TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 9, September
2014, pp. 65
6
6
~ 657
2
DOI: 10.115
9
1
/telkomni
ka.
v
12i9.459
7
6566
Re
cei
v
ed
No
vem
ber 1, 20
13; Re
vised
Apr 19, 201
4; Accept
ed Ju
ne 1, 2014
Resear
ch and Experiment on Electromagnetic Force
Properties of LPMBLDCLM for Electromagnetic Launch
Huilai Li*, Xi
aomin Li, Zhi
y
u
an Li
Dep
a
rtment of Unma
ned Aircr
a
ft Vehicle E
n
g
i
ne
erin
g, Mech
anic
a
l Eng
i
n
e
e
r
ing C
o
ll
ege,
Shiji
azh
u
a
ng, P.R.Chin
a, 050
003
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: oec_l
j
w
2
0
0
9
@
16
3.com
A
b
st
r
a
ct
In order to i
m
pr
ove th
e thrust characteri
st
ics of the movin
g
-mag
net
type lin
ear p
e
r
ma
ne
n
t
ma
gn
et br
ushl
ess DC
motor
(LPMBLDCM),
the structur
al
character
i
stics and
mag
netic
field
ar
e ana
ly
zed.
T
he influ
enc
e rule of el
ectric
al par
a
m
eters
and struct
ura
l
para
m
eters on
the el
ectro
m
a
gnetic pr
op
ertie
s
and thr
u
st perf
o
rmanc
e are r
e
searc
hed
by finite e
l
e
m
ent
a
nalysis (F
EA).
T
he effect regu
larity of structu
r
al
para
m
eters to
mov
e
r vel
o
cit
y
and thr
u
st are res
ear
ch
e
d
as w
e
ll
as
the el
ec
trical para
m
eters.
T
h
e
LPMBLDCM system
is
estab
lished, and s
o
me relevant tests were
taken to verify the correctness of
simulati
on
res
u
lts. Si
mul
a
tio
n
a
nd
exp
e
ri
mental
resu
lts s
how
that the
thrust a
nd v
e
l
o
city of
mover
are
affected by so
me k
e
y p
a
ra
meter. The r
e
sults w
ill
sure
l
y
provid
e the
referenc
e an
d
gui
danc
e for
the
opti
m
i
z
at
ion of
electro
m
agn
eti
c
and th
rust ch
aracteristics of
LPMBLD
CM.
Ke
y
w
ords
: LP
MBLDCM, structural par
a
m
et
ers, elec
trical param
e
ters,
velocity, thrust
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
Linea
r p
e
rm
a
nent
ma
gnet bru
s
hle
s
s DC
motor (L
PM
BDCM
) h
a
s t
he adva
n
tage
of high
flux density, large th
ru
st, high ene
rgy ef
ficien
cy
and simple stru
ct
ure,
which
i
s
very
suitable to
be applie
d to UAV laun
ch for ele
c
tromagn
etic catapult [1-2]. Different wi
th the chemi
c
al
laun
che
r
, line
a
r m
o
tor for
electroma
gne
tic lau
n
ch g
e
nerally
re
quire hig
h
voltag
e, high
curre
n
t,
high and
con
s
tant thru
st. The variation range of tran
si
ent vocility for mover is very large, and the
terminal vel
o
city ca
n rea
c
h several
ten
s
met
r
e
s
pe
r
se
con
d
. All th
ose
speci
a
l
condition
s
req
u
ire
high th
ru
st o
u
tput and
en
ergy effici
en
cy for LPMB
DCM. The
r
efo
r
e, the re
se
ar
ch of
st
ru
ct
u
r
al
para
m
eters a
nd
ele
c
tri
c
al
para
m
eters e
ffect
ru
le
are
ne
ce
ssary t
o
promote
p
e
rform
a
n
c
e f
o
r
LPMBDCM design.
The o
p
timiza
tion de
sign
o
f
stru
ctural
p
a
ra
m
e
ters an
d ba
ck-EMF
waveforms
a
r
e two
main ways to prom
ote th
rust p
e
rfo
r
m
ance fo
r LP
MBDCM. T
h
e re
sea
r
ch o
f
single
stru
ctural
para
m
eters t
o
improve th
e perfo
rman
ce con
c
e
r
n
s
p
o
le/arc co
efficient, slot wi
dth, thickn
ess of
iron
yoke,
wi
nding
mod
e
,
pole
s
stru
cture. Ho
weve
r, t
h
is
optimization m
e
thod
h
a
s it
s
sho
r
ta
ge.
For
the other non-obje
c
tive
para
m
et
ers,
they can’t b
e
optionally ch
ange
d in si
m
u
lated p
r
o
c
e
s
s
.
It must be recal
c
ul
ated
whe
n
othe
r para
m
et
ers h
a
ve been
ch
ange
d. The i
n
fluen
ce
s of gap
length, thickn
ess of mag
n
e
t, pole arc
coeffici
ent,
n
u
mbe
r
of pol
e pairs to th
e avera
ge th
rust
have b
een
alrea
d
y di
scussed
for a
dou
ble-sid
e
d
line
a
r sy
nch
r
on
ou
s
motor
used
for
electroma
gne
tic catap
u
lts i
n
[3], and so
me re
sult
s
are con
s
i
s
tent
with the com
puted results in
[4]. For LPM
B
LDCM, the
optimazati
on of
stru
ctural para
m
eters
can in
cre
a
se t
he flat width
o
f
trapezoidal back-EMF, so the th
rust output will be sm
ooth [5].
In this pap
er, the rule of
stru
ctural p
a
ra
m
e
ters a
nd ele
c
tri
c
al
para
m
eters to thrust
perfo
rman
ce
for LPMBLDCM are
re
se
arched by
si
mulation an
d
experiment.
The para
m
e
t
er
matchin
g
an
d optimizin
g
desig
n pro
b
l
em for LPM
B
LDCM are
explore
d
to apply small
and
middle scal
e UAV laun
che
r
, and also to pr
ovide
references to cata
pult desi
gn.
2.Electroma
gnetic Field
Model Analy
s
is
Figure 1
sho
w
s the
sim
p
lified di
agram
o
f
two
dimen
s
i
onal
sol
u
tion
area
for LPM
B
LDCM.
To cal
c
ul
ate
the flux de
nsity in the
air-
gap, the
symmetry bo
unda
ry co
ndi
tions b
e
twe
e
n
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Re
sea
r
ch an
d Expe
rim
ent on Electrom
agneti
c
Forc
e
Propertie
s
of
LPMBLDCL
M for… (Huil
a
i Li)
6567
perm
ane
nt magnet i
s
setted. The pe
rma
nent mag
net
regio
n
, air
ga
p regi
on, an
d
slotted
stato
r
regio
n
are di
vided in turn
along the y-a
x
is dire
ct
ion
with thre
e bo
xes alveola
r
l
a
yers
boun
d
a
ry.
The moving
-magnet move
r coul
d only sl
ip along x-axi
s
dire
ction.
Figure 1. The
Diagram of T
w
o Dim
e
n
s
io
nal Solution
Area
The dist
ributi
on equ
ation
s
of magnetic fi
eld for ea
ch l
a
yer are give
n as follo
ws [
6
]:
22
22
22
22
2
2
22
0
(1
)
(1
)
xm
m
m
xs
s
AA
Re
g
i
n
xy
AA
j
vn
A
J
R
e
g
i
n
xy
A
A
jv
n
A
R
e
g
i
n
xy
ⅠⅠ
ⅡⅡ
Ⅱ
0
Ⅲ
Ⅲ
Ⅲ
Ⅰ
Ⅱ
Ⅲ
(
1
)
Whe
r
e A is th
e curl of ma
g
netic vecto
r
;
s
and
m
resp
ectiv
e
ly denote
s
the co
ndu
ctivity
of perma
nent
magnet an
d i
r
on yo
ke;
de
notes the
pol
e pitch;
x
v
deno
tes the velo
city of mover
along
the x-a
x
is directio
n;
n is the m
u
ltiple
of th
e
sp
ace
ha
rmo
n
ics to fu
ndam
e
n
tal field.
m
J
is
derived from the formul
a (2
).
0
()
0
/
j
ta
t
mr
JM
B
e
(2)
Whe
r
e
r
B
is the
rema
nen
ce of
perma
nent
magnet,
0
is the equivale
nt angul
ar velo
city.
Magneti
c
obt
ained o
u
t of positio
n and
the windi
ng
curre
n
ts, a
c
cordin
g to Am
pere
'
s l
a
w, the
mover thru
st
have been
suffered a
s
formula (3
).
VV
AA
FJ
B
d
V
J
i
j
d
V
xx
(3)
Ho
wever, thi
s
m
e
thod
re
sults were
n
o
t accu
rate,
for th
e n
o
n
-
linea
r m
agn
etic flux
leakage
ha
s
not be
en ta
ke
n into th
e m
o
del. To
a
c
curately cal
c
ul
ate the fiel
d a
n
d
pa
ram
e
ters for
LPMBLDCM, the ideal choi
ce is the finite
element anal
ysis(FEA).
3. Finite Element An
aly
s
is and Res
u
lts
3.1. Model Analy
s
is And
Parameter Setting
s
The tra
n
si
ent
pro
c
e
ss
of el
ectro
m
ag
neti
c
lau
n
ch is
a
highly couple
d
pro
c
e
s
s. T
he ed
d
y
curre
n
t loss and hysteresi
s
loss a
r
e se
ri
ous at
high
speed. The m
u
tual indu
cta
n
ce of mover
and
windi
ng
will p
r
esent n
on-li
n
ear
ch
ange
s
with the diffe
rent po
sition a
nd velo
city, which i
s
difficul
t
to establi
s
h t
he exa
c
t mat
hematical mo
del. Maxw
ell Ans
o
ft
software c
an
au
tomatic
a
lly cons
ider
part shap
e, material pro
p
e
rties, relativ
e
po
si
tion a
n
d
other pa
ra
meters, whi
c
h co
uld
solve
th
e
probl
em of
d
i
rect
cal
c
ul
ation. The
tra
n
s
mit dy
na
mic pro
c
e
s
s of
electroma
gne
tic lau
n
ch fo
r
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 65
66 – 657
2
6568
LPMBLDCM
is
re
sea
r
ch
ed u
s
ing
Ma
xwell An
soft 2-D tra
n
si
e
n
t solve
r
in
this p
ape
r.
The
influen
ce of
stru
ctural pa
rameters a
n
d
electri
c
al
p
a
ram
e
ters to
the thru
st perfo
rman
ce
for
LPMBLDCM
are inve
stigat
ed, whi
c
h pr
o
v
ide foundati
on for optimal
desig
n.
Figure 2. Dyn
a
mic Simulati
ve Model of
LPMBDCM
Figure 3. Section of the Meshe
d
Model o
f
LPMBDCM
Figure 2.
sh
o
w
s the
mod
e
l
of LPMB
DCM, incl
uding
mover, stator,
windi
ng,
tra
n
slatio
n
domain
and
solution d
o
mai
n
. Figure 3.
shows th
e
me
she
d
mo
del o
f
LPMBDCM
by FEA. It can
be seen th
at the mover, d
r
ive coil a
nd transl
a
tion do
main a
r
e rela
tive intensive
so a
s
to e
n
sure
the a
c
curacy
of sim
u
lation.
The
wi
ndin
g
coil i
s
setted
as
se
rial
multi
l
ayer, a
total
of 4 l
a
yers. T
he
material
of
so
lution d
o
main
is vacuum,
a
nd the
exte
rn
al d
r
ive
circui
t is
esta
blishe
d, as
sho
w
n
i
n
Figure 4. The current in the windi
ng i
s
adde
d
by external driv
e circuit, and
the voltage and
resi
stan
ce
va
lues
can
be
setted
as req
u
ired.
The
si
mulation tim
e
is f
r
om
0 m
s
to 15
0 m
s
with
0.05 ms
step-size ch
ang
e.
Figure 4. The
External Driv
e Circuit
The pole pit
c
h (
τ
), length
of air-g
ap (g
s), size of sl
ot
dimensi
on (wc, lc, h
s
), si
ze of PM
dimen
s
ion
(wm, lm, hm), and the length
of prim
ary a
nd se
co
nda
ry sectio
n sh
ou
ld be adju
s
te
d
prop
erly. Some motor pa
rameters an
d PM ch
a
r
acte
ristic have b
e
e
n
sho
w
n in T
able 1.
Table 1. Moto
r De
sign
Data
and PM Cha
r
acte
ri
stic
Sym
bol
Item
Value
τ
p
ole
p
itch
60mm
g
s
len
g
th of ai
r
-
g
a
p
2mm
w
c
w
i
dth of
slot
15mm
l
c
len
g
th of slot
20mm
h
s
depth of slot
100mm
w
m
w
i
dth of
PM
50mm
l
m
len
g
th of PM
90mm
h
m
thickness of PM
12mm
L
p
primar
y
len
g
th
1.8m
L
s
secondar
y
length
235mm
PM materi
al
Nd-Fe
-
B
0
-
+
240V
Label
I
D
=
V
D
C
-
V
o
l
t
age
Label
I
D
=
V
C
1ohm
R3
1ohm
R2
1ohm
R1
Label
I
D
=
V
B
Label
I
D
=
V
A
Lw
i
ndi
n
g
A
Lw
i
ndi
n
g
B
1pH
L1
1pH
L2
1pH
L3
Lw
i
ndi
n
g
C
D3
D1
D5
D4
D2
D6
Mo
d
e
l
D
m
odel
V
S_
1
M
odel
V
sw
i
t
ch
D1
1
D3
3
D5
5
D6
6
D4
4
D2
2
V
S_
2
V
S_
3
V
S_
5
V
S_
4
V
S_
6
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Re
sea
r
ch an
d Expe
rim
ent on Electrom
agneti
c
Forc
e
Propertie
s
of
LPMBLDCL
M for… (Huil
a
i Li)
6569
3.2 Struc
t
ura
l
Parameters
Effec
t
And
Analy
s
is
3.2.1. The Influence of Po
le/Arc Co
efficient
For LPMBL
D
CM
，
the pol
e
/
arc
coefficie
n
t is defined
as:
m
w
(6)
To re
sea
r
ch the influen
ce
of pole/arc co
effi
cient on th
e cog
g
ing force, so
me con
d
itions
have be
en lo
cated
as follo
w: the initial
velocity of
the mover i
s
2
0
m/s; the at
uating voltag
e is 0
V; the mass
of mover and
load is 10
kg.
The eddy
cu
rre
nt effect has be
en ign
o
red. The coggi
ng
force re
sults unde
r
differe
nt
pole/
arc coefficient hav
e been o
b
tai
ned,
as
sh
own in Figure 5.
As
can
be
se
en
from Fig
u
re
5, the
cog
g
in
g force
almo
stly rem
a
in
s
stationa
ry an
d they a
r
e
q
u
ite
small
com
p
a
r
ed to th
ru
st force o
u
tput.
The ave
r
ag
e
of cog
g
ing
force
clo
s
ely e
q
ual to 0
N,
which
does not produce any effect to th
rust
output, but
it will
cause t
h
rust fluctuati
on and vel
o
city
fluctuation
s
.
Figure 5. Re
sult of Coggin
g
Force con
s
i
derin
g
Pole/Arc
Coeffic
i
ent
Figure 6. The
Relation
ship
of Back-EMF
and
Pole/Arc
Coeffic
i
ent
Figure 7. Re
sult of Coggin
g
Force Th
ru
st con
s
id
erin
g Pole/Arc Coefficient
Figure 6
sho
w
s the
ba
ck-EMF re
sult
s
unde
r
diffe
re
nt
pole/a
r
c coefficient, wh
en
the
atuating volt
age i
s
300V.
It can be
seen that a
ra
ther la
rg
e width of ba
ck-EMF
wave
sha
pe
occurs with a
100 de
gre
e
el
ectri
c
al an
gle
when p
o
le
/arc co
efficient i
s
0.9. As the decre
scen
ce
of
pole/arc coefficient, the top width of
back-EMF
waveshape will become
na
rrow down
and t
he
raised am
plitud disto
r
tion
will also app
ear. Figu
re
7
gives the ba
sic la
w of a
c
cele
ration
ch
ange
whi
c
h differe
nt pole/arc coeffici
ent are con
s
id
ere
d
.
The mover has differe
nt acceleration
respon
se
wh
en the pol
e/a
r
c
coeffici
ent
is re
spe
c
tive
ly assu
med
at 0.6,
0.7, 0.8 and 0.9. T
h
e
optimal accel
e
ration
re
spo
n
se o
c
curs when t
he pol
e/arc
coeffi
cien
t approximate
at 0.7.
0
10
20
30
40
50
60
-20
-10
0
10
20
30
Ti
m
e
(
m
s)
C
o
gg
i
n
g F
o
r
c
e
(
N
)
= 0
.
6
= 0
.
7
= 0
.
8
= 0
.
9
0
5
10
15
20
25
-200
-100
0
100
200
Dis
t
a
n
c
e
(
m
)
Fi
el
d I
n
t
ensi
t
y
(
t
esl
a
)
=0
.9
=0
.8
=0
.
7
=0
.
6
0
20
40
60
80
0
5
10
15
20
Ti
m
e
(
m
s
)
V
e
lo
c
i
ty
(m/s
)
=
0
.
8
=
0
.
7
=
0
.
9
=
0
.
6
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 65
66 – 657
2
6570
3.2.2. The Influence of
Air
g
ap Width
As some lit
eratu
r
e
have
been
repo
rted t
hat the
airg
ap
widt
h could i
m
p
a
ct the
distrib
u
tion
of magn
etic fiel
d, so th
e an
al
ysis of
airg
ap
length i
s
ve
ry important fo
r e
s
timating t
he
actual p
e
rfo
r
mance. The simulation con
d
itions
have
been lo
cated
as follo
w: the atuating voltage
is 2
40V; the
mass
of mov
e
r
and
load
i
s
1
0
kg. Th
e
airga
p
width
gs i
s
setted
from 2m
m to
4
m
m
with 1mm ste
p
-si
z
e
cha
n
g
e
, and the ed
dy current effect ha
s bee
n ignored.
Figure 8. Re
sults of Accele
ration con
s
id
ering Airgap
Width
The velocity cha
nge
s of mover are prese
n
ted
in accele
ration tim
e
with differe
nt airgap
width, a
s
sho
w
n in Fi
gu
re
8. The a
c
cel
e
ration
cu
rve
sho
w
s that the move
r ha
s a
rathe
r
larger
accele
ration
from
0m
s to
60ms an
d a
grad
ual
rollof
f
until to
ze
ro after 10
0m
s. Thi
s
ca
n
be
explained
th
at the sh
ort
airga
p
will
p
r
ovided
more
large
r
field
den
sity, which the maxim
a
l
velocity will b
e
com
e
large
and the a
c
cel
e
ration time
will also be
co
me sho
r
t.
3.3. Electrica
l Parameters
Effec
t
And
Analy
s
is
3.3.1. The Influence of
Ac
tua
t
ing Volta
g
e
Set the ele
c
trical p
a
ramete
rs
of extern
al
circuit
as foll
ow: the
re
si
stance R is 0.
1ohm;
the indu
ctan
ce L is
40m
H;
the voltage i
s
setted
from
60V to 24
0V with 60V
ste
p
-si
z
e
ch
ang
e.
Figure 9. sho
w
s the velo
cit
y
accel
e
ratio
n
results
con
s
ide
r
ing a
c
tu
ating voltage.
Figure 9. Re
sults of Accele
ration con
s
id
ering Volta
g
e
The tran
sie
n
t acceleration
resp
on
se d
oes n
o
t sho
w
linea
r rule
with the increa
se of
voltage. Th
e
mover
accel
e
ration
evident
ly pre
s
ent
in
crea
sing
at first and th
en
de
cre
a
se to
zero
in the process. Whe
n
the voltage
exce
e
d
s 180V, the addition of thrust
in the x-axis dire
ction a
nd
accele
ration
resp
on
se
are
not obvio
us.
The Am
pere
force
can
not
incre
a
se
s lin
early
due
to t
he
saturated fiel
d in the
air
gap. Th
e el
e
c
trom
agn
etic force i
s
d
e
cided by
cu
rrent, magn
et
and
positio
n a
c
co
rding t
he Am
pere
'
s l
a
w,
so the maximu
m thru
st doe
sn’t o
c
curs n
ear th
e maxi
mum
0
20
40
60
80
10
0
0
5
10
15
20
Ti
m
e
(
m
s
)
V
e
l
o
ci
ty (
m
/s)
g
s
= 2
m
m
g
s
= 3
m
m
g
s
= 4
m
m
0
25
50
75
100
125
150
0
5
10
15
20
Ti
m
e
(
ms
)
Ve
lo
c
i
ty
(
m
/s
)
U =
60V
U =
120V
U =
180V
U =
240V
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Re
sea
r
ch an
d Expe
rim
ent on Electrom
agneti
c
Forc
e
Propertie
s
of
LPMBLDCL
M for… (Huil
a
i Li)
6571
drive
cu
rre
nt
point. The
thrust flu
c
tuatio
n ap
pea
re
d
d
u
ring
the tran
sient
start
pro
c
ed
ure
an
d t
h
is
can b
e
expla
i
ned that the influence of
eddy curr
en
t resista
n
ce make
s the m
o
ver de
cele
rate,
whi
c
h
con
c
e
r
nes with th
e
cha
nge
rate
of coil
cu
rren
t. For the
gre
a
ter of th
e
chang
e rate, the
greate
r
of the
eddy
cu
rrent
re
sista
n
ce, b
u
t the Amp
e
re force
still pl
ays a
leadi
ng
role
in th
e
start
pro
c
ed
ure, so the thru
st increa
se
s with t
he en
h
ancement of
the voltage
despite
a little
fluctuation a
s
so
ciated.
3.3.2. The Influence of
Ac
tua
t
ing Curr
ent
Figure 10. Re
sults of Accel
e
ration con
s
i
derin
g
Cu
rren
t
Figure 10 sh
ows the mov
e
r velo
city curve
when t
he actu
ating
current alte
rs so
me
atypical valu
e. With the
enha
ncement
of actuat
ing
curre
n
t, the max velocit
y
incre
a
ses f
r
om
12.73m/s to
1
9
.52m/s, whe
n
the a
c
tuatin
g cu
rrent
is 5
0
A to 140A. It can
be
seen
that the move
r
velocity almo
st sh
ow lin
ea
r relatio
n
whe
n
the num
b
e
r of ampere tu
rns i
s
bel
ow
400. The
rate
of
mover a
c
cele
ration is
not evidently whe
n
the act
uatin
g cu
rre
nt exceed
s at given
value, such as
200A for
con
c
erned
model
machi
ne. According to
th
e
schemati
c
m
odel, a sub
s
cale long
prim
ary
LPMBDCM system are e
s
t
ablished, a
s
shown in Figu
re11.
Figure 11. Prototype mach
ine of a
sub
s
cale lo
ng- p
r
i
m
ary LPMBDCM.
Table 2. Experime
n
t and
Simulation Result
s for Me
asu
r
ed T
h
ru
st and Velocity
Voltage
()
UV
48 96 144
180
216
240
Average simulation thrust (
N
)
288
446
639
883
1265
1432
Average measur
ed thrust (
N
)
255
415
588
850
1186
1304
Max simulation velocity
(m/s
)
5.24
9.26
15.32
18.06
20.66
21.28
Max measur
ed v
e
locity
(m/s
)
4.95
8.84
13.85
17.09
19.25
20.06
0
50
100
150
200
250
-3
0
5
10
15
20
Ti
m
e
(
m
s
)
V
e
l
o
c
i
t
y
(m
/s
)
I
=
50A
I
=
80A
I
=
110A
I
=
140A
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 65
66 – 657
2
6572
Figure 12. Th
e Mover Velo
city chan
ge to Voltage
Simulation a
nd expe
rime
ntal re
sult
s o
b
tained
are shown in T
abl
e 2. Fig
u
re
1
2
sh
o
w
s
the comp
arative chart of m
easure
d
and
simulatio
n
velocity for the mover, and a
l
so the curve
of
averag
e me
a
s
ured th
ru
st and
simulatio
n
thru
st,
wh
e
n
the voltage
value chan
g
e
s from 48V
to
240V.
It c
a
n b
e
se
en
th
a
t
th
e th
ru
s
t
ou
tp
u
t
and
move
r velo
city almo
st
show line
a
r accretio
n
whe
n
the act
uating voltag
e is blew 2
0
0
V. Ho
weve
r, it does not appea
r exa
c
tly as a linear
relation.
Fo
r the m
agn
etic field
is saturated,
the
thru
st o
u
tp
ut ca
nn’t b
e
promoted
if the
enha
ncement
of the
voltag
e. The
variati
on
rule
s o
b
ta
ined from th
e
expe
riment
and
simul
a
tio
n
results
are b
a
si
cally con
s
i
s
tent. Mea
n
while, it ca
n b
e
se
en that th
e mea
s
u
r
ed
result
s a
r
e
a li
ttle
smalle
r tha
n
simulatio
n
re
sults
due to
the slidi
ng re
sista
n
ce, whi
c
h verify the
corre
c
tne
s
s
o
f
simulatio
n
an
alysis.
4. Conclusio
n
The impa
ct
rule research of st
ru
ctural
p
a
ra
met
e
r an
d ele
c
trical p
a
ra
m
e
ter on
electroma
gne
tic force pro
pertie
s
for L
P
MBLD
CLM
are the b
a
s
ic
wo
rk fo
r multi-pa
ram
e
ter
optimizatio
n desi
gn and
segmente
d
de
sign. The
the
o
retical
wo
rki
ng p
r
in
ciple
a
nd m
agn
etic f
i
eld
for LPMBDCM are
analy
z
ed i
n
this
pape
r. The i
n
fluen
ce rule
of stru
ctural
para
m
eter
and
electri
c
al
pa
rameter on
el
ectro
m
ag
neti
c
p
r
o
pertie
s
and th
ru
st p
e
r
forma
n
ce fo
r LPMBL
DCM
are
resea
r
ched
by FEA and experime
n
t. Simulation
and expe
rim
ent re
sults
sho
w
that the
enha
ncement
of actuating voltage
can a
pparently increase the move
r velocity wh
en the voltage
is le
ss tha
n
200V
and
the mini
mal
singl
e p
r
ima
r
y length fo
r
multi-sta
ge l
aun
che
r
can
be
desi
gne
d as
1.65m for p
r
o
posed LPMB
LDCM.
Referen
ces
[1]
MR Do
yl
e, DJ Samue
l
, T
Con
w
a
y
, RR Kl
im
o
w
ski. Electro
m
agn
etic aircr
a
ft launc
h s
y
st
ems- EMALS.
IEEE
T
r
ans. Magn., 19
95; 31(
1): 528–
53
3.
[2]
D Patterson, A Monti, C Brice,
T
Bertoncelli. Des
i
g
n
an
d simulati
on of
an electrom
a
gnetic a
i
rcraft
lau
n
ch s
y
stem. 37th IAS Annu.
Meeting. 2
0
02; 3: 195
0–1
9
57.
[3]
Li
yi Li, Ma Mi
ngn
a, Baoq
ua
n Kou, Qing
qu
an
Ch
en. Ana
l
ysis a
nd D
e
si
gn of Movin
g
-
M
agn
et-T
y
p
e
Lin
ear S
y
nc
hr
ono
us Motor f
o
r Electrom
ag
netic L
a
u
n
ch S
y
stem.
IEEE Trans. Plas
ma
Sci.,
2002;
39:
121
–1
26.
[4]
Kou Bao-Qu
an
, W
u
Hong-Xin
g
, Li Li-Yi, et al.
T
he
T
h
rust Characteristic
s Investigatio
n
of Doubl
ed-
Side
Plate
Per
m
ane
nt Mag
n
e
t Li
near
S
y
nc
hron
ous M
o
tor
for EML.
IEEE Trans. Magn.,
200
9; 4
5
:
501
–5
05.
[5]
Yong L
i
, Jibi
n Z
ou, Yong
pin
g
Lu. Opti
mu
m Desig
n
of Magnet S
hap
e in Per
m
ane
nt-Magn
e
t
S
y
nc
hro
n
o
u
s Motors.
IEEE Trans. Magn
., 2
003; 39(
6): 352
3–3
52
6.
[6]
K Ng, Z
Q
Z
h
u, D Ho
w
e
. Open-
Circu
it F
i
eld Distrib
utio
n
in a Brushle
ss Motor
w
i
th
Diametrica
l
l
y
Magn
etised P
M
Motor, Accountin
g for
Slotting a
nd Ed
d
y
Current Effects.
IEEE Trans.
Magn.,
19
96
;
32(5): 50
70
–50
72.
0
50
100
15
0
20
0
25
0
30
0
0
10
20
30
V
o
l
t
age
(
V
)
M
a
x Ve
l
o
ci
t
y
(
m
/
s
)
0
50
100
15
0
20
0
25
0
30
0
0
50
0
10
00
15
00
V
o
l
t
age
(
V
)
Th
r
u
s
t
Fo
r
c
e
(
N
)
S
i
m
u
l
a
ti
o
n
v
e
l
o
ci
ty
M
e
as
u
r
ed
v
e
l
o
c
i
t
y
S
i
m
u
l
a
ti
o
n
th
r
u
s
t
M
e
as
u
r
ed
t
h
r
u
s
t
Evaluation Warning : The document was created with Spire.PDF for Python.