energies

Article
Measurement Error Estimation for Capacitive Voltage
Transformer by Insulation Parameters
Bin Chen 1 , Lin Du 1, *, Kun Liu 2 , Xianshun Chen 2 , Fuzhou Zhang 2 and Feng Yang 1
1 State Key Laboratory of Power Transmission Equipment & System Security and New Technology,
Chongqing University, Chongqing 400044, China; chenbin@cqu.edu.cn (B.C.); yangfeng@cqu.edu.cn (F.Y.)
2 Sichuan Electric Power Corporation Metering Center of State Grid, Chengdu 610045, China;
wanglanzhu98@163.com (K.L.); fub95586@126.com (X.C.); zfzcqu@163.com (F.Z.)
* Correspondence: dulin@cqu.edu.cn; Tel.: +86-138-9606-1868

Academic Editor: K.T. Chau

Received: 1 February 2017; Accepted: 8 March 2017; Published: 13 March 2017

Abstract: Measurement errors of a capacitive voltage transformer (CVT) are relevant to its equivalent
parameters for which its capacitive divider contributes the most. In daily operation, dielectric aging,
moisture, dielectric breakdown, etc., it will exert mixing effects on a capacitive divider’s insulation
characteristics, leading to fluctuation in equivalent parameters which result in the measurement error.
This paper proposes an equivalent circuit model to represent a CVT which incorporates insulation
characteristics of a capacitive divider. After software simulation and laboratory experiments,
the relationship between measurement errors and insulation parameters is obtained. It indicates
that variation of insulation parameters in a CVT will cause a reasonable measurement error. From
field tests and calculation, equivalent capacitance mainly affects magnitude error, while dielectric
loss mainly affects phase error. As capacitance changes 0.2%, magnitude error can reach −0.2%.
As dielectric loss factor changes 0.2%, phase error can reach 50 . An increase of equivalent capacitance
and dielectric loss factor in the high-voltage capacitor will cause a positive real power measurement
error. An increase of equivalent capacitance and dielectric loss factor in the low-voltage capacitor
will cause a negative real power measurement error.

Keywords: capacitive voltage transformer (CVT); capacitive divider; insulation parameters;

magnitude error; phase error

1. Introduction
A capacitive voltage transformer (CVT) is widely used in high-voltage (HV) and extra high-voltage
(EHV) systems instead of a traditional voltage transformer (VT) [1,2]. A CVT’s structure can be mainly
divided into a capacitive divider and an electromagnetic unit. A capacitive divider is comprised of
an insulating cylinder and a series capacitors inside of it. An electromagnetic unit is comprised of a
compensation reactor, a step-down transformer and a damper. A capacitive divider can first lower the
system voltage as an input for an electromagnetic unit. By isolation and reduction of the step-down
transformer, the low voltage signals can be used for protection and measurement. Compared with a VT,
a CVT has the advantages of smaller size, lower cost, and refrain from ferroresonance [3–5]. However,
the structure of a CVT is more complicated than a traditional VT which has more measurement
accuracy influential parameters, therefore a CVT has more significant measurement error and a higher
probability of insulation faults.
Existing research involves CVT measurement errors concerning various factors. Reference [6]
uses PSCAD (power systems computer aided design) to simulate the measurement error of CVT under
different harmonic conditions. References [7,8] propose a CVT circuit model in harmonic conditions.
After simulations and experiments, a relationship between its equivalent parameters and harmonic

Energies 2017, 10, 357; doi:10.3390/en10030357 www.mdpi.com/journal/energies

Energies 2017, 10, 357 2 of 14

transfer characteristics is obtained. Reference [9] establishes the steady state CVT circuit model with
emphasis on the measurement error caused by system frequency and secondary burden change.
Reference [10] focuses on the measurement error caused by hysteresis characteristics of the core.
Energies 2017, 10, 357 2 of 14
It proposes an algorithm that estimates the voltage across the capacitor and the reactor by considering
the effect of the
transfer hysteresis characteristics
characteristics of the [9]
is obtained. Reference core and adds
establishes theitsteady
to thestate
measured secondary
CVT circuit model withvoltage to
obtain emphasis
the correct onvoltage. References
the measurement [11–13]
error causeduseby Maxwell to studyand
system frequency thesecondary
measurement burden error caused by
change.
Reference
space field. The [10] focuses onare
conclusions theuseful
measurement
for CVT error caused byoptimization.
parameters hysteresis characteristics of the core. It errors,
Beside measurement
plenty proposes
of studiesanare algorithm
focusedthat on estimates the voltage
CVT transient acrossReference
response. the capacitor and
[14] the ATP-EMTP
uses reactor by considering
(the alternative
the effect of the hysteresis characteristics of the core and adds it to the measured secondary voltage
transients program) to study CVT transient response under both system faults and ferroresonance
to obtain the correct voltage. References [11–13] use Maxwell to study the measurement error caused
conditions.
by spaceThefield.
resultTheshows CVT transient
conclusions are useful response is significantly
for CVT parameters relevant
optimization. withmeasurement
Beside CVT component
parameters
errors,and system
plenty conditions.
of studies are focusedReference [15] alsoresponse.
on CVT transient bases itsReference
study on[14] ATP-EMTP
uses ATP-EMTPsimulation
(the and
designsalternative
an effective electronic
transients program) to study CVT
ferroresonance circuit whichresponse
transient proves to effectively
under suppress
both system faultslowandvoltage
ferroresonance
transients and high conditions. The result shows
voltage ferroresonance CVT transient
oscillations. response [16]
Reference is significantly
proposesrelevant with
a new method to
CVT component parameters and system conditions. Reference [15] also bases
reduce the impact of the transient response which is based on a least squares technique. After the caseits study on ATP-EMTP
simulation and designs an effective electronic ferroresonance circuit which proves to effectively
study, it shows this method can minimize the impact of CVT transients.
suppress low voltage transients and high voltage ferroresonance oscillations. Reference [16] proposes
However, research on measurement errors of a capacitive voltage transformer caused by insulation
a new method to reduce the impact of the transient response which is based on a least squares
parameters variation
technique. After the arecase
rarely reported.
study, it shows The insulation
this method parameters
can minimize of a capacitive
the impact divider include
of CVT transients.
equivalent However,
capacitance, dielectric loss factor and surface leakage current. The
research on measurement errors of a capacitive voltage transformer caused by measurement error of
a CVT insulation
is directly relevant variation
parameters to a capacitive
are rarelydivider’s
reported. insulation
The insulationparameters.
parameters of Asa acapacitive
capacitive divider is
divider
exposedinclude equivalent capacitance,
to atmosphere, temperature, dielectric loss factor
moisture, and surface leakage
environmental current. The
contamination, andmeasurement
humidity it will
error of effects
exert mixing a CVT is ondirectly relevant to
its insulation a capacitive divider’s
characteristics, leading insulation parameters.
to equivalent As a capacitive
parameters fluctuation
divider is exposed to atmosphere, temperature, moisture, environmental contamination, and
and further resulting in the measurement error. In this paper, a comprehensive circuit model is
humidity it will exert mixing effects on its insulation characteristics, leading to equivalent parameters
proposed to characterize a CVT which involves both internal and external insulation characteristics.
fluctuation and further resulting in the measurement error. In this paper, a comprehensive circuit
After simulation
model is proposedand experiments,
to characterizethe relationship
a CVT between
which involves bothmeasurement errors insulation
internal and external and insulation
characteristics is obtained.
characteristics. After simulation and experiments, the relationship between measurement errors and
insulation characteristics is obtained.
2. Proposed CVT Circuit Model
2. Proposed CVT Circuit Model
The structure of a CVT is shown in Figure 1. CH is the high-voltage capacitor and CL is the
low-voltageThe structure of a CVT is shown in Figure 1. CH is the high-voltage capacitor and CL is the low-
capacitor. L is the compensation reactor which can be used to compensate the equivalent
voltage capacitor. L is the compensation reactor which can be used to compensate the equivalent
capacitance of a capacitive voltage divider. T is the step-down transformer and D is the damper.
capacitance of a capacitive voltage divider. T is the step-down transformer and D is the damper.
The equivalent model
The equivalent of aofCVT’s
model a CVT’scapacitive
capacitive divider
divider isisactually
actuallynotnot purely
purely capacitive
capacitive since
since it has it has
inner insulation dielectric loss and outer surface leakage current. Taking these aspects into
inner insulation dielectric loss and outer surface leakage current. Taking these aspects into account, account,
the proposed CVT CVT
the proposed circuit model
circuit is shown
model is shownininFigure
Figure 2.

Figure 1. Structure of capacitive voltage transformer (CVT).

Figure 1. Structure of capacitive voltage transformer (CVT).
Energies 2017, 10, 357 3 of 14
Energies 2017, 10, 357 3 of 14

Figure
Figure 2.
2. Proposed
Proposedcircuit
circuit model
model of
of aa CVT.
CVT.

It is named as first, second, and third capacitor from bottom to the top in the voltage divider
It is named as first, second, and third capacitor from bottom to the top in the voltage divider
respectively. C11, C12, C2, and C3 are the equivalent capacitance of each segment, where C11, C2, and C3
respectively. C11 , C12 , C2 , and C3 are the equivalent capacitance of each segment, where C11 , C2 ,
belong to the high-voltage capacitor, and C12 belongs to the low-voltage capacitor. The main factors
and C3 belong to the high-voltage capacitor, and C12 belongs to the low-voltage capacitor. The main
influencing equivalent capacitance are temperature and dielectric breakdown. Since a CVT capacitor
factors influencing equivalent capacitance are temperature and dielectric breakdown. Since a CVT
has negative temperature coefficient, equivalent capacitance will decrease when temperature
capacitor has negative temperature coefficient, equivalent capacitance will decrease when temperature
increases. Dielectric breakdown will cause a short circuit in series capacitors, which means equivalent
increases. Dielectric breakdown will cause a short circuit in series capacitors, which means equivalent
capacitance will increase. R11, R12, R21, and R31 are the insulation resistances of each segment. Increase
capacitance will increase. R11 , R12 , R21 , and R31 are the insulation resistances of each segment. Increase
of the dielectric loss factor represents a decrease in insulating resistance. The main factors influencing
of the dielectric loss factor represents a decrease in insulating resistance. The main factors influencing
insulation resistance are temperature and moisture in the insulation. When temperature increases,
insulation resistance are temperature and moisture in the insulation. When temperature increases, the
the dielectric loss of capacitor will first decrease and then increase. Therefore, the corresponding
dielectric loss of capacitor will first decrease and then increase. Therefore, the corresponding insulation
insulation resistance will first increase and then decrease. Moisture in the insulation will result in a
resistance will first increase and then decrease. Moisture in the insulation will result in a larger dielectric
larger dielectric loss, which means the insulation resistance will decrease. R13, R22, and R32 are the
loss, which means the insulation resistance will decrease. R13 , R22 , and R32 are the surface equivalent
surface equivalent resistances. Increase of leakage current represents decrease of surface resistance.
resistances. Increase of leakage current represents decrease of surface resistance. The main factors
The main factors influencing the surface resistance is environmental contamination and humidity.
influencing the surface resistance is environmental contamination and humidity. When contamination
When contamination and humidity in the atmosphere increases, the surface leakage current of a
and humidity in the atmosphere increases, the surface leakage current of a capacitive divider will
capacitive divider will increase, which represents the decrease in surface resistance. The
increase, which represents the decrease in surface resistance. The electromagnetic unit, Lk , is the sum
electromagnetic unit, Lk, is the sum of compensation reactor reactance and leakage reactance of the
of compensation reactor reactance and leakage reactance of the transformer, which can be used to
transformer, which can be used to resonate with equivalent capacitance2 of the voltage divider C0,
resonate2 with equivalent capacitance of the voltage divider C0 , Lk = 1/(ω C0 ). Rk is the sum of tuning
Lk = 1 / (ω C0 ) . Rk is the sum of tuning reactor resistance and short circuit resistance of the transformer;
reactor resistance and short circuit resistance of the transformer; Zm is the excitation impedance of
Zm is the excitation impedance of the transformer; Zd and ZL are the impedance of the damper and
the transformer; Zd and ZL are the impedance of the damper and load which are recalculated to the
load which are recalculated to the primary side of the step-down transformer.
primary side of the step-down transformer.
3.
3. Theoretical
Theoretical Analysis
Analysis of
of CVT
CVT Measurement
Measurement Error
Error
To simplifythe
To simplify theanalysis
analysismodel,
model, a high-voltage
a high-voltage capacitor
capacitor model
model and aand a low-voltage
low-voltage capacitor
capacitor model
model are applied. In Figure 3, CH and RH are the equivalent capacitance and resistance of the high-
are applied. In Figure 3, CH and RH are the equivalent capacitance and resistance of the high-voltage
voltage capacitor
capacitor and RCL are
and CL and andtheRL equivalent
are the equivalent capacitance
capacitance and resistance
and resistance of the low-voltage
of the low-voltage capacitor.
L
capacitor. The Thevenin model of the capacitive divider
The Thevenin model of the capacitive divider is in Figure 4.is in Figure 4.
Energies 2017, 10, 357 4 of 14
Energies 2017, 10, 357 4 of 14
Energies 2017, 10, 357 4 of 14

Energies 2017, 10, 357 4 of 14

Figure3.
Figure
Figure 3.3.Capacitive
Capacitivedivider
Capacitive dividerequivalent
divider equivalentmodel.
equivalent model.
model.

Figure 3. Capacitive divider equivalent model.

Figure4.
Figure
Figure 4.4.Thevenin
Theveninmodel
Thevenin modelof
model ofofthe
thecapacitive
the capacitivedivider.
capacitive divider.
divider.
Figure 4. Thevenin model of the capacitive divider.
Assumingsystem
Assuming systemvoltage
voltageisisUU s, the output voltage of the capacitive divider, U0, can be expressed
ss,, the output
output voltage
voltage of
of the
the capacitive
capacitive divider,
divider, U00,, can
Assuming
Assuming system voltage is U the
s, the output voltage of the capacitive divider, U0, can be be
can beexpressed
expressed
expressed
by
by Equation (1),
by Equation (1),
by Equation (1),
. . 1 1 11 1 1 .
U0 = U
U =1U s  1 1 /( / (1 1 1 ++ 11 ) ). (1)
U 0 s0=UU0 s=+
U1s jωC )..
/ ( / ( 1+ jωC ++ 1 1 + )jωC (1)
RL 1 1+ jω L C R1H 1 + jωC H 11 R+L jωC
+ CjωL C
L (1) (1)
R + jω R + +jωjω
CCH HH R ++ jωCLL L
L
L
L R LRL R RHH
H RRL L L
tan δ is the dielectric loss factor which can be used to characterize the real power loss of insulating
tan
material. δ tan δ
tanδ Substituting
is the
is dielectric
the tan δ =
is the dielectric loss
dielectric
loss factor
loss
1/(factor factor
ωCR)which which
which
into Equationcan
can
can be (1), be
be used
used
used to
U0tocanto characterize
characterize
be expressed
characterize thethe real
real
by
the power
power
equation
real power loss
(2),loss
of ofof
loss
insulatingmaterial.
material.Substituting
insulating Substituting tan
material. Substituting tan δ = 1 / (ω CR) ) into
tan δ = 1 / (ω CR ) intoEquation
into Equation(1),
Equation (1),UU
(1), U 0 can be expressed by equation
can be expressed by equation
. δ = 1 / (ω CR
0
insulating . 0 can be expressed by equation
(2), CH tan δH + jCH
(2),
(2), U0 = Us . (2)
CH tan δH + CL tan δL + j(CH + CL )
C tan δ + jC
CHHtanδ δ HH++jCjCHH
U 0 = U s
U 0==U
.
CH tan Cδ HH +tan
(2)
C0 and R0 can be calculated U U s H δ + Hj (C + C ) .
CL divider
tan L . (2)
from
0 the
s C capacitive
tan δ + C tan δL
+ j
CH tan δ H + CL tan δ L + j (CH + CL )
H H L L (C H
H +
equivalent
C L ) model, (2)
C0 and R0 can be calculated from the capacitive divider equivalent model,
2 model,
CC 0 and R0 can be calculated from the capacitive
(C tan
0 and R0 can be calculated from the capacitive
C0 = CH + CL + (HCH tan
divider
δH + C
divider equivalent
L tan δL ) model,
equivalent (3)
δ H + CL tan δ L )2
C0 = CH + CL + CH++ Ctan (3)
tanδ δC
(C(CHtan H +
H+ CCCLLtan L δ 2)2
δ )L
CC 0 = CH + CL + (3)
0 = CH + CL +
H H L L
C +CCL δ (3)
H +
CH tan δHδ ++CC
C tan CHLtan
tan
Lδ L
R0 = R0 = H H L L . (4)
2 2 . (4)
ω (CH ω(CδHHtan+
tan CδHtan δL Htan
C+LCtan
H tan +CδδCLL tan ωδ(Lω
))2 ++
L tan
CH(+CCHL )+ CL )2
C δ + δ
R0R0== ω(C tan δH + C Htan δL 2)2 + ω L
(C
.
+CC)L2)2. 5, which can be simplified
(4)
(4)
The circuit model
The circuit of the
model of the ω(CH Htan δH H+unit
electromagnetic
electromagnetic L tan δ
CunitL isisshown L + ω(C
L )shown inH+Figure
in
H
Figure
L 5, which can be simplified to to
FigureFigure
6, where Le and
6, where Re are
Le and separately
Re are separatelythetheequivalent
equivalent reactance
reactance andresistance
and resistance
of ofmZ
Zcanmd,, Z
,Z ZL. ZL . to
d , and
and
Thecircuit
The circuit model
model ofofthe
the electromagnetic
electromagnetic unitisisshown
unit shown ininFigure
Figure 5,5,which
whichcan besimplified
be simplified to
Figure 6, where L e and Re are separately the equivalent reactance and resistance of Zm, Zd, and ZL.
Figure 6, where Le and Re are separately the equivalent reactance and resistance of Zm, Zd, and ZL.

Figure
Figure 5. Circuit
5. Circuit model
model ofofthe
theelectromagnetic
electromagnetic unit.
unit.

Figure5.5.Circuit
Figure Circuitmodel
modelofofthe
theelectromagnetic
electromagneticunit.
unit.
Energies 2017, 10, 357 5 of 14
Energies 2017, 10, 357 5 of 14
Energies 2017, 10, 357 5 of 14

Figure 6. 6.Equivalent
Figure Equivalentmodel of the
model of theelectromagnetic
electromagnetic unit.
unit.
Figure 6. Equivalent model of the electromagnetic unit.

Combining Figures 4 and 6, the CVT equivalent model is shown in Figure 7.

Combining Figures 4 and
Combining Figures 4 and 6, 6,
thetheCVT
CVTequivalent
equivalent model
modelisisshown
showninin
Figure 7. 7.
Figure

U0
U0 U ''
U

Figure
Figure 7.
7. Equivalent
Equivalent model
model of
of the
the capacitor
capacitor voltage
voltage transformer.
transformer.
Figure 7. Equivalent model of the capacitor voltage transformer.
'
The
The output
output voltage
voltage of CVT U
of CVT which is recalculated to the primary side of the step-down
U ' which is recalculated to the primary side of the step-down
transformer
The can
outputcan
transformer be expressed
voltage of CVT
be expressed by Equation
by U 0 which(5),
Equation is recalculated to the primary side of the step-down
(5),
transformer can be expressed by UEquation
 = U ⋅ (5), R e + jω L e ..
'
R + jω L e (5)
U ' = U 00 ⋅ 1 / jω C 0 + R 0 + ejω L k + R + jω L + R (5)
1 / jω C 0 + R 0 + jω L k + R kk + jω L ee + R ee
. 0 . e e R + jωL
Since
Since the
the main
main focus
U =of
focus U 0this
of · paper
this paper is
is on
on the
the relationship
relationship between
between measurement
.
measurement error
error and
and (5)
1/jωC + R + jωL + R + jωLeas
+ constant.
Re
characteristics, R + jω L and R + jω L can be treated as constant. If R ++ jjω
R + jω L R + jω L R ω LLk is
insulation characteristics, k k and0 e
0 e can
k be k treated If k
insulation k k e e is k k
marked
marked
Since Zkk,, and
as Z
theasmain and R
focus + jjω
Ree +
of ω LLee paper
this
is
is marked
marked as Z
Zee,,relationship
is onasthe
Equation
Equation (5) can
can be
be simplified
(5) betweensimplified to
to Equation
measurementEquation (6),
error(6),
and insulation
characteristics, Rk + jωLk and Re' +jωLe can be treated Z
Zee as constant. If Rk + jωLk is marked as Zk , and
U ' =
U =UUss ⋅⋅(5) can be ((Z + ω δδL Equation
+ jωCL ) +1 ..(6),
Re + jωLe is marked as Ze , Equation Zk + Ze )(ωC
Z )(
simplified CLL tan
tanto L + jωCL ) + 1
(6)
Zkk +
Z +ZZee +
+ k ωe C tan δ + jωC
(6)
ω C H tan δH + jωCH
H
. 0 . Ze H H

Assuming the U =ratio

voltage U s ·of the step-down transformer . (6)
Assuming the voltage ratio ofZ the step-down e )(ωCL tan δL +is
( Zk + Ztransformer jωC
is N,
L )+
N, the
1 output
the output voltage
voltage can
can be
be
expressed k + Ze + ωC tan δ + jωC
expressed by
by Equation
Equation (7),
(7), H H H

 transformer
 '
U =
U =UU ' // N
Assuming the voltage ratio of the step-down N ..
(7)
is N, the output voltage can be expressed
(7)
by Equation
It can (7),be concluded from Equations (6)–(7) that measurement error of the CVT is relevant with
It can be concluded from Equations (6)–(7) . that
. 0 measurement error of the CVT is relevant with
tan δδ H ,, C
CHH,, tan
C
H tanδδL .. Increase
CLL,, tan L Increase of
of C
CHH and tanU
and tan δδ H=will
U /N.
will cause
cause aa positive
H positive magnitude
magnitude error
error while increase (7)
while increase
of CL, tanδ will cause a negative magnitude error. Increase of tan δ
of can tanδconcluded
CL, be L will cause a negative magnitude error. Increase of tan δ H will cause a negative phase will cause a negative phase
It L from Equations (6)–(7) that measurement Herror of the CVT is relevant with
error
error while
while increase
increase of
of tanδδLL of
tan
CH , tan δH , CL , tan δL . Increase
will
will causes aa positive
CHcauses
and tan positive
δH will
phase
phase error.
error.
cause
Increase of
of C
CHH and
and C
Increasemagnitude
a positive
will
will both
CLLerror both
while
cause
cause
increase
aa negative
negative phase
phase error.
error.
of CL , tan δL will cause a negative magnitude error. Increase of tan δH will cause a negative phase error
while4.increase of tanMeasurement
δL will causes a positive phase error. Increase of CH and CL will both cause a
4. Simulation
Simulation of of Measurement Error Error with
with Various
Various Insulation
Insulation Parameters
Parameters
negative phase error.
Since
Since the
the parameters
parameters of of aa capacitive
capacitive divider
divider change
change moremore significantly
significantly than
than those
those of of an
an
electromagnetic
4. Simulation unit in
of Measurement
electromagnetic daily operation,
Error with
unit in daily operation, the parameters of an
Various Insulation
the parameters electromagnetic
Parameters
of an electromagnetic unit and load are assumed
unit and load are assumed
unchanged
unchanged in in simulation.
simulation. Assuming
Assuming external
external leakage
leakage current
current is
is 0,
0, the
the relationship
relationship between
between measurement
measurement
Since
error theinsulation
parameters of a can
capacitive divider changeloss more significantly than thoseofof an
error and insulation parameters can be obtained as the dielectric loss factor and equivalent capacitance
and parameters be obtained as the dielectric factor and equivalent capacitance of
electromagnetic
each capacitorunit
each capacitor are in daily
are changed.
changed. operation,
Figure
Figure 88 shows
shows thethesimulation
the parameters
simulation model
modelofof
ofan
CVT
CVT electromagnetic
and
and Table showsunit
Table 11 shows the and load are
the equivalent
equivalent
parameters
parameters of the CVT, where parameters of the ferroresonance damping circuit and the load
assumed of
unchanged the CVT,
in where
simulation. parameters
Assuming of the ferroresonance
external leakage damping
current is circuit
0, the and the
relationship are
load between
are
recalculated
measurement to
recalculatederror the primary
to theand side of
insulation
primary the
side of the step-down
parameters transformer. And
can be obtained
step-down transformer. the voltage ratio
as the ratio
And the voltage N of
dielectric the step-down
N of theloss factor and
step-down
transformer
equivalent is
is 330.
capacitance
transformer 330. of each capacitor are changed. Figure 8 shows the simulation model of CVT
and Table 1 shows the equivalent parameters of the CVT, where parameters of the ferroresonance
damping circuit and the load are recalculated to the primary side of the step-down transformer. And
the voltage ratio N of the step-down transformer is 330.
Energies 2017, 10, 357 6 of 14
Energies 2017, 10, 357 6 of 14

Figure
Figure 8.
8. Simulation
Simulation model
model of
of CVT.
CVT.

Table 1. CVT
Table 1. CVT simulation
simulation parameters
parameters [6].
[6].

CVT Parts Parameters Values

CVT Parts Parameters Values
C11/nF 28.572
C11
C/nF12/nF
28.572
66.666
Capacitive Voltage Divider C12 /nF 66.666
Capacitive Voltage Divider C2/nF 20.000
C2 /nF 20.000
C3C/nF 3/nF 20.000
20.000
Lk/H
Lk /H
136.78
136.78
R
Rk /kΩk/kΩ 1.82
1.82
LmL/H m/H 23,045.64
23,045.64
RmR/kΩ
m/kΩ 16600
16600
Electromagnetic
Electromagnetic Unit
Unit Rf /kΩ 131.8
Rf/kΩ 131.8
Lf /H 639.576
Lf/H
rf /kΩ 639.576
6.0
Cfr/nF
f/kΩ 6.0
15.868
C
Ld /H f/nF 15.868
6908
Load
RdL d/H
/kΩ 6908
2892.258
Load
Rd/kΩ 2892.258
4.1. CVT Measurement Error with Different Dielectric Loss Factor
4.1. CVT Measurement Error with Different Dielectric Loss Factor
To avoid the measurement error caused by change of equivalent capacitance, the equivalent
To avoid
capacitance of the
the measurement erroriscaused
capacitive divider set as by change
constant of equivalent
according capacitance,
to Table 1. When the the individual
equivalent
capacitance of the capacitive divider is set as constant according to Table 1.
capacitor’s dielectric loss factor changes, the dielectric loss factor of other capacitors remainWhen the individual
0.05%.
capacitor’s dielectric loss factor changes, the dielectric loss factor of other capacitors
CVT phase and magnitude error with different dielectric loss factors are shown in Figures 9 and remain 0.05%.
10.
CVT Itphase and magnitude error with different dielectric loss factors are shown in Figures
can be concluded, that dielectric loss factor mainly affects the phase error. Measurement error 9 and 10.
It can be concluded, that dielectric loss factor mainly affects the phase error. Measurement
caused by dielectric loss factor change in the second and third capacitor is totally the same. Increase of error
caused byloss
dielectric dielectric
in the loss factor
second, change
third, andinupper
the second
half ofand
the third capacitor will
first capacitor is totally
causethe same. Increase
a negative phase
of dielectric loss in the second, third, and upper half of the first capacitor will cause a
error and positive magnitude error. Increase of dielectric loss in the lower half of the first capacitornegative phase
error and positive magnitude error. Increase of dielectric
will cause a positive phase error and negative magnitude error. loss in the lower half of the first capacitor
will cause a positive phase error and negative magnitude error.
Energies 2017, 10, 357 7 of 14
Energies 2017, 10, 357 7 of 14
Energies 2017, 10, 357 7 of 14

25
25
Uniformly change
Uniformly change
Third capacitor change
20 Third capacitor change
20 Second capacitor change
Second capacitor change
Upper half of first capacitor
Upper half of first capacitor
15 Lower half of first capacitor
15 Lower half of first capacitor
CVT(’)
errorofofCVT(’)
10
10
Phaseerror

5
5
Phase

0
0

-5
-5

-10
-10 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Dielectric loss factor(%)
Dielectric loss factor(%)
Figure 9. CVT
Figure 9. phase
CVT phase error
phase error with different dielectric loss factor.
error with
Figure CVT different dielectric loss factor.
-3
x 10-3
2 x 10
2

0
0

-2
-2
CVT(%)
errorofofCVT(%)

-4
-4

-6
-6
Magnitudeerror

-8
-8
Magnitude

-10
-10 Uniformly change
Uniformly change
Third capacitor change
-12 Third capacitor change
-12 Second capacitor change
Second capacitor change
Upper half of first capacitor
-14 Upper half of first capacitor
-14 Lower half of first capacitor
Lower half of first capacitor
-16
-16 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Dielectric loss factor(%)
Dielectric loss factor(%)
Figure 10. CVT magnitude error with different dielectric loss factor.
Figure
Figure 10.
10. CVT
CVT magnitude
magnitude error
error with
with different
different dielectric
dielectric loss
loss factor.
factor.
4.2. CVT
4.2. CVT Measurement
Measurement ErrorError with
with Different
Different Equivalent
Equivalent Capacitance
Capacitance
4.2. CVT Measurement Error with Different Equivalent Capacitance
To avoid
To avoid the
the measurement
measurement error error caused
caused byby change
change of
of the
the dielectric
dielectric loss
loss factor,
factor, the
the dielectric
dielectric loss
loss
factorToofavoid
each the measurement
capacitor is fixed error
to caused
0.05%. When by individual
change of the dielectricequivalent
capacitor’s loss factor,capacitance
the dielectric loss
varies,
factor of each capacitor is fixed to 0.05%. When individual capacitor’s equivalent capacitance varies,
factor of each capacitor
the equivalent
equivalent is fixed
capacitance to 0.05%.
of other
other When individual
capacitors capacitor’sCVT
remains unchanged.
unchanged. equivalent capacitance
measurement errorvaries,
with
the capacitance of capacitors remains CVT measurement error with
the equivalent
different capacitance
equivalent of
capacitance other
is capacitors
shown in remains
Figures 11 unchanged.
and 12. CVT measurement error with
different equivalent capacitance is shown in Figures 11 and 12.
different
It canequivalent
can capacitance
be concluded
concluded thatis shown in Figures
equivalent 11 and 12.
capacitance mainly affects
affects the
the magnitude
magnitude error. error.
It be that equivalent capacitance mainly
It can be
Measurement error concluded
error causedthat
caused by equivalent
by equivalent capacitance
equivalent capacitance mainly
capacitance change affects
change in the
in the magnitude
the second
second and error.
and third Measurement
third capacitor
capacitor is is
Measurement
error
exactlycaused
the by equivalent
same. Increase capacitance
of equivalent change in the second
capacitance in the and thirdthird,
second, capacitor
and is exactly
upper theofsame.
half first
exactly the same. Increase of equivalent capacitance in the second, third, and upper half of first
Increase
capacitorofwillequivalent
will cause aa capacitance
negative phase in the
phase second,
error and third, andmagnitude
positive upper half error.
of firstIncrease
capacitorofwill cause a
equivalent
capacitor cause negative error and positive magnitude error. Increase of equivalent
negative phase
capacitance in error
in the and
the lower positive
lower half
half of magnitude
of the
the first error.
first capacitor Increase
capacitor will of
will cause equivalent
cause aa negative capacitance
negative phase
phase error in the
error and lower half
and negative
negative
capacitance
of the first capacitor
magnitude error.
error. will cause a negative phase error and negative magnitude error.
magnitude
Energies 2017, 10, 357 8 of 14
Energies 2017, 10, 357 8 of 14
Energies 2017, 10, 357 8 of 14
0.05
0.05
0
0
-0.05
-0.05
-0.1

of CVT(’)
-0.1
-0.15

of CVT(’)
-0.15
-0.2

errorerror
-0.2
Phase
Phase -0.25
-0.25
-0.3 Uniformly change
-0.3 Third capacitor
Uniformly change change
-0.35
Second capacitor
Third capacitor change
change
-0.35 Upper half of first capacitor
-0.4 Second capacitor change
Lower
Upper half
half of
of first
first capacitor
capacitor
-0.4
-0.45 Lower half of first capacitor
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-0.45
0 0.1 Equivalent
0.2 capacitance
0.3 change
0.4 percentage(%)
0.5 0.6 0.7
Equivalent capacitance change percentage(%)
Figure 11. CVT phase error with different equivalent capacitance.
Figure 11. CVT phase error with different equivalent capacitance.
Figure 11. CVT phase error with different equivalent capacitance.
0.3
0.3
0.2
0.2
0.1
0.1
of CVT(%)

0
of CVT(%)

0
-0.1
-0.1
errorerror

-0.2
-0.2
Magnitude

-0.3
Magnitude

-0.3
-0.4 Uniformly change
-0.4 Third capacitor
Uniformly change change
-0.5
Second capacitor
Third capacitor change
change
-0.5 Upper half of first capacitor
-0.6 Second capacitor change
Lower
Upper half
half of
of first
first capacitor
capacitor
-0.6
-0.7 Lower half of first capacitor
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-0.7
0 0.1 Equivalent
0.2 capacitance
0.3 change
0.4 percentage(%)
0.5 0.6 0.7
Equivalent capacitance change percentage(%)
Figure 12. CVT magnitude error with different equivalent capacitance.
Figure
Figure 12.12.
CVTCVTmagnitude
magnitudeerror
error with
with different
differentequivalent
equivalentcapacitance.
capacitance.
4.3. CVT Real Power Measurement Error with Different Insulation Parameters
4.3. CVT Real Power Measurement Error with Different Insulation Parameters
4.3. CVTOnce Real Power
the phase Measurement Errorerror
and magnitude with under
Different Insulation
various Parameters
insulation parameters is obtained, the real
power Once the phase and
measurement errormagnitude
of the CVTerror under
can also be various insulation
calculated. Assuming parameters
Us and Uisare obtained, the real
the magnitude
Once measurement
power the phase and magnitude
error of the CVT error
can under
also be various insulation
calculated. Assuming U parameters
and U are is magnitude
the obtained, the
of the system voltage and the measuring voltage, the magnitude error can be expressed as
s
real of
power
( K nthe
U −U
measurement
system errorthe
voltage and of measuring
the CVT can also be
voltage, thecalculated. Assuming
magnitude error can beUsexpressed
and U are
s ) / U s × 100% . The phase error can be expressed as θ2 −θ1 , where θ1 is the phase difference
as the
magnitude
( K n U − of
U )the
/ U
s s and
system
× 100% .voltage
The and
phase the
error measuring
can be voltage,
expressed
s I, and θ is the phase difference between U andas the
θ −θmagnitude
2 I.1 , where θ error
is the can be
phase expressed
difference as
between U 2
1

( Kn U − Us )/U
between × 100%.and The
θ2 isphase error can be expressed Uas θ2 I.− θ1 , where θ1 is the phase difference
RealUpower
s sand I,measurementtheerror
phaseofdifference
the CVT canbetween and
be expressed by Equation (8),
Us and
between Real powerI, and θ2 is the phase
measurement error difference
of the CVT between U and I.by Equation (8),
can be expressed
Real power measurement error W of%the θ2 -Ube
KnUI coscan s I cosθ1
= CVT expressed
×100% . by Equation (8), (8)
Us Iθcos
KnUI cos 2 -Uθs1I cosθ1 ×100% . W% = (8)
U I
Kn U I cos θ2 − U
s cos θ1 s I cos θ
1
Substituting ε %=( K nUW% −Us= ) / U s × 100% and θ = θ 2 − θ1 ×into 100%.
Equation (8), the real power (8)
Us Iand
cos θθ1 = θ − θ into Equation (8), the real power
errorεof
Substituting
measurement %=( nU − U
theK CVT s ) /be
can × 100%
U s expressed by Equation 2 1(9),
measurement
Substitutingerror of (the
KnCVT can)/Ube expressed by Equation (9), θ into Equation (8), the real power
ε% = W−
U %U × 100%
= s[(1+ ε )s cos θ − (1+and tanθθ21 −
ε ) sinθθ = − 1]1× 100% . (9)
measurement error of the CVT W% = [(1+
can ε ) cos θ − (1+
be expressed byε )Equation
sin θ tan θ 1(9),
− 1] × 100% . (9)
It can be seen from Equation (9) that the power measurement error of the CVT is also relevant
with Itthe
can be seen
power fromofEquation
factor
W% the 1 + (9)
= [(system. that
If the
ε) cos θ the(power
−power measurement
1 + εfactor
) sin θ tan 1 1error
cosθ1θ is−set of the
to 100%.
]× 0.8 CVT is
lagging, thealso
realrelevant
power (9)
with the power factor of the system. If the power factor cos θ1 is set
measurement error under various insulation parameters can be calculated. CVT real power to 0.8 lagging, the real power
measurement
measurement error
It can be seenerror under
fromwith various
different
Equation insulation
(9)dielectric
that lossparameters
the power factor and can be error
calculated.
equivalent
measurement CVTare
capacitance
of the CVT isreal power
shown
also in
relevant
with the power factor of the system. If the power factor cos θ1 is set to 0.8 lagging, thein real
measurement
Figures 13 and error
14. with different dielectric loss factor and equivalent capacitance are shown
Figures 13 and 14.
power measurement error under various insulation parameters can be calculated. CVT real power
Energies 2017, 10, 357 9 of 14

measurement error with different dielectric loss factor and equivalent capacitance are shown in
Figures 13 and
Energies 2017, 14.
10, 357 9 of 14
Energies 2017, 10, 357 9 of 14
0.2
0.2

0.1

CVT(%)
0.1
CVT(%)
0
of of
0
error
error

-0.1
measurement

-0.1
measurement

-0.2
-0.2
power

-0.3 Uniformly change

power

-0.3 Uniformly change

Third capacitor change
Real

Third
Secondcapacitor
capacitorchange
change
Real

-0.4 Second capacitor change

-0.4 Upper half of first capacitor
Upper
Lower half
half of
of first
first capacitor
capacitor
Lower half of first capacitor
-0.5
-0.5 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 0.1 0.2 0.3
Dielectric 0.4 factor(%)
loss 0.5 0.6 0.7 0.8
Dielectric loss factor(%)
Figure 13. CVT real power measurement error with different dielectric loss factor.
Figure CVT
13. 13.
Figure real
CVT power
real powermeasurement errorwith
measurement error withdifferent
different dielectric
dielectric lossloss factor.
factor.
0.3
0.3
0.2
0.2
CVT(%)

0.1
CVT(%)

0.1
0
of of

0
error

-0.1
error

-0.1
measurement
measurement

-0.2
-0.2
-0.3
-0.3
power

-0.4 Uniformly change

power

-0.4 Uniformly change

Third capacitor change
-0.5
Real

Third
Secondcapacitor
capacitor change
change
-0.5
Real

Second capacitor
Upper half change
of first capacitor
-0.6 Upper
-0.6 Lower half
half of
of first
first capacitor
capacitor
Lower half of first capacitor
-0.7
-0.7 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
0 0.1 0.2 capacitance
Equivalent 0.3 0.4 percetage(%)
change 0.5 0.6 0.7
Equivalent capacitance change percetage(%)
Figure 14. CVT real power measurement error with different equivalent capacitance.
FigureFigure
14. 14.
CVT CVT
realreal powermeasurement
power measurement error
errorwith
withdifferent
differentequivalent
equivalent capacitance.
capacitance.
5. Experiment
5. Experiment
5. Experiment
Since the dielectric loss factor of a real CVT is hard to arbitrarily change, in order to better study
Since the dielectric loss factor of a real CVT is hard to arbitrarily change, in order to better study
the
Since relationship between
the dielectric CVT
lossCVT measurement
factor error and
of a real CVT insulation parameters with experiments,
in order to aabetter
the relationship between measurement error is
and hard to arbitrarily
insulation parameters change,
with experiments,
low-voltage CVT model is proposed. The structure of the CVT model is shown in Figure 15. Series
studylow-voltage
the relationship between
CVT model CVT measurement
is proposed. The structureerror
of theand
CVT insulation
model is parameters
shown in Figure with15.
experiments,
Series
connected capacitors are put inside an insulation cylinder to simulate the capacitive divider, and
a low-voltage
connected capacitors are put inside an insulation cylinder to simulate the capacitive divider,15.
CVT model is proposed. The structure of the CVT model is shown in Figure andSeries
series connected resistors are in parallel with the capacitors to represent the equivalent resistive
connected
currentcapacitors
series connected
by dielectric are putThe
resistors
loss. inside
are an insulation
in parallel with the cylinder
electromagnetic unit is equaledtobysimulate
capacitors to representthe the capacitive
the compensation reactordivider,
and the and
equivalent resistive
seriescurrent by dielectric
connected resistorsloss.are
Theinelectromagnetic
parallel with unit
the is equaled bytothe
capacitors compensation
represent the
step-down transformer both having equivalent parameters. In experiment, the dielectric loss factor
reactor and resistive
equivalent the
step-down
current transformer
by capacitive
dielectricdivider
loss. Theboth having equivalent parameters. In experiment, the dielectric loss factor
of the canelectromagnetic unit
easily be altered by is equaled
adjusting by the compensation
the resistors. Phase and magnitude reactor and the
error
of the capacitive
step-down transformer divider
bothcan easilyequivalent
having be altered by adjusting the
parameters. In resistors.
experiment,Phase and
the magnitude
dielectric losserror
factor of
of the CVT can be measured by a Lecory HDO6104 12-bit high precision oscilloscope by comparing
of the CVT can be measured by a Lecory HDO6104 12-bit high precision oscilloscope by comparing
the voltagedivider
the capacitive of the CVT
canmodel
easilyand a 3 kV high
be altered accuracy class
by adjusting the 0.1 reference
resistors. VT. and magnitude error of
Phase
the voltage of the CVT model and a 3 kV high accuracy class 0.1 reference VT.
the CVT can The real picture of low-voltage CVT model is shown in
be measured by a Lecory HDO6104 12-bit high precision oscilloscope Figure 16. CVT experiment wiring the
by comparing
The real picture of low-voltage CVT model is shown in Figure 16. CVT experiment wiring
diagram
voltage is shown
of theisCVT in
model Figure
and 17.17.
a 3 kV high accuracy class 0.1 reference VT.
diagram shown in Figure
The real picture of low-voltage CVT model is shown in Figure 16. CVT experiment wiring diagram
is shown in Figure 17.
Energies 2017, 10, 357 10 of 14
Energies 2017, 10, 357 10 of 14

Energies 2017,
Energies 2017, 10,
10, 357
357 10 of
10 of 14
14

Figure 15. Structure of low-voltage CVT model.

Figure 15. Structure of low-voltage CVT model.
Figure 15.
Figure 15. Structure
Structure of
of low-voltage
low-voltage CVT
CVT model.
model.

Figure 16. Capacitive divider of low-voltage CVT model.

Figure 16. Capacitive divider of low-voltage CVT model.

Figure 16.Capacitive
Capacitive divider
divider of
Figure 16. oflow-voltage
low-voltageCVT
CVTmodel.
model.

Figure 17. CVT experiment wiring diagram.

5.1. CVT Measurement Error with Various

Figure
Figure 17. High-Voltage
17. CVT
CVT experiment
experiment Capacitor Dielectric Loss Factor
wiring diagram.
wiring diagram.
Figure 17. CVT experiment wiring diagram.
In order to eliminate the measurement error caused by dielectric loss factor change in the low-
5.1. CVT
5.1. CVT Measurement
Measurement ErrorError with
with Various
Various High-Voltage
High-Voltage Capacitor
Capacitor Dielectric
Dielectric Loss
Loss Factor
Factor
voltage
5.1. CVT capacitor, Error
Measurement the low-voltage
with Various capacitor is fixed
High-Voltage to 100 nF
Capacitor with noLoss
Dielectric parallel
Factorresistance. The
In order
dielectric
In order to
lossto eliminate
factor the measurement
measurement
of low-voltage
eliminate the error
capacitorerror caused
is 0.012%
caused by dielectric
which
by dielectric loss factor
can be measured
loss factorwith
change in the
the low-
a dielectric
change in low-
loss
In order
voltage to eliminate
capacitor, the the measurement
low-voltage capacitor error
is caused
fixed to 100bynF dielectric
with no loss factor
parallel
voltage capacitor, the low-voltage capacitor is fixed to 100 nF with no parallel resistance. The change
resistance. Thein the
dielectric
low-voltage loss factor
dielectric capacitor, of
loss factor ofthe low-voltage capacitor
low-voltage
low-voltage is
capacitor
capacitor 0.012%
is 0.012% which
is which can
fixed can be
to 100 measured
nF withwith
be measured with a dielectric
no aparallel loss
dielectricresistance.
loss
The dielectric loss factor of low-voltage capacitor is 0.012% which can be measured with a dielectric
loss meter. The high-voltage capacitor is fixed to 25 nF. The dielectric loss factor of the high-voltage
capacitor with different parallel resistance is shown in Table 2.
Energies 2017, 10, 357 11 of 14
Energies 2017, 10, 357 11 of 14

meter. The high-voltage capacitor is fixed to 25 nF. The dielectric loss factor of the high-voltage
meter. 2017,
Energies The10,
high-voltage capacitor is fixed to 25 nF. The dielectric
loss factor of the high-voltage
capacitor with357
different parallel resistance is shown in Table 2. 11 of 14
capacitor with different parallel resistance is shown in Table 2.
Table 2. Dielectric loss factor of high-voltage capacitor with different resistance.
2. Dielectric
Table 2.
Table Dielectric loss
loss factor
factor of
of high-voltage
high-voltage capacitor
capacitor with
with different
different resistance.
resistance.
Capacitance CH/nF Resistance RH/MΩ Dielectric Loss Factor (%)
Capacitance CH/nF Resistance RH/MΩ Dielectric Loss
LossFactor (%)
25 CH /nF
Capacitance 150 RH /MΩ
Resistance Dielectric0.095 Factor (%)
25
25 150
150 0.095
0.095
25 120 0.115
25
25 120
120 0.115
0.115
25 90 0.151
25
25 90
90 0.151
0.151
25
25 6060 0.223
0.223
25 60 0.223
25
25 3030 0.434
0.434
25 30 0.434
25
25 2020 0.648
0.648
25
25 20
15 0.648
0.859
25 15 0.859
25 15 0.859
rated voltage
The testing voltage is 0.8, 1, and 1.2 of rated voltage (3 kV).
kV). Measurement
Measurement error under various
The testing voltage is 0.8, 1, and 1.2 of rated voltage (3 kV). Measurement error under various
dielectric loss of the high-voltage
high-voltage capacitor
capacitor is
is shown
shown in
in Figures
Figures 18
18 and
and 19.
19.
dielectric loss of the high-voltage capacitor is shown in Figures 18 and 19.
0
0

-10
-10

-20
of CVT(’)

-20
of CVT(’)

-30
error

-30
error
Phase
Phase

-40
-40
0.8Un
-50 0.8Un
Un
-50 Un
1.2Un
1.2Un
-60
-600 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
0 0.1 0.2 0.3loss of0.4
Dielectric 0.5 capacitor(%)
high-voltage 0.6 0.7 0.8 0.9
Dielectric loss of high-voltage capacitor(%)
Figure
Figure 18. CVT
CVT phase
phase error
error with
with different high-voltage
high-voltage capacitor dielectric
dielectric loss factor.
factor.
Figure 18.
18. CVT phase error with different
different high-voltage capacitor
capacitor dielectric loss
loss factor.
0.9
0.9
0.8Un
0.8 0.8Un
Un
0.8
Un
1.2Un
0.7 1.2Un
0.7
of CVT(%)

0.6
of CVT(%)

0.6
0.5
error

0.5
error

0.4
Magnitude

0.4
Magnitude

0.3
0.3
0.2
0.2
0.1
0.1
0
00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
0 0.1 0.2 0.3loss of0.4
Dielectric 0.5 capacitor(%)
high-voltage 0.6 0.7 0.8 0.9
Dielectric loss of high-voltage capacitor(%)
Figure 19. CVT magnitude error with different high-voltage capacitor dielectric loss factor.
Figure 19. CVT
Figure 19. CVT magnitude
magnitude error with different
error with different high-voltage
high-voltage capacitor
capacitor dielectric
dielectric loss
loss factor.
factor.

Plots in Figures 18 and 19 coordinate with regulations obtained from simulation. When the
Plots in Figures 18 and 19 coordinate with regulations obtained from simulation. When the
dielectric
Plotsloss factor of18high-voltage
in Figures capacitor
and 19 coordinate increases,
with the phase
regulations errorfrom
obtained is increasing in theWhen
simulation. negative
the
dielectric loss factor of high-voltage capacitor increases, the phase error is increasing in the negative
direction while
dielectric magnitude
loss factor error is increasing
of high-voltage capacitor in the positive
increases, direction.
the phase error is increasing in the negative
direction while magnitude error is increasing in the positive direction.
direction while magnitude error is increasing in the positive direction.
Energies 2017, 10, 357 12 of 14

5.2. CVT Measurement

Energies 2017, 10, 357 Error with Various Low-Voltage Capacitor Dielectric Loss Factor 12 of 14
Energies 2017, 10, 357 12 of 14
In order to eliminate the measurement error caused by dielectric loss factor change in the
5.2. CVT Measurement Error with Various Low-Voltage Capacitor Dielectric Loss Factor
high-voltage capacitor, the
5.2. CVT Measurement Errorhigh-voltage capacitorCapacitor
with Various Low-Voltage is fixedDielectric
to 25 nF,Losswith
Factorno parallel resistance.
In
The dielectricorder to eliminate
losstofactor the
of the measurement error caused by dielectric loss factor change inbythe high-
In order
voltage eliminate
capacitor, thehigh-voltage
measurement
the high-voltage
capacitor
capacitorerror is 0.012%
is caused
fixed toby25
whichloss
dielectric
nF, with
canfactor
be measured
change
no parallel in theahigh-
resistance.
dielectric
The
loss meter. The low-voltage
voltage capacitor, capacitor is fixed to 100 nF. The dielectric loss factor of the low-voltage
dielectric loss factorthe
of high-voltage capacitor
the high-voltage is fixed
capacitor to 25which
is 0.012% nF, with
can beno measured
parallel resistance. The
by a dielectric
capacitor withloss
dielectric different parallel
factor of resistancescapacitor
the high-voltage is shown is in Tablewhich
0.012% 3. can be measured by a dielectric
loss meter. The low-voltage capacitor is fixed to 100 nF. The dielectric loss factor of the low-voltage
loss meter.with
capacitor Thedifferent
low-voltage capacitor
parallel is fixed
resistances to 100 nF.
is shown The dielectric
in Table 3. loss factor of the low-voltage
Table
capacitor with 3. Dielectric
different lossresistances
parallel factor of low-voltage
is shown incapacitor
Table 3. with different resistance.
Table 3. Dielectric loss factor of low-voltage capacitor with different resistance.
Table 3. Dielectric
Capacitance CL /nF loss factor of low-voltage
Resistance capacitor with
RL /MΩ differentLoss
Dielectric resistance.
Factor (%)
Capacitance CL/nF Resistance RL/MΩ Dielectric Loss Factor (%)
100
Capacitance CL/nF Resistance 60 RL/MΩ Dielectric Loss0.064
Factor (%)
100 60 0.064
100100 30
60 0.0640.118
100100
100
30
20
30
0.118
0.1180.172
100 100 20
15 0.1720.224
100 20 0.172
100100 15
12 0.2240.276
100 15 0.224
100100 12
10 0.2760.330
100
100 12
10 0.276
0.330
100 10 0.330
PlotsPlots
in Figures 20 and 21 coordinate with regulations obtained from simulation. When the
in Figures 20 and 21 coordinate with regulations obtained from simulation. When the
dielectric Plots
dielectric in
loss loss
factorFigures
factor of20
of the theand 21 coordinate
low-voltage
low-voltage with
capacitor
capacitor regulations
increases,
increases, obtained
thephase
the phase from
error
error is simulation.
increasing
is increasing When the
inpositive
in the the positive
dielectric
direction while
direction loss factor
magnitude
while of the
magnitude low-voltage
error is is
error capacitor
increasing
increasingin increases,
in the the phase
negativedirection.
the negative error
direction. is increasing
However,
However, in
since
since the positive
the insulation
the insulation
direction
parameters inwhile
parameters the magnitude
simulations
in the error
simulationsand
andisexperiments
increasing
experimentsin are
the different,
are negative
different,direction. However,
thespecific
the specific value
value insince the insulation
in experiments
experiments is notis not
parameters
the same in the simulations and experiments are different, the specific value in experiments is not
the same as inasthe
in the simulations.
simulations.
the same as in the simulations.
22
22
0.8Un
20
0.8Un
Un
20
Un
1.2Un
18
1.2Un
18
16
of CVT(’)

16
of CVT(’)

14
14
error

12
error
Phase

12
Phase

10
10
8
8
6
6
4
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
4
0.05 0.1 Dielectric
0.15 loss0.2 of low-voltage
0.25 capacitor(%)
0.3 0.35 0.4
Dielectric loss of low-voltage capacitor(%)
Figure 20. CVT phase error with different low-voltage capacitor dielectric loss factor.
Figure CVT
20. 20.
Figure phase
CVT error
phase errorwith
withdifferent
different low-voltage capacitor
low-voltage capacitor dielectric
dielectric lossloss factor.
factor.
0.1
0.1
0.8Un
0.05
Un
0.8Un
0.05
1.2Un
Un
0 1.2Un
of CVT(%)

0
of CVT(%)

-0.05
-0.05
error
error

-0.1
Magnitude

-0.1
Magnitude

-0.15
-0.15

-0.2
-0.2

-0.25
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
-0.25
0.05 0.1 Dielectric
0.15 loss0.2 of low-voltage
0.25 capacitor(%)
0.3 0.35 0.4
Dielectric loss of low-voltage capacitor(%)
Figure 21. CVT magnitude error with different low-voltage capacitor dielectric loss factor.
Figure 21. CVT magnitude error with different low-voltage capacitor dielectric loss factor.
Figure 21. CVT magnitude error with different low-voltage capacitor dielectric loss factor.
Energies 2017, 10, 357 13 of 14

6. Conclusions
This paper proposes an equivalent circuit model to represent a CVT which incorporates the
insulation characteristics of a capacitive divider. This simulation model can reflect the measurement
error change caused by insulation parameters fluctuation. In order to verify the simulation results,
a low-voltage CVT experimental model is proposed. After software simulations and laboratory
experiments, the relationship between measurement error and insulation parameters is obtained.
Conclusions are as following. These conclusions are all straightforward in judging how the
measurement error of the CVT will change according to the variation of insulation parameters and
whether a certain CVT has the possibility to exceed the measurement error limit.

(1) Insulation parameters fluctuation of the CVT’s capacitive divider can contribute a fairly large
measurement error. Equivalent capacitance fluctuation mainly affects magnitude error, while
dielectric loss factor mainly affects phase error. When equivalent capacitance changes 0.2%,
magnitude error can reach −0.2%. When dielectric loss factor changes 0.2%, phase error can
reach 50 .
(2) An increase of equivalent capacitance in the high-voltage capacitor will cause a positive
magnitude error while that in the low-voltage capacitor will cause a negative magnitude error.
An increase of dielectric loss factor in the high-voltage capacitor will cause a negative phase error
while that in the low-voltage capacitor will cause a positive phase error.
(3) An increase of equivalent capacitance and dielectric loss factor in the high-voltage capacitor will
cause positive real power measurement error. An increase of equivalent capacitance and dielectric
loss factor in the low-voltage capacitor will cause negative real power measurement error.

Acknowledgments: The authors acknowledge the Funds for Innovative Research Groups of China (51321063).
Author Contributions: Lin Du, Bin Chen and Feng Yang developed the theory and performed the simulation.
Kun Liu, Xianshun Chen and Fuzhou Zhang carried out the experiments. Bin Chen and Lin Du analyzed the data
and wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. He, B.T.; Li, Y.Q.; Bo, Z.Q. An adaptive distance relay based on transient error estimation of CVT. IEEE Trans.
Power Deliv. 2006, 21, 1856–1861. [CrossRef]
2. Firous, B.A.; Majid, S.P.; Afshin, R.Z.; Reza, I. Analysis and suppression of the coupling capacitor voltage
transformer ferroresonance phenomenon. IEEE Trans. Power Deliv. 2009, 24, 1968–1977.
3. Yu, H.; Yang, G.; Shang, R.Y.; Chen, Y.T.; Lu, Z.M. Fault simulation and diagnostic test for abnormal voltage
of secondary coil of capacitor voltage transformer. In Proceedings of the 5th International Conference
on Electric Utility Deregulation and Restructuring and Power Technologies (DRPT), Changsha, China,
26–29 November 2015; pp. 1617–1623.
4. Wellings, J.G.; Mortlock, J.R.; Mathews, P. Capacitor voltage transformers. Electr. Eng. 1936, 79, 577–584.
[CrossRef]
5. Chakrabarti, R.; Basu, S.K. Steady state theory and design of capacitor voltage transformer. IET J. Mag. 1972,
10, 153–160. [CrossRef]
6. Yuan, Z.Y.; Zhao, W. Research on the capacitor voltage transformer measurement error under
harmonic environment. In Proceedings of the Chinese Automation Congress (CAC), Changsha, China,
7–8 November 2013; pp. 104–109.
7. Duan, X.B.; Zhu, M.X.; Hu, W.P.; Zhou, W. Key parameters affecting harmonic transfer characteristics of
capacitor voltage transformer. Power Syst. Technol. 2014, 38, 3153–3159.
8. Feng, Y.; Wang, X.Q.; Chen, X.M.; Wu, S.P.; Mao, A.L. Influence of circuit parameters of capacitor voltage
transformer on grid harmonic voltage measurement. Proc. CSEE 2014, 34, 4968–4975.
Energies 2017, 10, 357 14 of 14

9. Sun, J.T.; Wen, X.S.; Lan, L.; Li, X. Steady-state error analysis and digital correction for capacitor voltage
transformers. In Proceedings of the International Conference on Electrical Machines and Systems, Wuhan,
China, 17–20 October 2008; pp. 769–773.
10. Kang, Y.C.; Zheng, T.Y.; Choi, S.W.; Kin, Y.H.; Kim, Y.G.; Jang, S.L.; Kang, S.H. Design and evaluation of a
compensating algorithm for the secondary voltage of a coupling capacitor voltage transformer in the time
domain. IET Gener. Transm. Distrib. 2009, 3, 793–800. [CrossRef]
11. Li, Z. The Optimization of Structure Parameters of Electronic Voltage Transformer. Master’s Thesis,
Harbin Institute of Technology, Harbin, China, 30 June 2008.
12. Li, X.; Wang, X.Q.; Wu, S.P.; Yu, C.Y. Study of additional error for capacitor voltage transformer caused
by high voltage wire. In Proceedings of the Chinese Society of Electrical Engineering, Beijing, China,
21–24 November 2012; pp. 27–35.
13. Feng, S.Z.; Chen, X.S.; Luo, R.X.; He, N.; Zhou, K. Simulation on influence of external electric field on error
characteristics of CVT. In Proceedings of the 2016 International Conference on High Voltage Engineering
and Application, Chengdu, China, 19–22 September 2016.
14. Bakar, A.H.A.; Lim, C.H.; Mekhilef, S. Investigation of transient performance of capacitor voltage
transformer. In Proceedings of the IEEE International Power and Energy Conference, Putrajaya, Malaysia,
28–29 November 2006; pp. 509–515.
15. Jayachandra, S.; Yesuraj, D.J. Modeling and simulation of capacitor voltage transformer transients
using PSCAD/EMTDC. In Proceedings of the 2011 IEEE Trondheim PowerTech, Trondheim, Norway,
19–23 June 2011; pp. 1–8.
16. Pajuelo, E.; Ramakrishna, G.; Sachdev, M.S. Phasor estimation technique to reduce the impact of coupling
capacitor voltage transformer transients. IET Gener. Transm. Distrib. 2008, 2, 588–599. [CrossRef]

© 2017 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/).