TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 12, Decembe
r
2014, pp. 81
2
6
~ 813
2
DOI: 10.115
9
1
/telkomni
ka.
v
12i12.49
66
8126
Re
cei
v
ed Ma
y 9, 2014; Re
vised O
c
tobe
r
26, 2014; A
c
cepted
No
ve
m
ber 10, 201
4
New Research on MEMS Acoustic Vector Sensor Used
in Ground Marker of Pipeline
Liu Meng-r
a
n
1,2
，
Zhang Guo-jun
1,2
，
Jian Ze-min
g*
1
，
Liu Hon
g
1
，
Song Xiao-pen
g
1
,
Zhang Wen
-
dong
1,2
1
Ke
y
L
abor
ator
y of Instrument
ation Sci
enc
e & D
y
n
a
mic Me
asurem
ent, Ministr
y
of Ed
uca
t
ion,
North Un
iversit
y
of Ch
ina, T
a
iyu
an 0
3
0
051,
Chin
a;
2
Ke
y
L
abor
ator
y Of Science a
nd T
e
chnol
og
y on Electron
ic T
e
st & Measurement,
North Un
iversit
y
of Ch
ina, T
a
iyu
an 0
3
0
051,
Chin
a
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: jianz
emin
g
x
@16
3
.com
A
b
st
r
a
ct
Rece
ntly, grou
nd
mark
er bas
ed o
n
the
prin
cipl
e
of ac
oust
i
c detecti
on
ha
s beco
m
e the
trend
o
f
pip
e
li
ne tech
n
o
lo
gy. As to the difficulty a
n
d
low
a
ccuracy
of grou
nd
mar
k
er, this pap
er
introduc
es a
new
MEMS bion
ic
acoustic vect
or
sensor w
i
th hi
gh sens
itivity, goo
d perfor
m
a
n
ce in
low
freq
uency a
nd l
o
w
e
r
pow
er. As to the pr
obl
e
m
of
the port/starbo
a
rd bl
ur of
this
vector sens
or
in gro
u
n
d
mark
ing, it pr
opos
e
s
a
new
pl
an
an
d
proves
the
corr
ectness
of the
pla
n
th
ro
ug
h th
eoretic
an
alysi
s an
d ex
per
i
m
ental
verific
a
tio
n
.
T
he res
u
lts sh
ow
that the
me
an
error
of the
directi
ona
l a
n
g
l
e is w
i
thi
n
5 d
egre
e
me
etin
g
the n
e
e
d
s of th
e
eng
ine
e
ri
ng pr
oject.
Ke
y
w
ords
:
pi
p
e
lin
e gro
u
n
d
markin
g, MEMS vector sens
or,
the port/starbo
a
rd bl
ur, directi
on an
gl
e
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
Pipeline tra
n
s
po
rtation ha
s be
come o
n
e
of
five top
transpo
rtation
mode to ke
ep pa
ce
with aviatio
n
,
railway, ro
ad
and
wate
r tra
n
sp
ort a
nd
h
a
s
played
an
increa
singly l
a
rge
r
role
in t
h
e
national e
c
o
n
o
my and nati
onal defe
n
se
con
s
tru
c
tion
[1]. Meanwh
ile, the accid
ents of pip
e
li
ne
leak
hap
pen
freque
ntly with the age
of pipeline, t
he corro
s
ion
of pipeline
and the
artificial.
Pipeline in
sp
ection ga
uge
design
ed fo
r pipelin
e de
fect detectio
n
is an imp
o
rtant dete
c
tion
equipm
ent.
While
pipeli
n
e in
spe
c
tion
gaug
e
works in the
pipeli
ne, it would
gene
rate
abo
ut 1m
error
per
1km. The lon
g
e
r
the pi
peline
is, the g
r
eat
er the
cum
u
l
a
tive error
would b
e
ge
ne
rated
and the less accurate p
o
sition would
be got [2-4]. So, for long-di
stan
ce
pipeline d
e
fe
ct
detectio
n
, the
insp
ectio
n
g
auge
mu
st be
marked
on
ce
a kil
o
mete
r i
n
the p
r
o
c
e
s
s of moving. T
h
e
time wh
en th
e pipeli
ne in
spectio
n
ga
ug
e is
getting t
h
rou
gh th
e m
a
rki
ng
point i
s
a
s
certai
ned
by
the groun
d
marker of pi
peline. T
hen
by com
p
a
r
in
g to the
cou
n
t of the
mileage
wheel i
n
the
moment, the
cou
n
t value
of the Mile
a
ge wheel
wa
s
corre
c
ted t
o
elimin
ate the a
c
cumul
a
ted
error, a
nd
get
the a
c
curate
positio
ning
of the defe
c
t. In
re
cent ye
ars, with the
ra
pi
d develo
p
me
nt
of the pi
pelin
e ind
u
stry,
do
mestic a
nd fo
reign
co
mp
an
ies
are
a
c
tive
ly looki
ng fo
r
new pri
n
ci
ple
s
and meth
od
s of tracki
ng
markers. The
desig
n of
po
sitionin
g
syst
em ba
sed o
n
the prin
ciple
of
aco
u
sti
c
is si
mple, easy to
achieve, suitable fo
r real-t
ime pro
c
e
s
si
ng and
with higher
sen
s
itivity
and a wi
de ra
nge of the det
ection.
In this pape
r the MEMS bionic a
c
o
u
s
tic
vecto
r
sensor with
h
i
gh se
nsitivity,
good
perfo
rman
ce
in low fre
que
ncy and l
o
we
r po
wer
ha
s been u
s
e
d
in
grou
nd ma
rker. And when
it
wa
s u
s
e
d
to
accurate th
e
positio
n of
th
e pip
e
line
in
spectio
n
g
aug
e, a
ne
w
sch
e
me
eliminati
n
g
the probl
em o
f
the port/starboar
d blu
r
ha
s bee
n put forward.
2. The Worki
ng Principles and Mathe
m
atical Mod
e
l
2.1. The Wor
k
ing Principles [5
-7]
MEMS bioni
c acou
stic ve
ct
or
sen
s
o
r
ba
sed o
n
the
pie
z
oresi
s
tive p
r
incipl
e
can
m
easure
the low fre
q
u
ency till the Zero.
High
sensitiv
ity, good pe
rform
a
n
c
e in lo
w fre
quen
cy and l
o
w
power ma
ke
it a unique a
d
vantage in
measuri
ng
th
e wea
k
signa
ls. MEMS vector
sen
s
o
r
h
a
s
been
sho
w
n i
n
Figure 1.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Ne
w Re
se
arch on MEMS Acou
stic Ve
ctor Sen
s
or
Used in Gro
und
Marker of
… (Liu Meng
-ran
)
8127
Figure 1. Obj
e
ctive gra
ph
of ME
MS aco
u
stic ve
ctor
sensor
MEMS bioni
c acousti
c
vector
sensor i
s
based on
silicon and
co
mposed of
four-arm
silicon micro
s
tru
c
ture with
a standa
rd piezore
s
i
s
tive silicon mi
cro
m
achi
ning p
r
oce
s
s and ri
gid
cylindri
c
al
bo
dy fixed in the ce
nter of th
e beam.
Stru
cture
mod
e
l is sho
w
n in Fi
gure
2. The
r
e
are
eight eq
ual-v
alue
strai
n
varisto
r
, R1,
R2, R3, R4, R5,
R6, R7
and
R8, m
a
de by me
an
s of
diffusion i
n
th
e four-arm. R1, R2, R3 an
d R4
co
nstitu
te a Wh
eatst
one b
r
idg
e
, a
nd R5, R6,
R7
and R8 con
s
titute anothe
r. The dist
ribut
ion of pie
z
oresi
s
tor o
n
the micro-
stru
cture is
sho
w
n
in
Figure 3 and
comp
ositio
n of the Wheat
stone b
r
idg
e
is sh
own in Figure 4.
Whe
n
the vibration
sign
a
l
of X-dire
ction
act
s
on t
he micro
-
stru
cture, the b
e
a
m will
prod
uce a
s
ymmetrical stress dist
ributi
on, if R1 and
R3 corre
s
po
nd to the ten
s
ion, R2 and
R4
woul
d
co
rre
spond
to th
e p
r
essu
re
an
d
R5,
R6,
R7
a
nd
R8
wo
uld
corre
s
p
ond
to the
shea
r f
o
rce.
Whe
n
the
wi
dth of the
be
am is mu
ch
g
r
eate
r
th
a
n
th
e thickn
ess, t
he
shea
r
def
ormatio
n
can
be
ignored. In th
is situ
ation
we ca
n ba
si
cal
l
y treat
that R5, R6, R7
a
nd R8 chan
g
e
nothin
g
, wh
ile
the resi
stan
ce values of (R1, R3
) and
(R2, R4
)
cha
nge into the
oppo
site dire
ction; Wh
en the
vibration
signal of Y-di
rection
acts
on the micro-structure, the
beam will produce asymm
e
tri
c
al
stre
ss distri
b
u
tion, if R5 and R7 corre
s
pond to t
he tensi
on, R6 a
nd R8 would
corre
s
p
ond to the
pre
s
sure and
R1, R2, R3
and
R4 woul
d co
rrespond
to the sh
ear force.
Whe
n
the width of
the
beam
is mu
ch g
r
eate
r
tha
n
the
thickne
ss, th
e
sh
ear
deform
a
tion
can b
e
ig
no
red
.
In this situ
ation
we
ca
n
basi
c
ally treat th
at R1,
R2, R3
and
R4
chan
ge n
o
thing,
while
the
re
si
stan
ce val
u
e
s
of
(R5, R6) a
nd
(R7, R8)
cha
nge into the o
ppo
site dire
ct
ion.
Whe
n
a si
gn
al is ap
plied t
o
the se
nsor,
it can be d
e
comp
osed int
o
X-directio
n and Y-
dire
ction,
ca
ntilevers hav
e be
en
defo
r
med
an
d Whe
a
tston
e
bridg
e
s
have
bee
n cha
n
ged.
Acco
rdi
ng to
the output
s o
f
the wh
eatst
one b
r
idg
e
changi
ng, we
can
dete
r
min
e
the di
re
ctio
n of
sign
al so
urce
.
Figure 2. Dia
g
ram of the a
c
ou
stic
sen
s
o
r
microstructu
re model
Figure 3.Dia
g
r
am of co
nne
ction of dist
ribution
of piezatran
s
i
s
tor on th
e micro
stru
ctu
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 12, Decem
ber 20
14 : 8126 – 81
32
8128
Figure 4. Photo of wheatst
one br
i
dge fo
rmed by pie
z
atran
s
isto
r
2.2. Mathem
atical Model
Analy
s
is [8-10]
Whe
n
the cyl
i
nder of the sensor mi
cro
s
t
r
uctu
re
suffer the load force of X-dire
ction, the
force
Fx
will
gene
rate t
w
o
co
mpon
ents
in the
ce
nter
block of
the micro stru
cture, the h
o
ri
zo
ntal
force F
H
alo
ng the X-axis dire
ction an
d torqu
e
M a
r
oun
d the Y-axis. Since th
e mess of ce
nter
block of microstru
c
tu
re is
much
small
e
r than the
cylinder, so it ca
n be igno
red i
n
the cal
c
ulati
on.
The mechani
cal analy
s
is
model of the micros
tructu
re has be
en e
s
tabli
s
he
d an
d sho
w
n
in Figu
re
5-7
.
Figure 5
shows the
mo
vement of
a
c
cele
rom
e
ter
whe
n
subje
c
t
ed to a
no
rmal
hori
z
ontal
a
c
cele
ration.
Fi
gure
6
sho
w
s the
force
a
nalysi
s
a
bout
ce
ntral
blo
c
k, an
d Fi
gu
re 7
sho
w
s the
force
analy
s
is
a
bout
single
cantilever. If F
x
dire
ction i
s
in the
positiv
e directio
n of
the
X-axis, the fo
ur resi
stors
(R1, R2, R3
a
nd R4) in
th
e
cantileve
r in
turn sho
w
th
e pre
s
su
re (-)
,
tensio
n (+), p
r
essu
re
(-) an
d tens
i
on (+),
that is R1
de
cre
a
ses,
R
2
in
c
r
e
a
s
e
s
,
R
3
d
e
c
r
e
as
es
an
d
R4 i
n
crea
se
s.
If Fx dire
ctio
n is i
n
the
ne
gative di
recti
on of the
X-a
x
is, the fou
r
resi
stors
(R1,
R2,
R3 an
d R4
) in the cantilev
e
r in turn
sho
w
the t
ensi
o
n
(+), p
r
es
su
re
(-), ten
s
ion (
+
) an
d pr
es
s
u
re
(-), that is R1
increa
se
s, R2
decre
a
s
e
s
, R3 in
cre
a
ses
and R4 de
cre
a
se
s.
Figure 5. Movement of accelero
meter
when s
ubje
c
te
d to a normal
hori
z
ontal a
c
cele
ration
Figure 6. Photo of force an
alysis a
bout
center blo
c
k of the micro
structure
Figure 7. Photo of force an
alysis a
bout single cantilever
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Ne
w Re
se
arch on MEMS Acou
stic Ve
ctor Sen
s
or
Used in Gro
und
Marker of
… (Liu Meng
-ran
)
8129
By the mecha
n
ics of materi
al theory:
The stress
)
(
x
of any point in the singl
e cant
ileve
r unde
r the actions of ben
ding
moment
)
(
x
M
and hori
z
ontal fo
rce
H
F
can be ex
pre
s
sed a
s
:
bt
F
M
a
aL
L
bt
L
a
x
aL
L
H
x
)
3
3
(
3
2
)
(
3
3
2
2
2
2
)
(
(1)
In the form
ul
a: L(
μ
m
)
i
s
t
he len
g
th of
cantileve
r; b(
μ
m) i
s
the
wi
dth of the
ca
ntilever;
t(
μ
m) is
the thic
knes
s
of the c
antilever; a(
μ
m) is the hal
f-width of the
cente
r
blo
ck.
3. Ne
w
Sche
me of the Po
sition De
ter
m
ination
In the ca
se o
f
no stre
ss,
resi
stan
ce in t
he X-di
re
ctio
n brid
ge do
e
s
not chan
ge
, and th
e
bridg
e
b
a
lan
c
e
s
. When t
he st
re
ss
exist, the resi
st
ance of the
varisto
r
woul
d ch
ang
e. O
u
tput
voltage in X-d
i
rectio
n ca
n b
e
expre
s
sed
as:
in
outx
V
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
V
)
)(
(
)
)(
(
)
)(
(
4
4
3
3
2
2
1
1
4
4
2
2
3
3
1
1
(2)
Her
e
,
4
3
2
1
R
R
R
R
,
R
R
R
R
R
4
3
2
1
. In
the formula,
in
V
is the input voltage.
Formul
a (2
) can be ap
proxi
m
ately expre
s
sed a
s
:
in
outx
V
R
R
V
(3)
Whe
n
a vibra
t
ion sign
al (which
can
be
seen a
s
a
sinu
soid
al sig
nal) acts o
n
the
sen
s
o
r
,
the outp
u
t voltage
s in X
-
dire
ction
and
Y-direct
io
n
are Vx
and
Vy, resp
ectiv
e
ly. Due
to t
h
e
dire
ction of t
he pip
e
line b
u
ried
ha
s be
en kno
w
n, what only nee
d to do i
s
to
determi
ne t
he
orientatio
n from ze
ro to
1
80 de
gre
e
s.
Two
Whe
a
tst
one b
r
idg
e
s
have a
c
tually
measured t
h
e
voltage diffe
rence b
e
twe
e
n
the
re
si
stors
(R2 a
nd
R3)
and
(R6
a
nd
R7
), resp
ectively. And
the
final testin
g
voltage ha
s
been
output
after the
voltage
differe
nce
goin
g
b
y
the differe
ntial
amplifier
circuit. And if the
voltage acro
ss the
re
sisto
r
R2 is g
r
e
a
te
r than the voltage acro
ss
R3,
Vx is p
o
sitive
, otherwi
se V
x
is
negative;
the volt
ag
e
across the
re
sisto
r
R6
is g
r
eate
r
tha
n
t
he
voltage acro
ss R7, Vy is p
o
sitive,
on the other ha
nd,
Vy is negative.
We can get t
he dire
ction
angle of the
Pipeline in
sp
ection g
aug
e
(so
und
sou
r
ce)
θ
is
arcta
n
(Vy
/
Vx
) (0 °
<
θ
≤
90
°) by
calcul
ating. If the so
und
so
urce in the fi
rst q
uad
rant,
the
dire
ction a
ngl
e of the sou
nd source
θ
is arctan
(Vy / Vx); If
the sou
nd
sou
r
ce in the
se
co
nd
quad
rant, the
directio
n an
gle of the sound so
urce i
s
180
-
θ
. Det
e
rmin
ation of sound
sou
r
ce is
sho
w
n i
n
Fig
u
re
8. By an
alyzing th
e trend
s an
d
rel
a
tionship
s
of
the output vo
ltages
of the
Vx
and Vy, determine the sp
e
c
ific lo
cation
of the soun
d sou
r
ce (in the
first or se
co
n
d
quad
rant
).
Figure 8. Dia
g
ram of dete
r
mination of sound
sou
r
ce
Whe
n
the so
und sou
r
ce is in the first quad
rant, R2 and R6 are sufferi
ng the sa
me
norm
a
l stress or n
egati
v
e stre
ss (cha
ngin
g
sy
nch
r
on
ou
sly).
R3 and
R7 also
cha
nge
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 12, Decem
ber 20
14 : 8126 – 81
32
8130
synchro
nou
sl
y. But R2 a
n
d
R3 (R6
an
d R7)
ch
ang
e a
s
ynchro
no
usly. Outp
ut
voltage of
bri
dge
V
out
x
in X-direction goi
ng b
y
the differen
t
ial mode am
plifier ci
rcuit is Vx, and Ou
tput voltage of
bridg
e
V
out
y
in Y-dire
ction
going by the differential m
ode amplifie
r circuit is Vy. Vx and Vy
are
sufferin
g
the same no
rmal stre
ss
or neg
ative
stre
ss
(cha
ngin
g
syn
c
hrono
usly
); Whe
n
the
so
und
sou
r
ce is in the se
co
nd q
uadrant, R2
and R6 are resp
ectively suffering the n
o
rmal
stre
ss
and
the neg
ative stre
ss (cha
n
g
ing a
s
yn
chronou
sly)
. R3
and
R7
cha
nge a
s
yn
chronou
sly. And
R2
and
R3 (R6
a
nd R7)
cha
n
g
e
asyn
ch
ron
o
u
sly. Output
voltage of bri
dge V
outx
in X-directio
n goi
ng
by the diffe
re
ntial mod
e
a
m
plifier
circui
t is Vx, an
d
Output voltag
e of b
r
id
ge V
out
y
in Y-direction
going
by the
differe
ntial
mode
amplifi
e
r
circ
uit
is
Vy. Vx and
Vy cha
nge
asyn
chrono
u
s
ly.
Acco
rdi
ng to
the
relatio
n
s
hip
bet
wee
n
Vx an
d V
y
, we
can
make
sure t
he
sou
nd
so
urce
(distin
gui
shin
g
Ⅰ
and
Ⅱ
q
uadrant).
4. Analy
s
is a
nd Verificati
on of the Ex
perimental T
esting
Resul
t
In the exp
e
ri
ment, the
se
nso
r
i
s
ve
rtically do
wn
ward, and
sen
s
i
ng h
ead
bu
ri
ed in
the
soil tight and
heavy (in Fig
u
re 9
)
, ensuri
ng that the se
nso
r
is
coupl
ed with soil well.
In order to
si
mulate the
fri
c
tion
acou
stic si
g
nal
s of th
e pip
e
line
in
spectio
n
g
aug
e runnin
g
in the
pipeli
n
e bette
r, we
can
u
s
e th
e
way of
tappin
g
on
the
grou
nd to
sim
u
lat
e
it. By collecting
and an
alyzin
g the tappin
g
sign
als, we can find
the ta
pping
sign
als
are in 8
0
HZ-3
00HZ
(in Fig
u
r
e
10)
whi
c
h
a
r
e in
clud
ed in
the
spe
c
tral
ran
ge
of th
e fri
c
tion
sig
nal of th
e pi
peline
in
spe
c
tion
gaug
e ru
nnin
g
in the pip
e
line. To a
ce
rtain ext
ent, tappin
g
sig
nal
s could
simul
a
te the actu
a
l
a
c
ou
s
t
ic
s
i
gna
l.
0
200
400
60
0
0
0.
5
1
1.
5
2
率
频
Hz
幅
值
v
Figure 9. Method of the
M
E
MS acou
stic vector
s
e
ns
or
b
u
r
i
e
d
Figure 10. Spectru
m
of the beating
sign
a
l
First
of all,
whe
n
MEMS
vector a
c
ou
stic
se
nso
r
i
s
placed,
sem
i
circle i
s
p
a
in
ted with
aco
u
sti
c
se
nsor a
s
the cen
t
er and
r as t
he ra
diu
s
of the ci
rcl
e
.On
the semi
circl
e
, dra
w
a poi
nt
every 10° an
d dra
w
the po
int of zero to 180 deg
re
es. Senso
r
expe
rimental sche
matic diag
ra
m is
sho
w
n
in Fi
g
u
re
11. In
Fi
gure
11, X
di
rectio
n a
nd
0
°
ove
r
lap, Y
dire
ction
and
90
° ove
r
lap
,
O
stand
s fo
r th
e
se
nsor. T
he
sen
s
o
r
a
nd th
e soun
d
sou
r
ce
are
lo
cate
d in th
e
same
plan
e, an
d th
e
experim
ental
treatment an
gle is the ho
ri
zont
al a
ngle
and pitching
angle i
s
not calcul
ated.
Figure 11. Di
agra
m
of the experim
ent
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Ne
w Re
se
arch on MEMS Acou
stic Ve
ctor Sen
s
or
Used in Gro
und
Marker of
… (Liu Meng
-ran
)
8131
In the Experiment, tappin
g
the grou
nd
in
the equa
l diversio
n p
o
int, the sen
s
or
will
transl
a
te
aco
u
stic si
gnal
s
colle
cted
into
ele
c
tri
c
al
sig
nals. The
ele
c
tri
c
al sig
nals
are colle
cted
by
data acqui
sition ca
rd an
d p
r
ocesse
d in MATLAB.
Data
acqui
sition
ca
rd i
s
PXTe-10
71
of the
NI, a
nd the
termi
nal b
o
x is B
N
C211
0.
Experimental
sampli
ng ra
te is 10K, and coll
ectin
g
time is 10s. Tap equal
diversi
on p
o
int
respe
c
tively(r=3.5m
)
, and
captu
r
e an
d store the
exp
e
rime
ntal dat
a by the acq
u
isition
card and
comp
uter. Direction angl
e can be
g
o
t
by
data
p
r
o
c
e
ssing (in
Tabl
e
1). The
re
sult
s sho
w
that the
need
s of the proje
c
t will be
met within 5 degree
s of the mean e
rro
r
of the directio
n angle.
No
w
we
use t
he exam
ple
s
of 40°
an
d 1
4
0
° to
explain
the p
r
oble
m
of the p
o
rt/st
a
rbo
a
rd
blur. Tap 40
° and 140
°eq
u
a
l diversi
on p
o
int once re
spectively. Signals
colle
cted
by senso
r
are
sho
w
n in Fi
g
u
re 1
2
. Figure 12 (a
) sho
w
s that
Vx a
nd Vy, the output voltage
colle
cted, ch
ange
synchro
nou
sl
y, and the
so
und
so
urce
should
be i
n
t
he first qu
ad
rant
Ⅰ
; Fig
u
r
e
12
(b
)
sho
w
s
that Vx and
Vy, the outp
u
t voltage
co
llected,
ch
an
ge a
s
yn
chro
nou
sly, and
the
sou
nd
so
urce
should
be i
n
t
he first quadrant II. Sound
source m
e
asured are the
same
with
the real
station of
the sou
nd so
urce, whi
c
h furthe
r prove
s
the corre
c
tne
ss of the
s
e p
r
ogra
m
s a
nd theori
e
s.
Table 1. Re
sult of determi
nation of
MEMS acou
stic
vector
sen
s
o
r
Real
angle
（
°
）
Direction
angle
（
°
）
error
（
°
）
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
3.4236
12.3271
27.7468
36.9309
42.6213
48.9025
56.6354
73.2508
84.2464
85.3726
99.4331
105.8263
123.6328
134.4365
136.5586
156.4202
158.0138
166.2341
178.1342
3.4236
2.3271
7.7468
6.9309
2.6213
-1.0975
3.3646
3.2508
4.2464
-4.6274
-0.5669
-4.1737
3.6328
4.4365
-3.4414
6.4202
-1.9862
-3.7659
-1.8658
(a) Sign
al wh
en beatin
g 40
°
(b) Sign
al wh
en beatin
g 14
0°
Figure 12. Te
sting si
gnal
0
50
10
0
15
0
20
0
25
0
-0
.
2
-0
.
1
5
-0
.
1
-0
.
0
5
0
0.
0
5
0.
1
0.
1
5
T
h
e s
a
m
p
l
i
ng
p
o
i
n
t
s
T
he out
put
s
i
gnal
v
o
l
t
a
ge/
v
x-
d
i
r
e
ct
i
o
n
y-
d
i
r
e
ct
i
o
n
0
50
100
15
0
20
0
250
-0
.
1
-0.
0
8
-0.
0
6
-0.
0
4
-0.
0
2
0
0.
0
2
0.
0
4
0.
0
6
0.
0
8
T
h
e s
a
m
p
l
i
n
g
poi
nt
s
T
he out
put
s
i
gnal
v
o
l
t
a
ge/
v
x-
d
i
r
e
ct
i
o
n
y-
d
i
r
e
ct
i
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 12, Decem
ber 20
14 : 8126 – 81
32
8132
5. Conclusio
n
A MEMS bionic a
c
ou
stic
vector sen
s
o
r
with
high se
nsitivity, good perfo
rman
ce in low
freque
ncy a
n
d
lower
power h
a
s
been
use
d
in the
grou
nd m
a
rki
ng. As to the
pro
b
lem of t
he
port/sta
rbo
a
rd blur of this vecto
r
sen
s
or in
g
r
o
u
n
d
marking, a
new sch
e
m
e
has b
een
put
forwa
r
d. By analyzi
ng the
theory an
d
mathemati
c
al
model of th
e vector
se
n
s
or
structu
r
e,
it
con
c
lu
de
s that if Vx and
Vy chang
e synchrono
us
ly
, the soun
d sou
r
ce shoul
d be in the f
i
rst
quad
rant I; if
Vx and
Vy
cha
nge
a
s
yn
chrono
us
ly,
t
he sou
nd so
urce sho
u
ld be
in
the se
cond
quadrant II. The result
s sh
ow that the needs of the project
will be
met within 5
degrees of the
mean e
r
ror of
the dire
ction
angle. And t
he co
rrec
tn
ess of the prop
ose
d
ne
w pro
g
ram h
a
s
be
en
further valid
ated.
Referen
ces
[1]
Guo Min-zh
i, Yang Ji
a-Yu. St
atus an
d dev
el
opme
n
t trend o
f
contempor
ar
y transporti
on te
chno
log
y
fo
r
oil .
Chi
na Petr
ole
u
m
and C
h
e
m
ic
al Ind
u
stry
. 200
4; (7): 16~
2
0
.
[2]
Cui
Ya
o-Yao. Stud
y
on
the Ke
y
T
e
chno
lo
gies
in th
e Rackin
g
and L
o
c
ation S
y
stem
of Pipelin
e
Spectio
n
Gaug
e Based o
n
Ac
oustic Sens
or Arra
y
s
. PhD T
hesis. T
i
anjin : T
i
anjin Un
ivers
i
t
y
; 20
11.
[3]
W
u
Xi
ao. R
e
s
earch
on Ab
o
v
e Groun
d M
a
rkin
g an
d T
r
ackin
g
of Oil
and Gas P
i
pe
line I
n
terna
l
Inspectio
n
Instrument Base
d o
n
Geoph
on
e Arra
y
.
PhD T
hesi
s
.
T
i
anjin : T
i
anjin Un
iversit
y
; 2
011.
[4]
W
U
Xiao, JIN
Shi-ji
u, LI Yi-
b
o. Abov
e-gro
u
nd mark
er s
y
st
em of p
i
p
e
li
ne
i
n
terna
l
i
n
specti
on i
n
strume
nt
base
d
on g
e
o
p
hon
e arra
y.
Na
notech
n
o
l
ogy a
nd Precis
ion E
ngi
neer
in
g.
20
10; 8(6):55
4
.
[5]
GE Xiao-
ya
n
g
, Z
H
ANG Guo-
j
un, D
U
C
h
u
n
-
hui. A
n
e
w
ME
MS bi
onic
ac
o
u
stic vector
se
nsor
used
i
n
abov
e-gro
u
n
d
marker of pip
e
l
i
ne.
Pie
z
oelect
rics & Acoustooptics
. 201
2; 3
4
(6):88
2-88
5.
[6]
Che
n
Sh
ang,
Xue
Che
n
-
y
ang, Zh
ang
W
en-do
ng,
Fa
bricati
on
and
testing
of a
silic
on-b
a
se
d
piez
oresistiv
e
tw
o
-
a
x
is
acce
le
rometer.
Na
not
echn
olo
g
y a
n
d
Precisi
on E
n
g
i
ne
erin
g
. 20
08;
6(4): 27
2-
277.
[7]
Che
n
S
han
g.
Rese
arch
of a
Bion
ic Vect
or
H
y
dro
pho
ne
B
a
sed
o
n
Si
lico
n
. PhD
T
hesis. T
a
i
y
u
an:
Nort
h
univ
e
rsit
y
of Chin
a; 200
8.
[8]
Xu
Ji
ao, Z
h
an
g Guo-
jun,
Shi
Gui-
xi
ong.
Ad
vancem
ents i
n
enc
aps
ulati
o
n
of h
a
ir
vector
h
y
dro
pho
ne
.
Chin
ese Jo
urn
a
l of Sensors
a
nd Actuators
.2
011; 24(
4): 519
-520.
[9]
Liu
Li
n-
xia
n
, Z
han
g Gu
o=
jun,
Xu
Ji
ao. D
e
si
gn
and
test fo
r a
dou
bl
e T
-
shap
e MEMS
b
i
onic
vect
o
r
h
y
dro
pho
ne.
Jo
urna
l of Vibrati
on an
d Shock
.
201
3; 32(2): 13
0-13
1.
[10]
Xu J
i
a
o
, Li Ju
n
,
Z
hang Guo-
ju
n. Desig
n
of a
nove
l
vector h
y
drop
hon
e b
a
se
d on MEMS.
Pie
z
oel
e
c
tri
cs
& Acoustooptics.
2012; 34(
1): 90-9
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.