Precise Measurement of Characteristic Responses
for Unexploded Ordnance
Xiaoyan Liu
1
, Shuang Zhang
1
, Haofeng Wang
1
, Shudong Chen
1
*, Zhiwen Yuan
2
, Haiyang Zhang
2
,
Dong Fang
2
and Jun Zhu
2
*
1
College of Electronic Science and Engineering, Jilin University, Changchun City, Jilin Province, China
2
Science and Technology on Near-Surface Detection Laboratory, Wuxi City, Jiangsu Province, China
Keywords: UXO, characteristic responses, dipole model, transmitting coil.
Abstract: The equipment for measuring the characteristic responses of unexploded ordnance (UXO) is proposed in
this paper. It is composed of two parts: a solenoid and a pair of rectangular coils. The uniformity of the
primary field for solenoid reaches 98% and 97% for rectangular coils. The characteristic responses of six
targets with different size, length and outer diameters are measured. The results indicate that the amplitude
and decay rate of the characteristic responses change significantly with the length and outer diameter. The
longer the target, the greater the amplitude will be. The larger the diameter of the target, the slower the
decay rate will be. It means that the physical information can be well reflected by characteristic responses.
1 INTRODUCTION
Unexploded ordnance, left after wars or other
military activities, is an increasingly serious
international humanitarian and environmental
problem, which has caused many civilian
casualties. Thus, how to detect and clean up UXO
has attracted wide attention all over the
world(Miller R, 2008).
Transient electromagnetic(TEM) system has
been widely used for UXO detection(Laurens B et
al, 2013). The most frequently used model for
representing the EMI response of a metallic target
approximates the whole object with a set of
orthogonal co-located point dipoles(Shubitidze F,
2012). Based on the dipole model, Nagi Khadr has
studied the relationships between the characteristic
responses and the aspect ratio of UXO(Khadr N et
al, 1998). Leonard R. Pasion et al. have calculated
the EMI response of an axisymmetric target and
estimate the position and orientation of the target
based on the measured response(Pasion L R et al,
2007). Chen Shudong and other scholars studied
the influence of physical parameters on the
characteristic responses of targets (Shu-Dong C et
al, 2017). However, none of these studies have
talked the measurement of characteristic responses
in detail.
According to the dipole model, only the target
is excited by a uniformly field, can the
characteristic responses be measured accurately.
However, it is very difficult to produce a uniform
magnetic field for transmitting coil used in a TEM
system. Based on the dipole model, a specially
designed equipment for measuring the
characteristic responses accurately will be
proposed in this paper. The structure and
parameters of the equipment will be listed in detail.
We will measure and analyze the characteristic
responses of typical targets by this equipment.
2 DIPOLE MODEL
According to the dipole model, the magnetic field
(secondary field) H generated by the eddy current
of targets excited by the primary field H
p
can be
equivalent to the magnetic field generated by an
induction dipole m.
498
Liu, X., Zhang, S., Wang, H., Chen, S., Yuan, Z., Zhang, H., Fang, D. and Zhu, J.
Precise Measurement of Characteristic Responses for Unexploded Ordnance.
In 3rd International Conference on Electromechanical Control Technology and Transportation (ICECTT 2018), pages 498-502
ISBN: 978-989-758-312-4
Copyright © 2018 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

x
y
z
z
m
x
m
y
m
P
H
P
r
o
Figure 1: Orthogonal dipole model of UXO.
As shown in Figure 1, the secondary field H in
point P can be calculated as:
()
3
3
4
rr
r π
−
=
ee I m
H
(1)
Where r represents the position vector of point
P, r is the modulus of r, and e
r
is the unit vector of
r. The induction dipole m can be calculated as:
00
00
00
xxx
yyyp
zzz
β
β
β
⎡⎤⎡ ⎤
⎢⎥⎢ ⎥
==
⎢⎥⎢ ⎥
⎢⎥⎢ ⎥
⎣⎦⎣ ⎦
m
mm H
m
(2)
For an axisymmetric target, such as UXO, the
first and second dipole m
x
, m
y
are both
perpendicular to the major axis, and the β
xx
and β
yy
refer to the magnetic polarizations
correspondingly. The third dipole m
z
is parallel to
the major axis of the target, and the β
zz
refers to the
magnetic polarization parallel to the major axis of
targets. The characteristic responses of targets are
defined as the derivative of the magnetic
polarization.
()
z
z
p
d
lt
dt
β
=−
(3)
()
yy
xx
v
d
d
lt
dt dt
β
β
=− =−
(4)
According to the equations (3, 4), the
characteristic responses l(t) consists of two
components l
p
(t) and l
v
(t) for axisymmetric targets.
It can be expressed as follows:
()
() ()
()
v
v
p
lt
tlt
lt
⎡⎤
⎢⎥
=
⎢⎥
⎢⎥
⎣⎦
l
(5)
For an axisymmetric target, the characteristic
responses can be divided into two kinds. One is
l
p
(t) parallel to the major axis, and the other is l
v
(t)
perpendicular to the major axis.
The EMI response of a metal target can be
approximated to the response of a single dipole
only when the volume of the target is small enough
or the transmitting coils are far enough. The
primary field is substantially uniform over the
range of the target. In order to achieve the uniform
primary field, two kinds of transmitting coils will
be discussed in next section.
3 TRANSMITTING COILS
3.1 Transmitting Coil to Excite the
Target Parallel to the Major Axis
The long straight solenoid can generate the
uniform primary field along the major axis,
ensuring that all parts of targets can be excited
uniformly. The specific parameters of long straight
solenoid can be shown in Figure 2:
Figure 2: The parameters of a long straight solenoid.
As shown in Figure 2, the parameters of a
solenoid include the length l, the radius r and the
number of turns n
p
. The internal primary field of a
long straight solenoid is parallel to the major axis
of it. If the transmitting current through the long
straight solenoid is I, the amplitude of the primary
field H
pp
at any point along the z-direction can be
calculated as:
() ()
22
22
22
[]
2
22
p
pp
nI
lz lz
l
rl z rl z
H
+−
=+
++ +−
(6)
According to equation (6), when the length l is
1.2 meters, the radius r is 0.11 meters, the number
of turns n
p
is 30, the current I is 9 amps, the
distribution of the normalized primary field along
the major axis can be calculated, as shown in
Figure 3:
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
98.0
98.5
99.0
99.5
100.0
Relative error/%
Relative distance/m
Figure 3: The relative error of field along the axial.
Precise Measurement of Characteristic Responses for Unexploded Ordnance
499

It can be seen from Figure 3 that the amplitude
of the primary field at the center of the solenoid is
the largest. Within the range of -0.3 meters to 0.3
meters, the primary field along the major axis of
the solenoid is almost uniform. The uniformity of
the primary field can reaches 98%. According to
the equation (6), we can see that the amplitude of
primary field at the center of the long straight
solenoid H
pp
(0) can be calculated as:
()
22
0
4
p
pp
nI
H
rl
=
+
(7)
3.2 Transmitting Coil to Excite the
Target Vertical to the Major
Axis
According to the principle of the Helmholtz coils,
a pair of rectangular coils can be applied to
generate the uniform primary field being
perpendicular to the major axis. A pair of
rectangular coils has been designed by this way to
generate the uniform primary. The structure and
parameters of a pair of rectangular coils are shown
in Figure 4:
Figure 4:The parameters of a pair of rectangular coils.
As shown in Figure 4, the parameters of a pair
of rectangular coils include the coil length a, the
width b, the distance d and the number of turns n
v
.
For the currents in the two coils are the same, the
x-component and the y-component of the primary
field in the center plane of two coils cancel each
other and the z-component is doubled. The primary
field on the center plane can be expressed as:
0
22
222 222
{[ ]
2()
()() ()()
v
pv
nI
ax ax by
H
z
by
ax by z ax by z
μ
π
−++
=+
++
−+++ ++++
22
222 222
[]
()
()() ()()
by by ax
zax
ax by z ax by z
−+−
++
+−
−+−+ −+++
22
222 222
[]
()
()() ()()
ax ax by
zby
ax by z ax by z
+−−
++
+−
++−+ −+−+
22
222 222
[]}
()
()() ()()
by by ax
zax
ax by z ax by z
+−+
++
++
++++ ++−+
(8)
When the current intensity I is 7.2 amps and z
is 0.18 meters (d/2), n
v
is 12, the distribution and
the error rate of the primary field generated by a
series of rectangular coils on the central plane can
be calculated as:
9
3
9
3
9
4
9
4
9
4
9
4
9
4
9
4
9
4
9
4
9
5
9
5
9
5
9
5
9
5
9
5
9
5
9
5
9
6
9
6
9
6
9
6
9
6
9
6
9
6
9
6
9
7
9
7
9
7
9
7
9
7
9
7
9
7
9
7
9
7
9
7
9
8
9
8
9
8
9
8
9
8
9
8
9
8
9
8
9
9
9
9
9
9
9
9
9
9
9
9
1
0
0
100
Relative error of magnetic field(%)
Relative distanc e of lon
g
ed
g
e X
(
m
)
Relative distanc e of wide edge Y(m)
-0.3 -0.2 -0.1 0 0.1 0.2 0.3
-0.1
-0.05
0
0.05
0.1
0.15
Figure 5:The error rate of the primary field.
As shown in Figure 5, due to the two
rectangular coils are lied in opposite directions, the
primary fields of x-component and y-component
that are symmetric about the central plane almost
offset to zero each other. And only the amplitude
of the primary field along z-component is
superimposed on each other. The uniformity of the
primary field reaches 97% within a certain area,
which can ensure the accuracy of the measured
results of the characteristic responses.
The amplitude of primary field H
pv
(0,0) at the
central plane generated by rectangular coils can be
calculated as:
()
2222
222
16
11
0, 0
44
44
v
pv
nI
ab
H
dadb
abd
π
⎛⎞
=+
⎜⎟
++
⎝⎠
++
(9)
4 MEASUREMENT RESULTS
AND ANALYSIS
4.1 Targets Description
The characteristic responses of six targets will be
measured and analysis in this section. Three of
them are tubular targets with same outer diameter
and wall thickness and different length from 50mm
to 300mm. The other three targets are mortar shells
of different diameters. All these targets are shown
in Figure 6:
ICECTT 2018 - 3rd International Conference on Electromechanical Control Technology and Transportation
500

Figure 6: Six kinds of typical target body.
The parameters and numbers of the six targets
are listed in the table below.
Table 1: Parameters of targets.
Number Length
(m
m
)
Diameter
(m
m
)
Thickness
(m
m
)
Diameter
(m
m
)
N1 300
75 4
\
N2 200
\
N3 57
\
N4
\
\
\
120
N5
\
\
\
82
N6
\
\
\
60
4.2 Characteristic Responses Analysis
The characteristic responses l
p
(t) and l
v
(t)of three
tubular targets numbered as N1, N2 and N3 are
calculated, as shown in Figure 7:
0 5 10 15 20 25 30 35
1E-5
1E-4
1E-3
0.01
0.1
1
10
100
N1 _ l
p
N2 _ l
p
N3 _ l
p
Characteristic response(V/m
2
T)
Channel
0 5 10 15 20 25 30 35
1E-5
1E-4
1E-3
0.01
0.1
1
10
N1 _ l
v
N2 _ l
v
N3 _ l
v
Characteristic response(V/m
2
T)
Channel
Figure 7: The l
p
(t) and l
v
(t) of tubular targets.
The characteristic responses l
p
(t) and l
v
(t)of
three mortar shells numbered as N4, N5 and N6
are calculated, as shown in Figure 8:
0 5 10 15 20 25 30 35
0.01
0.1
1
10
100
Characteristic response(V/m
2
T)
Channel
N4 _ l
p
N5 _ l
p
N6 _ l
p
5 101520253035
1E-4
1E-3
0.01
0.1
1
10
N4 _ lv
N5 _ lv
N6 _ lv
Characteristic response(V/m
2
T)
Channel
Figure 8: The l
p
(t) and l
v
(t) of mortar shells.
As shown in Figure 7 and Figure 8, the
characteristic responses l
p
(t) decay more slowly
with larger amplitude compared to l
v
(t). For tubular
targets, the amplitude of l
p
(t) and l
v
(t) both increase
with the length of targets. When the length of the
tubular targets reduce to 50 mm, the characteristic
responses decay so fast that the signal-to-noise
ratio of the signal is too low to be used at late time.
For mortar shells, the larger the volume, the
greater the amplitude of l
p
(t) and l
v
(t) and the
slower the characteristic responses attenuation will
be.
5 CONCLUSIONS
An equipment which can generate uniform primary
field is proposed in this paper. The uniformity of
the primary field produced by the equipment
reaches 97%. The characteristic responses of six
targets have been measured and analyzed.
According to the dipole model, the
characteristic response l(t) of an axisymmetric
target can be accurately measured only when the
targets are excited uniformly by the primary field.
The calculated results show that uniformity of the
primary field reaches 98% for solenoid and 97%
for rectangular coils.
By the equipment designed in this paper, the
characteristic responses of six targets have been
measured. The results show that the amplitude of
the l
p
(t) is about ten times than that of l
v
(t), at the
early stage. The longer the length of the target, the
larger the amplitude of responses will be.
The equipment designed in this paper can be
used to measure the characteristic responses of
targets precisely. The physical information, such as
the shape, structure can be well reflected by their
Precise Measurement of Characteristic Responses for Unexploded Ordnance
501

characteristic responses. The work described in
this paper is important for accurate detection and
identification of unexploded ordnance.
ACKNOWLEDGEMENTS
This work was supported by the National Natural
Science Foundation of China under Grant No.
41704145 and the Science and Technology on
Near-Surface Detection Laboratory under Grant
No. TCGZ2017A003.
REFERENCES
Miller R. Introduction to this special section: Near-
surface geophysics [J]. The Leading Edge, 2008, 27
(11): 1423-1423.
Laurens B, Barry Z, Stephen B. Detecting and
Classifying UXO. The Journal of ERW and mine
action. Vol. 17 No. 1 2013.
Shubitidze F. A complex approach to UXO
discrimination: combining advanced EMI forward
models and statistical signal processing[R]. SKY
RESEARCH INC ASHLAND OR, 2012.
Khadr N, Barrow B J, Bell T H, et al. Target shape
classification using electromagnetic induction sensor
data[C]//Proceedings of UXO Forum’98. 1998.
Pasion L R, Billings S D, Oldenburg D W, et al.
Application of a library based method to time
domain electromagnetic data for the identification of
unexploded ordnance[J]. Journal of Applied
Geophysics, 2007, 61(3): 279-291.
Shu-Dong C, Shu-Xu G, Shuang Z, et al. Modeling and
analysis of transient electromagnetic responses of
finite conductors in the near-surface[J]. CHINESE
JOURNAL OF GEOPHYSICS-CHINESE
EDITION, 2017, 60(1): 403-412.
ICECTT 2018 - 3rd International Conference on Electromechanical Control Technology and Transportation
502