : 3 psi to 15psi.

• Psi is a unit of pressure.
• Psi stands for Pounds per Square Inch.
• 1 Psi = 6894.75729 Pascal.
• For large values, this range can be amplified to enough
pressure to move the valve stem.

• There are Current to Pressure Converters (I/P) as well as

Voltage to Pressure (V/P) Converters.
Problem: A process furnace heats a process stream from near ambient
temperature to a desired temperature of 300° C. The process stream outlet
temperature is regulated by manipulating the flow rate of fuel gas to the
furnace as shown below:
a) Discuss the objectives of
control strategies ?
b) What is the measured
output ?
c) What is the manipulated input
?
d) What are possible
disturbances ?
e) Is this a continuous or batch
process ?
f) Is this a feed-forward of
feedback controller ?
g) Should the control valve fail
open or fail close ? For the
strategy you choose is the
valve gain positive or
negative ? Why ?
h) Discuss Safety, Environmental
and Economic issues.
Fundamental models
of Process Control
 Fundamental Models
• A methodology is to be developed for any dynamic
process to be studied, such as Chemical Processes
etc.
• Reason for Modeling: Improvement and better
understanding of the process.
• Developed Model are used for following purposes:
– Operator Training
– Process Design
– Safety System Analysis
– Process Control
• Operator Training: Any person responsible for
operation of any process will be called Process
operator. The developed model can be used
for his training purpose so that he can learn to
take proper action to upset conditions, before
actual operation on process.
• Process Design: A dynamic process model can
be used to properly design chemical process
equipment for a desired production rate.
Ex- A model of batch reactor can be used
approximate its size for certain product.
• Safety : Dynamic process models can be used
to design safety systems.
Ex- It can be used to determine the duration,
after a valve fails, for a system to reach a
certain pressure.
• Process Control: Feedback systems are used
to maintain the process variables at desired
values.
Ex- A control system may sense output
temperature of the product to set input for
the steam flow.
 PROCESS MODELING
Assumption to develop Process model:
The process is lumped, linear and time
invariants dynamic process.
Two approaches for Process modeling:
1) Integral Balance: used for developing model of
distributed parameter systems .
2) Instantaneous Balance / Mass or Energy
balance : Time rate of change is written directly.
 Dynamic Balance of Mass or Energy
• It describes that rate of mass accumulation is
equal to rate of mass in by flow minus rate of
mass out by flow.

[Rate of Mass accumulation]=

[Rate of mass in by flow]-[Rate of mass
out by flow]
Mathematically
• V= Volume ρ= Density of Material
• q= volumetric flow rate (volume/time)
for perfect mixed system
M=Total mass M=ρV
For steady state (dv/dt =0)
qin - qout = 0
qin = qout

(Operating Condition of the System)

 Model of Gas Surge Drum
• Surge drum one often used as intermediate
storage capacity for gas streams that are
transferred between chemical process unit .
Consider a drum shown as qi inlet molar flow rate
qin P q

q – outlet molar flow rate

Control problem : Manipulate one flow rate to maintain a desired drum
pressure Therefore, develop a model that describe how the drum pressure
varies with inlet and outlet flow rates
V – volume of drum
N – the total amount of gas (moles ) contained in the drum
Assumption ideal gas pV = nRT
P – pressure, T – temperature R – ideal gas constant
Rate of accumulate of the mass of gas in the drum

MW – molecular weight (constant)

Continuously Stirred Tank Reactor (CSTR)/Heater or
Heat Mixing Tank
(Process model based on energy balance)
Consider a perfectly mixed stirred-tank heater,
with a single feed stream and a single product
stream shown in the figure. Assuming that
the flow rate and temperature of the inlet
stream can vary, that the tank is perfectly
insulated and that the rate of heat added per
unit time (Q) can vary, we develop a model to
find the tank liquid temperature as a function
of time.
Stirred- tank heater
Material balance eq.

Neglecting change in density due to

temperature

Energy Balance
 The total work done on the system as a
combination of the shaft work and energy
added to the system to get the fluid into the
tank and the energy that the system performs
on the surrounding to force the fluid out.

Substituting the value of total energy in eq.(i)

we, get
 Expanding the derivative term and assuming the
density constant , then cancelling the common term
gives ,
 Types of Systems
• Hydraulic systems: It involves usually oils as
medium for energy or power transmission in
the process. Also knwn as fluid system or
liquid level systems
• Pneumatic systems : It involve usually gases or
compressed air as medium of energy or power
transmission.
 LIQUID LEVEL SYSTEMS
Assumption : Fluid has a laminar flow i.e.
Reynolds number is less than 2000.
Fluid dynamics is described using Linear
Differential equations [LDE].
Otherwise it could be Turbulent flow/ Non
linear flow having Reynolds number beyond
3000-4000, for such systems Non-Linear
differential equation is used to describe them.
Consider a liquid level system
The fixed values denoted by capital letters (Q,H).
let say for a small duration there is some
change in the system all the varying quantities
are represented by small letters (q,h).
• Resistance ‘R’ of a liquid level flow through a
pipe is defined as the change in liquid level
difference necessary to cause a unit change
on flow rate.
R = difference in liquid level (meter)
change in flow rate(cubic meter/sec)
For a small deviation in coming flow rate say qi
during an interval of ‘ dt ‘
in outflow rate change ‘qo’
change in level of hank is ‘dh’ .

Capacitance ‘C’ is defined as change in quantity

of stored liquid because of unit change in head.
C = change in liquid stored (cubic meter)
change in head (m)
Applying Mass balance equation

V=Ah
A= Area is capacitance
 Mathematics Model of Process control
system (Fluid & Thermal)
• Hydraulic System – Fluid System Oil
• Pneumatic – Fluid system that uses air or gases.
• Liquid Level System
• Laminar Flow – Reynold number is less than 2000
• Turbulent flow – Reynold number > 3000 to 4000
System with laminar flow can be represented by linear
differential equation.
Consider the liquid level system
If flow through restriction is laminar, then the
relationship between steady state flow rate and
steady state head at the level of the restriction is
given by
Q = KH
Q = steady state liquid flow rate m3/sec
K = coefficient m2/sec
H= steady state head, m
For laminar flow, resistance is obtained:
• Laminar flow resistance is constant and
analogous to resistance of electrical circuit
However, for turbulent flow

• And resistance Rt for turbulent flow is

•
A liquid level system Linear differential
equation is analogues with a RC circuit
Block diagram of RC Circuit
( A First order process )
 INTERACTING PROCESS MODEL
Taking example of interaction liquid level system
Q = Steady state flow rate
H1= steady state level of liquid intake 1
H2= steady state level of liquid intake 2
C1= Capacity of tank 1
C2= Capacity of tank 2
R1= Resistance of control valve 1
R1= Resistance of control valve 2
qi = small change in flow rate of liquid
qo= deviation in outflow rate
q1= outflow rate of liquid flow tank 1
The second stage loading the first stage (driving
power from first stage )is called interacting
process

For Tank 1

For Tank 2
Applying Laplace Transformation in all the above
equations, we get
Block diagram Reduction of interacting
liquid level system
Non-Interacting System
Develop a State Space model of Process
• In processes with multiple input and multiple
out put (MIMO) the block diagram reduction
become cumbersome to use.
As for using transfer function w.r.t. first input all
the other inputs should be considered zero.
State space model is quit useful in MIMO
process, using state space it is quit easy to find
the relation between any set of input &
output.
Develop a state space model of interacting
liquid level process
Tank 1

Tank 2
 Hydraulic Systems
1. Transfer of force through a fluid(usually a
liquid)
2. Accuracy(high power to weight ratio)
3. Smoothness and simplicity of operation
Hydraulic systems mainly find applications in
the aircraft control systems and machine tool
applications.
 Properties of hydraulic fluids
• Viscosity:-It is the measure of internal friction
(the resistance of the fluid to flow).
• The resistance of the fluid to the relative
motion of its parts is called Dynamic or
absolute viscosity.
• The kinematic viscosity is the ratio of the
dynamic viscosity to the mass density.
ν=(μ/ρ)
 Hydraulic Circuits
• General components of a hydraulic circuit
are:-
• Reservoir
• Pumps to force the fluid through the
circuit(like a battery in electrical network).
• Valves to control fluid flow
• Lastly an actuator to convert the fluid pressure
into useful work.
A Typical Hydraulic circuit
Dashpots
Also called a damper a dashpot
acts as a differentiating element
 Transfer Function for dashpot
• Using net force balance on the
system(Newton Law)

A-piston area in inches square

K-Spring constant lb/inch
• Also flow rate q is given by

R is the resistance to flow

As the flow through the liquid should be equal
to the change in the mass of the oil(during
dt seconds)
Hence,
The above equation can be rearranged as

or,

Taking Laplace Transform(assuming zero initial

conditions)
We get,

The transfer function is given by,

Let us define ,
Then,
• The mathematical model of a dash pot system is
given by

So from above equation, we have,

Where time constant T is given by,

• Note that as

and

Therefore, the viscous friction coefficient b is

 Advantages and Disadvantages
Advantages : Disadvantages :

• Hydraulic power is not readily

• The fluid acts as a lubricant .
available(as in case of
• Small actuators can develop electrical).
large torques.
• Comparatively higher cost than
• Have fast response. thermal systems.
• Good speed regulation . • Fire and explosion risk exists.
• Can be operated under • Poor damping characteristics.
continuous, intermittent
conditions without damage.
 Comparison of Hydraulic & Pneumatic
systems
Pneumatic Systems Hydraulic Systems
• Air or gas is the medium • Water or fluid is the
and these are compressible. medium and these are in-
compressible.
• Air or gas lack lubricating • Oil or surge as fluid
property. medium as well as
lubricating element.
• Operating Pressure is • Operating pressure is more
usually low as compare to then the pneumatic
hydraulic systems. systems.
• Output power is • Output power is
considerably lower then considerably higher then
hydraulic system. pneumatic system.
• Accuracy is less and lower • Accuracy is better then
. pneumatic systems.
• Operating temperature • Operating temperature
commonly used is 50-60 commonly used is limited
deg Cel and can be as high to 20-70 deg Cel. only .
as 200 deg Cel.
• It is fire proof or
insensitive to • Vulnerable to fire and
temperature. sensitive to temperature.
• External leakage is • Internal leakage is
allowed. allowed.
 PNEUMATIC SYSTEM MODELLING
• Pneumatic system uses compressible fluid as
working medium and it is usually air.
• In pneumatic systems, compressibility effects of gas
cannot be neglected and hence dynamic equations
are obtained using conversion of mass.
• In pneumatic systems, change in fluid inertia energy
and the fluid’s internal thermal energy are assumed
negligible.
 DISCRIPTION OF PNEUMATIC SYSTEM
• Pneumatic devices involve the flow of gas or
air, through connected pipe lines and pressure
vessels.
• Hence, the variables of pneumatic system are
mass flow rate and pressure P.
• The mass flow rate is analogous to current. The
pressure variable is analogous to voltage.
• The two basic elements of a pneumatic system
are the resistance and capacitance
 Definition for Pneumatic Resistance
• The gas flow resistance, R is defined as the
rate of change in gas pressure difference for a
change in gas flow rate.
q – flow rate
- change in pressure
 Definition for pneumatic capacitance
• Pneumatic capacitance is defined for a
pressure vessel and depends on the type of
expansion process involved. The capacitance
of a pressure vessel may be defined as the
ratio of change in gas stored for a change in
gas pressure.
• 𝛒 – density
• C – Capacitance
• v – volume of vessel
 MASS BALANCE EQUATION
• Rate of mass accumulation = mass in – mass
out
• Pressure change is dp during interval dt.
• Applying Laplace transform and rearranging
the terms we get.

1st order process

 Pros and Cons of Pneumatic System
• Advantages
– The air or gas used is non inflammable and so it offers
safety from fire hazards.
– The air or gas has negligible viscosity, compared to
high viscosity of hydraulic fluids.
– No return pipelines are required and since air can be
let out at the end of work cycle.
Disadvantage
• The response is slower than that of hydraulic systems
because of the compressibility of working fluid.
 APPLICATION OF PNEUMATIC SYSTEM
• Guided Missiles
• Aircraft systems
• Automation of production machines
• Automatic controllers
• And many more……
 Pneumatic Control Elements
1) Nozzle flapper Amplifiers
(1st stage amplifier)
(displacement to pressure transducer)
2) Pneumatic Relay
(2nd stage amplifier)
Nozzle Flapper Amplifier
Pressure Displacement Characteristics
Pneumatic Relay
Pneumatic relay
Pneumatic Proportional Controller
 Pneumatic P+D controller

a)Pneumatic P+D controller

b)step change in e & corresponding changes in x & p
c)Block Diagram of the controller.
 Pneumatic P+I+D Controller

a) Pneumatic P+I+D Controller

b) Block diagram of controller
 DYNAMICS OF PROCESSES
• Time response : it is time domain behavior of
process when subjected to a prescribed input
like unit impulse, unit step, unit ramp etc.
Time response is combination of
a) Transient response
b) Steady state response
Example of 1 order process,
st

and finding out the response

Consider a First Order Process

u = Input Variable
y = Output Variable
Equitation (1) can be represented as
Say Time constant ‘T’ and
Steady state gain ‘K’

Where T & K are parameters of the process.

Now consider unit step input , the process
dynamics can be determined using the equitation
(3).
Taking Laplace Transformation on both side of
eq. (3), we get
s y(s) – y(0) + y(s) = K u(s)
Y(0) is the initial condition let Yo
(Ts+1) y(s) – T Yo = K (1/s)
(Ts+1) y(s) = K/s + T Yo
y(s) = K/s(Ts+1) + T Yo/ s(Ts+1)
Process dynamics
y(t) = Inverse Laplace[y(s)]
 y(t)=Inverse Laplace[K/s–KT/(sT+1)+TYo/(sT+1)]

Where the K is the Steady state constant

and the is the transient response.
Final valve of the response or ultimate response
Notes :
1. Steady state response/ultimate response of
the process is ‘K’ times the magnitude of
input.
2. The response attains after t = T
y(t) = 0.632 times the steady state value.
3.

It indicate that is initial rate of rise is

maintained then it will achieve steady state
value t = T.
 Effect of Time Constant on
the process dynamics

If
T - large, then process
response is slow.
T - small ,then faster
time response .
T = 0, then
instantaneous rise
of response.
 Inverse response of a process

• Processes where the initial direction of the

step response is opposite to the direction of
their steady state response are called INVERSE
RESPONSE PROCESS.
• Processes which are associated a zero in right
half s plane gives inverse response
characteristics.
Process with inverse response
where

Condition for inverse response

Also
Inverse response of the system to a unit step input

Yg is overall response
 Types of Input to system

• Step Input
• Ramp input
• Impulse input
• Sine input
Step Input

If A=1, Then function is called unit

step function
Laplace transform is given by :-
Ramp input

If A=1, Then function is called unit

ramp function
Laplace transform is given by :-
Sinusoidal function

Laplace transform is given by :-

 Impulse function
One of the more useful functions in the study of linear systems is the
"unit impulse function." An ideal impulse function is a function that
is zero everywhere but at the origin, where it is infinitely high

Laplace transform of Unit Impulse function is unity(1) and area under impulse
function is also unity
 Pure Gain system
Consider a first order system whose speed of response is very fast i.e. time constant
is very small.
For unit step input, output is given by

y(t)=Ku(t)
Taking Laplace transform
Y(s)=KU(s)
But U(s)=1
So transfer function=K
A process having such characteristics is called Pure gain process and transfer function
contain only one parameter “K”.
Such systems are always at steady state and no transient behavior occur. They
reach the final state instantly.
Physical example of pure gain system
There are very few process that exhibit pure gain characteristics
One example is Capillary system as shown below
There is a flow construction that that results in the pressure drop on the
incompressible fluid flowing in the capillary system.

H=head

Liquid Liquid
Flow (F) outflow
 h=R*F

Where R=Resistance in terms of deviation of initial steady state “ho”.

y=R*u
With
 Pure Capacity system
Consider 1st order differential equation used to model the dynamics of first order
system

If =0
A process defined by the above equation is called pure capacity
process.
Observe that regardless of the nature of input signal u(t), the solution
can be find out by direct integration method
 Example of Pure capacity system
The most common example of pure capacity system is storage tank or surge tank
with an outlet pump as shown in figure.
Such a tank is generally employed in industries as intermediate storage tank.

Inlet flow
at Fi
 Question-Dynamic Behavior of the liquid level in a storage tank

A 250 liter tank used for liquid storage is configured so that its mathematical
model and its dynamic behavior are discussed

The cross sectional area of tank is 0.25 square meter and it may be assumed
uniform. The outlet valve resistance, which we shall assume to be linear for
simplicity, has c=0.1 meter square/min.

1.What is initial value of the liquid level in tank?

2.If the inlet flow rate was suddenly changed to 87 liters/min, obtain an expression
for how the liquid level in the tank will vary with time.
3.When will the liquid level in tank be at 0.86m mark? And to what final value will
liquid level ultimately settle.
 THE LEAD LAG SYSTEM
• Its general transfer function is

• Where K is the system gain, ξ the lead time

constant, τ the lag time constant.
• If the phase of the system is positive then the
system is said is said to be a leading system.
On the other hand if its negative then it’s a
lagging system.
 Dynamic Response of Lead Lag system
• The transfer function can be manipulated to

• Where
• The dynamic response therefore is given by
Dynamic response of a lead lag system
• From the above equation the total response
can be divide into two parts:
• First due to a pure gain system – ρK to a
certain input function.
• Secondly to a first order system(Gain k and
time constant τ) to the same input function.
• So the response can be rewritten as
 Unit step response
• For unit step response substituting h(s)=u(s)

• Which can be rearranged to give

 Second Order Systems

• Where y(t) and x(t) are output and input

variables respectively.
• Also represented as

• Where , τ=( )^(1/2) and,

• Taking Laplace transform on both sides, we
get

so that the general transfer function is given by

where the characteristics parameters of the

system are K the steady state gain, ξ the
damping coefficient, τ the natural period
Example of second order system(The spring
shock absorber system)
Step Response to second order systems
• U(s)=A/s for a step input of magnitude A.

Rearranging it we get
• Solving the above equation for y(t) we get

where r1 and r2 are given by

 Nature of roots and response
• By examining the quantity under the radical
sign we infer that the roots can be either real or
complex depending upon ξ.
1) For 0< ξ<1: r1 and r2 are complex conjugates.
In this case therefore the response is a
damped sinusoid. It ultimately settles at AK.
2) ξ=1: r1=r2 real and equal roots critically
damped response.
3) ξ<1 : r1 and r2 are real and distinct implying an
over damped response.
Graphical representation of the various step
responses
 General Nth Order System
• The response of any order system depends
only the nature of its roots of its characteristic
equation. There can be mainly two
possibilities.
• The roots are real(either distinct or multiple)
• Some roots occur as complex conjugates while
others are real.
• The first possibility translates to the cascade
connection of N first order system in series.
• The second results in a combination of several
under damped systems included in the series
of the first order systems.
 Higher order systems with zeros
• General Equation of transfer function is given
by

• Parameters – K the steady state gain, p poles

and q zeros. The above system is often
represented as K (p,q).
Unit Step Response of general (p,q) system

• The unit step response of the general (p,q)

order system can be obtained as follows

• Above expression can be manipulated as

• Where

Hence the step response is given as (by taking

inverse Laplace transform)
Few note points Regarding the step response
1. When p-q=1, regardless of the value of q and
p, as long as q>=1, it can be shown that the
initial slope of the response will be non zero.
2. When p-q>1 slope of response is zero and
the shape of the response is sigmoidal.
3. In general the response of the system will be
swifter for q>p(more number of zeros) than
thee response of the corresponding system
with lower value of q(fewer zeros).
 the diagram a third order system with no
zero and a third order system with one zero
(p=3, q=1).

Note that the response of the system with

zero is sigmoidal, but it is less sluggish than
the response for pure third order system.
Process with delay time
Question : Two non interacting tanks are connected in
series as shown in Fig. .the time constant are 1
and 0.5 and R2=1. Sketch the response of the level in
tank 2 if a unit step change is made in the inlet
flow rate to tank 1.
If we consider height of tank 2 as output variable and input flow rate of tank tank
1 as our input variable , than transfer function would be

For a unit step change in input flow rate “Q” we obtain

Taking inverse Laplace transform after applying partial fraction, it gives

Plotting the time response we get
 Two Non interacting tanks

If we take two non interacting tanks, the transfer function

would be
If we consider time constants of both tanks equal to unity and

The transfer function would become

Factorizing the denominator, we obtain

For unit step change in input, applying partial fraction and
taking inverse Laplace transform, we obtain time response
expression of Output flow rate of tank 2

Plotting it and comparing with the response of Non interacting tank

Question-Determine the Transfer function H(s)/Q(s) for liquid
level system shown, Resistance R2 and R1 and linear. The flow
rate from tank 3 is maintained constant at “b” by means of a
pump; i.e., the flow rate from tank 3 is independent of head H.
The tanks are non interacting
 PID Controller Tuning
• Effect of Controller Gain : An increase in controller
gain (K) will speed up the closed loop response.
Expect for first & second-order processes with P-
only control, too large value of controller gain will
destabilize the system.
• Integral Time: An increase in integral time tends to
“slow down” the closed-loop response, while
decreasing the integral time speeds up the
response. Too small an time integral can cause the
closed-loop system to be unstable.
• Derivative Time: An increase in derivative
time tends to help stabilized the closed-loop
response. A disadvantage of the derivative
action is that it causes the control system to
be sensitive to measurement noise.
P-Only Control
The proportional only algorithm is , which has
the following transfer function relationship
between error and controller output.
PI-Control : The PI algorithm is

Which has the following transfer function

relationship between error and controller
output.
PID control : The ideal PID algorithm is

Which has following transfer function

relationship between error and controller
output.
Ziegler-Nichols Closed-loop Method

The Ziegler-Nichols closed loop tuning

technique was perhaps the first rigorous
method to tune PID controller. The technique
is not widely used today because of the
closed-loop behavior tends to be oscillatory
and sensitive to uncertainty. We study the
technique for historic reasons and because it
is similar to commonly used automatic tuning.
 Ziegler-Nichols methods is consists of
following steps:

1. With P-only closed-loop control, increase

the magnitude of the proportional gain until
the closed-loop is in a continuous oscillation.
For slightly larger value of controller gain, the
closed-loop system is unstable, while for
slightly lower value the system is stable.
 2. The value of controller proportional gain
that cause the continuous oscillation is called
the critical gain. The peak-to-peak period is
called critical period.
3. Depending on the controller chosen ,P ,PI or
PID use the values in the table for the tuning
parameters, based on the critical gain and
period.
Ziegler-Nichol Open-loop Tuning
Parameters
𝜏(integral 𝜏(derivative
time constant) time constant)
Controller Type K
(Controller gain)

P-Only 1/ k𝜭 - -

PI 0.9𝜏/k𝜭 3.3 𝜭 -

PID 1.2𝜏/k𝜭 2𝜭 0.5𝜭

In table 𝜭 is the delay element and k =Kp/Tp

where Kp is the process gain & Tp is the process integral
time constant.
 Cohen-Coon Method
This method is based on first-order + time-delay
process model. A set of tuning parameters was
empirically developed to yield a closed-loop response
with a decay ratio of ¼. The tuning parameters as a
function of the model parameters as shown in table.
The major problem with the Cohen-Coon parameters
is that they tends not to be very robust, that is a small
change in the process parameters can cause the
closed loop system to become unstable.
 Cohen-Coon Tuning Parameters
𝜏 (integral) 𝜏
(derivative)
Controller K
type (controller gain)

𝜏[1 + (𝜭/
3𝜏)]/k𝜭
P-Only - -

𝜏[0.9+𝜭/ 𝜭[30+3𝜭/𝜏]/
12𝜏]/k𝜭 [9+20 𝜭/𝜏]
PI -

𝜏[4/3+𝜭/4𝜏]/ 𝜭[32+6𝜭/𝜏]/ 4𝜭/ [11

k𝜭 [13+8𝜭/𝜏] +2𝜭/𝜏]
PID

In table 𝜭 is the delay element and k =Kp/Tp

where Kp is the process gain & Tp is the process integral time
constant.
Development of Process Control Strategy

For development of process control strategy the

following elements foe the process need to be
indentified.
1. Control Objective
2. Input variables
3. Output variables
4. Constrains
5. Operating Characteristics of the process
6. Safe Environment & Economics
7. Control Structures.
Input Variables
a) Manipulated variables : they are adjusted
to change the output variables.
b) Disturbance : these cannot be controlled but
they effect the output of the process.
Output Variables
c) Measured output : these are measured and
used as feedback loop parameters.
d) Unmeasured output : these are not
measured.
Constrains
a) Hard constrains : Fully open or Fully closed
positions of valve is hard constrains for it.
b) Soft constrains : The valve can have
intermediate positions, other then fully open
or closed.
Operating Characteristics
a) Continuous Process
b) Semi-Batch Process
c) Batch Process
 Control Structures
• Feedback control – Single feedback loop.
• Feed forward control – Anticipatory control.
• Cascade control – Multiple feedback loops.
• On-Off control – Simple control bur it is not
very precise, we can use it in band or range.
• Ratio control – Desired ratio is maintained.
• Sequential control – Implemented in process
through intelligent devices like PLC based.
 FEEDBACK CONTROL
Output of the process is consciously measured
and compared with the set point/reference
value, depending on the error the controller
generate an actuating signal which correct the
plant.
For example : Liquid level control system
1) Control Objective : Maintain a desired value
of height in tank.
2) Manipulated input : q1 inlet flow rates &
q2 outlet flow rates.
3) Output variable : ‘h’ height if the liquid in the
tank.
4) Constrains : Control valves hard constrain either
fully open or fully closed.
Control valve types
a) Air to open
b) Air to close
Fail safe operation : under abnormal condition it
should change its position from what it was under
normal condition.
CASE 1: Feedback control strategy with q1 as
manipulated variable ( here q2 is constant).
CASE 2: Feedback control strategy with q2 as
manipulated variable (here q1 is constant).
 Drawback of Feedback control
• Until unless there is change in the output
there is no correction, so the system has
always a time lag.
• They are complex and costlier.
Advantage of Feedback control
It got better accuracy in compare to non
feedback or open loop. Output is corrected
only after the feedback sensors measures the
changes in output.
FEED FORWARD CONTROL (FFC)
The disturbance input here is measured and
adjust the manipulated input before the
process output change. This is anticipator
control.
let now take the same example of Liquid Level
control and how the feed forward control work.
Here q1 is the disturbance
and q2 is the manipulated input.
Feed forward control Strategy
for Liquid Level control
Block Diagram for FFC Liquid level control
The entire loop of Feed Forward Controller
(FFC)should be fast enough so that before the
effect of disturbance appear on the output, it
should have nullified it. Otherwise the Feed
forward Controller take more time then the
process lag or process delay it will be as good
as a feedback control.
The performance of FFC is limited by model
uncertainly without a perfect process model
FFC can not compensate and there will be
unwanted effect .
 Summary of Feed-forward Controller

1. The feed-forward controller must be

physically realizable and stable ,as noted by
the following points
a) If the process has a RHP zero, it must be
factored out before designing the feed-
forward controller. Otherwise the feed-
forward controller will be unstable.
b) The disturbance time delay must be greater than
the process time delay for perfect feed-forward
compensation. If the process time delay is greater
than the disturbance time delay, the feed-forward
controller will have no time-delay compensation,
and perfect control cannot be achieved.
c) If the process is higher order than the
disturbance, fast time constants probably must
be neglected in designing the feed-forward
controller.
2. Often a static feed-forward controller does
not change the close loop stability of the
feedback system, assuming the feed-forward
controller is stable. Also, a feed-forward
controller does not change the set point
response if a closed loop system.
3. Feed-forward control can be implemented
with either the classical feedback or IMC
(internal model control).
FEED FORWARD AND FEEBBACK CONTROL

Feedback and feed forward both the control

strategies can be used in a process to get the best
control.
For example : A fixed furnace to heat a process
fluid at desired temperature .
Control objective – To maintain the
temperature of outlet process fluid.
1) In FFC strategy the temperature of outlet Process
fluid is not measured which needed to be
controlled. The disturbance is inlet rate of
 Process fluid is compensated by changing the
fuel gas flow rate (manipulated variable).
2) In Feedback (FB) strategy the temperature of
the outlet process fluid is measured and
compare with the set point and depending on
the error the controller change the flow rate of
the fuel gas inlet.
3) In FFC & FB control strategy the inlet flow rate
of process fluid and outlet temperature both
are measured and the fuel gas inlet rate is
controller by looking at both the
measurements.
Control Strategies for heating a process fluid
at desired Temperature
FEEDFORWARD CONTROL FEEDBACK CONTROL
FEED FORWARD AND FEEDBACK
CONTROL STRATEGY
Block Diagram for Fixed Furnace or Jacketed
Heat Exchanger
From Block Diagram, we get
The effect of feed Forward loop is only on
numerator of Transfer function.
The effect of feedback loop is only on
numerator of Transfer function.
It is noticed from equitation (4) that a stable
Feed Forward Control does not effect the
close loop stability of plant as it does not
appear in the closed loop characteristic
equitation. i.e.
Condition for eliminating the effect of
disturbance through Feed Forward Control is

Feed Forward Control is a negative

compensator.
 Cascade Control
• It consist of two loops Master loop (outer loop/ Primary
loop) & Slave loop ( inner loop/ Secondary loop) in
feedback position mostly. The slave loop should be faster
then the master loop.
• Cascade control involves multiple measurements of
output variables or controlled variable and single
manipulated variable.
• Control loops are nested inside the other.
• For effective control through cascade controller, the inner
loop dynamics must be faster then outer loop.
• The more sensitive parameters are placed in inner loop
and less sensitive in outer loop.
 Cascade control for jacketed heat
exchanges
• In same fixed furnace/Jacketed heat
exchanger the floe of fuel gas depend on the
pressure of the fuel gas.
• Consider pressure of fuel gas as a disturbance
variable we are regulating temperature of
process fluid and flow the fuel gas through
cascade control.
 Process instrumentation Diagram of
cascade control of jacketed heat exchanger
Block Diagram for Cascade Control of
Jacketed heat exchanger
In Cascade control strategy the secondary
controller rejects the fuel gas pressure
disturbance, even before it is even felt at the
process outlet temperature, when primary
process has much larger time constant and
secondary has smaller time constant then the
cascade control provide the satisfactory result.
 Ratio Control
It is used when ratio of two or more process streams
are to be held in proportion to each other. Ratio
control is needed for Blending operation, holding
the fuel-air ratio in furnance,combustion, Reactor
feed control.
There are three approaches to realize the ration
control.
1) Scaling Approach
2) Direct ratio control
3) Indirect ratio control
 Scaling Approach for Ratio Control
Suppose two streams A & C having flow rate of Fa & Fc
respectively. Consider Fa as Wild stream (i.e. no control
valve is there to control it) and Fc as manipulated stream.
There is a strict requirement of effective flow transmitter is
this kind of ratio control otherwise it will not work properly.
Fy is the Intermediate converter which convert the signal
compatible for Fc as it is needed signal in the form of
pressure (3- 15 psig) for the control valve.
FFy ratio station is used to apply a user defined scaling
factor ‘k’ to signal that stream Fc should be ‘k’ times the Fa
so to maintain the ratio at desired level.
Scaling approach with ratio station
 Direct ratio control
Both the flow rates
are measured a
ratio controller FFC
manipulates the
flow rate of stream
‘C’ to produce the
desired ratio of
stream ‘C’ to
stream ‘A’.
Block Diagram for Direct Ratio Control
Assume 4 – 20 mA signal,
the measured ratio is
calculation

To provide an actuating
signal to control valve Rth
is compared with derived
ratio Rr and an error signal
is generated which is
further processed by a PID
controller.
 Indirect Ratio Control

The measured value of

process stream a is Qa
is operated by the
ratio station.
Assuming 4 – 20 mA
signal the desired
value of Qr for the
flow rate of stream B
is determined by
equation
Qr = 4 + Rr ( Qa – 4 )
Block Diagram of Indirect Ratio Control
 Split-Range control
A split-range strategy is often used in
situations where one or more valves may be
used, depending on the operating scenario.
An example application is a batch reactor.
Batch reactor must often be heated from
ambient temperature to a desired operating
temperature. Particularly if the reaction is
exothermic, then cooling will be needed to
maintain the desired reactor temperature.
 In the split-range control strategy shown in the
figure, if the jacket temperature controller output is
between 0 and 50 %, the cold glycol valve is open.
if the jacket temperature controller output is
between 50 to 100 %, the hot glycol valve is open.
For safety reasons we wish the cold glycol valve fail-
open and the hot glycol valve to
fail-closed.
Batch reactor temperature controller has a split
range output, where the cold glycol valve is open
during cooling mode and the hot glycol valve is
open during heating mode
Batch reactor temperature controller using split
range control
 Selective and Override Control
Often a controller needs to select from a set of
possible measurement for control. For
example consider the packed-bed reactor. It is
important to maintain the “Hot spot”
temperature at a certain value to minimize
catalyst degradation. In this strategy the “high
selector” (represented as the block with >)
choose the highest of three bed temperature
to send to the temperature controller.
In algorithmic form let T1, T2 & T3 represent
the three temperature measurements and T
represent the temperature measurement used
by the controller , then
T = max(T1,T2,T3)
When manipulating the fuel gas flow to a fired
heater (furnace), it is important to make
certain that a minimum air/fuel ratio is
maintained. High/Low selectors can be used
to assure this.
High Temperature selector for a packed-bed
reactor
ON-OFF Control(Bang-Bang Control)

It is equivalent to proportional control with high

gain. Small finite error ‘e’ causes the output ‘u’ to
switch between maximum & minimum value of
output. Output will be either maximum Or
minimum, that will depend on the sigh of the error.
Figure 1 shows the error signal/input signal &
Figure 2 shows the output of the ON/OFF
controller
On/Off ex. Heater of process tank
Objective: To Keep the water at desired temperature.
Depending on temperature the liquid in the tank, the
power supply is connected or disconnected to the heating
element. Hence temperature cycles about the set point Ts
with in a narrow band dT.
If switch is On – Off then wear and tear occurs hence
reduce the life of switch.
Let ‘Tc’ cycles time,
Large ‘Tc’ means large deviation in temperature.
Small Tc means small deviation but frequent turn on- turn
off of the switch.
Block diagram of heated process fluid
The temperature of Fluid will follow the
curve shown in the figure below
 Measurement
and
Signal Conditioning elements
 Sensing Elements
• Resistive sensing Elements
i) Resistance Thermometer & thermistors
ii) Metal & Semiconductor strain Gauges
• Capacitive sensing elements
i) Variable separation displacement sensor
ii) Variable dielectric displacement sensor
iii) Variable area displacement sensor
• Inductive sensing element
i) Variable inductance/reluctance displacement
sensor
ii) LVDT Displacement sensor
• Electromagnetic sensing elements
• Thermoelectric sensing elements (thermocouple)
• Elastic Sensing elements
• Piezoelectric sensing elements
• Piezoresistive sensing elements
• Electrochemical sensing elements
 Signal Conditioning Elements
• Deflection bridges
deflection bridges are used to convert the
output of the resistive, capacitive and
inductive sensors into a voltage signal.
Any linear network can be represented by a
Thevenin equivalent circuit consisting of a
voltage source ‘E’ together with a series
impedance ‘Z’.
Thevenin equivalent circuit for a deflection
bridge
 Amplifiers
Amplifiers are necessary in order to amplify
low level signals e.g. thermocouple or strain
gauge bridge output voltage, to a level which
enable them to be further processed.
Operational Amplifier: it is a high gain,
integrated amplifier design to amplify signals
from D.C. up to many kHz. The transfer
characteristic depends entirely on the
feedback network.
Instrumentation amplifier: An
instrumentation amplifier is a high
performance differential amplifier system
consisting of several closed loop operational
amplifier. An instrumentation amplifier gives
output voltage which depends only on the
difference of the two voltages V1 & V2.
 Current Transmitters
A current transmission system reject series
interference more efficiently than a voltage
transmission system. Transmitters used in the
process industries gives a output d.c. current
signal in the standard range 4 to 20 mA
proportional to the input measured variable.
The output has a live zero of 4 mA this enables a
distinction to be made between a zero input
(output=4mA) and a fault situation(output=0mA).
Types of transmitters
1. Closed loop Transmitters: these system have
high gain- negative feedback. The overall
transmitter sensitivity depends mainly on the
sensitivity of the elements in the feedback path
and in largely unaffected by changes in the
characteristics of the forward path element for
eg. amplifier drift. The zero & span of these
transmitters can be adjusted but, in case of
force balance systems the adjustment have to
be made mechnaically.
2. Open loop Transmitters: The overall
transmitter sensitivity depends on sensitivity of
every element in the system , these element
sensitivity can change due to modifying input or
non linear effects. However by using precision
mechanical components and high performance
electronic integrated circuits, the above effects
are minimized. They have the advantage that
zero & span adjustment can be made
electronically which is more convenient than
mechanical adjustment.
 Closed loop-Differential pressure
transmitter
As shown in trailing figure a simplified block
diagram of a closed loop differential pressure
transmitter operating on the force balance or more
accurately torque balance principle. The
diaphragm force (P1 – P2)Ad produces a clockwise
torque Ad(P1 – P2) a on the lever arm which is
supported by a clockwise torque Fob due to the
zero spring. The output current I is fed back to coil,
inside a permanent magnet, this produces a force
Kmi on the lever arm
As shown in the figure below a simplified
block diagram of a closed loop differential
pressure transmitter operating on the force
balance or more accurately torque balance
principle. The diaphragm force ‘(P1 – P2)Ad’
produces a clockwise torque ‘Ad(P1 – P2)’ a on
the lever arm which is supported by a
clockwise torque Fob due to the zero spring.
The output current ‘I’ is fed back to coil, inside
a permanent magnet, this produces a force
‘Kmi’ on the lever arm and corresponding
anticlockwise torque ‘KmIb’. Any imbalance of
these torques causes the lever to rotate, the
resulting displacement ‘x’ is measured by a
sensor-bridge-amplifier system. The resulting
change in output current is fed back to the coil
and magnet to adjust the anticlockwise torque
until an approximate torque balance is again
obtained.
Block Diagram of Closed loop Differential
Pressure Transmitter
 Open-loop Differential Pressure
Transmitters
The input differential pressure (P1 – P2) acting
over the area ‘Ad’ of the sensed diaphragm
produces a deflecting force ‘Ad(P1 – P2)’
which is opposed by the elastic spring force
‘kx’. The resulting diaphragm deformation ‘x’
is small and approximately proportional to
(P1 – P2). This displacement can be measured
using a variable capacitance, variable
reluctance or resistive stain gauge sensors.
Another possibility is to use a piezoresistive
sensor consist of a silicon diaphragm. With a
strain gauge element, a d.c. four element stain
gauge bridge is used, the amplifier system
consist of an instrumentation amplifier and
output current regulator. The a.c. bridge is
used along with a carrier amplifier system
consist of an a.c. amplifier, phase sensitive
demodulator, d.c. amplifier and output
current regulator. Span & zero adjustment are
made electronically via the amplifier system.
Block Diagram of Open loop Differential
Pressure Transmitter