unit iv
DESCRIPTION
neural net 3rd unitTRANSCRIPT
-
1UNIT-4
FUZZY LOGIC CONTROL
www.Vidyarthiplus.com
-
Fuzzy Membership Functions
Fuzzy Operations
Fuzzy Union
Fuzzy Intersection
Fuzzy Complement
2
www.Vidyarthiplus.com
-
Some info for LAB
Work on an m-file (open m-file for each task,
write your programme, save the file (e.g.,
lab2task1), then execute the file. Now, this file
has become a function in MATLAB). (see the first
weeks slides - Week 1).
Use help (e.g., help newfis) if you
dont know how to use the function. It gives you
information about how to use the function and
what parameters it requires
3
www.Vidyarthiplus.com
-
Fuzzy Membership Functions
One of the key issues in all fuzzy sets is how to determine fuzzy membership functions
The membership function fully defines the fuzzy set
A membership function provides a measure of the degree of similarity of an element to a fuzzy set
Membership functions can take any form, but there are some common examples that appear in real applications
4
www.Vidyarthiplus.com
-
Membership functions can
either be chosen by the user arbitrarily, based on the users experience (MF chosen by two users could be different depending upon their experiences, perspectives, etc.)
Or be designed using machine learning methods (e.g., artificial neural networks, genetic algorithms, etc.)
There are different shapes of membership functions; triangular, trapezoidal, piecewise-linear, Gaussian, bell-shaped, etc.
5
www.Vidyarthiplus.com
-
Triangular membership function
a, b and c represent the x coordinates of the three vertices
of A(x) in a fuzzy set A (a: lower boundary and c: upper
boundary where membership degree is zero, b: the centre
where membership degree is 1)
=
cxifcxbif
bcxc
bxaifabax
axif
xA
0
0
)(
a b cx
A(x)1
0
6
www.Vidyarthiplus.com
-
Gaussian membership function
c: centre s: width m: fuzzification factor (e.g., m=2)
A(x)
=
m
As
cxmscx
21
exp),,,(
0 1 2 3 4 5 6 7 8 9 100
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
x
c=5
s=2
m=2
7
www.Vidyarthiplus.com
-
c=5
s=0.5
m=20 1 2 3 4 5 6 7 8 9 10
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 100
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
c=5
s=5
m=2
8
www.Vidyarthiplus.com
-
c=5
s=2
m=0.2
c=5
s=5
m=5
0 1 2 3 4 5 6 7 8 9 100.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 100
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
9
www.Vidyarthiplus.com
-
Fuzzy Operations
(Fuzzy Union, Intersection, and Complement)
Fuzzy logic begins by barrowing notions from crisp logic, just as fuzzy set theory borrows from crisp set theory. As in our extension of crisp set theory to fuzzy set theory, our extension of crisp logic to fuzzy logic is made by replacing membership functions of crisp logic with fuzzy membership functions [J.M. Mendel, Uncertain Rule-Based Fuzzy Logic Systems, 2001]
In Fuzzy Logic, intersection, union and complement are defined in terms of their membership functions
This section concentrates on providing enough of a theoretical base for you to be able to implement computer systems that use fuzzy logic
Fuzzy intersection and union correspond to AND and OR, respectively, in classic/crisp/Boolean logic
These two operators will become important later as they are the building blocks for us to be able to compute with fuzzy if-then rules
10
www.Vidyarthiplus.com
-
Logical AND () Logical OR (U)Truth Table
A B A B
0 0 0
0 1 0
1 0 0
1 1 1
Truth TableA B A UU B0 0 00 1 11 0 11 1 1
Crisp UnionCrisp Intersection
AAB
B
Classic/Crisp/Boolean Logic
11
www.Vidyarthiplus.com
-
Fuzzy Union
The union (OR) is calculated using t-conorms t-conorm operator is a function s(.,.) Its features are
s(1,1) = 1, s(a,0) = s(0,a) = a (boundary) s(a,b) s(c,d) if a c and b d (monotonicity) s(a,b) = s(b,a) (commutativity) s(a,s(b,c)) = s(s(a,b),c) (associativity)
The most commonly used method for fuzzy union is to take the maximum. That is, given two fuzzy sets A and B with membership functions A(x) and B(x)
))(),(max()( xxx BAAUB =12
www.Vidyarthiplus.com
-
Fuzzy Intersection
The intersection (AND) is calculated using t-norms. t-norm operator is a function t(.,.) Its features
t(0,0) = 0, t(a,1) = t(1,a) = a (boundary) t(a,b) t(c,d) if a c and b d (monotonicity) t(a,b) = t(b,a) (commutativity) t(a, t(b,c)) = t(t(a,b),c) (associativity)
The most commonly adopted t-norm is the minimum. That is, given two fuzzy sets A and B with membership functions A(x) and B(x)
))(),(min()( xxx BABA =
13
www.Vidyarthiplus.com
-
Fuzzy Complement
To be able to develop fuzzy systems we also have to deal with NOT or complement.
This is the same in fuzzy logic as for Boolean logic
For a fuzzy set A, A denotes the fuzzy complement of A
Membership function for fuzzy complement is
)(1)( xx AA
=
14
www.Vidyarthiplus.com
-
15
4.2 Knowledge Base
1. The database provides necessary
definitions that are used to define
linguistic variables and fuzzy data
manipulation in the FLC.
2. The rule base characterizes the
control goals and control policy by
means of a set of linguistic control
rules.
www.Vidyarthiplus.com
-
16
5.2 Knowledge Base
Fig. 4(a) Membership functions for Error
- 2 - 1 .5 - 1 - 0 .5 0 0 .5 1 1 .5 2
0
0 .2
0 .4
0 .6
0 .8
1
E R R O R
D
e
g
r
e
e
o
f
m
e
m
b
e
r
s
h
i
p
N L N M N S Z E P S P M P L
www.Vidyarthiplus.com
-
17
5.2 Knowledge Base Cont.
Fig. 4(b) Membership functions for Change in Error
- 0 . 5 - 0 . 4 - 0 . 3 - 0 . 2 - 0 . 1 0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5
0
0 . 2
0 . 4
0 . 6
0 . 8
1
C H A N G E I N E R R O R
D
e
g
r
e
e
o
f
m
e
m
b
e
r
s
h
i
p
N L N M N S Z E P S P M P L
www.Vidyarthiplus.com
-
18
5.2 Knowledge Base Cont.
Fig. 4(c) Membership functions for Change in Output
- 0 .0 3 - 0 .0 2 -0 .0 1 0 0 .0 1 0 .0 2 0 .0 3
0
0 .2
0 .4
0 .6
0 .8
1
C H A N G E IN O U T P U T
D
e
g
r
e
e
o
f
m
e
m
b
e
r
s
h
i
p
N L N M N S Z E P S P M P L
www.Vidyarthiplus.com
-
19
4.3 Decision Making Logic
Based on the fuzzified inputs and rule
base, fuzzy output is determined by
applying the rules of Boolean Algebra
(Union and Intersection).
www.Vidyarthiplus.com
-
20
Genetic Algorithms (GA) OVERVIEW
A class of probabilistic optimization algorithms
Inspired by the biological evolution process
Uses concepts of Natural Selection and Genetic Inheritance (Darwin 1859)
Originally developed by John Holland (1975)
Particularly well suited for hard problems where little is known about the underlying search space
Widely-used in business, science and engineering
www.Vidyarthiplus.com
-
21
Classes of Search TechniquesSearch Techniqes
Calculus Base
TechniqesGuided random search
techniqes
Enumerative
Techniqes
BFSDFS Dynamic
Programming
Tabu Search Hill Climbing Simulated
Anealing
Evolutionary
Algorithms
Genetic
Programming
Genetic
Algorithms
Fibonacci Sort
www.Vidyarthiplus.com
-
22
A genetic algorithm maintains a population of
candidate solutions for the problem at hand,
and makes it evolve by
iteratively applying a set of stochastic
operators.
www.Vidyarthiplus.com
-
23
Stochastic operators
Selection replicates the most successful solutions
found in a population at a rate proportional to
their relative quality
Recombination (Crossover) decomposes two
distinct solutions and then randomly mixes their
parts to form novel solutions
Mutation randomly perturbs a candidate solution
www.Vidyarthiplus.com
-
24
Simple Genetic Algorithm{
initialize population;evaluate population;while TerminationCriteriaNotSatisfied{
select parents for reproduction;perform recombination and mutation;evaluate population;
}}
www.Vidyarthiplus.com
-
25
The Evolutionary Cycle
selection
population evaluation
modification
discard
deleted
members
parents
modified
offspring
evaluated offspring
initiate &
evaluate
www.Vidyarthiplus.com
-
26
GA Cycle
www.Vidyarthiplus.com
-
27
Crossover
www.Vidyarthiplus.com
-
28
Mutation
www.Vidyarthiplus.com