Process control
Systems
30 November 2017 By Eng. Hesham A. Selim
AUTOMATIC
PROCESS
CONTROL
30 November 2017 By Eng. Hesham A. Selim
DEFINITION
Any operation that regulates some internal physical
characteristic that is important to a living organism
Example:
natural regulation in humans include body
temperature, blood pressure, and heart rate
30 November 2017 By Eng. Hesham A. Selim
Process Control Variables
DISTURBACES
MANIPULATED CONTROLLED
VARIABLES PROCESS VARIABLES
30 November 2017 By Eng. Hesham A. Selim
Element of process control
The purpose of this control
system is to maintain the
liquid level at some
prescribed height (H)
above the bottom of the tank
Element of process control
1-Process
2-Measurement
3-Evaluation
4-Control
30 November 2017 By Eng. Hesham A. Selim
General Requirements of a
Control System
The primary requirement of a control system is that it be reasonably stable.
1. SYSTEM ERROR
The system error is the difference between the value of
the controlled variable set point and the value of the
process variable maintained by the system
e(t) = PV(t) SP(t)
30 November 2017 By Eng. Hesham A. Selim
SYSTEM RESPONSE
-under damped (cyclic response)
-damped.
30 November 2017 By Eng. Hesham A. Selim
Intuitive Approach to Process
Control Concepts
The practice of process control arose long
before the theory or analytical methods
underlying it were developed. Processes
and controllers were designed using
empirical methods that were based on
intuition (feel) and extensive process
experience. Most of the reasoning involved
was nonmathematical. This approach was
unscientific trial and error, but it was a
successful control method.
30 November 2017 By Eng. Hesham A. Selim
Intuitive Approach to Process Control Concepts
we can obtain the desired temperature, Td.
A further process requirement is that the level of the tank must
neither overflow nor fall so low that it exposes the heater coils.
30 November 2017 By Eng. Hesham A. Selim
On/Off Control
This on/off type of control can be
expressed mathematically as follows:
e = PV SP
30 November 2017 By Eng. Hesham A. Selim
Proportional Control
30 November 2017 By Eng. Hesham A. Selim
Proportional Control (continue)
Proportional control can be described mathematically as follows:
V = Kce + m
where
V = is the control valve position
Kc = is the adjustable proportional gain of a typical process
controller
m = is a constant, which is the position of the control valve when
the system error (e) is zero.
Proportional control can be illustrated by using the three graphs
in Figure 1-12 and setting the proportional constant to three
different values (i.e., Kc = 1, Kc <1, and Kc >1).
30 November 2017 By Eng. Hesham A. Selim
Proportional Control (continue)
30 November 2017 By Eng. Hesham A. Selim
Proportional Control (continue)
This definition can be related to proportional gain Kc by
noting the following equation
PB=(1/Kc) .100
30 November 2017 By Eng. Hesham A. Selim
Proportional Control (continue)
Process response with proportional control
30 November 2017 By Eng. Hesham A. Selim
Proportional & Integral
Control
On what basis should the manual reset
be automated?
30 November 2017 By Eng. Hesham A. Selim
Proportional & integral Control (continue)
After the initial upset the reset action returns the error to zero and there is no offset.
Recognizing that the reset action moves the control valve at a rate proportional
to error, this control mode is described mathematically as follows:
where
dV/dt = the derivative of the valve position
with respect to time (t)
Ki = an adjustable constant
30 November 2017 By Eng. Hesham A. Selim
Proportional & integral Control (continue)
Valve equations:-
30 November 2017 By Eng. Hesham A. Selim
Proportional & integral Control (continue)
One problem with PI control bears
mentioning. If a control loop is using PI
control, the possibility exists with the
integral (reset) mode that the controller
will continue to integrate and change the
output even outside the operating range of
the controller. This condition is called
reset windup.
For example, the heat exchanger
30 November 2017 By Eng. Hesham A. Selim
Proportional & integral Control (continue)
30 November 2017 By Eng. Hesham A. Selim
Reset windup control
30 November 2017 By Eng. Hesham A. Selim
Proportional& Derivative
(PD) Control
30 November 2017 By Eng. Hesham A. Selim
Proportional-Integral-
Derivative Control
-Deciding which control action (i.e., PD, PID, etc.) should be used in a control
system will depend on the characteristics of the process being controlled.
-By understanding these concepts you will be better able to select the proper
control action type for effective control.
30 November 2017 By Eng. Hesham A. Selim
Table 1-1. Guidelines for
Selecting Controller Modes
30 November 2017 By Eng. Hesham A. Selim
What Is Good Control?
Purpose Of Automatic Control
Single-loop Feedback Control
Process Controllers
Time Elements of a Feedback Loop
Sensor Time Response
Types of Control
Combinations of Control Functions
Tuning Control Loops
Ziegler-Nichols Tuning Methods
Open Loop Method I.E Process Reaction Curve Method :
Advanced Control Loops
30 November 2017 By Eng. Hesham A. Selim
Purpose Of Automatic Control
The purpose of automatic control is
to provide a means by which any
process operating condition within a
production system can be maintained
in a stable and consistent manner.
30 November 2017 By Eng. Hesham A. Selim
Single-loop Feedback
Control
In a feedback control
loop, the variable to be
manipulated is
measured. This
measured process value
(PV) is then compared
with a set point (SP) to
generate an error signal
(e = PV - SP).
30 November 2017 By Eng. Hesham A. Selim
Process Controllers
There are three types of
controllersmechanical,
pneumatic, and electronic
and they all serve the same
function.
The four common controller
functions are proportional,
proportional plus integral (PI),
proportional plus derivative (PD),
and proportional plus integral
plus derivative (PID).
30 November 2017 By Eng. Hesham A. Selim
Single-loop Feedback
Control
the sensor must be converted into the
same units
30 November 2017 By Eng. Hesham A. Selim
Time Elements of a Feedback
Loop
The time of response of the control loop is the
combination of the responses of the sensor, the
transmitter, the controller, the final control
element, and the process.
it is necessary to understand the concept of
time delays or lags in process control
systems.
30 November 2017 By Eng. Hesham A. Selim
Time Lags
In process control, the term lag means
any relationship in which some result
happens after some cause.
We will start our discussion of time
response and time lag with sensor time
response.
30 November 2017 By Eng. Hesham A. Selim
Sensor Time Response
where
mi = the initial sensor output
measurement
mf = the final sensor output
value
= the sensor time constant
If mi=0
30 November 2017 By Eng. Hesham A. Selim
Sensor Time Response
(continue)
If we wish to find the value of the output exactly
seconds after a sudden change occurs, then
Thus, we see that one time constant (1)
represents the time at which the output value
has changed by 63.2 percent of the total
change. If we solve the Equation for time equal
to 5 or five time constants, we find that
This means that the sensor reaches 99.3 percent
of its final value after five time constants
30 November 2017 By Eng. Hesham A. Selim
Sensor Time Response
The following example illustrates a
typical sensor response application.
The first-order lag is the most common
type of time element encounteredin
process control
30 November 2017 By Eng. Hesham A. Selim
Sensor Time Response (continue)
The system response is called a first-order lag
because the output lags behind the input, and the
differential equation for the system shown in
Figure is a linear first-order differential equation.
Laplace transform method
30 November 2017 By Eng. Hesham A. Selim
Response of a first-order
system to step
30 November 2017 By Eng. Hesham A. Selim
Dead Time Lag
td = t 1 - to .
In this process, the maximum rate of change for the output does not occur
at time zero (to) but at some later time (t1). This is called dead time in
process control:
Dead time is the most difficult condition to overcome in process control.
During dead time, there is no process response and
therefore no information available to initiate corrective action.
30 November 2017 By Eng. Hesham A. Selim
Dead Time Lag(continue)
EXAMPLE
Problem: Determine the dead time for the process shown in Figure 1-9 if the
temperature detector is located 50 meters from the heat exchanger and the
velocity of the process fluid in the discharge pipe is 10 m/s.
Solution: The dead time is given by td = D/v. Since D = 50 m and v = 10 m/s
30 November 2017 By Eng. Hesham A. Selim
Types of Control
There are four basic types of control:
Two-step action,
Proportional action,
Integral action,
Derivative action.
30 November 2017 By Eng. Hesham A. Selim
Combinations of Control Functions
proportional only (P).
proportional plus integral (P + I),
proportional plus derivative (P + D),
proportional plus integral plus derivative (P + I + D).
Deciding which control action (i.e., PD, PID, etc.) should be used in
a control system will depend on the characteristics of the process
being controlled
30 November 2017 By Eng. Hesham A. Selim
Guidelines for Selecting Controller
Modes
There are three important factors to consider when
tuning a PID controller:
1-The characteristics of the process,
2-The selection of controller modes,
3-The performance criteria of the control loop
30 November 2017 By Eng. Hesham A. Selim
Tuning Control Loops
What Is Good Control?
the system's response curve has an amplitude ratio or decay ratio of one-quarter.
A decay ratio of one-quarter means that the ratio of the overshoot of the first
peak in the process response curve to the overshoot of the second peak
is four to one
30 November 2017 By Eng. Hesham A. Selim
What Is Good Control?
In some cases, it is important to tune the
system so that there is no overshoot; in other
cases, a slow and smooth response is
desired; some cases warrant fast response
and significant oscillations are no problem,
etc. The point is you must determine what
control is good for each specific loop.
30 November 2017 By Eng. Hesham A. Selim
The Tuning Concept
The feedback controller is only one
piece of hardware in the entire loop;
there are many other hardware items
connected to form the balance of the
loop. For the purposes of adjusting
the feedback controller, it is
convenient and sufficient to view
everything else
30 November 2017 By Eng. Hesham A. Selim
The Tuning Concept
why controllers cannot be preset at the factory but.
instead, must be individually tuned for individual
loops.
Each of the individual blocks around the feedback
loop represents an algebraic or differential
equation, i.e., it represents a mathematical
statement for that particular piece of hardware
Since all of these blocks are coupled together, all
of these equations represent a simultaneous set
of mathematical equations. If you are able to
determine what you think good control is.
30 November 2017 By Eng. Hesham A. Selim
Advanced Control Loops
Cascade Control Loops
Ratio Control Loops
Override Control
30 November 2017 By Eng. Hesham A. Selim
Cascade Control Loops
In cascade control in Figure
the output from one controller
, the master, is the set point for
another controller commonly
referred to as the slave.
The master will have an independent
plant measurement.
30 November 2017 By Eng. Hesham A. Selim
Ratio Control Loops
ratio is maintained between two or more variables.
Ratio control is often associated with process operations
in which two or more streams must be mixed together
continuously to maintain a steady composition in the resulting mixture
30 November 2017 By Eng. Hesham A. Selim
Override Control
30 November 2017 By Eng. Hesham A. Selim
Ziegler-Nichols Tuning
Methods
This gain is called the
ultimate gain or sensitivity (Su).
The period of these sustained oscillations
is called the ultimate period (Pu).
30 November 2017 By Eng. Hesham A. Selim
Ziegler-Nichols Tuning
Methods
P CONTROL
30 November 2017 By Eng. Hesham A. Selim
Ziegler-Nichols Tuning
Methods
EXAMPLE
Problem: The Ziegler-Nichols ultimate method was used to determine an
ultimate sensitivity of 0.3 psi/ft and an ultimate period of 1 min for a level
control loop. Determine the PID controller settings that are needed for good
control.
Solution: Using the equations for PID control,
30 November 2017 By Eng. Hesham A. Selim
OPEN LOOP METHOD
i.e Process Reaction Curve Method
30 November 2017 By Eng. Hesham A. Selim
OPEN LOOP METHOD
i.e Process Reaction Curve Method
30 November 2017 By Eng. Hesham A. Selim
Thank you.
Questions ?