TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.5, May 2014, pp
. 3825 ~ 38
3
1
DOI: http://dx.doi.org/10.11591/telkomni
ka.v12i5.5098
3825
Re
cei
v
ed
No
vem
ber 1
2
, 2013; Re
vi
sed
De
cem
ber 3
1
,
2013; Accep
t
ed Jan
uary 1
4
, 2014
Fuzzy Control Analysis with Back-EMF in Washing
Machin
es
Min-
y
a
n DI
Heb
e
i Norm
al
Univers
i
t
y
, Sh
ij
iazh
uan
g He
be
i 050
02
4, Chin
a,
T
e
l.: 0086-0
3
1
1
-80
787
94
2, fax: 00
86-0
3
1
1
-8
078
79
42
e
m
a
i
l
:
mi
ny
andi
@
1
26
.co
m
A
b
st
r
a
ct
T
h
is pa
per
put
s forw
ard a
method
of detec
ting
mate
r
i
al
a
nd q
u
a
n
tity of the cloth
in
w
a
shin
g
mac
h
i
ne bas
e
d
on detecti
on
motor back E
M
F
,
not
only can obta
i
n the
pulse n
u
m
ber
of reporting t
h
e
infor
m
ati
on q
u
a
lity an
d q
uanti
t
y of the cl
oth, but als
o
can c
o
llect infor
m
atio
n of puls
e
w
i
dth, the puls
e
cy
cle,
hel
p to
i
m
prov
e the
detecti
on
accur
a
cy of th
e
materi
al
an
d
qua
ntity of th
e
cloth. On th
is b
a
sis, T
h
is
articl
e
also
gives
a s
pecific c
ontro
l
rules
of the fu
zz
y
c
ont
rol
of
auto
m
atic w
a
s
h
in
g
mac
h
in
e, W
i
th this fu
zzy
control r
u
le ta
b
l
e, off-lin
e F
u
zzy control r
e
spo
n
se ca
n
b
e
obt
ain
ed. F
i
na
lly
e
nqu
iry for
m
us
ed to Pro
g
ra
m
of
asse
mb
ly lan
g
uag
e, deter
mi
n
ed
materi
al a
n
d
qua
ntity
an
d
dirty degr
ee o
f
the clot
h, to achi
eve the b
e
s
t
effect of clothe
s w
a
shing
by s
e
lecti
ng w
a
ter l
e
vel, fl
ow
i
n
ten
s
ity, put the
a
m
o
unt of
deter
gent a
nd w
a
s
h
i
n
g
tim
e
.
Ke
y
w
ords
:
fu
zz
y
co
ntrol w
a
shin
g machi
ne,
det
ectio
n
lo
ad, back EMF
of motor;
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
To
ma
ke
the fuzzy co
ntrol
rule of
the wa
shin
g ma
chi
n
e. First u
s
e
a
variety of
se
nso
r
s to
detect the
st
atus of
wa
shi
ng related, a
s
the
c
ont
rol
basi
s
. Fo
r ex
ample, the
re
sea
r
ch u
s
e t
he
load se
nsor,
temperatu
r
e
sen
s
or, water level
se
n
s
ors, photo
e
l
e
ctri
c
se
nsors, and provide
obje
c
tive informatio
n ne
cessary for a
u
tomatic de
cision
-ma
k
ing
level, curre
nt intensity and
wa
shin
g time
. The load
se
nso
r
is o
ne o
f
the key in the fuzzy
control wa
shin
g m
a
chi
ne, its m
a
in
task i
s
to det
ect the imp
e
d
ance (i.e. cl
oth materi
al an
d cloth q
uanti
t
y), and so th
is pa
per m
a
in
ly
studie
s
the washi
ng ma
chi
ne load
sen
s
or [1-3].
2.
The Traditio
nal Load Se
nsor for a
Washing Mac
h
ine
As early a
s
t
w
entieth
Cen
t
ury ago,met
hod for m
e
a
s
uring th
e mat
e
rial a
nd qu
a
n
tity
o
f
cloth ,
dete
c
tion of
com
m
o
n
ly used
so
-called d
r
y
te
st: put the l
aun
dry into th
e
washi
ng m
a
chi
ne
in water
befo
r
e the
loa
d
t
e
st, the
digit
a
l tachomete
r
recordi
ng t
he moto
r
sp
eed, me
mory
the
transi
ent
re
cordin
g
of mo
tor, ele
c
tro
n
i
c
cou
n
ter record
s
wavefo
rm processin
g
after
analy
z
ed.
Logi
c analyzes the ch
ang
e of stat
e. The operating state paramet
ers a
r
e dete
r
mined a
c
cord
ing
to the test results of wa
sh
ing machine.
But t
here will
be
variou
s
in
fluences on t
he load test i
n
the detectio
n
pro
c
e
ss [4
-5]:
(1) Effect of
motor
sp
eed:
moto
r
spee
d have
a g
r
e
a
t effect o
n
t
e
st. As
moto
r spee
d
increa
se
s, the load te
sts
data rai
s
e, p
r
ese
n
ts a
n
o
n
linear relatio
n
shi
p
.
The
re
sults also sh
ow
that in the low
speed to a cert
ain value, the detecti
on m
e
thod
will be fail
ed.
This is the m
a
in
influen
ce of d
r
iving sp
eed
stabili
zation
circuit.
(2)
Effect
s of
startin
g
an
d
bra
k
in
g cha
r
acteri
stics:
risein
g loa
d
chara
c
te
risti
c
s have a
good
pe
rform
ance in
singl
e pha
se
capa
citor m
o
tor,
b
u
t for the i
n
fluen
ce of
se
ri
es ex
cited m
o
to
r
itself charact
e
risti
c
s an
d
perip
he
ral d
r
i
v
ing ci
rc
uit ef
fects
app
ear
messy an
d rules are h
a
rd
to
find, espe
cial
ly in the light load and he
avy load co
n
d
itions is p
a
rt
icula
r
ly promi
nent, It is mo
re
difficult to distingui
shing th
e cha
r
a
c
teri
st
ics
wh
e
n
there is cloth
e
s fri
c
tion an
d a light load.
(3)
Affect o
f
the mecha
n
ical
pro
p
e
rt
ies:
clot
hing an
d
room b
ody
friction
and drive tran
smissio
n
sy
st
em have a
relatively
larg
e impa
ct on
the
test,
the forme
r
chan
g
e
s
along
with
th
e differe
nt produ
cts,
the
latter
hav
e t
he
effect on
whe
n
the
cl
othes
to a
ce
rt
ain
extent. The two shoul
d be
con
s
ide
r
ed i
n
the test method.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3825 – 38
31
3826
(4)
Sup
p
ly voltage effects:
the power
sup
p
ly voltage have a gre
a
t effect on the test
results, the te
st data ch
ang
e al
ong
with the voltage flu
c
tuation.
In view
of the
above
all
so
rts of effe
cts
a
nd
co
mpl
e
x
sen
s
o
r
circuit
(m
ainly by t
he d
r
ive
circuit, stea
d
y
spee
d ci
rcu
i
t, measu
r
ing
circuit an
d di
gital pro
c
e
s
si
ng unit) m
a
ke
s the me
asuri
n
g
s
p
ee
d
me
th
od
lo
ad
s
e
nso
r
o
f
h
i
gh
co
s
t
an
d
t
he
testing
accu
racy n
o
t hi
gh
, therefo
r
e, it
is
repla
c
e
d
by the mea
s
u
r
ed
EMF method.
3.
The Method
of Mea
s
ure
m
ent Load
Based on
Detection Mo
tor
Back Emf
After addi
ng
a small
amou
nt of water to
the
wa
shin
g
machi
ne, the
motor
ru
ns wi
th load
for a
sh
ort ti
me. Wh
en th
e po
we
r
supp
ly is cut off sudde
nly, the
motor
rotatio
n
will
rem
a
in
for a
sho
r
t pe
riod o
f
time due to the ine
r
tia effe
ct. At this poi
nt, the rotor
remane
nce wil
l
cut stato
r
an
d
then ind
u
ce
potential, this potential
can
be ta
k
en
out
of the moto
r windi
ng. Un
der th
e da
mp
ing
effect of clot
hing, motor
spe
ed falls
and ba
ck
EMF also i
n
creases
with time and fina
lly
disa
ppe
ars sho
w
in
g a trend of da
mping o
sc
ill
ation.
The attenuation and
o
scill
ation
cha
r
a
c
teri
stics of
ba
ck E
M
F si
gnal
a
r
e relate
d
to
the qu
antity and
materi
al
of the
wa
sh
ed
clothe
s. In general, the mo
re cloth
e
s a
r
e, the fast
er the co
unter p
o
tential atten
uation will be
[6-
8].
3.1. The Gen
e
ral Meth
od of Ba
ck EMF
Dete
ction
We a
c
hi
eve the re
sult of lo
ad testin
g by det
ectin
g
the
motor’
s ba
ck
EMF, and the
motor
back EM
F v
o
ltage i
s
m
u
ch
more la
rg
er tha
n
singl
e chip mi
cro
c
omp
u
ter. S
o
we mu
st
use
isolatio
n but not dire
ct sa
mpling. Usin
g a linea
r ph
otoele
c
tric
co
upler
can i
s
o
l
ate high voltag
e
and interfe
r
e
n
ce, and o
b
tain satisfa
c
to
ry detection
signal. So we turn the ba
ck EMF AC signal
after a po
we
r outage to
DC sig
nal by h
a
lf wave
recti
f
ier, and to p
u
lse
sign
al af
ter photo
e
le
ctric
isolatio
n a
n
d
sh
apin
g
am
plification, th
en
we
se
nd
them to th
e
microcounte
r
to me
asure
the
pulse numb
e
r (as
sho
w
n in
Figure 1
)
.
Figure 1. The
Circuit of Pulse Detectio
n
Figure 2. The
Analysis of Cloth Quantity
Acco
rdi
ng to
the coll
ecte
d numb
e
r
of pulse
s,
we
can
determin
e
the am
oun
t of cloth
quantity. We
con
s
id
er that
the pulse n
u
mer i
s
le
sser, the lau
n
d
r
y is mo
re, a
nd vice ve
rsa(a
s
s
h
ow
n
in
F
i
gu
r
e
2)
.
Figure 3. The
Analysis of Cloth Material
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Fuz
z
y
Cont
ro
l A
nalysi
s wit
h
B
a
ck
-E
MF i
n
Wa
shin
g Machi
n
e
s
(Min
-
y
an
DI
)
3827
As for the cloth materia
l
, it can be di
vided into cot
t
on, blended,
synthetic fibe
r this thre
e
categ
o
rie
s
. Cotton and
sy
nthetic fibe
r
clothe
s h
a
ve
different da
mping u
nde
r different wa
ter
levels.
Due
to
the
synthetic fiber cl
othe
s’
se
nsit
ive
deg
ree
to
water l
e
vel is not
as goo
d a
s
cott
on
clothe
s, so
we can ju
dge t
he garment
material
by shutting off the motor an
d measuri
ng th
e D-
value of pul
se numb
e
r u
n
der different
water l
e
vel. The bigg
er the
pulse
numb
e
r is, mo
re li
kely
the cloth i
s
cotton. While t
he sm
aller th
e pulse num
b
e
r is, mo
re li
kely the
c
l
oth
is
s
y
nthetic
fiber
(as sho
w
n
in Figure
3).
3.2 The Improv
ed Method of Ba
ck EMF De
tec
t
io
n
From th
eo
retical a
nalysi
s
a
nd expe
rime
n
t
al observatio
n
, we fin
d
tha
t
this meth
od
is on
e-
side
d for it only use
s
th
e pulse num
ber info
rm
ati
on, and its
ability to distingui
sh the cloth
material
is no
t high,
and
its dispe
r
si
on i
s
relatively
la
rg
e. Ma
king
u
s
e of
a n
e
w ki
nd of
cl
oth fa
bri
c
quantity interf
ace
circuit(as sho
w
n in Fi
g
u
re 4
)
,
we
ca
n not only det
ect the nu
mb
er of pul
se
s, but
also
colle
ct pulse width,
pulse
cycle
(frequ
en
cy).
We ca
n see the attenual o
scill
ation
cha
r
a
c
teri
stics of ba
ck EM
F ideally an
d
comp
re
hen
si
vely. In the condition of th
e sam
e
nu
m
ber
of pulse, we
can furth
e
r di
stingui
sh bet
wee
n
cloth farbi
c
and
clo
t
h quantity, then imp
r
ove the
detectio
n
accura
cy.
Figure 4. The
Circuit Princi
ple Dia
g
ra
m of Load Dete
ction
3.2.1. The Te
sting of
Cloth Quanti
t
y
After the
clot
hes are p
u
t i
n
to the la
und
ry, we
drain and start
th
e motor,
u
s
in
g th
pul
se
voltage which
is on
for 0.3
s
and off for 0.
7s to d
r
ive th
e motor for 3
2
s. In thi
s
pro
c
e
ss,
we u
s
e
a
photoel
ectri
c
co
uple
to send
and
re
ceive pul
se
s f
o
r m
e
a
s
uri
n
g
the rotation
al ine
r
tia
circle
number
when the roulette
are
s
h
ut off. At this
poin
t, t
he c
l
othing is much more,
the rotation time
is sh
orte
r, the pulse n
u
mbe
r
is
less
er. More is
the oppos
i
te.
Cou
n
t Pulse
and Cl
othing
Weig
ht’s Rel
a
tion Cu
rve h
a
s Sho
w
n in
Figure 5.
Figure 5. Analysis of Clothi
ng Qua
n
tity
Fi
gure 6. The
Discrimi
natio
n Curve of Cotton
and Synthetic Fiber Pro
d
u
c
ts
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3825 – 38
31
3828
3.2.2.
The T
e
s
t
i
n
g
o
f
C
l
o
t
h
Material
The follo
wing
includ
es th
e
disting
u
ish
b
e
tw
ee
n cotton and
synthe
tic fiber p
r
od
ucts
as
well a
s
the differen
c
e bet
ween soft fabric and thi
ck
cl
oth. Definite method
s are as follo
ws:
Based
on th
e
test loa
d
, we
drai
n a
w
ay li
ttle water,
an
d u
s
e the
pul
se
wate
r
whi
c
h i
s
o
n
for 0.3
s
and
off for 0.7s t
o
actu
ate the
motor fo
r 32
s, then
re
cord the pul
se
n
u
mbe
r
N; If the
detectio
n
p
u
l
s
e l
oad
num
ber is M, the
n
we
can
de
termine
the
gene
ral
situa
t
ion of m
a
terial
distrib
u
tion d
epen
ding o
n
the value of M
-
N. The
cotto
n is mu
ch mo
re, the value
of M-N i
s
larg
er,
the smalle
r is the oppo
site.
Whe
n
the
p
r
o
portion
of
cot
t
on an
d
synth
e
tic p
r
o
d
u
c
ts
is diffe
rent, th
e twi
c
e
curve
of the
measured pul
se num
be
r ha
s sh
own in Figure 6.
Cotton g
ood
s are the
sam
e
, but for the h
a
rd a
nd thi
ck
cloth to
wel like denim
and t
he soft
fabric,
the
wa
shin
g meth
od
s a
r
e
differe
n
t. How to di
sti
ngui
sh?
It ne
eds the
wate
r level
sen
s
o
r
to
measure. The spe
c
ific me
thod is: After the pulse
dri
v
e for 32s by
water inje
cti
on, we co
mp
are
the chan
ge o
f
water qu
anti
t
y.If the variat
ion i
s
sma
ll, i
t
prove
s
the
cloth i
s
easy
to ab
sorb
wat
e
r
and ten
d
s to
be to
wel
clo
t
h, or mayb
e
denim
fabri
c
. Figure 7 i
s
the wate
r-level cu
rve in
b
o
th
ca
se
s.
Soft fabric o
r
hard thi
c
k cl
o
t
h also
ca
n b
e
identified
b
y
the wavefo
rm chan
ge
s o
f
analog
cur
r
e
n
t
(as
sh
own in Fig
u
r
e
8).
Figure 7. The
Water L
e
vel Curve of Soft
Fabri
c
an
d Hard Thi
c
k Clo
t
h
Figure 8. The
Compa
r
i
s
on
of Two Typical
Current Wav
e
form in Different Hardness
Cloth
4.
The De
sign of Soft
w
a
r
e
and Ha
rd
w
a
r
e
in Fuzzy
Control Washi
ng Machine
Fuzzy control
wa
shin
g ma
chin
e refers t
o
the
wa
shin
g ma
chin
es judge th
e
wei
ghts, the
texture and t
he deg
re
e of
pollution of t
he cl
othes
th
roug
h sen
s
ors, to automat
ically dete
r
mi
ne
the wate
r level of high an
d low, the do
sag
e
of
detergent, temperature a
nd wa
ter flow inten
s
ity,
and
dete
r
min
e
the
optimu
m
cl
eanin
g
seque
nce. As
a
result of th
e control o
b
je
ct of th
e
wa
shing
pro
c
e
s
s i
s
dif
f
icult to
use
accurate m
a
thematic
al m
o
del to
de
scri
be, so th
e traditional
cont
rol
method is diff
icult to obtain
ideal cle
anin
g
effe
ct, whil
e the fuzzy control meth
o
d
can
well so
lve
the p
r
obl
em. In the
p
r
o
c
ess of
washi
ng
clothin
g
,
how mu
ch
of
the
cloth
e
s,
fabri
c
s
su
ch
as
hard
n
e
s
s an
d softness a
r
e fuzzy, so
a
t
first, we
m
u
st d
o
a l
o
t
of experi
m
en
ts, co
ncl
ude
the
artificial washing metho
d
, thus formin
g the fuzzy
control rule
s.
Acco
rdin
g to the inform
ation
received from
the sensor, the wa
shin
g machi
ne det
e
r
mine
s ho
w much of the clothe
s, hard
n
e
ss
and
smu
dgy
degree
of the
fabri
c
, and
refers to a fu
zzy de
cisi
on, t
hus, fini
she
s
water inje
ctio
n
stren
g
th, wa
shin
g time, the flow stren
g
th, wash
ing
way, dehydration time, draina
ge, etc,
all
function
s
and
automati
c
ally
com
p
lete
s
such
a
s
, aut
o
m
atically
com
p
lete the "
w
at
er", "wa
s
hi
ng
”, ”
draini
ng
”, “de
h
ydration
” an
d other fun
c
tions.
4.1. Hard
w
a
r
e
Design o
f
Fuzz
y
Control Washing
Machine
Fuzzy cont
rol
princi
ple dia
g
ram of the intelligent wa
shing ma
chine
refers to Figure 9, it
is a multipl
e
input and
multiple outp
u
t control sy
stem. Multipl
e
input in
clu
des th
e water
temperature,
water level, t
he lig
ht tra
n
smittance
of
water
and the tes
t
data
of motor’s rotational
prop
ertie
s
aft
e
r do
wntim
e
. The data
accept fu
zzy
p
r
oce
s
sing
after input to the
control sy
ste
m
,
the system a
c
cordi
ng to the pro
c
e
s
sin
g
re
sult
s a
n
d
the control rules
summ
ari
z
ed by ma
nu
al
wa
shin
g, ca
rries on fu
zzy control op
era
t
ion, obt
ains t
he co
ntrol d
e
c
isi
on, and
control the
out
put
by cont
rolling the actuat
ors.
During t
he proce
ss,
the system
uses the test data of water
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Fuz
z
y
Cont
ro
l A
nalysi
s wit
h
B
a
ck
-E
MF i
n
Wa
shin
g Machi
n
e
s
(Min
-
y
an
DI
)
3829
temperature
and the light transmi
ttance
of water as
basi
s
to anal
yze and jud
g
e
the dissolut
ion
con
d
ition of
water on
det
erge
nt an
d
smudgy d
egr
e
e
of the
cloth
i
ng, dete
r
min
e
s th
e am
ou
nt of
deterg
ent a
n
d
wa
shi
ng ti
me. It use
s
t
he ine
r
tia te
st data of the
wa
shin
g mot
o
r in
low run
n
ing
spe
ed a
s
the basi
s
to ana
lyze and jud
g
e
the mate
ria
l
material an
d the amount
of the clothing,
determi
ne
s the amo
unt o
f
water
sup
p
l
y
and the
water flow
mo
de when
wa
shin
g. The la
tter
achi
eves
the process
by controlling
the motor’
s sp
eed and it
s posi
tive and
negative operation. In
addition, the system al
so
adopt
s testin
g water le
vel
to ensure a
u
tomatic water su
pply as
the
pred
efined l
e
vel. By testing the light transmitta
n
c
e of
water wh
e
n
cle
aning (i.
e
.,
rinse
)
,
the
system ju
dge
s the cl
ear
condition of water to ma
ke
sure the ri
nse pro
c
e
s
s is
end, switch
e
s
to
dewaterin
g o
peratio
n auto
m
atically [9-1
1].
Figure 9. The
Fuzzy Control Princi
ple Di
agra
m
of the Intelligent Wa
shin
g Machin
e
4.2. Soft
w
a
r
e
Design o
f
Fuzz
y
Rule
Ware
room
There a
r
e m
a
inly thre
e in
put varia
b
le
s of fuzzy con
t
rol washing
machi
ne: the
materi
al
material
of cl
oth, the am
o
unt of
cl
oth,
smud
gy de
gree; The
r
e
ar
e mainly fou
r
output va
ria
b
les:
the level of the wa
shin
g machi
ne, wa
shin
g time
, tempe
r
ature a
nd detergent
dose, so it is a 3
input 4 output
stru
cture. Inp
u
t and output
variable
s
hav
e respe
c
tively several
kind
s of states:
a) The Inp
u
t
Cloth qu
antity: more, many, lest, rare
ly (discourse dom
a
in is 0-25 kg).
Cloth mate
ria
l
: cotton, blen
ding, syntheti
c
fi
ber; (disco
urse dom
ain i
s
1% - 10
0%
cotton
conte
n
t).
Smudgy d
e
g
r
ee: ve
ry di
rty, generally dirty,
not
too di
rty. (aft
er
qua
ntizati
on, the
discou
rse do
main is 0
-
10
0
)
.
b) The O
u
tpu
t
Wa
shin
g time: short, sh
orte
r, middling, lo
ng, longe
r;
Was
h
ing dose; rarely, less, many, more, mos
t;
Wate
r-flo
w st
rength: wea
k
, medium an
d stron
g
;
Water level: very low, low,
high, very high.
The buil
d
ing
of a fuzzy
co
ntrolle
r
rule f
o
rm a
nd fu
zzy inferen
c
e
rule table i
s
b
a
se
d on
the
expert knowl
edge a
n
d
a manual
operator’
s
lo
ng-
te
rm exp
e
rien
ce a
c
cu
mulation; it
is a
langu
age
of repre
s
e
n
tation
reasonin
g
base
d
on
intuit
ion.
Usually it is conn
ecte
d
by a series
of
words, ta
king
the
wa
shi
ng pro
c
e
s
s h
a
s
th
ree
m
a
in fuzzy
inf
e
ren
c
e a
n
tecedent
and f
our
fuzzy rea
s
oni
ng into acco
u
n
t, using IF-T
HEN
rule
s, and its form
s o
f
expression f
o
r:
IF (the amo
u
n
t of clothin
g
)
IS (the am
o
unt of
clothi
n
g
of a certain
lingui
stic val
ue) a
n
d
(the material
material of
clothin
g
) IS (the materi
al material of a certain lingu
istic value) a
n
d
(sm
udgy de
gree) IS (sm
u
d
g
y degre
e
of a certai
n ling
u
istic valu
e).
THEN
(water
level) IS (water level of a
certai
n lin
gui
stic value
)
an
d (water flo
w
int
ensity
)
IS (water flow intensity of a certain linguisti
c
value) and (wa
s
hin
g
time) IS (washi
ng time of a
certai
n lingui
stic value)a
nd
(wa
s
hi
ng do
se) IS
(wa
s
hin
g
dose of a certain ling
u
isti
c value).
In orde
r to
re
pre
s
ent fu
zzy
rules
co
nci
s
ely, we use d
i
gital to stand
for the fuzzy control
output. For
e
x
ample: wa
sh
ing
time (ve
r
y
sh
ort, sh
or
t,
medium, l
ong
, very lon
g
)
= (1,
2,
3,
4,
5),
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 3825 – 38
31
3830
the method of the remain
ing 3
output
are si
mila
r to
this,
when th
e output dom
ains a
r
e
3, they
sho
u
ld be rep
r
esented by 3
numbe
rs, As sho
w
n in Ta
ble 1:
Table 1. Outp
ut (digital re
prese
n
tation)
Output
(digital rep
r
esentation)
1
2
3
4
5
w
a
shing time
short
shorter
middle
long
longer
w
a
shing dose
rarel
y
less
man
y
more
most
w
a
ter
-
f
low intensity
w
e
ek
medium
strong
--
--
w
a
ter
level
very
low
low
high
ver
y
high
--
Acco
rdi
ng to the expert
s
’ e
x
perien
c
e a
n
d
com
b
in
ing
with the actu
al situation of
washing
clothe
s, the fuzzy control rule
s ca
n be
obtai
ned a
s
sho
w
n in Table 2. The fuzzy rul
e
s a
r
e
rep
r
e
s
ente
d
by four digits,
the first digit is the
height
of the water level, the secon
d
is the flow
intensity, the third is th
e a
m
ount of det
erge
nt, f
ourth
is the washi
ng time, su
ch
as "43
55"me
ans
high water, st
rong
wate
r flow, many dete
r
gent, long time wa
shin
g ‘control way.
Table 2. Fu
zzy Control Rul
e
Table of Ful
l
-autom
atic M
a
chi
n
e
cloth
very
dir
t
y
generall
y
dirt
y
not too dirt
y
cloth material
cloth quantit
y
sy
n
t
hetic fiber
more
4355
4234
4223
man
y
3244
3233
3222
lest
2233
2222
2111
rarel
y
1123
1122
1111
blending
more
4355
4334
4224
man
y
3345
3234
3223
lest
2234
2223
2111
rarel
y
1223
1122
1111
cotton
more
4355
4334
4324
man
y
3345
3234
3223
lest
2334
2234
2212
rarel
y
1223
1222
1111
Usi
ng the fu
zzy co
ntrol
rul
e
s tabl
e,we
can get
the fa
zzy
cont
rol resp
on
se tabl
e offline.
Finally, we
use the a
s
semb
ly langua
ge l
ook-up
t
able
pro
c
ed
ures to
determi
ne a
ppro
p
ri
ate wa
ter
level, water stream, d
e
t
ergent q
u
a
n
tity and washi
ng time
in the correspon
ding
cloth
quantity,cloth
material an
d dirty degree condition
s
to a
c
hieve the b
e
s
t cloth washi
ng effects.
5
.
Conclu
sion
The co
unte
r
electromotive
force d
e
tecti
on me
thod of
cloth qua
ntity and materi
al in this
pape
r
can
n
o
t only g
e
t the info
rmatio
n of
cloth
m
a
terial
and
q
uantity, but
also
can
coll
ect
comprehensi
ve informatio
n of pulse
wi
dth and cy
cle.It helps
improve the measur
e preci
s
ion of
cloth q
uantit
y and mate
ri
al, and p
r
ovi
des
more ob
jective ba
si
s to best
de
cision
s of
wat
e
r
level,water te
mpretu
re, wa
ter flow inten
s
ity
and washing do
se. B
a
se
d on it,we
have offered
the
spe
c
ific
cont
rolling re
gulati
on of automa
t
ic wa
shin
g m
a
chi
ne which
is fuzzy-cotrolled. Autom
a
tic
wa
shin
g ma
chin
e of fuzzy co
ntrol
g
r
eatly re
du
ces the
dep
e
nden
ce
on
artifical
cath
arsi
s
experie
nce, whi
c
h save
s time, worry, water, el
ect
r
icit
y and improv
es washi
ng rate and intelli
gent
degree comp
rehe
nsively.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Fuz
z
y
Cont
ro
l A
nalysi
s wit
h
B
a
ck
-E
MF i
n
Wa
shin
g Machi
n
e
s
(Min
-
y
an
DI
)
3831
Ackn
o
w
l
e
dg
ements
This
wo
rk
was fina
nci
a
lly sup
porte
d b
y
the Suppo
rt Program of
Heb
e
i Scie
n
c
e a
nd
Tech
nolo
g
y (Grant
No. 1
1
2139
02
D), th
e Appli
c
ation
Develo
pmen
t Found
ation
of Heb
e
i
Normal
University (Grant No. L201
0K07)
,
and the Tea
c
he
r speci
a
l re
sea
r
ch
of Heb
e
i Normal University
in edu
cation reform in 20
13
Year.
Referen
ces
[1]
Z
hang d
a
o
de, Yang g
u
a
n
g
y
o
u
.
T
he desig
n
of fu
zz
y c
ontro
ller w
a
shi
ng
machi
ne use
d
in
Industrial
,
Chin
a, Hub
e
i Industr
y Un
iver
sit
y
. 20
09; (1): 22–
25.
[2]
Liu h
e
, Yu che
ngb
o, Z
hang f
angfa
ng.
T
he
ana
lysis of fu
z
z
y
contro
l for auto
m
atic w
a
s
h
in
g machi
n
e
.
Cho
ngq
in
g Institute of T
e
c
hnolog
y. 20
02; 29(
3): 24-25.
[3]
Che
ng
xi
ngg
uo
, Chen
g jia
ng
h
ong.
T
he d
e
si
g
n
of F
u
zz
y
C
o
ntrol w
a
shi
ng
mac
h
i
n
e
. Hu
na
n Institute of
T
e
chnolog
y. 2
009; 23(
5): 33-
35.
[4] Joachim
Holtz.
Pulse w
i
dth M
odu
latio
n
for El
ectronic Pow
e
r
Convers
i
o
n
.
IEEE Transactions on P
o
wer
Electron
ics
. 19
94; 82(8): 1
194
-121
4.
[5] Karl
H
Edelmoser,
Feli
x A
Hi
mmelstoss.
Analysis
of a
New High-Efficiency DC-to-AC I
n
verter
.
IEEE
T
r
ansactio
n
s o
n
Pow
e
r Electronics
. 19
99; 14
(3): 454-4
60.
[6]
Hua
ng Z
h
ong
li
n, Hua
ng J
i
ng.
Power Electronics MATLAB P
r
actice
. Bei
jin
g, Natio
nal
Defe
nce Ind
u
st
r
y
Press. 2009: 2
03–
23
5.
[7]
Nag
abh
ush
a
n
a
Katte, Naga
b
hush
an R
a
j
u
Kond
uru, Bh
a
skar Pob
bath
i
, Parvathi S
i
da
radd
i.
Fu
z
z
y
Log
ic Appl
ie
d to an Oven T
e
mp
eratur
e Con
t
rol System
. Sensors
& T
r
ansducers. 20
11;
133(1
0
): 65-
73.
[8]
D Le
nin
e
, Ch
Sai Ba
bu, G Shank
arai
ah. Pe
rformanc
e Ev
al
uatio
n of F
u
zz
y a
nd PI C
ontr
o
ller for B
oost
Converter
w
i
th Active PFC.
T
E
LKOMNIKA Indon
esia J
o
urna
l of Electri
c
al Eng
i
n
eeri
n
g
. 201
2; 2(4):
445-
453.
[9]
Vija
ya
ba
la
n R, S Ravivarma
n
.
Source Invert
er fo
r Photovo
l
taic S
y
stem
w
i
th F
u
zz
y
L
ogic
Control
l
er.
T
E
LKOMNIKA Indon
esi
a
Jour
nal of Electric
al
Engin
eeri
n
g
. 2
012; 2(4): 3
71-
379.
[10]
Abub
akkar S
i
d
d
ik A, Sh
an
ge
et
ha M. Impl
e
m
entatio
n of F
u
zz
y Lo
gic
c
o
n
t
roller
in P
hoto
v
oltaic P
o
w
e
r
gen
eratio
n
usi
ng B
oost
Conv
erter a
n
d
Boos
t Inverter.
T
E
L
K
OMNIKA Ind
ones
ia J
our
nal
of El
ectric
a
l
Engi
neer
in
g
. 2012; 2(3): 2
49-
256.
[11]
Mridul
Jha, S
P
Dub
e
y
. N
e
u
r
o-F
u
zz
y bas
e
d
Co
ntro
l
l
er fo
r a T
h
ree- Ph
ase F
o
ur-W
ire
Shu
n
t Activ
e
Po
w
e
r Filter.
T
E
LKOMNIKA Indo
nesi
a
Jour
nal of Electric
al
Engin
eeri
n
g
.
2
011; 1(2): 1
48-
155.
Evaluation Warning : The document was created with Spire.PDF for Python.