Process Control

Process Controllers

The University of Jordan

Chemical Engineering Department

Fall 2022

Prof. Yousef Mubarak

1
 Process Controllers

“Standard block diagram of feedback controller with one disturbance”

Y. Mubarak Process Control 2
 Example.

Find the transfer function that relates the output y with input u :

Solution:

Steady-state Eq.:

Deviated variables:

 Dynamic model in terms of deviation variables:

Y. Mubarak Process Control 3

 Originally input and output signals are time varying:

 Controlled variable set point:

 Servo control: (Manipulate the system input to keep the output close to a given reference trajectory)
o Set point variable:

o Set point deviated variable:

 Regulatory control: (Manipulate the system input to counteract the effects of disturbances)
o Constant set point:
o Set point deviated variable:

0 Note that:
Small letter : actual variable
Capital letter: deviated variable

Y. Mubarak Process Control 4

 Measured controlled variable:

 Measured controlled variable (deviated value):

 Controller output variable: p(t)

 Controller output variable (deviated value):

where is called Bias or nominal value of controller output signal which is usually adjusted, during calibration,
to be half of the span of the range of the controller signal output.

Y. Mubarak Process Control 5

 Definition of Span and Zero:
– Span: magnitude of the range of the output signal.
– Zero: lower limit of the output signal

 According to Instrumental Society of America (ISA), Controller output signal has the
following standard ranges:
Pneumatic signal: 3 – 15 psi Zero = 3 psi ; Span=12 psi
Electrical Signal: 4 - 20 mA Zero = 4 mA ; Span=16 mA
Electrical Signal: 0–10 VDC Zero = 0 VDC ; Span=10 VDC

 Error in controlled variable: (deviated error): Watch units!

Y. Mubarak Process Control 6

 Controller is the “brain” of the control loop. It decides “what to do” based on the error between the
desired set point value (SP) of controlled variable and its measured value.

 Basic types of controllers:

 On-Off controller.
 Different modes of proportional (P), Integral (I) and derivative (D) Controllers:
 Proportional (P) controller
 Proportional Integral (PI) controller
 Proportional Derivative (PD) controller
 Ideal Proportional Integral Derivative (PID) controller
 Actual (PID) controller with filter.

Y. Mubarak Process Control 7

 On-Off controllers:

 The on-off control is the simplest form of a controller, which switches ON when the error is
positive and switches OFF when the error is zero or negative.
 An on-off controller doesn’t have intermediate states but only fully ON or fully OFF states.

 Due to the switching logic, an on-off controller is often called a

bang-bang controller or a two-step controller.
 Regardless of the size of the error, the output of the on-off
controller can only be fully ON or fully OFF, it is not proportional
with the error.

Y. Mubarak Process Control 8

Advantages of On-Off controllers:
 simple & easy to design
 inexpensive
 easily accepted among operators

Pitfalls of On-Off controllers:

 not effective for “ good ” set-point control (the controlled variable cycles)
 produce wear on the final control element (it can be attenuated by a large dead band, at the
expense of a loss of performance)

Y. Mubarak Process Control 9

Examples of On-Off controllers
 Batch process control.
 Residential heating and domestic refrigerators.
 Solenoid in home heating unit
 Sprinkler systems.

Y. Mubarak Process Control 10

 On-Off controllers:

 Synonyms: “two-position” or “bang-bang” controllers.

 Controller output has two possible values.
 Ideal case:

pmax is the “on” value

pmin is the “off” value.

 Practical case (dead band):

δ = tolerance

Y. Mubarak Process Control 11

Proportional (P) Controller

The control variable is manipulated according to:

where: is the controller bias

is the controller gain

 The controller gain can be adjusted ( “ tuned ” ) to make the manipulated variable changes as
sensitive as desired to the deviations between set-point and controlled variable

 The sign of can be chosen to make the controller output u increase or decrease as the error
increases

Y. Mubarak Process Control 12

Proportional (P) Controller

 The bias is the value of the controller output which, in manual mode, causes the measured
process variable to maintain steady state at the design level of operation [e(t) = 0] when the
process disturbances are at their expected values

 The bias value is assigned at the controller design level, and remains fixed once the controller is
put in automatic

 is constant at the nominal steady state

Y. Mubarak Process Control 13

1. Proportional (P) controller

 In P controller, the controller output signal

p(t) varies linearly with error e(t)

where Kc is called controller proportional

gain.

 When error (offset) has zero value, the controller output signal reaches its steady state Bias value

 Action of controller: as absolute error increases, the controller output signal must increase to give
large change in manipulated variable, X(t).

Y. Mubarak Process Control 14

1. Proportional (P) controller:

 Reverse or Direct action controller:

 Direct acting controller (Kc < 0): choose negative Kc value to increase the controller output signal as the
measured controlled variables increases above the set point.

 Reverse acting controller (Kc  0): choose positive Kc value to increase the controller output signal as the
measured controlled variables decreases below the set point.

Advantages of Proportional (P) controller:

 There conceptually simple easy to tune (a single parameter is needed, Kc ; the bias is determined from steady
state information).

Pitfalls of Proportional (P) controller:

 Pitfalls cannot remove offset (offset is enforced by the controlled)

Y. Mubarak Process Control 15

1- Proportional (P) controller:
 Therefore, regardless of possible negative or positive errors, controller output signal, p(t)
should be always positive.

“ideal behavior” “actual behavior”

As error changes, p(t) will change immediately (fast corrective action in very simple form).

Y. Mubarak Process Control 16

1. Proportional (P) controller:

 Proportional band (PB):

 Proportional band is defined only for dimensionless Kc.

 As will be shown later, the final error “offset” can be reduced by increasing the absolute
value of Kc (reducing PB%).

 Introducing very high values of Kc will lead to oscillatory response or even unstable
situation.
 Proportional Controller Transfer Function:

Y. Mubarak Process Control 17

1. Proportional (P) controller:

 Effect of Kc on the controlled variable response under step change in disturbance variable

Y. Mubarak Process Control 18

1. Proportional (P) controller:

 Example: How is the response of proportional controller to unit step change in e(t)?

 Exercise 1: How is the P controller response to ramp change in e(t)?

Y. Mubarak Process Control 19

20
21
22
23
 2.Proportional Integral (PI) controller:

Synonyms: "reset", "floating control"

 The P controller cannot remove offset because the only way to change the controller bias during
non-nominal operations is to cause e(t) = 0.

 The rationale behind a PI controller is to set the “ actual ” bias different from , thus letting the error
be zero.

 The control variable is manipulated according to (PI Controller signal output):

reset time or integral time (adjustable parameter)

Y. Mubarak Process Control 24

 2.Proportional Integral (PI) controller:

 The integral mode will change the bias value to eliminate the steady state error (offset).
 The action is not immediate until the integral becomes significant.
 The integral mode tends the system to be more oscillatory or even unstable.
 Note that until e(t) = 0, the manipulated input keeps on changing because of the presence of the
integral term.
 The change in p(t) will stop only when e(t) = 0
 Net result: the integral action can eliminate offset.

Y. Mubarak Process Control 25

2.Proportional integral (PI) controller:
 Reset rate:
 Infinite integral time or zero reset rate P
controller

 Advantages (Combined benefits)

 Fast action
 Eliminate the offset
 Disadvantages
 Oscillatory or unstable with integral control
 Two parameters to tune (Kc and )

Y. Mubarak Process Control 26

2. Proportional integral (PI) controller:

Reset Time:
 The bias value will be reset every by the amount of action taken by the P control. This is called
“reset time”

 The integral mode eliminates the

steady state error (offset)

Y. Mubarak Process Control 27

 2. Proportional integral (PI) controller:

 PI controller Transfer Function :

Y. Mubarak Process Control 28

 2. Proportional integral (PI) controller:

 Example: How is the response of PI controller to unit step

change in e(t)?

 Integral action  Kc /  I

 Exercise 2: How is the PI controller response to ramp

change in e(t)?

Y. Mubarak Process Control 29

 2. Proportional integral (PI) controller:

 Effect of integral time and gain Kc of PI controller on the response of controlled deviated
variable Y(t):

 As Kc increases or decreases: less offset, the response will be faster, more overshooting,
less oscillation.
Y. Mubarak Process Control 30
Summary for PI control

Advantages:

 Steady state offset can be eliminated

 Steady state offset can be eliminated the process response can be considerably speeded up with
respect to open-loop

Pitfalls:

 Tuning is harder (two parameters must be specified, Kc and )

 The process response becomes oscillatory; bad tuning may even lead to instability

 The integral action may “ saturate ”

Y. Mubarak Process Control 31

 3. Proportional Derivative (PD) controller:
 PD Controller signal output:

derivative or preact time (adjustable parameter)

 Zero derivative time P controller

 Derivative mode is used to improve dynamic
response of the controlled variable.

 Derivative mode it does NOT eliminate the

offset.
 Two parameters to tune (Kc and

Y. Mubarak Process Control 32

3. Proportional Derivative (PD) controller:

 PD controller Transfer Function :

 Example: How is the response of PD controller to unit step change in e(t)?

(Same as P controller)

 Exercise 3: How is the PD controller response to ramp change in e(t)?

Y. Mubarak Process Control 33

4. Proportional Integral Derivative (PID) controller:

i. If the error if increasing very rapidly, a large deviation from the set-point may arise in a short time.

ii. Sluggish processes tend to cycle

 The rationale behind derivative action is to anticipate the future behavior of the error signal by
considering its rate of change

 The control variable is manipulated according to:

34
 4. Proportional Integral Derivative (PID) controller:

 PID controller signal output:

 Infinite integral time and zero derivative time P

controller
 zero derivative time PI controller
 Infinite integral time PD controller
 Advantages (Combined benefits)
 Fast action
 Integral mode will eliminate the offset.
 Derivative mode will make the process output to land on set point smoothly.
 D mode tends to reduce the oscillation and enhance the stability.

Y. Mubarak Process Control 35

4. Proportional Integral Derivative (PID) controller:

 Disadvantages
 If there is noise in the process variable, noise will be amplified by the derivative. If the
measurement is noisy, use the measurement after smoothing out (filtering).
 The derivative requires information on error in the future: impossible use
approximation of derivative
 Three parameters to tune (Kc, and . It is quite complicated for three tuning parameters.
 PID controller Transfer Function :

Y. Mubarak Process Control 36

4. Proportional Integral Derivative (PID) controller:

 Effect of derivative time of PID controller on the response of controlled deviated variable Y(t):

 As increases: the response will be

slower, less oscillatory, less
overshooting (when there is no noise).

 Example: How is the response of PID controller to unit step change in e(t)

(Same as PI controller)

 Exercise 4: How is the PID controller response to ramp change in e(t)?

Y. Mubarak Process Control 37
4. Proportional Integral Derivative (PID) controller:

Parallel form of
PID controller

 This PID controller is ideal nointeracting. In many cases it is physically unrealizable.

 Integral time must be equal or greater than the derivative time: Typically .

Y. Mubarak Process Control 38

 5. Actual Interacting PID controller:

 Transfer function:

Where  is derivative filter parameter

 This controller is physically realizable.

 Any transfer function which has only zeros is not realizable, realizability requires less zeros than poles

Y. Mubarak Process Control 39

 Comparison between ideal and actual PID without filter:

 Actual PID without filter ( = 0):

 Ideal PID controller:

Compare:

 In this form,

Y. Mubarak Process Control 40

 Key characteristics of commercial PID controllers:

Y. Mubarak Process Control 41

 Important remark:

 Basically, the controller gain, Kc, is dimensional quantity. Its physical unit is the ratio of the unit of
controller output signal (mA, psig) to the unit of the controlled variable ( oC, m, Pa, gpm, mol/L, …etc).
 Kc can be dimensionless quantity if the set point and measured controlled variable are transmitting to a
signal as that of controller output:

Y. Mubarak Process Control 42