Professional Documents
Culture Documents
Fuzzy Logic1
Fuzzy Logic1
Introduction
• Introduced in 1965 by Lofti A. Zadeh, a professor in the University of
California, Berkeley
• Got accepted as emerging technology since 1980s
• Dictionary meaning of fuzzy is not clear, noisy etc.
• Antonym of fuzzy is crisp
• Having a fundamental trade-off between precision and cost which can be
called “principle of incompatibility”.
• Generalizes classical two valued logic for reasoning uncertainty.
• Helps us to tackle the uncertainty and vagueness associated with the
event.
• The factors influencing humans learning ability are:
– Generalization
– Association (memory mapping)
– Information loss (memory loss or forgetfulness).
• Undecidability
– Ambiguity that originates from the inability to distinguish between various states of an
event is termed as undecidability.
• Probability Theory versus Possibility Theory
– Possibility measures the degree of ease for a variable to take a value, while probability
measures the likelihood for a variable to take a value.
– So they deal with two different types of uncertainty.
– Possibility theory handles imprecision and probability theory handles likelihood of
occurrence.
– fuzzy set theory can define set membership as a possibility distribution.
– Also, fuzzy logic measures the degree to which an outcome belongs to an event while
probability measures the likely hood of event to occur.
Concept of Fuzzy System
Classical Sets and Fuzzy Set
• ‘Crisp set’ is a collection of distinct (precisely defined) elements. In
classical set theory, a crisp set can be a superset containing other crisp
sets.
• Fuzziness is a property of language. Its main source is the imprecision
involved in defining and using symbols.
• Representation of a Classical Set
– Classical set as already mentioned is a collection of objects of any kind. They can be
represented as list method, rule method and characteristic function method.
Method Examples
A x 1 A x
~
Example: Union
Example: Intersection
Example: Complement
A real-life example
Important Terminologies in Fuzzy Set
Operations
• Empty Fuzzy Set:
– A fuzzy set A is said to be an empty set if it has no members and its membership
function is zero everywhere in its universe of discourse U.
if A(x) = 0, x U
( A) A
Double negation law
A A A A A A
Idempotency
Commutativity
A B B A A B B A
Associative Property ( A B) C A ( B C ) ( A B) C A ( B C)
Distributive Property A ( B C ) ( A B) ( A C) A ( B C ) ( A B) ( A C)
Adsorption
A ( A B) A A ( A B) A
De Morgan’s laws
A B A B A B A B
A B
Law of Contradiction
A A U
Law of Excluded Middle
A A (x) 0 0
Intersection of a Fuzzy Set with an Empty Set
Some of the properties that are not valid for fuzzy sets but valid for crisp sets are
Natural Language and Fuzzy Interpretations
• A collection of these primary terms will form phrases, of our natural
language. Examples of some atomic terms are slow, medium, young,
beautiful, angry, cold, temperature etc.
• The collection of atomic terms can form compound terms. Examples of
compound terms are very cold, medium speed, young lady, fairly beautiful
picture, etc .
Membership Functio n M odel fo r Atomic Values
1
0.9
0.8
Membership Value of Temperature
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 10 20 30 40 50 60 70 80 90 100 110 120
Temperature(T) in Degrees
Linguistic Modifiers
Primary Terms with
Modification Membership Function
Modifiers
[ ( x )] 2
Very 2
I
x
[ ( x )] 4
Very very 4 I
x 1
More or less 1 [ ( x )] 2
Slightly 2
I
x
[ ( x )]1.25
Plus 1.25 I x
[ ( x )] 0.75
Minus 0.75 I x
(x)
1 , x x max 1 , x x min
Over I
x
0, x x max 0, x x min
(x)
1 , x x min 1 , x x max
Under I
x
0, x x min 0, x x max
(x) (x)
2
2 , 0 0.5
2 , 0 0.5 x
2
I
x I
Indeed 2
1 2[1 ] 2 , 0.5 1.0 ( x )
1 2 1
(x)
, 0.5 x 1.0
I x I
Linguistic Modifiers
• Fuzzy Concentration
– Concentrations tend to concentrate or reduces the fuzziness of the elements in a fuzzy
set by reducing the degree of membership of all elements that are only ‘‘partly’’ in the
set.
Concentration
1.2
1
A ( Low, T )
1
Low Before
2
Membership Value Concentration
1
A ( Very Low, T )
0.6
2.4
1 0.0005(T 30) 0.4
0.2 After
Concentration
Very
Low
0
-0.2
0 20 40 60 80 100 120
Temperature(T) in Degrees
• Fuzzy Dilation
– Dilation tend to dilate or increase the fuzziness of the elements in a fuzzy set by
increasing the degree of membership of all elements that are only ‘‘partly’’ in the set.
Dilation
1
0.9
1
A ( Low, T ) 0.8 After
0.7
Low
Membership Value
0.6
0.5
1 Slightly
A (Slightly Low, T ) 2.4
0.5 Low
1 0.0005(T 30)
0.4 Before
Dilation
0.3
0.2
0.1
0
0 20 40 60 80 100 120
Temperature(T) in Degrees
• Fuzzy Intensification:
– This property of the linguistic modifier is found to have a property which is a
combination of both fuzzy concentration and fuzzy intensification.
Intensification
1
0.9
Before
Intensification
1
A ( Low, T ) 0.8
0.7 After
Intensification
Membership Value
0.6
Indeed
1
2
1 Low
2 2.0
, 0 2.0
0.5 0.5
A ( Indeed Low, T ) 1 0.005(T 30) 1 0.005(T 30)
1 2 1
2
1 0.4
, 0.5 2.0
1.0
1 0.005(T 30)2.0 1 0.005(T 30)
0.3
0.2
0.1
0
0 20 40 60 80 100 120
Temperature(T) in Degrees
Logical Operations using Linguistic
Modifiers
• It is possible to perform logical operations using the linguistic modifiers,
however the order of precedence need to be followed to avoid confusion.
This problem can be resolved by properly using the parentheses by the
rule that use “association to the right”.
• The precedence of operation should be “NOT” in the First, “AND” in the
second and “OR” in the third.
• For example, if we have a compound terms “plus very minus very low”,
then following the rule “association to the right” while using the
parentheses we get “Plus (very (minus (very (low))))”.
Structure of Fuzzy Inference System
Fuzzification
• The process of converting the crisp values to fuzzy variable is called
‘fuzzification’.
Fuzzy Connectives
• The linguistic variables are combined by various connectives such as
negation, disjunction, conjunction and implication etc
- Negation 1 T ( P)
P
max((u i )) 1
1.
Peak
0 point
0.
(u i )
9
0.
8
0.
7
0.
6
0.
5
0.
4
0.
3
0.
2
0.
1
u* ui
Centre of Area (COA) Defuzzification
OUT (u i ) N
1.
u i OUT (u i )
0
0.
u * i 1
N
9
0.
i 1
OUT (u i )
8
0.
7
0.
6
0.
5
0.
4
0.
3
0.
2
0.
1
0
1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16
1
ui
u * 6.65
Weighted Average Method
Z
z
j1
j OUT ( z j )
u* Z
OUT (u i )
z2
j1
OUT (z j ) (d) Mean of Maxima (MOM)
z1
1.0
1 M
u um
0.9 *
0.8
M m 1
0.7
0.6
0.5
(e) Centre of Sums (COS)
0.4
0.3
N n
u
0.2
0.1 i C' ( u i )
k
u
* i 1
N n
k 1
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ui C' ( u i )
k
i 1 k 1
u * 5.86
MoM : Example 1
MoM : Example 2
COS Example
(f) Centre of Largest Area (COLA)
(g) First of Maxima
OUT (u i )
N
max( C m ( u )) max( C i ( u )), i 1....N C
1.
0 C1
C2
u i Cm (u i )
0.
9 u* i 1
N u * min( u i )
0.
8
0. Cm (u i )
7 i 1
0. (h) Last of Maxima
6
Cm = C1
0.
5
0.
4
0. u * max(u i )
3
0.
2
0.
1
0
ui
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
u*= 4.06
Find the crisp value using first of maxima
Find the crisp value using last of maxima