Dr.-Ing. Erwin Sitompul President University Lecture 7 Introduction to Neural Networks and Fuzzy Logic President University Erwin Sitompul NNFL 7/1 http://zitompul.wordpress.com
Jan 12, 2016
Dr.-Ing. Erwin SitompulPresident University
Lecture 7
Introduction to Neural Networksand Fuzzy Logic
President University Erwin Sitompul NNFL 7/1
http://zitompul.wordpress.com
President University Erwin Sitompul NNFL 7/2
Meaning of “fuzzy”IntroductionFuzzy Logic
Covered with fuzz;Of or resembling fuzz;Not clear; indistinct
A fuzzy recollection of past events.Not coherent; confused
A fuzzy plan of action.Unclear, blurred, or distorted
Some fuzzy pictures from a Russian radar probe.
President University Erwin Sitompul NNFL 7/3
4 Seasons
0
0.5
1
Time of the year
Mem
bers
hip
Spring Summer Autumn Winter
IntroductionFuzzy Logic
President University Erwin Sitompul NNFL 7/4
Tall Persons
0 : A person is not tall
1 : A person is tall
IntroductionFuzzy Logic
President University Erwin Sitompul NNFL 7/5
Incorporation of human’s perception
0 : room is not warm1 : room is warm
IntroductionFuzzy Logic
Room Temperature
President University Erwin Sitompul NNFL 7/6
Characteristic function:
young
1, age( ) 20( )
0, age( ) 20
xx
x
young = { x P | age(x) ≤ 20 }
Set DefinitionFuzzy Logic
Classical Sets
A=“young”1
0
A ( )x
yearsx20x
President University Erwin Sitompul NNFL 7/7
Classical Logic
Element x belongs to set Awith a certain “degree of membership”:
(x)[0,1]
Element x whether belongs to set A or not at all:
(x){0,1}
Fuzzy Logic
A=“young”1
0
A ( )x
yearsx21x
A=“young”1
0
21x
A ( )x
yearsx
Set DefinitionFuzzy Logic
Fuzzy Sets
President University Erwin Sitompul NNFL 7/8
yearsx
A=“young”1
0
21x
A ( )x
0.4A
Fuzzy Set A = {(x,A(x)) | x X, A(x) [0,1]}is defined by a universe of discourse x where0 ≤ x ≤ 100 and a membership function A where A(x) [0,1]
Definition:
Fuzzy SetsSet DefinitionFuzzy Logic
President University Erwin Sitompul NNFL 7/9
x
(x)
1
0
Set DefinitionFuzzy Logic
Some DefinitionsSupport of a fuzzy set A
supp(A) = { x X | A(x) > 0 }Core of a fuzzy set A
core(A) = { x X | A(x) = 1 }α-cut of a fuzzy set A
Aα = { x X | A(x) α}
α = 0.6
President University Erwin Sitompul NNFL 7/10
Fuzzy Logic Control (FLC)Fuzzy Logic ControlFuzzy Logic
Fuzzy Logic Control (FLC) may be viewed as a branch of intelligent control which serves as an emulator of human decision-making behaviour which is approximate rather than exact.
FLC uses the IF-THEN rules, similar to binary control (Programmable Logic Controller, PLC).
Rule Format:Ri: IF x is Aj AND y is Bk THEN z is Cl
Ri: IF x is Aj OR y is Bk THEN z is Cl
President University Erwin Sitompul NNFL 7/11
Logic Operators
A B A B A B
Fuzzy Logic OperatorsFuzzy Logic
President University Erwin Sitompul NNFL 7/12
p q p q 0 0 0 0 1 1 1 0 1 1 1 1
p q p q 0 0 0 0 0.4 0.4 0 1 1
0.4 0 0.4 0.4 0.4 0.4 0.4 1 1 1 0 1 1 0.4 1 1 1 1
Boolean OR Fuzzy OR
Fuzzy Logic OperatorsFuzzy Logic
Boolean OR and Fuzzy OR
President University Erwin Sitompul NNFL 7/13
p q p^q 0 0 0 0 1 0 1 0 0 1 1 1
p q p^q 0 0 0 0 0.4 0 0 1 0
0.4 0 0 0.4 0.4 0.16 0.4 1 0.4 1 0 0 1 0.4 0.4 1 1 1
Boolean AND Fuzzy AND
Boolean AND and Fuzzy ANDFuzzy Logic OperatorsFuzzy Logic
President University Erwin Sitompul NNFL 7/14
Example: Air FanFuzzy Logic ControlFuzzy Logic
Conventional (On-Off) Control:IF temperature > X °C,
THEN run fan,ELSE stop fan.
Fuzzy Control: IF temperature is hot,
THEN run fan at full speed;
IF temperature is warm, THEN run fan at moderate speed;
IF temperature is comfortable, THEN maintain fan speed;
IF temperature is cool, THEN slow fan;
IF temperature is cold, THEN stop fan.
President University Erwin Sitompul NNFL 7/15
Example: Stopping A Car
13600 N 0 N1500 kg
(0) 25 m(0) 10 m s
Fm
yy
Break force
Mass of the car
Initial position
Initial velocity
FF my y
m
Fuzzy Logic ControlFuzzy Logic
President University Erwin Sitompul NNFL 7/16
Example: Stopping A Car
( )p dK e T ey
m
pF K e
, 0e w y w
p pK e K yy
m m
With Kp = –240, the car will stop at the traffic light after 10 s.
P-Control PD-Control
2
p
p d p
K my
w s K mT s K m
Choosing ζ = 1, Td = 1, Kp = 6000, the car will stop at the traffic light after 5 s.
Fuzzy Logic ControlFuzzy Logic
2
2 22n
n ns s
President University Erwin Sitompul NNFL 7/17
Example: Stopping A CarFuzzy Logic ControlFuzzy Logic
Fuzzy Logic Control: IF distance is long AND approach is fast,
THEN brake zero; IF distance is long AND approach is slow,
THEN brake zero; IF distance is short AND approach is fast,
THEN brake hard; IF distance is short AND approach is slow,
THEN brake zero.
President University Erwin Sitompul NNFL 7/18
Fuzzy Membership FunctionsFuzzy Logic ControlFuzzy Logic