Process Control

Method

Muhammad Waqiuddin Bin Tajudin

Awang Lutfan Bin Haji Awang Lamat
Iskandar Zulkarnain anak Tuah
Mohammad Hazwan Bin Idris
Muhamad Abdul Raziq Bin Abdullah
Table Of Content

● Introduction
● Open-Loop Control
● Closed-Loop Control
● Single variable control loop
● Type of Control Modes
● Tuning the controller
● Advanced Control Techniques
Introduction
● Many different operations are performed in an industrial machine to manufacture a
product.
● For example, fluids flow through pipes at a certain rate, ingredients fill a vat to a required
level, heat is applied to a vessel to cause a chemical reaction, or a vacuum pressure is
applied to a confined tank to extracts its contents.
● Each one of these operations is referred to as a process.
● Many of these individual processes are combined and run simultaneously to produce a
finished product of a desired quality as rapidly and inexpensively as possible.
● To satisfy these requirements, each process must be precisely controlled, often by some
type of automatic control device.
● The automatic operations performed by an industrial manufacturing machine are
referred to as process control.
Example of Process
Open-Loop Control
● Process control operations are performed automatically by either open-loop or closed-
loop systems. If the process is controlled only by setpoint commands, without feedback
measurements signal, the system is referred to as open-loop system.
● Open-loop control is used in application where simple processes are performed. Timing
functions are often the key factor used to control the operation.
● Examples of an open-loop process machines are cafeteria dishwashers, commercial
laundry machines, and printed circuit board burn-in chambers.
● This equipment runs through a series of timed cycles, which are activated by controller
devices such as relay ladder logic hardware, sequential drum controllers, programmable
controllers, or computers.
● The advantage of open-loop system is that they are relatively inexpensive. Their main
drawback is that without a feedback loop, there are no control capabilities to make
corrections if the process deviates from its required state.
Example of an open-loop system
Closed-Loop Control

● A closed loop control system is a set of mechanical or electronic devices that

automatically regulates a process variable to a desired state or set point without human
interaction.
● Closed-loop automatic control systems are more effective than open-loop systems. With
the addition of a feedback loop, they become self-regulating.
● Closed loop control systems are widely used in various industry applications including
agriculture, chemical plants, quality control, nuclear power plants, water treatment plants
and environmental control.
Example of a closed-loop system
● The operation of a temperature-type process control closed-loop system.
● It shows a heat exchanger, which is used to heat a liquid to a temperature of 100°F.
● A source of system supplies thermal energy to the exchanger.
● The amount of steam that passes through a control valve determines the temperature at
which the liquid is heated.
● The primary element (sensor) is used to detect the condition of controlled variable,
which is the temperature of the liquid leaving the exchanger.
● The sensor’s output, which is called the measured variable, is conditioned by a
transducer/transmitter into a standard signal before it is sent to the controller.
● The controller compares the feedback signal to the set point, and an error signal is
developed if there is a difference.
● The controller uses the error signal to make computations to determine which type of
control signal to produce at its output.
● The control signal is sent to the final control element (which is the control valve).
● This valve varies the steam flow into the exchanger.
● The steam is the manipulated variable that causes a change in the controlled variable.
● These actions of measuring, comparing, computing, and correcting go on continuously.
Process Behavior

● The primary objective of process control is to cause a controlled variable to remain at a

constant value at or near some desired set point.
● The term variable refers to the fact that an element varies when an influence to which it
is exposed causes the variable to change.
● A change can happen when only one the following conditions occurs:
1. A disturbance appears
2. Load demands vary
3. Set points are adjusted
● When a change does occur, the objective of the process control system is to return the
controlled variable to the set point as quickly as possible.
Example of process behavior
● The behaviour of a process system can be examined by observing the controlled variable
response after one of these influences abruptly changes.
● Referred to as a step change, it takes place over a small-time interval and is plotted on
time graph (Figure 15-2(b)) as a vertical line.
● Figure 15-2(a) shows how step changes develop in an actual application.
● The flow of fluid through a pipe system is the process.
● The fluid flow rate leaving the valve is the controlled variable.
● The position at which the valve is set is considered the set point.
● The flow rate governed by the position of the flow restrictor inside the valve is the
manipulated variable.
● The demand for fluid downstream from the valve is the load, and the variance in
upstream pressure is considered the disturbance.
● Figure 15-2(b) is a time graph that shows how a set point change, disturbance and a load
variance affect the controlled variable.
● Flow, the controlled variable, is plotted on top by a measurement device such as a strip
chart recorder.
● At time 1, the flow rate increases when the valve position is opened wider, while the
upstream pressure remains constant.
● At time 2, with the valve position unchanged, the flow rate decreases because the
upstream pressure drops.
● The flow rate would decrease in a similar fashion if the demand for fluid increased
downstream, while the upstream pressure and the valve position remained constant.
Single Variable Control Loop

What is single variable control loop?

● The act of controlling one specific variable within a process through the use
of either an analog or digital control method.
Element of a single variable control loop

A single variable control loop consists of 4 elements, which are:

● Measuring device.

● Transducer/Transmitter.

● Controller.

● Final control element.

● These elements can also be referred to as instrumentation devices.

● The dynamic response of the control loop can be defined as a measure

of corrective action of the loop as a function of time to the deviation.

● When a step change occurs, the change is not immediate and any
corrective actions made by the loop takes time.
Response time of the instrument

● All instruments have a time lag.

● Time lag can be defined as the time duration in which a change is received at

the input until an output response is produced.

● Time lag also includes the time it takes for a signal to pass from one instrument

to the other.
There are 6 factors which contribute to instrument time delay, which are:

● Response time of a sensor.

● Time lag of a transducer.
● The distance a signal must travel from transducer to the controller.
● The time lag of final control element.
● The time required for the controller to process information
● The distance a control signal travel from controller to final control
element
Pure lag of the control variable

● The control variable may contribute to the delay in the loop.

● When a step change occurs, the loop reacts by altering the manipulated variable.

● This results in an increase or a decrease in the energy applied by the actuator.

● However static a change is opposed by the static inertia of the controlled variable

creating a delay.

● Over time the energy will overcome the resistance and produce the desired state.
● This delay reaction is referred to as pure lag.
● A common method to analyse pure lag is to introduce a
step change and observe the response.
DEAD TIME
Dead time is the time between the instant a deviation of the controlled variable occurs and
the instant corrective action begins.
Ways to reduce time reaction in a system

There are 2 ways to reduce reaction time of a control loop:

1. Select a controller with features that provide the appropriate control action for a
particular process.
2. Properly tune the controller to optimize the process regulation.
Factor which affect pure lag
The 4 main factors which can affect pure lag:

● The capacity (physical size) of the controlled variable.

● The physical properties of the controlled variable.

● The chemical properties of the controlled variable.

● Dead time, which is the time between the instant a deviation of the
controlled variable occurs and the instant corrective action begins.
Control Modes:

● On-Off
● Proportional
● Integral
● Derivative
On-Off

● The type of control mode often used for slow acting operations.
● This kind of action controls a final control element that has only two conditions,fully on or fully off the controller
cannot move the final control element to any intermediate position between the two extremes.
● The advantages of the on-off control are that it is the least expensive closed-loop system and the easiest to
design.
● Its limitation is that it cannot vary the controlled variable with precision.

Figure 15-5(a) Figure 15-5(b)

On-Off

● The controller compares the temperature (controlled variable) to the set point. When the temperature increases
above the set point, the controller turns a compressor (final control element) fully on. As the temperature lowers
below the set point, the compressor is turned off. The controlled variable cycles above and below the set point, as
shown in Figure 15-5(a).
● The drawback of this system is that the rate at which the final control element switches on and off is very high.
This condition can result in excessive wear to equipment. To reduce the switching rate, an on-off differential, or
hysteresis, is programmed into the controller. Also referred to as a deadband, it causes the controller to produce
its on and off signals at different values around the set point.
● For example, a home heating thermostat may have a deadband of 4 degrees. If the temperature setting is 70
degrees, the furnace turns on at 68 degrees and turns off at 72 degrees, as shown graphically in Figure 15-5(b).
The drawback of using a deadband is that the controlled variable will deviate from the set point by a larger amount
than a system that does not use this method.
Proportional Mode

● A proportional controller produces an output signal with a magnitude that is proportional

to the size of the error signal it is correcting.

● The error signal is the difference between the measured variable and the set point.

● A small error will cause the output to change by a small amount and a large error will
cause a larger output change. The output of the proportional controller moves the final
control element to a definite position to attain a desired value of the controlled variable.

● The proportional action can be accomplished in two different ways, by the time
proportioning method, and by the amplitude proportional method
Time Proportioning

● The output at the controller is continually switched fully on and fully off, the average voltage produced is varied by
changing the ratio of signal on to signal off.
● The ratio produced by the controller is determined by how much the measured variable differs from the set point

At set point

Below set point

Above set point

Amplitude Proportional

● The magnitude of the signal is proportional to the size of the error signal. The analog signal produced by a
controller is either a variable voltage or a variable current.
● The controller also has the capability of amplifying the amount at which its output changes in proportion
to the change applied to its input.
● There are two ways to refer to the amplification of a proportional controller:
● Proportional gain which is the ratio of change in output to the change in input
● Proportional band is the percentage change in the controlled variable that causes the final control
element to go through 100 percent of its range
Integral mode
● Integral control mode is a mode which allows the PV(process variable) to return to SP(set point).
● This action is accomplished by repeating the proportional mode so long as the error is still present.
● The main purpose of the integral mode is to eliminate offsets or steady state error.
● The integral mode accomplishes this task by integrating the error over a period of time intervals.
● This process can be defined by the formula:
● KC/TI ∫edt
● KC = Controller gain
● TI = integral constant or reset time
● t = Time
● The integral control action will occur until no errors exist, as the function occurs over a period of time.
● The integral action can also be expressed as:
● Repeats per minute = the number of times the proportional action is repeated per minute.
● Minute per repeats = the number of minutes it takes for a single repeat.
Derivative mode
● Though integral mode is effective at eliminating offset, it is far slower then proportional mode.
● A faster mode compared to proportional mode is derivative mode (rate mode).
● The main function of derivative mode is to speed up controller action, compensating for delays in the
feedback loop.
● It can be expressed by the formula:
● KC*TD*dERR/dt
● Where TD is the derivative or rate time.
● The derivative mode only acts when the error change with time.
● An advantage of derivative mode is that it can reduce the time taken to return PV to SP, however it can
also amplify the affects of noisy signals.
Tuning the controller

● to minimize the dynamic response in a process system.

● optimal balance of mode adjustments is achieved, the system will minimize the size of
the initial deviation and return the controlled variable to set point as quickly as possible
if a disturbance or load change occurs.
Tuning the controller

● 1. Trial and error tuning method.

● 2. ziegler -Nichols Tuning method.
● 3. continuous cycling method.
● 4. Reaction curve method.
● 5. Direct synthesis method.
● 6. Controller autotuning.
 Tuning the controller

Important step that need to be considered:

1. Study the diagram control loop.

2. Obtained proper clearance for tuning activities.
3. Confirm that each component are working properly.
 Trial and error tuning method

Steps:

● Place the controller at manual mode.

● Turn the derivative mode off.
● Adjust the integral mode to minimum gain setting.
● Place the controller in the auto mode.
● Increase proportional mode until a cycling action start begin.
● Adjust the setting into one half value. This should stop the cycling.
● Produce a step change.
● 5 to 10% of span.
● Observed the system response.
● If it is sluggish increase the gain.
● If the system cycle continuously reduce the gain.
Problem with trial and error method

● Time consuming.
● One mode affects another mode.
● Re adjust component oftenly.
● Time.
● Patience.
● Optimal balance.
● Achieve desired output.
 ziegler -Nichols tuning method
In early 1940s,

Two engineers

● John zieglers and Nathaniel Nichols

Develope two formal procedure for tuning control loop

● Continuous cycling method.

● reaction curve method.
 ziegler -nichols continuous cycling method

Graph

● Forcing to oscillate in even continuous cycle

● By adjusting the proportional that cause cycling to take place
● In diagram time duration of one cycle called an ultimate period
Process identification procedure for
Continuous cycling method

● It is perform in close loop in auto mode

● Its objective is to obtained ultimate period and the ultimate proportional value
● Step involve two ways:

1.Controller with proportional gain

2.Controller has proportional band

Procedure for controller with proportional gain
Procedure proportional band setting
Formula for calculation

1. Formula for a Proportional-Only Controller.

2. Formula for a Proportional-integral Controller.
3. Formula for a Proportional-integraI-Derivative Controller.
Calculations for a Proportional-Only Controller

In order to obtain proportional setting

Formula for a Proportional-integral Controller
Reset time and reset rate(PIC)
Formula for a Proportional-IntegraI-Derivative Controller
Reset rate and reset time (PIDC)
Direct curve method
The main objective of direct curve method is to avoid oscillating condition such as excessive temperature in nuclear water
plant

The graph provides three different values to determine the proper controller settings:

l. The effective delay (D), which is the time that expires from when the step change is made until the process variable
begins to react. This delay is caused by the process lag, dead time, or both.

2. The process reaction rate, which is defined as how much the process changes per unit of time. This value is
obtained by calculating the slope of the process reaction curve. A curve with a steep slope indicates a faster
reaction rate than a curve with a gradual slope.

3. The unit reaction rate, which is a measure of how much the process reacts for each percent of actuator change. To
determine this value, the size of the step change must be taken from the graph. The size is read as a percentage of
the span.
The reaction curve tuning method performs an open-loop step change with the controller in the manual mode to
obtain process identification information.
Direct curve method

● A process reaction curve produced

by a step change.

● The step change information is used

because its size affects how quickly the
process.
Step 1: Put the controller in the manual mode.

Step 2: Produce a Step change by changing the

controller output 5 to 10 percent and observe the
rate at which the process responds on the chart
recorder.

Step 3: Find the maximum slope Of the reaction

curve and draw a tangent line at this point, as
shown in Figure .

Step 4: Calculate the slope of the tangent by

drawing two lines on the graph. Line A is a
horizontal line that begins at the starting point on
the tangent, as shown in figure.
Step 4: Calculate the slope of the
tangent by drawing two lines on
the graph. Line A is a horizontal
line that begins at the starting
point on the tangent, as shown in
Step 5: Determine the process reaction
rate, which is indicated by the slope of
the tangent, by using the formula,

where,
R = Process Reaction Rate
A = Time in minutes
B = Percentage of the process variable
change
Step 6: Calculate the unit reaction rate (RI).
Divide the reaction rate (R) by the percentage of
the actuator change (X).

Step 7: Determine the effective delay (D). This delay

is shown on the graph as the time from which the
step change is made to where the tangent line crosses
the line Of initial controlled variable status.
Direct synthesis method

● It use the same procedure as the direct curve method

● involves making a known step change to the final control element, and observing the set point
and the controlled variable
● The value in graph are used in formulas
● controller is properly tuned, it directly synthesizes (matches) the process.
● that the controlled variable does not overshoot when it reaches set point.
 steps

● bump test by making a setpoint change with the

controller in the manual mode.
● result is that the controller alters the final control
element and causes the process to change to a
different steady state
● time from which the controlled variable changes from
its original state to a new steady-state condition is
referred to as the dynamic settling time
● shows four common response signals, called process
models, that develop in industrial systems.
Common process model
Pure Gain Process: Pure gain is developed if the feedback signal is exactly the same as the setpoint
change signal applied to the final control element, Pure gain processes are uncommon in
industry.
● First Order Process: occurs when the process lags behind the signal applied to the final control
element.
- The amount Of lag is due to the static inertia physical and chemical properties of the controlled
variable
- The lag is directly proportional to the process time constant. As the time constant increases, the
time the process takes to settle at a new steady-state value also increases.

● First Order Plus Dead Time: The first order reference, pertains to lag time due to the time
constant of the process.
-The dead time refers to the movement from when the final control element begins to change
until the time at which the sensor begins sending a feedback signal to the controller.

● Second Order Process: There are two time constant delays in a second order process. One is due
to the lag time of the process and the response time of instruments in a control loop, such as a
sensor or a transmitter.
Controller autotuning

● Modern microprocessor-based controllers are designed to calculate the appropriate

mode settings automatically. When the autotuning function is it causes the process to
cycle. By measuring the time duration of a cycle, it either adjusts the settings
automatically for each mode or displays the recommended values for the operator to
enter.
Advanced Control Techniques

-Incorporation of PID modes into single loop. To gain adequate control of the system.

-There are 4 techniques:

1) Cascade Control
2) Feedforward Control
3) Ratio Control
4) Adaptive Control
Cascade Control
Cascade Control
Cascade Control

- Cascade control is a system that uses two additional components, a sensor, and a controller
to form a second feedback loop.

- Often results in faster. more precise performance than with single loop PID control systems.
Feedforward Control
Feedforward control
Feedforward Control
Feedforward Control

-By measuring a variable that enters a process and by taking corrective action if it is affected by
a disturbance, a deviation of the controlled variable for set point is reduced or eliminated.

-feedback systems determine a correction that needs to be made after the control variable
deviates from the set point.

-only used when no variation from set point can be tolerated in a process or a system is very
slow in responding to corrective action.
Ratio Control
Ratio Control

-Proportionally control the flow of one ingredient based of the amount of flow of another
ingredient.

-The ratio control flow method is used for continuous blending application, or can be
integrated to provide a specified volume for a batch process
Adaptive Control
Adaptive Control

-a device or controller (which has adaptive capabilities) that converts non-linear signal to linear
signal.