Process

Control

Introduction to
Feedback
Control Loops
Chapter 7&8: Marlin
Chapter 7: Seborg
Chapter 6-1: Smith & Corripio
 Learning Objectives
• Introduce feedback control and important terms
• Understand a feedback block diagram and relationship to real world
• Intro to PID control

2
 where are we up to?

• So far, we have studied system responses

• A disturbance (step change,sine wave) enters our process
• Shown how process responds, depending on 1st order, 2nd order, dead time, etc
• Ultimately, we want process control
• Incoming disturbances have MINIMAL effect on process outputs

Next step… on to controlling the responses!

3
 This
Control is Good Control variability
benefits:Reducing

Time (sec) Time (sec)

0 2 3 4 6 7 8 10
0 1 2 3 4 5 6 7 8 9 10

-2.0
-2.0

-1.5 -1.5

-1.0
-1.0

-.5
-.5

.5 .5

1.0
1.0

1.5 1.5

2.0
2.0

Time 
Plot
Plot

Good
0
1 5 9
control
4
 This is impossible
Control benefits:Reducing variability

Time (sec) Time (sec)

0 2 3 4 6 7 8 10
0 1 2 3 4 5 6 7 8 9 10

-2.0
-2.0

-1.5 -1.5

-1.0
-1.0

-.5
-.5

.5 .5

1.0
1.0

1.5 1.5

2.0
2.0

Time 
Plot
Plot

Impossible
Good
1 5 9
Control
control
0

5
Control of Tank Level
• Want to keep constant tank level
• Flowrate of inlet stream is fluctuating
• Query … what happens to mass out?

Controlled Manipulated “Random”

Variable Variable Variable
Accumulation Mass out Mass in
(Level) (Liquid (Liquid
Leaving) Entering)

6
Denn. Chem Eng Intro
Control of Tank Level
• Inlet flowrate initially steady at q*
• Temporary “step change” in inlet
flowrate occurs
• Inlet flowrate increases by Q* then
returns to original value
• No Control: Tank level increases
then plateaus
• With Control: level increases but by
less and returns to close to original
level

7
Denn. Chem Eng Intro
 Constant inlet fluctuations
Inlet flowrate to tank
• Inlet varying constantly
• As a result, tank level will
also vary (without control)
• Simple control (solid line)
Tank level without and with control
significantly reduces tank
level changes

8
Denn. Chem Eng Intro
 A FEEDBACK Control System
Temperature control of a heat exchanger
Cold Temperature, flow changes are
Steam or fluid DISTURBANCES Controller
condensate
Set point
exhaust

Feedback loop
Hot fluid
Sensor & Steam
Transmitter Valve

9
 Piping and Instrument Diagram (P&ID)
Heat exchanger temperature control
Cold
Steam or fluid Panel mounted
Computer control
condensate controller
exhaust SP

TC
101

Hot fluid
TT Steam
Field mounted
101
instrument

10
 Feedback Control System
Block Diagram
All variables are functions of
time. On the diagram we
represent them as Laplace D DISTURBANCE
transforms of CHANGES

Ts E U F
+ T
Controller VALVE
PROCESS +
+ (Heat exchanger)
-
U/E = Gc(s)
Summing
Tm point
Measuring element
(sensor)
Error
detector

11
 Definition of “Open Loop”

Open loop = no controller

Controller is either “off”,
disconnected, or in “manual” D Gd
Controller +
SP E U F + Y
+
Gc Gv Gp
-

Ym
Gm

12
 Auto and Manual Control modes
• Each controller can be set to “auto” (on) or “manual” which
means “off”
• Auto mode: Controller output depends on e(t), controller
constants, and type of controller used. ( PI vs. PID etc.)
• Manual Mode: Controller output is adjusted manually.
• Manual Mode is very useful when unusual conditions exist:
• plant start-up
• plant shut-down
• emergencies

13
 Definition of “Closed Loop”
This is CLOSED-LOOP
D Gd
With control
+
SP E U F + Y
+
Gc Gv Gp
-

Ym
Gm

14
 Closed Loop Block Diagram

TF of the final control D(s) TF of the

element– e.g. a 1st
Controller TF – disturbance
order valve
decides what to do
about the error
TF of the GD (s)
Error=SP-MC Process

Setpoint
SP CV(s)
(or Y(s)
T(s)etc)
GM (s)

TF of the Measuring
device
15
 The Sensor and the Valve

• A level measuring sensor usually has a fast response so its time constant is
small and can be neglected (ie., τ = 0)
• The sensor TF is just a gain: Tm
K m
• Km = Measurement Gain (or Ks) T
• A control valve is usually a pneumatic valve & can be represented by a 1st
order TF

Typical τV = 1 sec-1 min

F K
 v
U 1 s (depends on valve size)
v
16
 Typical V = 1 sec – 1 min
(depends on the valve size

https://i.ytimg.com/vi/yHgTRyfkXxs/maxresdefault.jpg

17
https://sc01.alicdn.com/kf/HTB1cOLrJVXXXXchXVXXq6xXFXXXN/Big-size-butterfly-valve-DN3600.jpg_350x350.jpg
 Closed loop feedback control
D DISTURBANCE
Gd
Controller
+ CV
SP E U F
Controller VALVE PROCESS +
+ Gc Gv Gp
-
Process equipment

Tm
Measuring sensor
Gs

See Tute 4 Q5 for how to derive Shortcut: CV = forward

these two TF responses SP (1 + around the loop) 18
Marlin
 Setpoint Change Closed Loop
“Servo” Control

D Gd

+
SP E U Fs + CV
+
Gc Gv Gp
-

Tm
GS

19
 Disturbance Change Closed Loop
“Regulatory” Control
Regulates any disturbances
Disturbance upsets T D Gd
Results in difference between T and TSP
Controller sees error, adjusts valve to fix it
+
SP E U Fs + CV
+
Gc Gv Gp
-

Tm
Gm

20
 Temperature Control of a
Heat Exchanger – Block Diagram
oC/oC

%/oC TD Gd
oC/(kg/min)

+
TSP E U Fs + T
+
Kc Gv Gp
-

oC/oC (kg/min)/%

Tm
Km
Fast response
Exercise: Confirm that the product of all gains around the loop is dimensionless
21
 Controllers

• A controller calculates an output signal based on the measured error and a

control algorithm
• Error = Setpoint – Measured value
• Proportional Control
• Output is proportional to the error
• u(t) = Kc*e(t) + u(0)
U
• Kc is called the (proportional) gain  Kc
• u(0) is the output when error = 0 E
• Positive Kc = “direct acting”
• Negative Kc = “reverse acting”
22
Proportional Temperature Control using an
FODT Model
Cold fluid
T change TD • Use steam flow Fs to
control T at TSP
TSP SP
• Incoming temp
TC
disturbance TD
U 101 • Assume FOPDT
models for:
Hot fluid
F • T/Fs
TT
Steam • T/TD
101 T • Start with a
Proportional
Controller for TC101

23
 PID response matches
“common sense”
1 de
u(t)  K [e(t)   e(t)dt   d ]  u(0)
c Ti dt
• If a big error occurs:
• You need a big response Proportional P
• But there is still a remaining error:
• Adjust until you eliminate error Integral I
• Rapid change
• rapid response required Derivative D

24
 PID algorithm

• PID algorithm considers the weighting of the following types of corrections:

• Proportional P - Corrections based on the present error e(t)
• Integral I - Corrections based on the sum of the past errors (integral of e(t))
• Derivative D - Corrections based on future predictions of errors, using the rate of change of errors
over time (de/dt)

1 de
u(t)  K [e(t)   e(t)dt   d ]  u(0)
c T dt
i
25
 PID Transfer Function
1 de
u(t)  K [e(t)   e(t)dt   d ]  u(0)
c T dt
i
U is the controller output
E is the difference between SP L Laplace transform
& the measured value

U  1 
 K c 1    d s 
E  TI s 
Kc = controller gain TI = Integral time d = derivative time

26
Typical responses

27
Seborg
 Summary

• Controlling setpoint changes and disturbances

• Open / closed loop block diagrams
• Intro to feedback PID control

• Next lecture: more PID + calculating and eliminating offset

28