Professional Documents
Culture Documents
.in
rs
de
ea
yr
ab
or
ty
,w
.m
ha
kr
www.myreaders.info
Return to Website
Fuzzy Systems
Soft Computing
Fuzzy systems, topics : Introduction, fuzzy logic, fuzzy system
elements - input vector, fuzzification, fuzzy rule base, membership
function,
fuzzy
inferencing,
Classical
Logic
functions
defuzzyfication,
statement,
symbols,
and
output
tautology,
vector.
membership
approximate
generalized
reasoning,
modus
tollens
generalized
(GMT).
modus
Fuzzy
rule
ponens
based
fo
.in
rs
de
ea
,w
.m
yr
Fuzzy Systems
ab
or
ty
Soft Computing
ha
kr
Topics
35, 36
2 hours)
Slides
(Lectures
03-05
1. Introduction
Fuzzy Systems : Fuzzy logic and Fuzzy set theory; Fuzzy system
elements : Input vector, Fuzzification, Fuzzy Rule Base, Membership
function, Fuzzy Inferencing, Defuzzyfication, Output vector.
06-19
2. Fuzzy Logic
3. Fuzzification
4. Fuzzy Inference
Approximate
reasoning;
Generalized
Modus
Ponens
(GMP);
28
Example
6. Defuzzification
29
Centroid method.
7. References
02
30
fo
.in
rs
de
ea
yr
ha
kr
ab
or
ty
,w
.m
Fuzzy Systems
that exists in
Subjective
knowledge
that
exists
in
linguistic
form,
usually
impossible to quantify.
Fuzzy Logic can coordinate these two forms of knowledge in a logical way.
Fuzzy Systems
can
handle
linguistic knowledge.
problems
have
been
modeled,
simulated,
and
Expert Systems design have become easy because their domains are
inherently fuzzy and can now be handled better;
examples : Decision-support systems, Financial planners, Diagnostic
system, and Meteorological system.
03
fo
.in
rs
de
ea
Any system that uses Fuzzy mathematics may be viewed as Fuzzy system.
ab
or
ty
,w
.m
yr
1. Introduction
ha
kr
The Fuzzy Set Theory - membership function, operations, properties and the
have
been
described
in previous
lectures.
relations
input information,
obtained
from
the
Fuzzification
with
the
fuzzy
rule
base
and
fo
.in
rs
de
ea
ab
or
ty
,w
.m
yr
Fuzzy System
ha
kr
Fuzzy
Rule Base
Input
variables
X1
X2
output
variables
Fuzzy
Inferencing
Fuzzification
Y1
Y2
Defuzzification
Ym
Xn
Membeship Function
Fig. Elements of Fuzzy System
Fuzzy System elements
Input Vector :
X = [x1 , x2, . . . xn ]
comes out
from
the
collection
of
propositions
containing
linguistic
......
THEN (z is C)
with the rule base and conducts the Fuzzy reasoning process.
Defuzzyfication: Translate results back to the real world values.
05
fo
.in
rs
de
ea
or
ty
,w
.m
yr
2. Fuzzy Logic
A simple form of logic, called a two-valued logic is the study of "truth tables"
ha
kr
ab
and logic circuits. Here the possible values are true as 1, and false as 0.
This simple two-valued logic is generalized and called fuzzy logic which treats
"truth" as a continuous quantity ranging from 0 to 1.
Definition : Fuzzy logic (FL) is derived from fuzzy set theory dealing with
reasoning that is approximate rather than precisely deduced from classical
two-valued logic.
FL is the application of Fuzzy set theory.
FL allows set membership values to range (inclusively) between 0 and 1.
FL is capable of handling inherently imprecise concepts.
FL allows in linguistic form, the set membership values to imprecise concepts
fo
.in
rs
de
ea
or
ty
,w
.m
yr
ha
kr
ab
Examples :
Sentence
Truth value
Is it a Proposition ?
"Grass is green"
"true"
Yes
"2 + 5 = 5"
"false"
Yes
No
No
"x = x"
No
(don't
mean;
to air"
has no
Logic defines
the ways
of putting
(c) 12 * 12=144
07
Sentence
"Grass is green";
Truth value
true;
Proposition
yes
fo
.in
rs
de
ea
yr
or
ty
,w
.m
kr
ab
ha
or
connective :
"p is true"
p ~ !
conjunction
pq
disjunction
P v q ||
implication
p q
equivalence
&& &
NOT
"p is false"
AND
OR
if . . then
if and only if
"either p is true,
or q is true,
or both "
"if p is true, then q is true"
" p implies q "
"p and q are either both true
or both false"
08
fo
.in
rs
yr
ea
de
~p
ha
kr
ab
or
ty
,w
.m
Truth Value
pvq
and
09
p q p q p v q pq
pq
qp
fo
.in
rs
de
ea
yr
or
ty
,w
.m
Tautology
kr
ab
p, q,
ha
r, . . . .
and
(pq) (p) q
A proof of these tautologies, using the truth tables are given below.
Tautologies
(pq) [p (q)]
and
(pq) (p) q
(p) q
Note :
1. The entries of two columns pq
and
[p (q)]
are identical,
fo
.in
rs
yr
ea
de
ab
or
ty
,w
.m
Equivalences
ha
kr
Boolean
the correspondence
given below.
Logic
Set theory
T
F
1
0
x
+
ie complement
=
a, b, c
,
, U
()
p, q,
11
fo
.in
rs
yr
ea
de
or
ty
,w
.m
ha
kr
ab
(pq) [p (q)]
and
(pq) (p) q
Using these facts and the equivalence between logic and set theory, we
can obtain membership functions for p q (x , y) .
From 1st fact :
pq (x , y) = 1 - p q (x , y)
= 1 min [ p(x) , 1 - q (y)]
Uq
Eq (1)
(x , y)
Eq (2)
max [ 1
- p (x) ,
q (y)]
1 min [ p(x)
1 - q (y)]
Note :
1. Entries in last two columns of this table-2 agrees with the entries in
table-1 for pq , the proof of tautologies, read T as 1 and F as 0.
2. The implication membership functions of Eq.1 and Eq.2 are not
the only ones that give agreement with pq.
12
pq (x , y) = 1 - p (x) (1 - q (y))
Eq (3)
Eq (4)
fo
.in
rs
de
ea
yr
or
ty
,w
.m
In
kr
ab
ha
Modus Ponens
fo
.in
rs
de
ea
or
ty
,w
.m
yr
kr
ab
ha
Note : The fuzzy variables in fuzzy sets, fuzzy propositions, fuzzy relations
etc are represented usually using symbol
~ as
fo
.in
rs
de
ea
ty
,w
.m
yr
Recaps
01 Membership function
of
kr
ab
or
ha
and B with the membership functions A (x) and B (x) based on min/max
operations is A
(Eq. 01)
and B with the membership functions A (x) and B (x) based on algebraic
product is
A (x) B (x) ,
(Eq. 02)
with the membership functions A (x) and B (x) based on algebraic sum is
A
UB
x X
(Eq. 04)
(Eq. 06)
or
(Eq. 07)
h-m
x
G
0.5
0.3
0.4
0.2
0.0
(Eq. 08)
fo
.in
rs
de
ea
or
ty
,w
.m
yr
Fuzzy Propositional
A fuzzy proposition is a statement P which acquires a fuzzy truth
kr
ab
value T(P) .
ha
Example :
16
: Ram is honest
T(P) = 0.8 ,
T(P) = 1 ,
means P
is absolutely true.
fo
.in
rs
de
ea
or
ty
,w
.m
yr
Fuzzy Connectives
The fuzzy logic is similar to crisp logic supported by connectives.
ha
kr
ab
Symbols
Usage
Definition
Nagation
Disjuction
P Q
Max[T(P) , T(Q)]
Conjuction
P Q
min[T(P) , T(Q)]
Implication
P Q
1 T(P)
Here P , Q are fuzzy proposition and T(P) , T(Q) are their truth values.
the P and Q are related by the operator are known as antecedents
IF x
is A THEN y
is B,
is equivalent to
R = (A x B) U ( A x Y)
IF x is A THEN y is B, ELSE y is C
is equivalent to
R = (A x B) U ( A x C)
fo
.in
rs
yr
ea
de
P : Mary is efficient ,
T(P) = 0.8 ,
Q : Ram
T(Q) = 0.65 ,
ha
kr
ab
or
ty
,w
.m
: Mary is efficient ,
is efficient ,
i.e.
i.e.
fo
.in
rs
de
ea
yr
.m
w
w
Let
ha
kr
ab
or
ty
,w
= {a, b, c, d}
= {(a, 0)
(b, 0.8)
(c, 0.6)
(d, 1)}
= {(1, 0.2)
(2, 1)
(3, 0.8)
(4, 0)}
= {(1, 0)
(2, 0.4)
(3, 1)
(4, 0.8)}
= { 1, 2, 3, 4}
If x is A THEN y is B
(ii)
If x is A THEN y is B
Else y is C
Solution
To determine implication relations (i) compute :
The operator represents
IF x is A THEN y is B,
is equivalent to R = (A x B) U ( A x Y) and
A
a
AxB =
1
0
2
0
3
0
0.2
0.8
A x Y =
Therefore R = (A x B) U ( A x Y) gives
x
a
R =
0.2
0.8
0
0
fo
.in
rs
de
ea
yr
.m
w
w
,w
ty
or
ab
kr
ha
C
C
= {(a, 0)
(b, 0.8)
(c, 0.6)
(d, 1)}
= {(1, 0.2)
(2, 1)
(3, 0.8)
(4, 0)}
= {(1, 0)
(2, 0.4)
(3, 1)
(4, 0.8)}
IF x is A THEN y is B Else y is C,
is equivalent to
R = (A x B) U ( A x C) and
1
0
2
0
3
0
0.2
0.8
y
A
a
A x C =
0.4
0.8
R = (A x B) U ( A x C)
x
a
R =
0.2
0.8
gives
0
0
fo
.in
rs
de
ea
or
ty
,w
.m
yr
Fuzzy Quantifiers
In crisp logic, the predicates are quantified by quantifiers.
kr
ab
ha
Examples :
21
Absolute quantifiers
Relative quantifiers
almost
about
most
fo
.in
rs
de
ea
or
ty
,w
.m
yr
3. Fuzzification
The fuzzification is a process of transforming crisp values into grades of
kr
ab
ha
Low
.8
Medium
.6
.4
.2
0
20
40
60
80
Speed
100
120
140
X0 = 70km/h
V Low
Low
Medium
High
V High
.8
to fuzzy medium.
.6
.4
.2
10
20
30
40
50
60
70
80
90
00
Speed X0 = 40km/h
fo
.in
rs
de
ea
or
ty
,w
.m
yr
4. Fuzzy Inference
Fuzzy Inferencing is the core element of a fuzzy system.
kr
ab
Fuzzy Inferencing combines - the facts obtained from the fuzzification with the
ha
fo
.in
rs
de
ea
or
ty
,w
.m
yr
ha
kr
ab
If
x is A THEN y is B
x
is
is
B = A R(x, y)
where
fo
.in
rs
de
ea
or
ty
,w
.m
yr
kr
ab
If
x is A THEN y is B
ha
is B
x is A
where A , B , A , B are fuzzy terms.
Note : Every fuzzy linguistic statements above the line is analytically known
and what is below the line is analytically unknown.
To compute the membership function A , the max-min composition
of fuzzy set B with R(x , y) which is the known implication relation
(IF-THEN) is used. i.e.
A = B R(x, y)
where
fo
.in
rs
de
ea
yr
Example :
or
ty
,w
.m
Apply the fuzzy Modus Ponens rules to deduce Rotation is quite slow?
kr
ab
Given :
ha
(i)
HxS=
26
H x Y)
10
20
30
40
50
60
10
20
30
40
50
60
30
30
40
40
50
50
60
0.8
0.6
60
H x Y = 70
70
80
0.8
0.6
80
90
100
90
100
fo
.in
rs
de
ea
yr
,w
.m
30
40
50
60
30
40
50
60
70
0.8
0.6
80
0.8
0.6
ha
kr
ab
or
ty
10
R(x,Y) =
90
100
R (x, y)
= [0 0 0 0 0 0 0.9 1]
10
20
30
40
50
60
30
40
50
60
70
80
90
100
= [1 1 1 1 1 1 ]
27
fo
.in
rs
de
ea
or
ty
,w
.m
yr
kr
ab
ha
x2 is A2,
y2 is B2,
yn is Bn)
where linguistic variables xi, yj take the values of fuzzy sets Ai and Bj
respectively.
Example :
IF
THEN
C2 . . . Cn
where
y Y
is universe of discourse.
On the other hand, if the conclusion C to be drawn from a rule base R is the
disjunction of the individual consequents of each rule, then
C = C1 U
C2 U . . . U Cn
where
c (y ) = max ( c1 (y ), c2(y ) , cn (y )) ,
Y
28
is universe of discourse.
y Y where
fo
.in
rs
de
ea
or
ty
,w
.m
yr
6. Defuzzification
In many situations, for a system whose output is fuzzy, it is easier to take a
kr
ab
crisp decision
if the output
is
represented
as
ha
Centroid method
It is also known as the "center of gravity" of area method.
It obtains the centre of area (x*) occupied by the fuzzy set .
For discrete membership function, it is given by
n
x* =
i=1
n
i=1
xi (xi)
where
(xi)
and
fo
.in
rs
de
ea
1. "Neural
ha
kr
ab
or
ty
,w
.m
yr
References : Textbooks
2. "Soft Computing and Intelligent Systems Design - Theory, Tools and Applications",
3. "Fuzzy Sets and Fuzzy Logic: Theory and Applications", by George J. Klir and
Bo Yuan, (1995), Prentice Hall, Chapter 12-17, page 327-466.
5. "Fuzzy Logic: Intelligence, Control, and Information", by John Yen, Reza Langari,
(1999 ), Prentice Hall, Chapter 8-13, page 183-380.
6. "Fuzzy Logic with Engineering Applications", by Timothy Ross, (2004), John Wiley
& Sons Inc, Chapter 5-15 , page 120-603.
7. "Fuzzy Logic and Neuro Fuzzy Applications Explained", by Constantin Von Altrock,
(1995), Prentice Hall, Chapter 3-8, page 29-321.
30