Fuzzy Logic Control

Lect 6 Fuzzy PID Controller

 Contents
• PID
• PID Fuzzy
• Example
• Supervisory PID Fuzzy Control

2
 PID

▪ PID: Proportional Integral Derivative

▪ More than 90% of controllers used in industries are
PID or PID type controllers (the rest are PLC)
▪ PID controllers are simple, reliable, effective
▪ For lower order linear system PID controllers have
remarkable set-point tracking performance and
guaranteed stability.

3
Convectional PID Controller

4
 Convectional PID Controller
▪ Time Domain

▪ Frequency Domain

5
PID Controller

6
 PID Controller
Time Domain

Frequency Domain

7
A comparison of different controller types

8
General tips for designing a PID controller

• Obtain an open-loop response and determine what needs to be

improved
• Add a proportional control to improve the rise time
• Add a derivative control to improve the overshoot
• Add an integral control to eliminate the steady-state error
• Adjust each of Kp, Ki, and Kd until you obtain a desired overall
response. You can always refer to the table shown to find out which
controller controls what characteristics
• you do not need to implement all three controllers (proportional,
derivative, and integral) into a single system, if not necessary. Keep
the controller as simple as possible.

9
General tips for designing a PID controller

10
 Tuning of PID Controller
There are methods for tuning PID controllers, for
example:
▪ hand-tuning,
▪ Ziegler–Nichols tuning,
▪ optimal design,
▪ pole placement design, and
▪ auto-tuning (A° stro¨m and H¨agglund 1995).
There is much to gain, if these methods are carried
forward to fuzzy controllers.

11
 Why use fuzzy with PID
• Although PID controllers are able to provide
adequate control for simple systems, they are unable
to compensate for disturbances.
• We will use Fuzzy Logic controllers to improve the
PID controllers ability to handle disturbances.
• PID Control works well for linear processes
• PID control has poor performance in nonlinear
processes.
• Fairly complex systems usually need human control
operators for operation and supervision

12
 Types of Fuzzy Controllers:
- Direct Controller -
The Outputs of the Fuzzy Logic System Are the Command Variables of the Plant:

Command
IF temp=low
AND P=high
Variables
THEN A=med

IF ...
Plant

Fuzzification Inference Defuzzification

Measured Variables

Fuzzy Rules Output

Absolute Values !

Slide 13
 Types of Fuzzy Controllers:
- PID Adaptation -
Fuzzy Logic Controller Adapts the P, I, and D Parameter of a Conventional PID Controller:
Set Point Variable

IF temp=low
AND P=high P
THEN A=med
I Command Variable
D
IF ...
PID Plant
Fuzzification Inference Defuzzification

Measured Variable

The Fuzzy Logic System

Analyzes the Performance of the
PID Controller and Optimizes It !
Slide 14
 Types of Fuzzy Controllers:
- Fuzzy Intervention -
Fuzzy Logic Controller and PID Controller in Parallel:
Set Point Variable

IF temp=low
AND P=high
THEN A=med Command Variable

IF ...
Plant
PID
Fuzzification Inference Defuzzification

Measured Variable

Intervention of the Fuzzy Logic

Controller into Large Disturbances !

Slide 15
 Supervisory Control Systems

Most controllers in operation today have been developed

using conventional control methods. There are, however,
many situations where these controllers are not properly
tuned and there is heuristic knowledge available on how to
tune them while they are in operation. There is then the
opportunity to utilize fuzzy control methods as the supervisor
that tunes or coordinates the application of conventional
controllers.

16
 Fuzzy PID Control
Because PID controllers are often not properly tuned (e.g., due
to plant parameter variations or operating condition changes),
there is a significant need to develop methods for the automatic
tuning of PID controllers. While there exist many conventional
methods for PID auto-tuning, here we will strictly focus on
providing the basic ideas on how you would construct a fuzzy PID
auto-tuner.

17
 Fuzzy PID Control
▪ A fuzzy PID controller is a fuzzified proportional-integral-
derivative (PID) controller. It acts on the same input signals, but
the control strategy is formulated as fuzzy rules.
▪ If a control engineer changes the rules, or the tuning gains, it is
difficult to predict the effect on rise time, overshoot, and settling
time of a closed-loop step response, because the controller is
generally nonlinear and its structure is complex.
▪ In contrast, a PID controller is a simple, linear combination of
three signals: the P action proportional to the error e, the I-
action proportional to the integral of the error ‫𝑡𝑑𝑒 ׬‬, and the D-
action proportional to the time derivative of the error de/dt, or ˙e
for short.

18
 Fuzzy PID Control
Fuzzy PID controllers are similar to PID controllers under
certain assumptions about the shape of the membership
functions and the inference method (Siler and Ying 1989,
Mizumoto 1992, Qiao and Mizumoto 1996, Tso and Fung
1997).
A design procedure for fuzzy controllers of the PID type, based
on PID tuning, is the following:
Procedure Design fuzzy PID
1. Build and tune a conventional PID controller first.
2. Replace it with an equivalent fuzzy controller.
3. Fine-tune it.

19
 Fuzzy PID Control
The procedure is relevant whenever PID control is possible, or
already implemented. Our starting point is the ideal continuous
PID controller

The control signal u is a linear combination of the error e, its

integral and its derivative. The parameter Kp is the proportional
gain, Ti is the integral time, and Td the derivative time.

20
 Fuzzy PID Control
To implement fuzzy PID control on the computer, one first
needs a digital version of analog one.
Discretization of PID controller:
To digitize the analog controller, the following can be used:

21
 Fuzzy PID Control
In digital controllers, the equation must be approximated.
Replacing the derivative term by a backward difference and the
integral by a sum using rectangular integration, and given a
constant – preferably small – sampling time Ts , the simplest
approximation is,

Index n refers to the time instant. By tuning we shall mean the

activity of adjusting the parameters Kp, Ti , and Td in order to
achieve a good closed-loop performance.

22
Example

23
Example

24
Example

25
Example

26
 Example

Simulation result are shown , where red is system output, and green is error signal

27
 Supervisory Control Systems
▪ Human operators in the process industry are faced with
nonlinear and time-varying behaviour, many inner loops, and
much interaction between the control loops. Owing to sheer
complexity it is impossible, or at least very expensive, to
build a mathematical model of the plant, and furthermore the
control is normally a combination of sequential, parallel, and
feedback control actions.
▪ Operators, however, are able to control complicated plants
using their experience and training, and thus fuzzy control is
a relevant method within supervisory control.

28
 Supervisory Control Systems
Supervisory control is a multilayer (hierarchical) controller with
the supervisor at the highest level, as shown in Figure

29
 Supervisory Control Systems
▪ The supervisor can use any available data from the control system
to characterize the system’s current behavior so that it knows how
to change the controller and ultimately achieve the desired
specifications.
▪ In addition, the supervisor can be used to integrate other
information into the control decision-making process. It can
incorporate certain user inputs, or inputs from other subsystems.
▪ Supervisory control is a type of adaptive control since it seeks to
observe the current behavior of the control system and modify the
controller to improve the performance

30
 Supervisory Control Systems
For example, in an automotive cruise control problem, inputs from
the driver (user) may indicate that she or he wants the cruise
controller to operate either like a sports car or more like a sluggish
family car. The other subsystem information that a supervisor could
incorporate for supervisory control for an automotive cruise control
application could include data from the engine that would help
integrate the controls on the vehicle (i.e., engine and cruise control
integration). Given information of this type, the supervisor can
seek to tune the controller to achieve higher performance
operation or a performance that is more to the liking of the
driver.

31
 Supervisory Control Systems
Conceptually, the design of the supervisory controller can then
proceed in the same manner as it did for direct fuzzy
controllers: either via the gathering of heuristic control
knowledge or via training data that we gather from an
experiment. The form of the knowledge or data is, however,
somewhat different than in the simple fuzzy control problem.

32
 Supervisory Control Systems
the type of heuristic knowledge that is used in a supervisor
may take one of the following two forms:
1. Information from a human control system operator who
observes the behavior of an existing control system (often a
conventional control system) and knows how this controller
should be tuned under various operating conditions.
2. Information gathered by a control engineer who knows that
under different operating conditions controller parameters
should be tuned according to certain rules.

33
 High-level control configurations

Fuzzy controllers are combined with other controllers

in various configurations. The PID block consists of
independent or coupled PID loops, and the fuzzy
block employs a high-level control strategy. Normally,
both the PID and the fuzzy blocks have more than one
input and one output.

34
 Supervisory Fuzzy Control
There are four types of Fuzzy supervisory control:
1. Fuzzy replaces PID
2. Fuzzy replaces operator
3. Fuzzy adjusts PID parameters
4. Fuzzy adds to PID control

35
 Fuzzy replaces PID

In this configuration, the operator may select between a high-

level control strategy and conventional control loops. The
operator has to decide which of the two most likely produces
the best control performance.

36
 Fuzzy replaces operator

This configuration represents the original high level control

idea, where manual control carried out by a human operator is
replaced by automatic control. Normally, the existing control
loops are still active, and the high-level control strategy makes
adjustments of the controller set points in the same way as the
operator does. Again it is up to the operator to decide whether
manual or automatic control will result in the best possible
operation of the process, which, of course, may create
conflicts.

37
Fuzzy replaces operator

38
 Fuzzy adjusts PID parameters

In this configuration, the high-level strategy adjusts the

parameters of the conventional control loops. A common
problem with linear PID control of highly nonlinear processes
is that the set of controller parameters are satisfactory only
when the process is within a narrow operational window.
Outside this, it is necessary to use other parameters or set
points, and these adjustments may be done automatically by a
high-level strategy.

39
Fuzzy adjusts PID parameters

40
 Fuzzy adds to PID control

Normally, control systems based on PID controllers are capable

of controlling the process when the operation is steady and
close to normal conditions. However, if sudden changes occur,
or if the process enters abnormal states, then the configuration
may be applied to bring the process back to normal operation as
fast as possible. For normal operation, the fuzzy contribution is
zero, whereas the PID outputs are compensated in abnormal
situations, often referred to as abnormal situation management
(ASM).

41
Fuzzy adds to PID control

42
 Homework
13.2, 13.4, 13.5

Due 20/11/2011

43