sensors

Article
Study on the Application of Optical Current Sensor
for Lightning Current Measurement of
Transmission Line
Jin-ming Ge 1 , Yan Shen 1, *, Wen-bin Yu 1 , Yue Han 2 and Fang-wei Duan 2
1 Department of Electrical Engineering, Harbin Institute of Technology, Harbin 150001, China;
gejinming90@126.com (J.-m.G.); ywb_hit@126.com (W.-b.Y.)
2 State Grid Liaoning Electric Power Research Institute, Shenyang 110006, China;
hanyuehaha1@126.com (Y.H.); dfw8906@163.com (F.-w.D.)
* Correspondence: yanshen@hit.edu.cn




Received: 2 September 2019; Accepted: 18 November 2019; Published: 22 November 2019


Abstract: Accurate and reliable lightning current data are the basis of lightning protection design.
To measure lightning current data at different measurement locations in a transmission system,
the limitations of traditional lightning current sensors are analyzed, and optical current sensing
technology is adopted, which has the advantages of no magnetic saturation and no bandwidth
limitation. Compared with traditional application environments, the sensing technology is used in
special environments in transmission systems. This paper analyzes the influence of environmental
factors on sensors, and combines the extreme environmental requirements, such as temperature and
insulation requirements, to study the sensor. Starting from the sensitivity, the sensing characteristics
of the sensor are analyzed. The sensor is designed according to three aspects: sensing material
selection, spatial measuring position, and sensing material size optimization, such that it can satisfy
the different measurement requirements of towers, overhead ground wires, and transmission lines,
respectively. The experiments indicate that the developed sensors can meet the measurement
sensitivity requirements of different types of lightning strikes. The experimental results of sensors
exhibit a reasonable amplitude measurement accuracy, linearity, and waveform measurement
capability. These results provide important theoretical and experimental bases for the application of
optical current sensing technology to the measurement of the lightning current of transmission systems.

Keywords: optical current sensing technology; lightning current measurement; transmission system

1. Introduction
Lightning is a powerful natural discharge phenomenon that occurs between cloud-to-cloud or
cloud-to-ground, of which the discharge process releases enormous energy and produces powerful
electromagnetic pulses. Direct lightning strikes or lightning electromagnetic pulses not only pose a
serious threat to human life safety, but also have a huge impact on the power grid, which, with wide-area
distribution, leads to direct or indirect economic losses. The magnitude, steepness, and waveform
of the lightning current play an important role in lightning protection design of the power grid [1].
As the distribution of lightning in different regions has a strong relationship with the local atmospheric
conditions, topography, soil resistivity, and other factors [2], it is possible to investigate the lightning
current parameter measurement statistics in different regions by installing lightning current measuring
devices in a widely distributed transmission system. The lightning current parameter measurement
statistics of areas are used to develop scientific differential lightning protection according to the
lightning distribution and parameter characteristics of different regions [3]. Furthermore, lightning

Sensors 2019, 19, 5110; doi:10.3390/s19235110 www.mdpi.com/journal/sensors

Sensors 2019, 19, 5110 2 of 15

Sensors 2019, 19, x FOR PEER REVIEW 2 of 15

location systems (LLSs) are applied in lightning measurement and location in many areas; however,
Furthermore, lightning location systems (LLSs) are applied in lightning measurement and location
the direct measurement data of lightning current are necessary in order to calibrate the LLS using the
in many areas; however, the direct measurement data of lightning current are necessary in order to
lightning current sensor of a transmission system [4]. Therefore, it is very important to study lightning
calibrate the LLS using the lightning current sensor of a transmission system [4]. Therefore, it is very
current sensors suitable for power transmission systems.
important to study lightning current sensors suitable for power transmission systems.
Some researchers have conducted extensive research in this field, and currently mainly use the
Some researchers have conducted extensive research in this field, and currently mainly use the
following devices: magnetic steel bars [5], magnetic tape [6,7], and Rogowski coils [8,9]. Ping et al. [5]
following devices: magnetic steel bars [5], magnetic tape [6,7], and Rogowski coils [8,9]. Ping et al. [5]
started using magnetic steel bars to measure the Zhejiang Xin-Hang line for several decades, on
started using magnetic steel bars to measure the Zhejiang Xin-Hang line for several decades, on which
which the relevant industry standards in China are based. However, the measurement error of the
the relevant industry standards in China are based. However, the measurement error of the magnetic
magnetic steel bars is large; thus, it is not repeatable, and the installation and data reading method are
steel bars is large; thus, it is not repeatable, and the installation and data reading method are
inconvenient. The magnetic tape method significantly improves the measurement accuracy, stability,
inconvenient. The magnetic tape method significantly improves the measurement accuracy, stability,
and ease of measurement compared with magnetic steel bars [6,7], but still cannot measure the lightning
and ease of measurement compared with magnetic steel bars [6–7], but still cannot measure the
current waveform. The structure of the Rogowski coil is simple, convenient to install, and can be used
lightning current waveform. The structure of the Rogowski coil is simple, convenient to install, and
to measure large pulse current signals with a large amplitude and rapid change. It has been a primary
can be used to measure large pulse current signals with a large amplitude and rapid change. It has
research target of lightning current measuring devices for transmission systems in recent years [8,9].
been a primary research target of lightning current measuring devices for transmission systems in
However, the Rogowski coil has a bandwidth limitation in the measurement principle, which contains
recent years [8,9]. However, the Rogowski coil has a bandwidth limitation in the measurement
low frequency information with large amplitude in lightning current. As a new type of current sensing
principle, which contains low frequency information with large amplitude in lightning current. As a
technology, the optical current sensor (OCS) is only used in intelligent substations, because it has no
new type of current sensing technology, the optical current sensor (OCS) is only used in intelligent
iron core, no band limitation, and natural insulation, among others. It is also very suitable for lightning
substations, because it has no iron core, no band limitation, and natural insulation, among others. It
current measurement of transmission systems.
is also very suitable for lightning current measurement of transmission systems.
In this study, the application of optical current sensing technology, which can meet the waveform
In this study, the application of optical current sensing technology, which can meet the
measurement requirements of lightning current, is taken as the research background. The different
waveform measurement requirements of lightning current, is taken as the research background. The
measurement requirements of the sensing system for the lightning current measurement of the
different measurement requirements of the sensing system for the lightning current measurement of
transmission system are analyzed. Starting from the perspective of factors influencing the sensing
the transmission system are analyzed. Starting from the perspective of factors influencing the sensing
characteristics of optical sensing materials, two types of OCSs suitable for power transmission systems
characteristics of optical sensing materials, two types of OCSs suitable for power transmission
were developed. Experimental platforms were also built to verify the sensing characteristics of
systems were developed. Experimental platforms were also built to verify the sensing characteristics
the OCSs.
of the OCSs.
2. Design Idea of Lightning Current Sensor for the Transmission System
2. Design Idea of Lightning Current Sensor for the Transmission System
2.1. Sensing Model of OCS
2.1. Sensing Model of OCS
The Faraday magneto-optical effect sensing principle, applied by OCS for lightning current
The Faraday
measurement, magneto-optical
is shown effect
in Figure 1. The sensing
Faraday principle,
rotation angleapplied
ϕ (rad)by OCS for lightning
is expressed as current
measurement, is shown in Figure 1. The Faraday rotation angle φ (rad) is expressed as
dϕd=VH
VH·dl
dl (1)
(1)

whereVVisisthe
where theVerdet
Verdetconstant (rad/T·m),HHisisthe
constant(rad/T·m), themagnetic
magneticfield
fieldstrength
strength(A/m), andl lisisthe
(A/m),and thedistance
distance
light
lightpasses
passesthrough
throughaamagneto-optical
magneto-opticalmaterial
material(m).
(m).

Lightning
Polarizer Current I
Analyzer
Magnetic
Field StrengthH
Light source


Magneto-optical
Material
Output light

Figure1.1.Faraday
Figure Faradaymagneto-optical
magneto-opticaleffect.
effect.

The relationship between φ and H in the light propagation direction can be expressed as

.  =VHL =V  H ( L)dl
L
(2)
Sensors 2019, 19, 5110 3 of 15

The relationship between ϕ and H in the light propagation direction can be expressed as
Z
ϕ = VHL = V H (L)dl (2)
L

According to Malus’ law, the transmitted light intensity that passes through the analyzer can be
described as follows:
Jo = Ji cos2 α, (3)

where Ji is the intensity of the polarized light (µw), α is the angle between the linearly polarized light
and the analyzer (rad), and Jo is the output light intensity of the analyzer (µw).
Furthermore, the light intensity, which passes through the magneto-optical material and the
analyzer, can be described as follows when linearly polarized light is rotated by a magnetic field:

1
Jo = Ji cos2 (α − ϕ) = Ji (1 + cos 2(α − ϕ)). (4)
2
The sensitivity with which output light intensity can be used to estimate the Faraday rotation
angle is as follows:
dJo
= Ji sin 2(α − ϕ). (5)
dϕ
Because ϕ is very small, let α = ±45◦ to obtain larger responsivity to obtain the output light
intensities JO1 , JO2 . The output of photoelectric conversion is obtained by dividing the difference
between the electrical signals by the sum of them.

Jo1 − Jo2
uo = = sin 2ϕ (6)
Jo1 + Jo2

As seen in Equation (6), the influence of the light intensity fluctuation can be removed by
processing the light path information, as described earlier.
According to the Biot–Savart law, for a straight wire of infinite length, the relationship between
the magnetic field strength H at a certain point in the vicinity and the wire current I can be expressed as

B I
H= = , (7)
µ 2πr

where B is the magnetic induction (T), µ is the permeability (H/m), and r is the distance between the
point and the wire (m).
Substituting (7) into (2), Z
1
ϕ = VHL = VI dl = KI, (8)
2πr(l)
L

where K is the sensitivity of the OCS, which contains all the factors that affect the rotation from the
lightning current to the Faraday rotation angle. Therefore, by analyzing the sensitivity, we also analyze
the sensing characteristics of the OCS. It can be seen from Equation (4) that K is related to the optical
path length L; the magnetic field distribution, that is, the spatial position of the conduct and the OCS;
and V, that is, the sensing material. Therefore, K can be expressed as

K = f (a, V, L). (9)

When the above three influencing factors are determined, the magnitude of the measured lightning
current can be solved by the magnitude of the Faraday rotation angle ϕ. This paper will present the
OCS suitable for lightning current measurement in transmission systems from the above three factors.
Sensors 2019, 19, 5110 4 of 15

2.2. Environmental Requirements of Lightning Current Measurement

At present, the electronic transformers represented by OCS are mainly used in intelligent
substations, and the installation locations are mainly located in GIS (gas insulated switchgear). The
GIS equipment is completely filled with sulfur hexafluoride gas. All components in OCSs are in a fully
enclosed state such that the influence of environment factors can be isolated, such as external water
vapor and temperature. In the intelligent substation GIS, the current sensor is in an environment with
good insulation performance; its overall space is small; and the distance between the sensor and the
conductor is small, only about 0.2 m.
When the optical current sensing technology is applied in the lightning measurement of the
transmission system, the measurement environment has no advantage in the intelligent substation.
Therefore, it is necessary to study the following aspects: (1) Temperature. Because the lightning current
measurement environment of the transmission system is in the wild, it is necessary to analyze the effect
of temperature on magneto-optical materials. (2) Sensitivity. The wires in the transmission system are
exposed in the air, and are suspended by the insulator on the tower. Therefore, when measuring the
current on a three-phase transmission line, a certain safety distance needs to be met. In summary, in
order to achieve the measurement of the waveform characteristics of the lightning current, when the
OCS is applied to the environment of the transmission system, it is necessary to study and optimize
the above aspects of the sensor for the actual situation of the extreme measurement environment.

2.3. Lightning Current Measurement Form in the Transmission System

In different types of lightning strikes, the main path that the lightning current flows through also
differs in the transmission system. Therefore, lightning current measuring devices must be installed at
different positions in the transmission system to measure the current of different types of lightning
strikes. Table 1 presents a brief comparison of different measurement methods.

Table 1. Comparison of different measurement methods.

Measured Lightning Strike Type Type of Fault Caused Sensor Installation Position
Lightning strike at the top of a Tower top, tower body, tower foot,
Flashover caused by back striking
transmission tower centers of the intervals on the
Lightning strike on a ground wires, and ground wire
Flashover caused by shield failure
transmission line bracket [10–14]
Lightning strike on overhead Shielding failure or flashover
ground wires caused by back striking Transmission line [15,16]
Inductive lightning Invading wave of substation

Table 1 indicates that, from the perspective of the lightning current sensor requirements on a
transmission line, the sensor can be divided into two types: sensor type A represents lightning current
measurement in a tower and in ground wires. All the measured currents of this type are at the ground
potential and only need to be installed according to the use method of the measuring device at different
positions where the lightning current passes. The insulation requirement for sensor A is not high;
thus, it is installed close to the measured current using a magnetic steel bar or magnetic tape, and is
directly installed on the conductors to be measured by a Rogowski coil. Sensor type B represents the
lightning current measurement of phase conductors and must meet the insulation requirements. The
use of such sensors should not damage the safe operation of phase conductors. Therefore, the sensor
is required to realize noncontact lightning current measurement over a certain distance, such as a
differential loop [15] or space measurement coil [17]. According to different measurement requirements,
the installation position of the two types of sensors is shown in Figure 2.
Sensors 2019, 19, 5110 5 of 15
Sensors 2019,
Sensors 2019, 19,
19, xx FOR
FOR PEER
PEER REVIEW
REVIEW 55 of
of 15
15

Sensor A
Sensor A

Sensor B
Sensor B

Figure 2.
Figure 2. Sensor
Sensor installation
installation location
location diagram.
diagram.

According
According to to the above analysis,
the above when the
analysis, when the OCS
OCS isis applied
applied to to the
the lightning
lightning current
current measurement
measurement
of
of the
the transmission system, it
transmission system, it is
is necessary
necessary to to meet
meet the
the lightning
lightning current
current measurement sensitivity
measurement sensitivity
requirements
requirements of different
different lightning
lightning strike
strike types
types for
for the
the object to be measured.
requirements of different lightning strike types for the object to be measured. Compared with current Compared with current
transformers for substations, the lightning current measurement accuracy
transformers for substations, the lightning current measurement accuracy requirements are low and requirements are low and
the
the measuring environment is
measuring environment is complex,
complex, requiring
requiring high
high reliability
reliability of the sensor.
of the sensor. InIn order
order toto simplify
simplify
the sensing structure as much as possible, the closed-loop feedback is discarded
the sensing structure as much as possible, the closed-loop feedback is discarded to ensure sufficient to ensure sufficient
measurement accuracy,
measurement
measurement accuracy,and
accuracy, and the
and the sensing
the sensing performance
sensing performance is
performance is optimized
is optimized from
optimized from the
from the design
the design of
design of the
of the sensing
sensing unit.
unit.
Therefore, the straight-through
straight-through optical
optical current
current sensing
sensing structure
structure developed
developed
Therefore, the straight-through optical current sensing structure developed by our research group is bybyour
ourresearch
research group
group is
used
is to
used research
to research the
thesensing
sensing performance
performance of
ofthe
thesensors
sensors ininthis
this study
study [16].
[16].
used to research the sensing performance of the sensors in this study [16]. Figure 3 illustrates the Figure
Figure 3 illustrates the
structure
structure ofof optical
optical current
current sensing
sensing cellscells (OCSCs), which has
(OCSCs), which has aa simple
simple design, high reliability,
design, high reliability, uses
uses
dual-optical
dual-optical outputs
outputs to
to remove
remove thethe influence
influence of of light
light intensity
intensity fluctuations,
fluctuations, and
dual-optical outputs to remove the influence of light intensity fluctuations, and is suitable for the and
is is suitable
suitable for thefor the
harsh
harsh environment
environment of of
lightninglightning
current current measurement
measurement in in transmission
transmission
harsh environment of lightning current measurement in transmission lines. lines. lines.

Figure
Figure
Figure 3. A
3. A photograph
photograph
3. A photograph of
of the of the optical
the
opticaloptical current
currentcurrent sensing
sensingsensing cell (OCSC)
cell
cell (OCSC) (OCSC) structure
structurestructure of the
of thecurrent
of the optical opticalsensor
optical current
current
(OCS).
sensor (OCS).
sensor (OCS).
3. Magnetic Field Distribution of the Sensing Unit and Conductor Position
3. Magnetic
3. Magnetic Field
Field Distribution of of the
the Sensing
Sensing Unit andand Conductor
Conductor Position
Position
According to Distribution
the basic principle of opticalUnit
current sensing, the external magnetic field is a key
According
factorAccording to magneto-optical
causing the
to the basic
the basic principle
principle of optical
effect.
of optical current
If we set sensing,
the optical
current the
paththe
sensing, external
length
external magnetic
to L,magnetic
the relativefield is aa key
position
field is key
of
factor
the causing
wire
factor and the
causing thesensing
the magneto-optical effect.in
unit is as shown
magneto-optical effect. If Figure
If we set
we set 4,
the
the optical path
expressed
optical path length to). L,
= a(p, qto
as a length L, the
the relative
relative position
position
of the wire and the sensing unit is as shown in Figure 4, expressed as 
of the wire and the sensing unit is as shown in Figure 4, expressed as a  a( p, q) .
a a ( p , q ) .
Sensors
Sensors 2019, 19, x FOR PEER REVIEW 66 of 15
2019,19,
Sensors2019, 19,5110
x FOR PEER REVIEW 6of
of15
15

yy

pp
i(p,q)
i(p,q) BB

qq

x
A(0,0) (x,0)
A(0,0) (x,0) B(L,0)
B(L,0) x

Figure 4.
Figure Relativeposition
4. Relative positionof
ofthe
thewire
wire conductor
conductor to
to the
the sensor.
sensor.
Figure 4. Relative position of the wire conductor to the sensor.
In Figure 4, p is the distance of the wire from one end of the sensing unit and q is the vertical
In Figure 4, p is the distance of the wire from one end of the sensing unit and q is the vertical
In Figure
distance 4, p isconductor
of the wire the distance
fromofthe
thesensing
wire from unit.one Theend of the sensing
magnetic unit and at
field component q is thepoint
any vertical
on
distance of the wire conductor from the sensing unit. The magnetic field component at any point on
distance
the sensing of optical
the wirepathconductor from the
that coincides withsensing unit. The
the direction ofmagnetic
the optical field component
path at any point
can be expressed as on
the sensing optical path that coincides with the direction of the optical path can be expressed as
the sensing optical path that coincides with the direction of the optical path can be expressed as
I I q q
H (x) H = ( x) · I  2 q2 . 2 . . (10)
(10)
( x) (2p− (xp) +x)q22 q 2
H 2π (10)
2 ( p  x)  q
Corresponding
Corresponding todifferent
different effective magnetic
fields,fields, the obtained rotation angle can be
Correspondingtoto effective
different magnetic
effective magnetic the obtained
fields, rotation angle
the obtained can be
rotation expressed
angle can as be
expressed as
expressed as Z L
VI
ϕ =  VIVI ·  LL qq
2 00 ( (pp−x)x2 )22q+
q 2 dx2. dx. (11)
(11)
2π dqx . (11)
2 0 ( p  x)  q 2

It
It is
is assumed
assumed that
that the
the material
material isis selected
selected fromfrom densedense flint flint glass
glass model
model ZF-7,
ZF-7, which
which hashas been
been
It is assumed that the material is selected from dense flint glass model ZF-7, which has been
widely used in the
widely used in the OCS OCS in our group; its Verdet constant V = 17.8 rad/Tm, = 10 kA, and L = 10cm.
Verdet constant V = 17.8 rad/Tm, I = 10 kA, and L = 10 cm.
widely used in the OCS in our group; its Verdet constant V = 17.8 rad/Tm, I = 10 kA, and L = 10 cm.
When thethe temperature
temperature is constant
constant and
and the
the linear
linear birefringence
birefringence caused caused byby stress
stress isis not
not considered,
considered, the the
When the temperature is constant and the linear birefringence caused by stress is not considered, the
Faraday
Faraday rotation
rotation angles in different relative positions are as shown in Figure Figure 5.5.
Faraday rotation angles in different relative positions are as shown in Figure 5.

Figure 5. Faraday rotation angles in different relative positions.

Figure 5. Faraday rotation angles in different relative positions.
It can be seen that,Figure 5. Faraday
at a fixed rotation
distance, and angles in different
as p increases, therelative
rotationpositions.
angle tends to decrease, and
the Faraday
It can berotation angle
seen that, at is
a at a maximum
fixed distance, when
and as the
p wire is at the
increases, thecenter of the
rotation sensing
angle tendsunit. With the
to decrease,
It can be seen that, at a fixed distance, and as p increases, the rotation angle tends to decrease,
q, the influence of the change in p
and the Faraday rotation angle is at a maximum when the wire is at the center of the sensing0.4
increase in on the rotation angle weakens. After exceeding m,
unit.
and the Faraday rotation angle is at a maximum when the wire is at the center of the sensing unit.
it does
With thenot vary with
increase in q,p,the
and the response
influence of theof the sensor
change gradually
in p on decreases.
the rotation angle weakens. After exceeding
With the increase in q, the influence of the change in p on the rotation angle weakens. After exceeding
Sensor A has no insulation requirements and can achieve
0.4 m, it does not vary with p, and the response of the sensor gradually decreases. a large response by using a small
0.4 m, it does not vary with p, and the response of the sensor gradually decreases.
installation distance. Therefore, the conductor should be
Sensor A has no insulation requirements and can achieve a large response installed at the center of bythe using
sensor,a where
small
Sensor
L A has no insulation requirements and can achieve a large response by using a small
a = a( 2 , q), distance.
installation to get theTherefore,
maximumthe magnetic field.should
conductor SensorbeB installed
must maintain a certain
at the center of safe distance
the sensor, from
where
installation distance. Therefore, the conductor should be installed at the center of the sensor, where
the conductor to be measured. On the basis of the satisfaction of q, the value range of p is large, which
a  a( LL , q) , ,totoget the maximum magnetic field. Sensor B must maintain a certain safe distance from
a  aprovides
also ( 2 , q) a large get the maximum
elastic marginmagnetic field. Sensor B
for the long-distance must maintain
installation a certain
accuracy of thesafe distance
sensor, whichfrom
is
2
the conductor
applicable to
to the be measured. On the basis of the satisfaction of q, the value range of p is large, which
the conductor to actual measurement
be measured. On theof lightning
basis current in the
of the satisfaction of transmission system.
q, the value range of p is large, which
Sensors 2019, 19, 5110 7 of 15

4. Research on the Sensing Unit of OCS

4.1. Determination of Sensing Material

It can be seen from the above analysis that the sensitivity of the sensor required by different
measurement methods also differs. Equation (8) indicates that the sensitivity of the sensing unit can be
satisfied by selecting different sensing materials. The magneto-optical materials can be divided into
two types: diamagnetic and paramagnetic. The different responses when different materials are in
the same magnetic field are described by the Verdet constant, which can reflect the sensitivity of the
magneto-optical effect rotation angle to the external magnetic field, the value of which is related to the
electronic layer structure of atoms or ions in a magneto-optical medium.
The ions in the diamagnetic material exhibit an inert gas electron layer structure when subjected
to an applied magnetic field, and thus have no permanent electron orbital magnetic moment. Under
the action of the magnetic field, only Zeeman splitting of the ion level can occur, causing a change in
the electron orbital motion. The atom produces a small magnetic moment in the opposite direction,
rotating the plane of polarization of the light wave passing through the medium in the magnetic
field [18]; thus, its Verdet constant value is relatively small. Borrelli et al. derived the expression of the
Verdet constant of the diamagnetic material based on quantum theory [19]:

2
X
V = (4πNv2 ) [An /(v2 − v2n ) ], (12)

where N is number of carriers per unit volume (mol/L); An is a parameter related to the strength of the
transition; and v and vn are frequencies of the incident light and electronic transition, respectively (Hz).
In paramagnetic materials, because of the presence of ions that are prone to transition, such as
iron ions in the OCSahedral position in a Y3 Fe5 O12 (YIG) magneto-optical crystal, the rare earth ions
Pr3+ , Ce3+ , Tb3+ , Dy3+ , and so on, contained in the paramagnetic glass electrons in the medium, will
migrate when an external magnetic field is applied to the medium, indicating that linearly polarized
light will produce a larger optical rotation angle in the medium. The Verdet constant of paramagnetic
materials is larger than that of diamagnetic materials, and the sensitivity of the magneto-optical effect
is higher. On the basis of quantum theory, the expression of the paramagnetic Verdet constant is [20]

2
X
V = (K/T )(Nn2e f f /g) [Cn /(v2 − v2n ) ], (13)
n

where K = 4π2 µB v2 /3chk, ne f f = g[ J ( J + 1)]1/2 , N is the number of paramagnetic ions per unit volume
(mol/L), g is the Lande splitting coefficient, J is the total angular momentum quantum number (mol),
Cn is the transition probability, µB is the magnetic moment (A/m2 ), c is the speed of light (m/s), h is
Planck’s constant (J·s), k is Boltzmann’s constant (J/K), and T is the absolute temperature (K).
For OCS type A, because there is no installation limitation, a large sensitivity can be obtained
by a small installation distance. Therefore, diamagnetic magneto-optical glass with better stability is
preferred as the sensing material. Model ZF-7 magneto-optical glass is used in this study as a sensing
material for sensor A, for which the typical Verdet constant is 17.8 rad/Tm.
In the lightning current measurement of transmission lines, the OCS prioritizes the use of a
larger-sensitivity paramagnetic magneto-optical material. In this study, a Tb3 Ga5 O12 (TGG) crystal is
used as the sensing material. The crystal structure can achieve cation replacement in a larger ionic
radius. Compared with representative paramagnetic materials such as Tb3+ -doped magneto-optical
glass and a YIG crystal, the crystal has a higher Verdet constant, as the typical Verdet constant is
134 rad/Tm in 633 nm; has a low absorption coefficient; and is highly transparent in the visible and
near-infrared spectral range [21,22].
Sensors 2019, 19, 5110 8 of 15

4.2. Temperature Compensation Method for OCS Type B

Equations (12) and (13) indicate that the Verdet constant of the diamagnetic material is a
temperature-independent quantity, while the paramagnetic material is temperature-dependent. The
lightning current measurement of the transmission system is operated outdoors, of which the ambient
temperature will have a certain impact on the response of sensor B. In the case of high requirements
for measurement accuracy, it is necessary to compensate for the material response fluctuation caused
by temperature. The relationship between the Verdet constant and temperature can be obtained
through the Verdet constant model of paramagnetic materials, and the corresponding temperature
compensation value can be obtained according to the measured temperature.
When the wavelength of incident light is constant, the Verdet constant of the paramagnetic
material as a function of ambient temperature T can be expressed as [23]

B
V (T ) = A · (1 + ), (14)
T − Tc

where Tc is the Curie temperature of the material and A and B are constants related to the properties of
the material.
The response output signal of the OCS can be expressed as ϕ = K(T )I. The sensor was calibrated
to temperature T0 , and the ratio error caused by the change in temperature is

K(T )I − K(T0 )I0

ε= . (15)
K(T0 )I0

When T = T0 is considered, and ε = 0, the temperature-induced ratio error can be obtained as

" #
K (T )
∆ε = −1 . (16)
K(T0 )

The temperature fluctuation problem caused by the Verdet constant is equivalent to superimposing
a temperature-dependent interference signal on the Faraday rotation angle ∆ϕ(T ). The temperature
compensation method superimposes a compensation signal with the same value and negative relation
with a sign on the sensing signal in the signal processing unit. This process can be expressed as

ϕ(T ) = ϕ0 + ∆ϕ(T ) + ϕc . (17)

The error in response to temperature fluctuation is

∆K(T ) = K0 ∆ε. (18)

The temperature compensation signal can be obtained as

ϕc = −∆ϕ(T ) = −K0 I∆ε. (19)

5. Research on Sensing Unit Size of OCS

It can be seen from Equation (8) that, the longer the length of the sensing material, the larger
the Faraday rotation angle. It should be noted that the above formula was obtained under an ideal
environment. Linear birefringence occurs with changes in fields such as stress and temperature, so
linear birefringence must be considered when analyzing the size of magneto-optical materials.

ϕ = η(L)[α(L) + k(L)·KI ] (20)

L · tan β L·a
η(L) = e−L·α0 · [1 − 0.85 ] · [1 − 0.5 2 ] (21)
f · (NA) f · (NA)
Sensors 2019, 19, 5110 9 of 15

Equation (21) is the optical coupling efficiency of the optical current sensing unit, where α0 is
Sensors 2019, 19, x FOR PEER REVIEW 9 of 15
the percentage of luminous flux absorbed by the white light passing through every centimeter of the
magneto-optical material and
and initial
initialluminous flux,aaisisthe
luminousflux, thefiber radius, β isisthe
fiberradius, thetilt
tilt angle
angle of
of the fiber
magneto-optical material  the fiber
axis, f is the lens focal length, and NA is the numerical aperture of the optical fiber.
axis, f is the lens focal length, and NA is the numerical aperture of the optical fiber.
1 22 δ L
α(L( )L)= 1 sin ( 00 L ) sin 4θ
22 sin ( 22 ) sin 4
(22)
(22)

Equation
Equation (22)(22) is
is the
the zero
zero drift
drift angle,
angle, which
which is is independent
independent of of the
the Faraday
Faraday rotation
rotation angle.
angle. When
When
there is linear birefringence in the OCS, even if the current is zero, the Faraday
there is linear birefringence in the OCS, even if the current is zero, the Faraday angle will also angle will also exist,
exist,
where δδ00 is
where isthe
thebirefringence
birefringenceangle
angleperperunit
unitlength
length and and θ the
θ is is the incident
incident azimuth
azimuth of polarized
of polarized light.
light. By
By
appropriately setting the incident azimuth of polarized light, the influence of the zero drift on the
appropriately setting the incident azimuth of polarized light, the influence of the zero drift on the
Faraday
Faraday rotation
rotation angle
angle can
can be
be eliminated.
eliminated.
sin((δ00LL）
sin )
kk((LL) )= (23)
(23)
(
δ00LL))

Equation
Equation (23)
(23)isisthe birefringence
the birefringence factor, which
factor, is theisresult
which of theof
the result presence of birefringence
the presence directly
of birefringence
acting
directlyonacting
the Faraday angle. The
on the Faraday larger
angle. Thethe value,
larger thethe higher
value, thethe sensitivity
higher of the sensor.
the sensitivity of the sensor.
Simulate
Simulate the
the response
responseof ofsensors
sensorswithwithdifferent
differentlengths.
lengths.Let Lettilt
tiltangle
angleofofthe
thefiber axisβ β=0.3°.
fiberaxis = 0.3◦ . The
two types
typesofofOCSs
OCSsareare placed at positions
placed at positions 0.050.05
and 1andm away from the
1 m away conductor,
from respectively,
the conductor, through
respectively,
which
through 10which
kA of 10
current
kA offlows. The
current simulation
flows. curve of curve
The simulation the relationship between between
of the relationship the Faraday rotation
the Faraday
angle and
rotation the sensing
angle and theunit size unit
sensing is shown
size isinshown
Figurein 6.Figure 6.

Sensor A in 0.05m
Sensor B in 1m
Sensor B in 3m

Length of Magneto-optical Material (m)

Figure 6. Relationship between the Faraday rotation angle and sensing material size.

It
It can
can bebe seen
seen that,
that, considering
considering the the optical
optical coupling
coupling efficiency
efficiency and and linear
linear birefringence,
birefringence, the the
Faraday
Faraday rotation
rotation angle
angle increases
increases with
with the
the length
length of
of the sensing material
the sensing material and and reaches
reaches aa maximum
maximum after after
saturation. After that,
saturation. After that, the
the rotation
rotation angle
angle decreases
decreases with
with the
the length. Considering the
length. Considering the error
error caused
caused by by
linear birefringence and the utilization of magneto-optical materials, the material
linear birefringence and the utilization of magneto-optical materials, the material length is selected length is selected
from
from the
the optimal
optimal length,
length, that
that is,
is, the portion where
the portion where the rotation angle
the rotation angle increases
increases fastest
fastest with
with the
the length
length
before reaching its saturation. As the measuring distance increases, the optimal
before reaching its saturation. As the measuring distance increases, the optimal length also changes, length also changes,
but
but after
after the
the distance
distance increases
increases to to aa certain
certain extent,
extent, the
the response
response change
change has has aa weak
weak effect.
effect. This study
This study
selects
selects two
two types
types ofof magneto-optical
magneto-optical materials
materials with
with aa length
length ofof 55 cm.
cm.
It should be noted that the rotation angle of sensor
It should be noted that the rotation angle of sensor B, which uses B, which usesthethe magneto-optical
magneto-optical crystal,
crystal, is
is close
close to to
thethe amplitude
amplitude of of sensor
sensor A. A. However,
However, because
because of of
thethe
long long distance
distance fromfromthethe conductor
conductor to
to be
be measured,
measured, thethe length
length of of
thethe magneto-opticalmaterial
magneto-optical materialrequired
requiredfor forthe
therotation
rotation toto reach
reach saturation
saturation
increases.
increases. In order to emphasize this influencing factor, we set the measuring point 33 m
In order to emphasize this influencing factor, we set the measuring point m from
from the
the
conductor.
conductor. It It can
can be
be seen
seen from
from the
the rotation
rotation angle
angle curve
curve that,
that, when
when the the measuring
measuring distance
distance is is further
further
increased, the curve becomes flat. The effect of material length on the degree of saturation of the
rotation angle becomes weak. In addition, the effect of the length of the material caused by the change
in distance on the saturation of the rotation angle also becomes weak, and the range of optimal lengths
Sensors 2019, 19, 5110 10 of 15

increased,
Sensors thex FOR
2019, 19, curve becomes
PEER REVIEWflat. The effect of material length on the degree of saturation 10 ofofthe
15
rotation
Sensors angle
2019, 19, x becomes
FOR PEER weak.
REVIEWIn addition, the effect of the length of the material caused by the change
10 of 15
in distance
that can be on the saturation
selected of theTherefore,
is the same. rotation angle also becomes
the principle weak,
of the andselection
length the rangeofofmagneto-optical
optimal lengths
that
that can
can be
materialsbe selected
selected
has is the
is the
universal same. Therefore,
same. Therefore,
applicability the
the principle
principle
under different of the length
of the
measurement length selection
selection
distance of magneto-optical
of magneto-optical
requirements.
materials
materials has
has universal
universal applicability
applicability under
under different
different measurement distance requirements.
measurement distance requirements.
6. Experimental Verification
6.
6. Experimental
ExperimentalVerification
Verification
6.1. Sensitivity Experiment of Magneto-Optical Materials
6.1. Sensitivity
6.1. SensitivityExperiment
Experiment of of Magneto-Optical
Magneto-Optical Materials
Materials
In order to verify the sensitivity characteristics of the two types of magneto-optical materials,
ZF-7In In order
order
and TGG, to averify
to verify
Faradaythe magneto-optical
the sensitivity characteristics
sensitivity characteristics of the
of
effect platformthe twotwo
is types
types
used, of
as of magneto-optical
magneto-optical
shown materials,
in Figure 7.materials,
After the
ZF-7
ZF-7 and
and TGG,
TGG, aa Faraday
Faraday magneto-optical
magneto-optical effect
effect platform
platform is
is used,
used,
light source passes through the collimator and polarizer, it becomes polarized. The polarized as
as shown
shown in Figure 7. After
After the
lightthe
is
light source
light sourcepasses
transmitted passesthrough
through through thecollimator
the collimatorand
the magneto-optical andpolarizer,
materialpolarizer,
and isitit becomespolarized.
becomes
deflected polarized.
by The
The
the magnetic polarized
polarized light
light
field, which is
is
is transmitted
transmitted
generated through
bythrough
the thethe
current magneto-optical
magneto-optical
flowing throughmaterialmaterial
the andiswhich
and
solenoid, isdeflected
deflected
has aby by themagnetic
the
winding magnetic
number field,
field, which
which
of 630 andis is
a
resistance of 3.85 Ω. The uniformity of the magnetic field at the center of the solenoid is good. Thea
generated
generated by
by the
the current
current flowing
flowing through
through the
the solenoid,
solenoid, which
which has
has aa winding
winding number
number of
of 630
630 and
and a
resistance of
resistance
modulated ofpolarized
3.85 Ω.
3.85 Ω. The
The uniformity
uniformity
light is detected ofby
of thethe
the magnetic
magnetic field
field
polarizing at the
at
light the center of
center
analyzer of the
by the solenoid
solenoid
which The
is good.angle
the rotation The
modulated
modulated polarized
can be directly detected. light
lightWeis detected
is apply
detected by the polarizing
by thecurrent
a direct polarizing light
of 0–2 light analyzer
A to analyzer by which
by which
the solenoid the rotation
the rotation
and obtain angle can
angle
ten voltage
be directly
can detected.
be directly
response values. TheWe
detected. apply a direct
We apply
experimental current
a direct
results are of 0–2 of
current A to
presented 0–2theA
in solenoid and obtain
to the8. solenoid
Figure andten voltage
obtain tenresponse
voltage
values. The
response experimental
values. results are
The experimental presented
results in Figure in
are presented 8. Figure 8.

Light Source
Light Source Solenoid Coil
with Magneto-
Solenoid Coil
optical Material
with Magneto-
Current Source optical Material
Current Source

Polarized Light
Analyzer
Polarized Light
Analyzer

Figure 7. Schematic of experimental setup.

Figure 7.
Figure Schematic of
7. Schematic of experimental
experimental setup.
setup.

Figure 8. Response
Figure 8. Response of
of two
two materials
materials under
under different
different magnetic
magnetic fields.
fields. TGG,
TGG, Tb
Tb33Ga O12
Ga55O 12. .
Figure 8. Response of two materials under different magnetic fields. TGG, Tb3Ga5O12.
From the figure, the response of both materials increases with the magnetic field, and both
From the figure, the response of both materials increases with the magnetic field, and both
demonstrate
From theaa good linearity.
figure, The increase
the response in the
of both response
materials reflects the
increases sensitivity
with of the magneto-optical
thesensitivity
magnetic field, and both
demonstrate good linearity. The increase in the response reflects the of the magneto-
demonstrate a good
optical material. By linearity. The increase
function fitting, in the response
the relationship reflects
between the the sensitivity
response of the
of the twomagneto-
different
optical material. By function fitting, the relationship between
materials and the applied current in the experimental system is obtained: the response of the two different
materials and the applied current in the experimental system is obtained:
 ZF-7  0.199 I  0.04
 ZF-7  0.199 I  0.04
(24)
TGG  1.436 I  0.02 (24)
TGG  1.436 I  0.02
Sensors 2019, 19, 5110 11 of 15

material. By function fitting, the relationship between the response of the two different materials and
the applied current in the experimental system is obtained:

ϕZF−7 = 0.199I + 0.04

(
Sensors 2019, 19, x FOR PEER REVIEW (24)
11 of 15
Sensors 2019, 19, x FOR PEER REVIEW ϕTGG = 1.436I + 0.02 11 of 15

It can
It can be
be seen
seen that
that the
the slope
slope ofof the two
two sets of
of curves
curves can
can reflect
reflect the
the sensitivity
sensitivity of
of the
the system
system
can be seen that the slope of thethe
two setssets
of curves can reflect the sensitivity of the system output
output to
output to the
the excitation
excitation current
current when
when using
using different
different sensing
sensing materials.
materials. When
When using
using TGG,
TGG, thethe slope
slope
to the excitation current when using different sensing materials. When using TGG, the slope of the
of
of the function
the function is about
is about seven times that of ZF-7. The experimental results indicate that both materials
function is about sevenseven
timestimes that
that of of ZF-7.
ZF-7. The experimental
The experimental results
results indicate
indicate that that
bothboth materials
materials can
can
can reflect
reflect the magnitude
the magnitude of the magnetic
ofmagnetic
the magnetic field through the rotation angle. The sensitivity is higher
reflect the magnitude of the field field through
through the rotation
the rotation angle.angle. The sensitivity
The sensitivity is higher
is higher when
when using
when using the TGG TGG crystal,
crystal, which
which is is advantageous
advantageous when
when applied to to the
the lightning
lightning current
current
using the TGGthe crystal, which is advantageous when applied to theapplied
lightning current measurement of
measurement of
measurement of transmission
transmission lines
lines with
with the
the measurement
measurement distance
distance requirements.
requirements.
transmission lines with the measurement distance requirements.

6.2. Temperature
6.2. Temperature
6.2. Characteristics Experiment
Temperature Characteristics
Characteristics Experiment of
Experiment of Magneto-Optical
of Magneto-Optical Materials
Magneto-Optical Materials
Materials
The
The temperature experiment platform
temperature experiment platform isis set
set up
up as
as shown
shown in in Figure
Figure 9. 9. The
The magneto-optical
magneto-optical material
material
is placed
placedin
is placed in a temperature
inaatemperature controller,
temperaturecontroller, and
controller,and the
andthethesensor
sensor
sensor is calibrated
is is calibrated
calibrated to room
to room
to room temperature
temperature
temperature of 298 K.
of 298
of 298 We
K. We set
K. We
set
the
the experimental
set the experimental
experimental temperature
temperature
temperature range
range to
range 273–323
to 273–323
to 273–323 K, and
K, and
K, and with
with every
with every
every 10 K
10 10 increase in
K increase
K increase temperature,
in in temperature,
temperature, it is
it
it is
kept
is keptwarm
warm for 1
for h
1 htotoensure a
ensure uniform
a uniform temperature
temperature effect
effecton
on the
the material,
material,
kept warm for 1 h to ensure a uniform temperature effect on the material, and a magnetic field and
and aa magnetic
magnetic field
field is
is
generated
generated by
generated by
by aa power
power frequency
power frequency current
frequency current with
current with
with anan amplitude
an amplitude
amplitude of of 600
of 600 A.
600 A. The
A. The effect of
The effect of the
the temperature
temperature
temperature
measured at each temperature point on the ratio error is shown
measured at each temperature point on the ratio error is shown in Figure in Figure 10.
Figure 10.
10.

Conductor
OCSC Conductor
OCSC
Temperature
Temperature
Controller
Controller

Data Collection
Data Collection
Unit
Unit

Current
Current
Source
Source
Figure 9. Schematic of temperature experimental setup.
Schematic of
Figure 9. Schematic of temperature
temperature experimental setup.

Figure 10. Effect of temperature on measurement error of magneto-optical materials.

Figure 10. Effect of temperature on measurement error of magneto-optical materials.
Figure 10. Effect of temperature on measurement error of magneto-optical materials.
From the figure, the absolute value of the ratio difference of the ZF-7 magneto-optical glass
From
From the figure,
the figure, the
the absolute value
value of the
the ratio difference
difference of
of the
the ZF-7
ZF-7 magneto-optical
magneto-optical glass
glass in
in
in the temperature rangeabsolute
is maintained of withinratio
0.04. The measurement result is only affected by
the temperature
the temperature rangerange is is maintained
maintained within
within 0.04.
0.04. The
The measurement result result is
is only
only affected
affected by
by the
the
temperature-induced linear birefringence, where themeasurement
fluctuation is sufficiently small to meet
temperature-induced linear birefringence, where the fluctuation is sufficiently
temperature-induced linear birefringence, where the fluctuation is sufficiently small to meet the small to meet the
measurement requirement. The Verdet constant and linear birefringence of the TGG
measurement requirement. The Verdet constant and linear birefringence of the TGG magneto-optical magneto-optical
crystal are
crystal are affected
affected by
by temperature,
temperature, the the ratio
ratio difference
difference isis approximately
approximately linear
linear with
with temperature,
temperature,
and the fluctuation is relatively large. The temperature in thunderstorms during
and the fluctuation is relatively large. The temperature in thunderstorms during different seasons different seasons
varies from
varies from region
region to
to region,
region, butbut all
all are
are seasonally
seasonally distributed
distributed [24,25].
[24,25]. The
The temperature
temperature change
change in
in the
the
same region is relatively small in the short term, and the ratio error will decrease further after
 Sensors 2019, 19, 5110 12 of 15

measurement requirement. The Verdet constant and linear birefringence of the TGG magneto-optical
crystal are affected by temperature, the ratio difference is approximately linear with temperature, and
the fluctuation is relatively large. The temperature in thunderstorms during different seasons varies
Sensors 2019, 19, x FOR PEER REVIEW 12 of 15
from region
Sensors 2019, 19,to region,
x FOR PEERbut all are seasonally distributed [24,25]. The temperature change in the12same
REVIEW of 15
region is relatively small in the short term, and the ratio error will decrease further after calibration.
calibration of the LLS. If there is a further requirement for measurement accuracy, the temperature
The error isofmuch
calibration smaller
the LLS. thanisthe
If there averagerequirement
a further errors of thefor
LLS (20%), whichaccuracy,
measurement can achieve
the calibration
temperature of
compensation method proposed in this paper can be used to compensate the temperature error of
the LLS. If there
compensation is a further
method requirement
proposed for measurement
in this paper can be used accuracy, the temperature
to compensate compensation
the temperature error of
the Verdet constant of the TGG crystal. The ratio curve of each temperature point after compensation
method
the Verdetproposed
constant inof
this
thepaper
TGG can be used
crystal. The to compensate
ratio the temperature
curve of each temperatureerror ofafter
point the Verdet constant
compensation
is similar to the fluctuation range of the ZF-7 ratio curve, where the temperature stability significantly
ofsimilar
is the TGG crystal.
to the The ratio
fluctuation rangecurve ofZF-7
of the eachratio
temperature pointthe
curve, where after compensation
temperature is similar
stability to the
significantly
improves.
fluctuation range of the ZF-7 ratio curve, where the temperature stability significantly improves.
improves.
6.3. Impulse Current Measurement Experiment
6.3. Impulse
6.3. ImpulseCurrent
Current Measurement
Measurement Experiment
Experiment
We set up the impulse current measurement experimental platform (Figure 11). The lightning
We set
We set up
up the
the impulse
impulse current
current measurement experimental platform
measurement experimental platform (Figure
(Figure 11).
11). The
The lightning
lightning
current source generates an impulse current with a waveform of 8/20 μs. The measurement results
current source generates an impulse current with a waveform of 8/20 µs. The measurement
current source generates an impulse current with a waveform of 8/20 μs. The measurement results results
were calibrated using a Pearson current sensing head. The measurement locations of the two sensor
were calibrated
were calibrated using
using aa Pearson
Pearson current
current sensing
sensing head.
head. The
The measurement
measurement locations
locationsof
of the
the two
two sensor
sensor
types were set to 0.05 and 0.5 m, respectively. The measurement result of sensor A is taken as an
types were set to 0.05 and 0.5 m, respectively. The measurement result of sensor A
types were set to 0.05 and 0.5 m, respectively. The measurement result of sensor A is taken as is taken as an
an
example to illustrate the impulse current waveforms of the OCS, which is shown in Figure 12.
example to
example to illustrate
illustrate the
the impulse
impulse current
current waveforms
waveforms of
of the
the OCS,
OCS, which
which is
is shown
shown inin Figure
Figure 12.
12.

Impulse Current Generator

Impulse Current Generator
Pearson Coil
Pearson Coil

Oscilloscope OCSC
Oscilloscope OCSC

Signal Process
Signal Process

Figure 11. Impulse current measurement experimental platform.

Figure
Figure 11.
11. Impulse
Impulsecurrent
current measurement
measurement experimental
experimental platform.
platform.

Pearson Coil
Pearson Coil

OCS
OCS

Figure 12. Current waveform measured by OCS.

Figure 12. Current waveform measured by OCS.
Figure 12. Current waveform measured by OCS.
From the figure, the output waveform is consistent with the lightning current source waveform,
From the figure, the output waveform is consistent with the lightning current source waveform,
indicating thatfigure,
From the the sensor proposed
the output in thisispaper
waveform can accurately
consistent reflect key
with the lightning information
current such as the
source waveform,
indicating that the sensor proposed in this paper can accurately reflect key information such as the
amplitude,that
indicating steepness, and proposed
the sensor waveform inof this
the lightning
paper cancurrent flowing
accurately through
reflect the conductor.
key information suchIt should
as the
amplitude, steepness, and waveform of the lightning current flowing through the conductor. It
be noted that the shape and current path of the conductor may affect the measurement
amplitude, steepness, and waveform of the lightning current flowing through the conductor. results; it can be
It
should be noted that the shape and current path of the conductor may affect the measurement results;
calibrated according to different conductor conditions in the actual application. The lightning
should be noted that the shape and current path of the conductor may affect the measurement results; current
it can be calibrated according to different conductor conditions in the actual application. The
it can be calibrated according to different conductor conditions in the actual application. The
lightning current of different amplitudes was measured multiple times at the same position. The
lightning current of different amplitudes was measured multiple times at the same position. The
response of the two sensor types is presented in Table 2.
response of the two sensor types is presented in Table 2.
As can be seen from the above figure, the measurement waveform of OCS is consistent with that
As can be seen from the above figure, the measurement waveform of OCS is consistent with that
of the person coil, which can meet the requirements of type A for accurate full waveform
of the person coil, which can meet the requirements of type A for accurate full waveform
measurement, and can also meet those of type B measurement of key information such as amplitude
measurement, and can also meet those of type B measurement of key information such as amplitude
Sensors 2019, 19, 5110 13 of 15

of different amplitudes was measured multiple times at the same position. The response of the two
sensor types is presented in Table 2.

Table 2. Response of the optical current sensor (OCS).

Measuring Distance (m) I (kA) U (V) S (V·kA−1 )

1.75 1.11 0.634
2.13 1.358 0.638
0.05 (Sensor A)
4.26 2.726 0.64
5.37 3.43 0.639
1.73 0.817 0.472
2.52 1.207 0.479
0.5 (Sensor B)
3.85 1.828 0.475
4.96 2.35 0.473

As can be seen from the above figure, the measurement waveform of OCS is consistent with that
of the person coil, which can meet the requirements of type A for accurate full waveform measurement,
and can also meet those of type B measurement of key information such as amplitude and steepness.
The amplitude of the impulse current is changed under the same experimental condition, and the
responses of the two types of sensors are as shown in Table 2.
Table 2 indicates that, with the increase in the peak value of the impulse current, the outputs
of sensors have linearity and stability. The mean sensitivity values are 0.637 and 0.475, respectively,
although the measuring distances are quite different between the two sensor types. The response of the
sensor is comparable and consistent with the theoretical analysis. Referring to the existing waveform
peak test results, using the optical current sensing principle, that is, in the current measurement of
hundreds kA amplitudes, also has a good measurement effect [26]. Combined with the better linearity
of the impulse current measurement, it can be illustrated that optical current sensing technology is
applicable in high amplitude peak current measurement.

7. Conclusions
Accurate lightning current parameters form the basis of lightning protection designs of power
systems. On the basis of the analysis of characteristics of traditional lightning current measuring
devices in the transmission system, this study applies the optical current sensing technology to
lightning current measurement. Two types of lightning OCSs were studied, and the results were
verified by experiments.

1. On the basis of the analysis of the measurement of the existing lightning current measuring
device of the transmission system, the sensing characteristics of the OCS were analyzed from the
perspective of sensitivity. The ZF-7 and TGG crystals were used as the sensing materials to meet
the transmission measurement requirements of different types of lightning strikes.
2. It is verified by experiments that the sensitivity characteristics of the two magneto-optical
materials are suitable for the lightning current measurement in the transmission system. With
the temperature compensation method proposed in this paper, the two types of OCSs can
maintain a reasonable measurement accuracy under the lightning temperature environment of
the transmission system.
3. When the two types of sensors are at different measuring distances, and the impulse currents of
different amplitudes flow through the conductor to be measured, the measured amplitudes of the
two types of OCSs can maintain a good linearity, which illustrates the applicability to different
measurement distance requirements.

In this study, the sensing characteristics of different magneto-optical materials were utilized, which
provides a reference for the application of new sensing materials. In addition, the results presented in
Sensors 2019, 19, 5110 14 of 15

this paper provide research ideas for the design of lightning current waveform measuring devices that
meet different measurement requirements such as wind power and aviation. Subsequent research can
be carried out in the practical direction of two types of OCSs:

1. For sensor A, the distortion of the measured lightning current is small and can be used to estimate
the original parameters of the lightning current; however, the measuring current is affected by
factors such as the material and height of the tower. The distortion factors can be researched to
obtain the true lightning current parameters.
2. For sensor B, the measurement target is the lightning current transmitted from the strike position;
thus, the characteristics of the measuring current can be studied to realize the functions of
lightning strike point location and lightning strike fault recognition.

Author Contributions: Conceptualization, J.-m.G. and Y.S.; methodology, W.-b.Y.; formal analysis, Y.S. and
W.-b.Y.; investigation, Y.H. and F.-w.D.; writing—original draft preparation, J.-m.G.; writing—review and editing,
J.-m.G.; supervision, W.-b.Y., Y.H., and F.-w.D.; project administration, Y.S.; funding acquisition, Y.H. and F.-w.D.
Funding: This work was funded by the National Natural Science Foundation of China, grant number 51877047,
and the Science and Technology project of the State Grid Corporation of China, grant number 0711-17OTL135.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Lu, J.; Zhou, T.; Wu, C.; Li, B.; Tan, Y.; Zhu, Y. Fault statistics and analysis of 220 kV and above power
transmission line in province-level power grid. High Volt. Eng. 2016, 42, 200–207.
2. Li, R.; Cao, X.; Chen, K.; Yin, C. Effects of geographical and climatic conditions on lightning parameters.
High Volt. Eng. 2016, 52, 63–68.
3. Chen, J.-H.; Tong, X.-F. Distribution characteristics of lightning current magnitude measured by lightning
location system. High Volt. Eng. 2008, 34, 1893–1897.
4. Nag, A.; Mallick, S.; Rakov, V.A.; Howard, J.S.; Biagi, C.J.; Hill, J.D.; Uman, M.A.; Jordan, D.M.; Rambo, K.J.;
Jerauld, J.E.; et al. Evaluation of U.S. national lightning detection network performance characteristics using
rocket-triggered lightning data acquired in 2004–2009. J. Geophys. Res. Atmos. 2011, 116, 347–360. [CrossRef]
5. Ping, S. Statistical analysis on lightning current amplitude measurement on 220 kV Xin-Hang line. Electr.
Power 2000, 33, 72–75.
6. Wang, J.-F.; Qi, C.; Liang, X.-B.; Che, Y.-Y.; Fan, L.-L.; Wang, Y.L. Use of magnetic links to measure the
maximum rate-of-rise of lightning currents. IEEE Trans. Power Deliv. 2007, 22, 2445–2449.
7. Wang, J.F.; Lu, J.J.; Peng, Y.N.; Liu, Y.; Huang, J.J. Research on lightning parameter measurement and
measuring devices. Power Syst. Technol. 2009, 33, 162–165.
8. Yao, C.; Long, Y.; Wu, H.; Wang, Q.; Ma, Y.; Shen, Y.; Wang, L. Study of magnetic fields from different types of
lightning faults on a multi-tower system. IEEE Trans. Dielectr. Electr. Insul. 2014, 21, 1866–1874. [CrossRef]
9. Yang, B.; Fu, Y.-P.; Li, Y.-X.; Yao, M.-L.; Shi, L.-H. Optically powered lightning current measurement system
for high-tower observation. IEEE Access 2018, 6, 17022–17028. [CrossRef]
10. Gu, S.; Chen, S.; Feng, W.; Zhao, C.; Fang, Y.; Li, T. Development and application of lightning protection
technologies in power grids of China. High Volt. Eng. 2013, 39, 2329–2343.
11. Cheng, Y.; Li, C.; Zhang, F. Lightning current and flashover path measurement on high voltage transmission
lines. In Proceedings of the 2006 IEEE Conference on Electrical Insulation and Dielectric Phenomena, Kansas
City, MO, USA, 15–18 October 2006; pp. 425–428.
12. Du, L.; Chen, H.; Chen, S.; Luo, T.; Li, J. Identification method of shielding failure and back striking for
overhead transmission line. High Volt. Eng. 2014, 40, 2885–2893.
13. Hussein, A.M.; Janischewskyj, W.; Milewski, M.; Shostak, V.; Chisholm, W.; Chang, J.S. Current waveform
parameters of CN tower lightning return strokes. J. Electrost. 2004, 60, 149–162. [CrossRef]
14. Takami, J.; Okabe, S. Characteristics of direct lightning strokes to phase conductors of UHV transmission
lines. IEEE Trans. Power Deliv. 2007, 22, 537–546. [CrossRef]
15. Yao, C.; Long, Y.; Wu, H.; Mi, Y.; Wang, Q.; Ma, Y.; Shen, Y.; Yang, Z. A novel lightning current monitoring
system based on the differential-integral loop. IEEE Trans. Dielectr. Electr. Insul. 2013, 20, 1247–1255.
Sensors 2019, 19, 5110 15 of 15

16. Ge, J.-M.; Shen, Y.; Yu, W.-B.; Liu, R.-T.; Fan, W.; Yang, Y.-X. A lightning current measurement method based
on optical sensing technology. IEEE Sens. J. 2019, 19, 4250–4258. [CrossRef]
17. Wang, L.; Yan, S.; Li, X.; Mei, H.; Xing, L.; Qian, D. Development of lightning current sensor coil based on
lightning space magnetic field. High Volt. Eng. 2016, 42, 3392–3398.
18. Yang, Z.; Xu, S.Q.; Jiang, Z.H. Overview of research on magneto-optical glass applied to all-optical fiber
sensor. J. Chin. Rare Earth Soc. 2003, 21, 115–123.
19. Borrelli, N.F. Faraday rotation in glasses. J. Chem. Phys. 1964, 41, 3289–3294. [CrossRef]
20. Pye, L.D.; Cherukuri, S.C.; Mansfield, J.; Loretz, T. The Faraday effect in some non-crystalline fluorides. J.
Non-Cryst. Solids 1983, 56, 99–104. [CrossRef]
21. Zhuang, N.; Song, C.; Guo, L.; Wang, R.; Hu, X.; Zhao, B.; Lin, S.; Chen, J. Growth of terbium gallium garnet
(TGG) magneto-optic crystals by edge-defined film-fed growth method. J. Cryst. Growth 2013, 381, 27–32.
[CrossRef]
22. Yasuhara, R.; Tokita, S.; Kawanaka, J.; Kan, H.; Yagi, H.; Nozawa, H.; Yanagitani, T.; Fujimoto, Y.; Yoshida, H.
Development of cryogenic TGG ceramic based Faraday rotator for inertial fusion driver. J. Phys. Conf. Ser.
2008, 112, 032059. [CrossRef]
23. Slezák, O.; Yashura, R.; Lucianetti, A.; Mocek, T. Temperature-wavelength dependence of terbium gallium
garnet ceramics Verdet constant. Opt. Mater. Express 2016, 6, 3683–3691. [CrossRef]
24. Liou, Y.A.; Kar, S.K. Study of cloud-to-ground lightning and precipitation and their seasonal and geographical
characteristics over Taiwan. Atmos. Res. 2010, 95, 115–122. [CrossRef]
25. Collier, A.B.; Hughes, A.R.W.; Lichtenberger, J.; Steinbach, P. Seasonal and diurnal variation of lightning
activity over southern Africa and correlation with European whistler observations. Ann. Geophys. 2006, 24,
529–542. [CrossRef]
26. Blake, J.; Rose, A.H.; Glasow, C. An Optical Current Transducer having Linearity Better than 0.4% from 4 to
108,000 A. In Proceedings of the 2002 15th Optical Fiber Sensors Conference Technical Digest, Portland, OR,
USA, 6–10 May 2002.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).