Automatic Control System and Control Loops
Automatic Control System and Control Loops
For reliable operation of a process or system there are certain parameters, which
needed to be monitored . If there is any deviation in these parameters than a
corrective action has to be taken so that it does not affect the stability of the process
or system .
For this we need a control system to control these parameters with in a safe limit .
This control can be manual or automatic.
Manual control is having a possibility of human error . To avoid the human error we
use Automatic control system.
Automatic control is used in Power stations for safe and efficient operation of the
plant.
The type of automatic control used varies from Simple Pneumatic control systems on
auxiliary plant to Microprocessor based control systems on main plant loops.
Although the two systems are much different in their complexity, to understand either
system we must first understand basic automatic control theory.
3
CONTROL SYSTEM TERMINOLOGY
5
CONCEPT OF CONTROL SYSTEM
6 1
5 Automatic controller 2
When there are any disturbances in a process the out put of the process
deviates from its actual value. This causes instability in the process. To over
come these deviations we need to know how much is the deviation and than
take the corrective action.
Deviation (Error) = Set point – Measured value.
Measuring unit measures the Present value of the output or the controlled
condition.
Comparing element compares it with Set point and generates an error signal
(deviation).
Controlling element generates a control signal with respect to Error signal.
7
This control signal is fed to the Actuating Unit. The Actuating Unit changes
the Regulator position in accordance to the control signal.
Thus by changing the regulator position we can remove the error generated
in the output of the process or the controlled condition.
Process in now disturbance free and stable on removal of error.
Thus control system is very much essential for making a process or system
stable and error free.
8
TYPES OF CONTROL LOOPS
Disturbances
Regulating
Control unit Manipulated Plant Controller
signal variable condition
9
OPEN LOOP CONTROL SYSTEM
10
CLOSED LOOP CONTROL SYSTEM
Error
E(s) Amplifier
Input +/- Output
(Gain A)
R(s) C(s)
Feedback
F(s)
Feedback
(Gain B)
In close loop control system Amplifier gives an output which is equal to Gain of
amplifier and Error signal.
11
This Output is feedback to compare it with the input or set point.
The Resultant of comparing the set point and feedback signal gives error.
C(s)= A * E(s)……….i
F(s)=B * C(s)……… ii
E(s)=R(s)-F(s)……...iii
on solving these equations :
C(s)/R(s) = A/(1+A*B)…..iv
Thus by comparing the output with the set point we can easily reduce the
error to zero.
12
Control
signal output
Error
Input comparator controller # Regulating plant
( set-
point) unit
Disturbances
Auto / manual control
Measuring
Visual Unit
indication
DISCONTINUOUS ACTIONS :
•Two Step Actions: The output has two positions e.g. on-off or high-low or forward-
reverse. If the deviation is positive the output is one. if negative it is the other i.e.
Zero.
•Two Step Action with Overlap : If two step action causes frequent changeover of
the correcting unit, excessive wear may take place. This can bet reduced by
introducing an overlap.
•Multi step Action : This is similar to two-step except that the controller output has
several positions. An example is the automatic control of voltage by means of a tap-
changer on a transformer.
•Floating Action : The controller output changes when the deviation changes sign
and continues to change at a steady rate until the deviation again changes sign. An
example is a valve operated by a motor, which in turn is controlled by a reversing
switch. The switch will change over when the deviation changes sign.
14
• Two step, multi step and floating action are cheap forms of control and are used where some
variation of the controlled condition can be tolerated.
• On main power station plant loops hunting, sustained oscillation of the controlled condition,
Cannot be tolerated. Other forms of control must be used and these are referred to as
continuous control actions.
CONTINUOUS ACTIONS:
1. Proportional Action:
• The action of a controller whose output changes by an amount that is proportional to the
deviation.
• The amount by which the output changes for a fixed deviation is adjustable so that the
controller response can be adjusted to suit the particular control loop.
• This adjustment is referred to as proportional band.
Proportional wand (P.B.) : The change in deviation required causing the output
to change from one extreme to the other.
15
•
To explain these actions it is necessary to look at a practical example of a typical
controller and control loop. A typical pneumatic controller will receive two inputs :
•Desired Value: This is normally a red pointer which can be positioned at any point ran
the chart or scale of the instrument containing the controller. It also applies an input to
the controller as the desired value.
•Measured Value : This is normally a black pointer or pen, which is positioned on the
scale or chart and indicates the measured value of the controlled condition. It also
applies an input to the controller as the measured value.
•A typical pneumatic controller will give an output that will vary between 3 p.s.i.g. and
15 p.s.i.g. (0.2 to 1 Bar) if we consider a proportional only controller and there is no
difference between the desired and measured values (zero deviation) it is normally
arranged that the output pressure is 9 p.s.i.g.which is the mid point of the 3 - 15 p.s.i.g.
output range.
•Depending on the control loop the controller can be adjusted to be either direct or
reverse active.
16
• Direct Acting :If the measured value increases the controller output also
increases.
• Reverse Acting : If the measured value increases the controller output
decreases.
P.B. = 100
GAIN (Kp)
• OFFSET : Offset is the name given to a sustained deviation between desired and
measured values.
• Offset will occur on a control system controlled by the proportional only controller
following a-load change on the control system.
17
(Temperature)
For the P-action a “Proportional band” o
C P-Band PV
(PB) is made around the Set point. Process Value
95oC
Lets use as an example PB=10oC, this 92.5oC
SV
o o
means a “P-band / PB” from 85 C to 95 C.
90oC
If the Process Temperature changes the
Setpoint: Set Value
following will happen: o 87.5oC
90 C
•If the Temperature is below the PB the 85oC
TC output will be (100%) ON.
•If the Temperature is above the PB the
TC output will be (completely) OFF.
•In the P-band the output capacity of the
controller will be reduced from 100% to
0%. This is done Proportional to the TC
measured Temperature. This means: Output
1.1.At 85.0oC the output will be on for
100%
2.At 87.5oC the output will be on for
75% ON Proportional control OFF
3.At 90oC the output will be on for 50%
4.At 92.5oC the output will be on for
25% This means:
o
5.At 95 C the output will be on for 0% Depending on the deviation of the Set point the ouput will be
controlled PROPORTIONALLY
18
• The amount of offset can be reduced by decreasing the proportional band.
• There is however a limit to the lowest value of proportional band that can be used
for a particular loop. This limit is when the loop becomes so sensitive that
continuous oscillations of the measured value occur . This is called as Hunting.
• Sensitivity of the controller is inversely proportional to Proportional Band.
• Proportional action therefore gives us a continuous control action which will not
result in continuous oscillations of the controlled condition if used correctly.
• But the problem with purely proportional action is that offset will occur if the load
on the system changes.
• Output of Proportional controller = Kp * Error(deviation)
19
Effect of Kp on the response
20
2.Integral Action (I): The action of a controller whose output changes at a rate
proportional to the amount of deviation.
•The rate at which the output changes for a fixed deviation is adjustable to suit the
particular control loop. This adjustment is referred to as the Integral Action Time.
•Integral Action Time (I.A.T.): The time taken for the integral action to change the
output by the same amount as the proportional action.
•Due to this integral action Offset error is removed from the system but the response of
the system becomes slow.
•The response is slow because of the integral action time.
•As long as the deviation remains the output will continue to rise due to integral action.
•This slow response of the controller makes the system unstable.
21
Effect of Ki:
Control system using only a Ki term.. Ki = integral Gain.
in the figure, even for very small Ki, the system is difficult to
stabilize
22
3.Derivative Action : The action of a controller whose output changes by an
amount, which is proportional to the rate of change of deviation.
•The amount by which the output changes for a fixed rate of change of deviation is adjustable to
suit the particular control loop. This adjustment is referred to as the derivative action time (D.A.T.)
•Derivation Action Time (D.A.T.) : The time taken for the proportional action to reproduce the
same change in output as derivative action.
•In Derivative action:
•Output of the controller = Kd * Derivative of the Error(deviation)
•In derivative action type control, controller works always in anticipatory mode. This means it
always searches in advance if there is any error in the output.
•This type of action increases the response speed. This action is suitable to use in systems
which are having large time delays present in the system.
•The draw back of this system is that it can not remove the offset.
23
4.(Proportional + Integral ) Action: When we are employing Proportional
action along with Integral Action , We can reduce the offset error to zero.
•Proportional controller works up to the proportional band and increases the regulator
position to reduce the deviation.
•Beyond proportional band the control is done by Integral action . The integrator
Checks the error and continuously increases the regulator position to the extreme so
that the Offset error is reduced to zero.
Temperature
• Overshoot Offset
95oC
Undershoot
o
90 C
85oC
Integral time
Activation of:
Time
I-Action
24
• With the correct setting of the PB, it is normal to get an “Overshoot”, followed by
an “Undershoot“ and than “Stabilization” with an “Offset” from the Set point.
• Making the Integral time shorter will give more intense control with a quicker
response to eliminate the offset. But a too short Integral time would result in
“oscillation” (hunting)
• Making the Integral time too long will reduce the possibility of hunting but will slow
down your overall Process response. So the RIGHT setting is very important.
• Integral action is used in conjunction with proportional, action to remove offset. As
we have seen proportional action by itself gives rise to offset due to load changes.
Using integral action, as long as a deviation exists (Offset) the controller will cause
its output to change at a rate, proportional to the deviation, until it is removed
• Output of the controller = Kp * E(t) + Ki * Integral of E(t).
• E(t)= Error signal
• Thus in proportional + Integral action the control is greatly influenced by the
integral action (IAT).
25
Kp and Ki combined:
The combined Ki and Kp terms tend to improve the
slower Kp responses from before but does less to
improve the stability of the faster Kp response.
26
The setting of the right I-Time is very important !
°C
140
PV @
I=80s
80
A too long I-Time slows
down the whole Process
60
Example of
behaviour after a PV @
I=38s
disturbance
The RIGHT I-Time will 40
enable the TC to reach
the Set point quickly and
20 PV @
to eliminate the Offset I=20s
correctly.
0
27
5.Proportional + Derivative action : In proportional + derivative action the
proportional controller changes the regulator position to remove the error but still
the Offset is present in the output.
•The derivative action only enhances the speed of response of the system.
•Derivative action also increases the stability of the system .
•In (P + D) action :
Output of the Controller = Kp * E (t) + Kd * Derivative of E (t).
As we can see that P+D only increases the response of the system .Offset error is
still present in the system. This type of controller is good to use where we are more
concerned about the Lags present in the system.
28
100% With LONG D-Time
Output
Capacity
50%
0%
Time
When the Process Temperature suddenly drops, the Output capacity
is increased very rapidly
29
6. Proportional+Integral+Derivative action:
•When we get a “Disturbance” in our process the PI controller responds “rather
slow”
•That is exactly why the D-action is added: The D-Action looks at the speed of
change and if the output drops rapidly, it increases the output capacity of the
controller very rapidly.
In PID Control action:
•Output= Kp* Error +Ki * Integral of Error + Kd * Derivative of Error
•Thus in PID Control the error in the system as well as the speed or response of
the system both are improved .
•For proper control through PID we need to select the values of Kp , Ki and Kd
appropriately so that :
1.No Hunting in system
2.Error is Reduced to Zero
3.System stability is increased
30
PID CONTROLLER BLOCK DIAGRAM
31
EFFECT OF Ki ON PID RESPONSE
32
Effects of increasing parameters
Small
Kd Decrease Decrease None
Decrease
33
COMPARISON
Action Advantages Disadvantages
On-Off Cheap & simple Controlled condition will
oscillate around set point
On-Off with overlap Cheap & simple Controlled condition will hunt
at a lower frequency but
greater amplitude
Proportional Can give steady control Offset. Too Narrow PB can
cause hunting
Integral Elimination of offset Long recovery time , slow to
respond to rapid changes
Derivative Can anticipate what the Offset
controlled condition is
going to do
P+I Elimination of Offset Recovery time longer than
with P alone
P+D Shortens recovery time Offset still present
over p alone
P+I+D Offset removed. Recovery
time short. 34
SELECTION OF CONTROLLER
36
LPH-1 LEVEL CONTROL LOOP
37
LPH-1 Level Control Loop explanation
.
38
LPH-2 CONTRO LOOP
39
LPH-2 Level Control Loop explanation
40
LPH-3 CONTROL LOOP
41
LPH-3 Level Control Loop explanation
43
HPH-5A Level Control Loop Explanation
44
HPH-6A LEVEL CONTROL LOOP
45
HPH-6A CONTROL LOOP EXPLANATION
46
HOTELL LEVEL ONTROL LOOP
47
HOTELL Level Control Loop Explanation
48
DEAERATOR LEVEL CONTROL LOOP
49
DEAERATOR LEVEL CONTROL LOOP
50
DEAERATOR Level control loop
Explanation
51
DEAERATOR Level control loop
Explanation
• The condensate flow signal along with extraction steam flow and HP heaters 5A
& 5B drain flow to deaerator is a feed back signal in the control circuit which shall
confirm that new demand is actually being supplied. This level signal plays a
minor role and trims the control signal to maintain required level.
• When deaerator level rises 100 mm above high level set point, it opens the
deaerator overflow control valve DRV-48 to dump the excess condensate from
deaerator to LP drain flash tank. The overflow valve is closed when the level falls
100 mm below the high level set point. Maximum passing capacity of the
overflow valve DRV-48 shall be about 10% BMCR, i.e. 170 T/Hr.
• If the level reaches high - high level set point than impulse is generated :
i) To close motorised isolating valve EXV-13 on IPT steam extraction to deaerator line.
ii) Repeat command for opening of deaerator overflow valve DRV-48.
iii) Over-riding command for closing of deaerator pegging valves ASV-8/CRHV-6 & their
bypass valves ASV-10 / CRHV-8 on steam supply line to deaerator from low temperature
aux. steam header/cold reheat line.
iv) Closing of HPH-5A & HPH-5B drain control valves DRV-15 & DRV-22 to deaerator,
respectively.
52
FURNACE PRESSURE CONTROL LOOP
53
FURNACE Pressure Control Loop
Explanation
• Furnace Pressure control is required for efficient operation of Boiler. The Set
Point of the pressure is normally kept at -5mmwc (-5mmwc to -10mmwc).
Three Pressure transmitters are employed for the measurement of the
Furnace Pressure. Individual transmitter signal is passed through a first order
filters to remove the high frequency noise present in the signal. The steady
and correct value is selected using selection logic.
• A PI controller is employed which runs a control algorithm to generate
appropriate control signals. The controller output is applied to the final control
elements i.e. ID fans A and B regulating dampers. A position transmitter
monitors the precise operation of the dampers.
• If the measurement signal or the process value is more than the Set Point,
then PI controller generates more output and consequently ID fans regulating
dampers will open further to let out more flue gases to bring back the furnace
pressure near set point. If the furnace pressure drops then PI controller
produces output to regulate the ID fan dampers in appropriate direction.
• For better controls during load changes “Air Flow Demand” signal is employed
as a feed forward action. An increase in air demand will open damper and
maintain the furnace pressure in a better way.
54
BOILER DRUM LEVEL CONTROL LOOP
55
DRUM Level Control Loop Explanation
• Drum level control is important for smooth operation of any thermal power
generating plant. Now only three differential pressure transmitters are used for
measurement of drum level, these signals are compensated for variations in
density of drum water, drum steam and water in reference column.
• Compensated drum level is compared with drum level set point which is set by
the operator. Steam flow and feed water flow are also measured using
differential pressure transmitters and properly compensated.
• The difference of steam flow and feed water flow are added with the difference
of drum level set point and drum level actual (level compensated). These output
signals is a three element error. This 3 element error is fed to a PID controller
for generation of appropriate output for proper control of drum level. Now it is
clear from the figure that controller will try to nullify the three elements to zero
• If the steam flow and feed water is same and drum level set point and drum
level actual is also same, then three element errors is zero.
• Now if drum level actual is less than the set point then the set point three
element error becomes positive and PID controller will generate more output
and consequently control valve will open, admit more water flow and raise the
drum level to the set point value. If drum level is more than set point operation
will be in a reversed manner. 56
DRUM Level Control Loop Explanation
• When turbine load is increased by operator, steam flow will increase to meet
the required load demand, hence the difference between steam flow and feed
water flow will be positive; hence PID controller will generate more output to
admit more feed water and three element errors will become zero and drum
level will remain stable. In case load demand decreased by operator the same
process will happen in reverse manner.
57
FEED WATER CONTROL STATION DP
CONTROL LOOP
DPT DPT
1 2
Tx Selector
-
+
SP
PID
BFP SPEED
# HMI
E/H
ZT
58
FCS DP Control Loop Explanation
• The drum level loop on its own cannot maintain the drum level until the
differential pressure across the Feed Control Valve is maintained by the Feed
Control Section differential pressure controller.
• Two differential pressure transmitter measures the differential pressure across
the Feed Control Valve, the final value is selected using selection logic.
• The set point of the differential pressure across the FCS is kept at 8kg/cm2.
The process value is compared with the set value and the error is fed to the
PID controller for controlling the speed of the Boiler Feed Pumps, which in turn
will control the discharge pressure of the Boiler Feed Pumps.
• If the FCS differential pressure is less than the set point, then the PI controller
generates increasing output to increase the speed of BFP’s which in turn will
increase the discharge pressure of the BFP’s and bring back the differential
pressure near the set point.
• In case the FCS differential pressure is more then the loop behaves in reverse
manner.
59
Thank
You