Intelligent Control
1
Lecturer: VU VAN PHONG
Dr. Van-Phong Vu-Department of Automatic Control
2
Instructor: Dr. Vu Van Phong
Email: phongvv@[Link]
Websites: [Link]
Dr. Van-Phong Vu-Department of Automatic Control
3
Chapter 4: Fuzzy Implication
Dr. Van-Phong Vu-Department of Automatic Control
4 4.1 Linguistic Variables
Consider the sentence: ”Today’temperature is hot”: If we consider “
Today’s temperature” is a variable, then word “hot” is described the
value of variable “Today’s temperature”.
Definition: A variable can use the word in natural language to
describe its value is called linguistic variable. The word is
characterized in term of fuzzy sets.
For example: Temperature is a variable. The range of temperature is
[-50, 50], its values are “cold”, ”cool”, and “hot” words
Dr. Van-Phong Vu-Department of Automatic Control
5
Cold Cool Hot
0 20 40
Dr. Van-Phong Vu-Department of Automatic Control
6 Linguistic variable consist of :
Name: e.g. Temperature
Values: e.g. “hot”, “cool”, ”cold”
Range: Range from minimum temperature to max temperature
Dr. Van-Phong Vu-Department of Automatic Control
7 4.2 Fuzzy Rules
IF (…) THEN (...)
Example:
IF “Temperater is hot” THEN “ Speed is Fast”
Fuzzy rule includes:
(…) fuzzy proposion
IF-THEN Operation
Dr. Van-Phong Vu-Department of Automatic Control
8 4.2.1 Fuzzy Proposition
Consider a fuzzy rules: “ IF (…), THEN (...) ”
(...) is called a fuzzy proposition.
Classification:
Atomic fuzzy proposition:
“ x is hot”; ”y is hight”; ...
Compound fuzzy proposition
“x is hot and hight and light”
“ y is rich and beautiful and white skin”
Compound fuzzy proposition can consist of several atomic fuzzy proposition:
“x is fast” or “y is slow”
Dr. Van-Phong Vu-Department of Automatic Control
9 Operation on Fuzzy Proposition
Fuzzy proposition is described under the fuzzy relation or
fuzzy sets.
Operation “AND”
Compound fuzzy proposition:
FP= ”x is A” and “y is B”
A and B are two fuzzy sets on universe sets X and Y.
This fuzzy proposition is expressed the fuzzy relation:
¿
t is the intersection
Dr. Van-Phong Vu-Department of Automatic Control
10
Operation “Or”
Compound fuzzy proposition:
FP=”x is A” or “y is B”
A and B are two fuzzy sets on universe sets X and Y.
This fuzzy proposition is expressed the fuzzy relation:
¿
s is the union
Dr. Van-Phong Vu-Department of Automatic Control
11 Operation “NOT”
Compound fuzzy proposition:
FP=”x is A” or “y is Not B”
A and B are two fuzzy sets on universe sets X and Y.
This fuzzy proposition is expressed the fuzzy relation:
¿
s is the union and c is complement
Dr. Van-Phong Vu-Department of Automatic Control
12
Example: Describe the following compound fuzzy
proposition under the fuzzy relation.
Not(“ x is A” and “y is Not B”) or ”z is C”
Dr. Van-Phong Vu-Department of Automatic Control
13 4.2.2 Fuzzy Implication
Fuzzy Implication is described the “IF-THEN” operation.
Fuzzy implication: “IF p THEN q” is represent “
““or “(
∨
Dr. Van-Phong Vu-Department of Automatic Control
14 “ IF FP1 THEN FP2”
Fuzzy Implication is represented in term of Fuzzy relation.
Operations of a fuzzy relationship definition
Complement:
Intersection:
Union: ⋁
Noted: Because there many operations of complement, intersection
and union cause many calculation methods for fuzzy relation
Dr. Van-Phong Vu-Department of Automatic Control
15 Formula for fuzzy implication
“ IF FP1(x)THEN FP2(y)”
Dienes-Rescher Implication ()
=;=
=1- where
is max
max(1-, )=
Dr. Van-Phong Vu-Department of Automatic Control
16
Lukasiewicz Implication ()
=;=
=1- where
is Bounded Sum
min(1,(1- )=
Dr. Van-Phong Vu-Department of Automatic Control
17
Zadeh Implication ()
= ; = ()
=1- where
is max
Intersection is min
max(min(, ),(1- )=
Dr. Van-Phong Vu-Department of Automatic Control
18
Godel Implication
Dr. Van-Phong Vu-Department of Automatic Control
19
Mamdani Implication
Noted: This is the most important and popular
is defined
”
is Min
min( )=
is Product
=
Dr. Van-Phong Vu-Department of Automatic Control
20 Example 1: Consider two fuzzy set “small” on universe set X={x1,
x2, x3} and “Big” on universe set Y={y1, y2, y3}
0.1 1 0.8
𝑠𝑚𝑎𝑙𝑙={ + + }
𝑥1 𝑥 2 𝑥 3
big
Fuzzy rule:
IF x is “small” THEN ” y is Big”.
Find fuzzy implication: a) Dienes-Rescher Implication , b) Zadeh
Implication , c) Mamdani Implication (min)
Dr. Van-Phong Vu-Department of Automatic Control
Example 2: Consider two fuzzy set “small” on universe set X={x1,
21 x2, x3} and “big” on universe set Y={y1, y2, y3} , “ high” on
universe set Z={z1,z2}
0.1 1 0.8
𝑠𝑚𝑎𝑙𝑙={ + + }
𝑥1 𝑥 2 𝑥 3
big
high
Fuzzy rule:
IF x is “small” and “y is big” THEN ” z is high”.
Find fuzzy implication: a) Dienes-Rescher Implication , b) Zadeh
Implication , c) Mamdani Implication (min)
Use standard intersection.
Dr. Van-Phong Vu-Department of Automatic Control
22
Thanks for Listening
Dr. Van-Phong Vu-Department of Automatic Control