Page 1

5. Control structures
___________________________________________
Regulation systems based on the continuous feedback of the error signal,
have a great advantage, since the permanent comparison between the desired value of
the controlled variable and its current value, allows them to correct the effect of
disturbances, even without full knowledge of the characteristics of the
process.
The main disadvantage of feedback control is the need to
that there is an error signal for the controller to act by initiating the correction, it is
In other words, the more delay a process has, the more difficult it will be to control with
the
classic feedback system.
In other words, the feedback control corrects after the variables
disruptors have diverted the controlled from its benchmark.
For the above reasons, other control techniques that are
variants of PID controllers or have different approaches to classical control.
5.1 Cascade Control
Cascade control consists of two feedback loops, one of which is
internal to the other. Its objective is to improve the performance of a control loop
feedback that doesn't work satisfactorily, even though your driver is fine
tuned, due to the slow response of its controlled variable, which enters into
different points of the loop and whose effect on the controlled variable cannot be
detect quickly, impairing controllability.
One of the main applications of cascade control is in processes where the
manipulated variable is a service fluid (such as water or steam) subjected to
disturbances.

Page 2
In cascade control the output of the external or main loop, called the control
master, sets the reference point of the internal or secondary loop, called the
slave controller.
Example: Temperature control of an oven
FIGURE 5.1. Feedback control of a fuel furnace
In the feedback control system of an oven shown in figure 5.1 when
the measured temperature deviates from the set point, the controller varies the
position of the fuel valve, and if all the characteristics of the same (
Pressure. viscosity, etc.) are kept constant, the control will generally be
good.
However, if one of the characteristics, for example pressure, changes shape
sudden, the flow through the valve will follow the same variation even though its
stem remains fixed. Then the temperature will change and, after a certain
time (depending on the time constant and the system delay), the
Temperature variations will reach the controller and the controller will readjust the
position of the
valve according to the actions available to you. A may not be achieved
good regulation and can even totally prevent process control.
Oven
TIC
product
fuel
Controller
from
temperature
Transmitter
Valve
Oven
-
+
product
fuel
PROCESS

Page 3
It should be noted that the temperature is regulated more by the fuel flow than
by the position of the valve (if the fuel quality is constant). Without
However the flow is not controlled, as it is of secondary interest, but its
Variations affect temperature, which is the main variable in controlling the
process.
From the point of view of speed in the control of the process, it would be very
convenient to
quick adjustment of the valve position, as soon as the disturbance occurs
in fuel pressure, while slower temperature variations
that may occur due to other causes must be corrected to maintain the
temperature at set point.
This can be achieved using a cascaded two controller configuration,
one of which acts as primary and the other as secondary.
FIGURE 5.2. Cascade control of a fuel furnace
Tref
Controller
from
temperature
Controller
from
flow
Valve
Oven
-
+
product
fuel
PROCESS
-
+
Transmitter
flow rate
Transmitter
temperature
SECONDARY
PRIMARY
TIC
FC
TT
fuel
valve
transmitter
temperature
Controller
flow
Controller
temperature
Oven
product

Page 4
In this case, the primary controller controls the temperature and acts as a set point.
setpoint of an instrument that controls the flow and whose output signal adjusts the
valve position.
The second controller allows you to quickly correct for flow variations
caused by disturbances in fuel pressure, keeping the system
at all times with the ability to control the temperature on the instrument
primary.
In this arrangement the temperature controller (master) commands and the flow
controller
obedece (slave).
For cascade control to be effective, it is necessary to properly choose the
secondary variable taking into account the disturbances that may arise and
the response speeds of the different components, according to the
following requirements:
1. That the system under control can be divided into two simpler processes, to
close around these the control loops, primary and secondary.
2. That the secondary loop includes the greatest number of disturbances without
reaching
decrease your response time too much.
3. The secondary loop must be faster than the primary loop. Typically p must
be greater than 3 s (time constant of the secondary process).
Some typical combinations are shown below:
PRIMARY
SECONDARY
Temperature
Pressure
Temperature
Flow
Temperature
temperature
level
Flow
Composition
Flow
The advantages of cascade control are several:
1. The disturbances in the internal or secondary loop are corrected by the
secondary controller, before they can affect the primary variable.
2. Any variation in the static gain of the secondary part of the process is
compensated by its own tie.

Page 5
3. The time constants associated with the secondary process are reduced
drastically for the secondary loop.
4. The primary controller is assisted by the secondary controller to achieve a
large reduction in the variation of the primary variable.
Because the secondary loop exists as an element of the primary loop, the
Secondary controller must be properly adjusted before controller
primary, placing it in manual.
The setting for the secondary controller must be made for changes in the reference or
for changes in disturbance if severe changes are expected in this
last.
The golden rule of cascade control should also be taken into account: "If the loop
slave disappears, the master loop must remain stable "
Temperature control in a reactor
If the temperature of the cooling water changes, it becomes a disturbance
whose effect on the main loop can be corrected with a cascade control like
the one shown in the figure:
FIGURE 5.3 Cascade control of a reactor
TIC
TT
TIC
TT
Ref. Temperature
Cooling water
Ref. Temperature
in the tank
CONTROLLER
SLAVE
CONTROLLER
TEACHER
departure from
product
water from
cooling

Page 6
The block diagram for this control is shown below:
FIGURE 5.4. Block diagram of the cascade control of a reactor
Level control of an accumulator tank
In this case, variations in flow can affect the level in the tank before
that the level controller can correct them. For this reason a control is implemented
cascade, in which disturbances in the tank outlet flow are
controlled by the flow controller (slave controller) and the level controls by the
level controller (master).
FIGURE 5.5. Level control in an accumulator tank.
CONTROLLER
TEACHER
CONTROLLER
SECONDARY
VALVE
FROM
CONTROL
FLOW
FROM
STEAM
TEMPERAT.
OF THE
SHIRT
TEMPERAT.
OF THE
REACTOR
TRANSMITTER
OF VARIABLE
HIGH SCHOOL
TRANSMITTER
OF VARIABLE
PRIMARY
Ref.
+
+
-
-
TEMPERATURE
OF THE REACTOR
TEMPERATURE
SHIRT
LT
LC

Page 7
FIGURE 5.5. Cascade control of an accumulator tank.
5.2 Feed-forward control
In this type of control, the information related to one or more conditions that
can disturb the controlled variable, they feed back to minimize the deviation
of the controlled variable.
In systems with significant delay times with deviations of magnitude
duration, the error signal is detected long after it has been
produced the load change, therefore, the corresponding correction is delayed
and it sometimes happens that the controller acts when it is not needed because it has
eliminated the load change that led to the correction.
The anticipatory control (feed-forward) is based on the measurement of one or more
variables
input and acts simultaneously on the manipulated variable that produces the
desired output from the process. With look-ahead control, effects are canceled
undesirable effects of measurable disturbances by compensating for them before they
are perceived in
the exit.
This type of control requires an exact and complete knowledge of the characteristics
statics and dynamics of the process, as well as the way the disturbances
affect the output of the process.
Its design is based on a computer system that has as inputs the signals that
come from the measurement of disturbances and as an output the modification that
LT
FC
FT
LC

Page 8
must be done on the manipulated variable so that the controlled variable does not
deviate
from your reference point. In this way, the disturbing variable enters
simultaneously with corrective action thereby preventing deviation from occurring
would produce in the controlled variable. This fix before the error occurs
gives the name of anticipative to this control action.
In Figure 5.6 you can see a comparison between feedback controls, in
cascade and anticipatory applied to a heat exchanger.
This particular case requires knowledge of the relationship between the flow of the
product and the outlet temperature, the influence of disturbances on the
vapor pressure, in the input product temperature, in the performance of the
exchanger, etc.
That is, the relationship between outlet temperature and inlet flow constitutes
a model of the process and is used to find the transfer function of the
anticipatory control system.
The effectiveness of anticipatory control depends on the precision achieved in the
measurement of
the input variable and the precision of the calculated model. Moreover, you can
be costly or impossible to model the process.
From a strict point of view, anticipatory control can be considered as a
open loop control and its isolated application will result in a significant offset, that is
that is, the output variable (temperature in this case) will deviate significantly
than desired.
to. Feedback control
Exchanger
TIC
TT
Controller
from
temperature
Transmitter
from
temperature
Steam
Entry of
product
Temperature of
departure
Valve
Controller
Transmitter
Process

Page 9
b. Cascade control
c. Anticipatory Control
FIGURE 5.6. Comparison between feedback, cascade and anticipatory control
As a conclusion, it can be stated that the feedback control can control
well in steady state, but not sufficiently under conditions
dynamics of process operation.
On the other hand, anticipatory control is able to quickly govern changes
dynamic, but presents a considerable offset at the output.
Exchanger
FRC
FT
Controller
flow
Transmitter
flow
Steam
Entry of
product
Temperature of
departure
TIC
TT
Controller
flow
Controller
from
temperature
Transmitter
from
temperature
Transmitter
flow
Steam
Temperature of
departure
FIC
FT
Valve
Controller
slave
Controller
teacher
Transmitter
Transmitter
Process
Valve
Transmitter
flow of
product
Controller
anticipatory
Process

Page 10
Main applications :
1- In processes difficult to control by feedback due to the presence of
considerable downtime and delays.
2- In processes that receive uncontrolled flows from other parts of the
plant and that can affect the controlled variable.
3- Processes in which the controlled variable cannot be measured with precision or
continuous mode.
4- Processes in which the controlled variable is not fixed and is determined by others
variables.
Anticipatory controller design:
Consider the block diagram of anticipatory control. It is assumed that the
plant transfer function Gp (s) and the transfer function of the plant
Gn (s) disturbance are known:
FIGURE 5.7. Anticipatory Control Block Diagram
Where:
Gn (s)
: Disturbance transfer function
Gp (s)
: Plant transfer function
Gv (s)
: Valve transfer function
Hff (s)
: Transfer function of the primary detector element and the
anticipatory loop transmitter
Gc ff (s)
: Function of the pre-control counting unit
From the block diagram:
C (s) = Gc ff .Gv.Gp.E (s) + Gn.N (s)
Hff
Gn (s)
Gp (s)
Gv (s)
Gc ff (s)
N (s)
C (s)
R (s)
+
-
+
+
It is)

Page 11
where: E (s) = R (s) - Hff. N (s)
replacing in the previous equation:
C (s) = Gc ff .Gv.Gp.  R (s) - Hff. N (s)  + Gn.N (s)
C (s) = Gc ff .Gv.Gp.R (s) - Gc ff .Gv.Gp Hff. N (s) + Gn.N (s)
C (s) = Gc ff .Gv.Gp.R (s) -  Gc ff .Gv.Gp Hff + Gn  .N (s)
Since we want to eliminate the disturbance, then  Gc ff .Gv.Gp Hff + Gn  .N (s) = 0:
Then the transfer function of the Feed-Forward controller is:
)(
)(
) (G
= Gc
n
ff
sGp
sGv
Hff
s
⋅
⋅
This equation shows that the Feed-Forward controller design requires a
complete knowledge of the process that allows to deduce Gp (s) and Gn (s). The
Gv (s) and Hff functions can be obtained from the specifications of the manufacturers of
instruments.
Advantages of anticipatory control:
- Detects disturbing variables and takes corrective action before the
controlled variable deviates from its reference point.
- Useful for processes with dead time and very slow dynamic response.
Disadvantages :
- It requires measuring all the disturbing variables.
- Requires exact knowledge of the process.
- The model may be physically unfeasible. (If the polynomial of the numerator
of the transfer function of the controller is of greater degree than the polynomial
denominator)
- Does not correct unmeasured disturbances.
- It is insensitive to variations in the parameters of the control loop elements.

Page 12
5.3 Anticipatory control + feedback:
Although early control could lead to perfect control, the disadvantages before
noted makes it advisable to combine anticipatory control with feedback
for it to correct for disturbances in the plant model and changes in the
parameters of the control loop elements.
Temperature control in a heat exchanger
FIGURE 5.8. Anticipatory control + feedback
In this case, if the flow is constant, the signal from the flow controller
temperature passes unchanged to the valve. On the other hand, if there are variations
in flow, the signal from the corresponding Feed Forward controller is
Add or subtract, depending on the direction of variation in temperature.
Exchanger
FIC
FT
Controller
temp.
Controller
derivative
Transmitter
flow
Transmitter
from
temperature
Steam
Entry of
product
Temperature of
departure
TIC
TT
set
point
Transmitter
from
flow
Transmitter
from
Temperature
-
+
+
+
Process
Valve
Controller
from
temperature
Controller
Feed Forward
Dynamic
of the
disturbance
Temperature
from
departure

Page 13
In this way, load changes in the product flow rate are detected and
corrected immediately and compensate for the anticipated changes, which for this
reason,
could occur in temperature.
Feedforward + Feedback controller layout:
FIGURE 5.9. Anticipatory + Feedback Block Diagram
According to the block diagram, it can be shown that the function of
Feedforward controller transfer remains:
)(
)(
) (G
= Gc
n
ff
sGp
sGv
Hff
s
⋅
⋅
5.4 Ratio control (ratio control)
Ratio control is a control system in which a process variable is
controlled relative to another variable in a fixed proportion. Generally the
variables to be kept in a fixed relationship are the flow rates of two
currents, one of which, the uncontrolled one is subject to disturbances
frequent because it comes from the production unit.
Gn (s)
Gv (s)
Gc fb (s)
N (s)
C (s)
R (s)
+
-
-
+
+
+
Hff
Gc ff (s)
Gp (s)
Hff

Page 14
Ratio control can be done by a splitter or ratio station .;
in the first case, a non-linear element is introduced to the control loop so that
it is not recommended. The link station is a gain amplifier
adjustable.
In Figure 5.10. An example of two fluid ratio control is shown:
FIGURE 5.10. Ratio control using a divisor
FIGURE 5.11. Ratio control using a ratio station
Flow
variable
Flow
checked
FY

REF = F A / F B
F B
F A
FT
FT
FY
Flow
variable
Flow
checked
FY
REF = F A / F B
F B
F A
FT
FT
FC
Station
Of relationship

Page 15
While cascade control is a method that improves regulation in a
variable, the ratio control satisfies a specific need, the control of
relationship between two quantities.
The signal from the signal transmitter is multiplied by a factor set manually or
automatically. The multiplier output signal is the set point of the
controller whose output signal acts directly on the control valve
FIGURE 5.12. Ratio control block diagram
Here you have a G C slave controller and the master controller is open, there is no
master tie. For example, you want to have 3 times X or KX, the master is a command
fixed to the slave and the slave is the only control that remains.
This type of control can be considered in degenerate cascade control.
When using the ratio control, keep in mind:
1. Express the flows in the same units
2. the signals reaching the divisions, link station and controller must
have the same linear or quadratic characteristics.
3. Adjust the station ratio taking into account the range of the transmissions
C
1
1
C
C
F
C
F
=
Relationship
∆
∆
⋅
⋅
Where:
F C = manipulated flow ratio
Gn (s)
Gp (s)
Gv (s)
Gc (s)
N (s)
X'
X
+
-
+
+
K

Page 16
F 1 = free or uncontrolled flow ratio
∆C C = Range of controlled flow transmitter
∆C 1 = Range of uncontrolled flow transmitter
Applications:
1. Keep constant the ratio of two streams that mix to guarantee the
composition of the mixture.
2. Maintain an optimal ratio between fuel and air flow rates in a
boiler
3. Store the liquid flow rate at the vapor flow rate (L / V) in a tower.
absorption
Example 1: Ratio control of a mixing system when both are manipulated
flows.
FIG: 5.13 Ratio control of a mixing system
TO
I/P
TO
A/B
B
I/P
B
FY
FY
FY
FY
FY
FY
FY

Page 17
5.4 Override control
The override control is a system that is used to limit the variable of
processes to a high or low value in order to avoid damage to the process, in the
personal or equipment.
For its application it is required to apply control over two variables in a process,
related to each other in such a way that one or the other can be controlled by the same
manipulated variable.
Since a manipulated variable can only be controlled by one variable, it must exist
the possibility of transferring command from one of the control loops to the other when
the
operating complications require it. The transfer of command is achieved
connecting the output of the two controllers to a selector switch of the lowest
or the highest, of two signals whose output is connected to the final element of
control.
FIGURE 5.13. Block diagram of the override control
Two controllers G C1 and G C2 are used in the override control which control
the variables C 1 and C 2 respectively. The outputs of the controllers are connected to
the inputs of a signal selector switch, the switch output drives the
final element of control. Controller 2 reference point subdivides the range
of values of its controlled variable in range of accepted values and ranges of
unacceptable or risk values.
Controller 1 keeps the value of controlled variable 1 at its point of
reference, if the condition is satisfied that the value of the controlled variable 2 is
Gv (s)
Selector
Gc 2 (s)
C 1 (s)
R 2 (s)
+
-
+
+
Gp (s)
N (s)
R 1 (s)
-
+
C 2 (s)
+
Gn 2 (s)
Gn 1 (s)
Gc 1 (s)

Page 18
within the range of non-risk values. For controlled variable 2,
deviations from your reference point if they occur in the no-risk range. If the
controlled variable 2 enters the range of risk values the switch operates by
transfer final element handling to controller 2 to carry its variable
controlled to the no risk range. When this occurs the switch transfers the
Knob switch 1.
Example: Pipeline pumping
In this application there are two pressure controllers, one on the suction and one on the
impulse whose output signal is selected by a communication selector relay
with the control valve.
FIGURE 5.14. Override control in a pipeline
The control is carried out under normal operating conditions in the
impulsion and when due to any fault the pump suction pressure drops
below the safety limit, the speed controller must start operating.
suction instead of drive to achieve this, the suction controller is
reverse acting to the nominal working values and selector relay, selects the
minimum of the two signals that reach it.
In this way, the normal control will be carried out with the impulsion and
when the suction pressure drops too low and falls below its set point.
setpoint, the output signal decreases and becomes lower than the controller output
of impulse, with which the selector relay selects it and the valve becomes
controlled directly by suction pressure.
Aspiration
PIC
PT
Impulsion
Bomb
XY
PIC
PT

Page 19
5.5 Split range control
It is a control system in which there is a single controlled variable and two or more
manipulated variables that must have the same effect on the controlled variable.
To carry out this system it is required to share the controller output signal with
various final elements of control.
FIGURE 5.15. Split Range Control Block Diagram
Example: Temperature control of an electrolytic bath
In these baths the heat produced by the passage of the electric current is removed
by a controlled flow of cooling water.
When it is required to coat large pieces the bath temperature
descends, so coils are used to transport regulated flows of
steam to get you to your landmark.
When the electrolyte solution is at the set point, the flows of water and
steam should be null. The above actions must be performed with a controller
split range, whose output goes to the positioners of the water and steam valves.
Gv 1 (s)
Gc (s)
R 2 (s)
-
+
+
C (s)
+
Gn (s)
Gv 2 (s)
Gp (s)

Page 20
FIGURE 5.16. Starting range temperature control of an electrolytic bath
The following figures illustrate the operation of the controller and valves versus
to disturbances
FIGURE 5.17. Split Range Controller Operation
TRC
T REF
WATER
STEAM
cathode
anode
TT
Electrolytic bath
psig
fifteen
12
9
6
3
Talta
Tlow
exit sign
controller
water valve
100% open
open water valve
open steam valve
exit sign
controlle