energies
Article
An Accurate Evaluation of Switching Impulse Voltages for
High-Voltage Tests
Peerawut Yutthagowith
Citation: Yutthagowith, P. An
Accurate Evaluation of Switching
Impulse Voltages for High-Voltage
Tests. Energies 2022, 15, 4760.
https://doi.org/10.3390/en15134760
Academic Editor: Adam Smoliński
Received: 27 May 2022
Accepted: 20 June 2022
Published: 29 June 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the author.
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 (https://
creativecommons.org/licenses/by/
4.0/).
Energies 2022, 15, 4760. https://doi.org/10.3390/en15134760
School of Engineering, King Mongkut’s Institute of Technology Ladkrabang, 1 Chalongkrung Rd., Ladkrabang,
Bangkok 10520, Thailand; peerawut.yu@kmitl.ac.th; Tel.: +66-(0)2-329-8330
Abstract: For assessment of the insulation performance of high-voltage (HV) equipment installed
in extra-high-voltage (EHV) systems, switching impulse voltage tests are performed in an HV
testing laboratory. The waveform parameters of the switching impulse voltages are defined by
peak voltage (Up ), time to crest (Tp ), and time to half (T2 ) according to IEC 60060-1. In this paper,
a new, simplified, and accurate approach used for determination of the waveform parameters of
the switching impulse voltages is presented. The formula used in the evaluation of Tp was derived
from analytically simulated two-exponential waveforms, where Tp and T2 are in the ranges of
20 µs to 300 µs and 1000 µs to 4000 µs, respectively. The accuracy of the proposed approach was
validated by the waveforms collected from the test waveform data generator (TDG) provided by
IEC 61083-2, simulations, and experiments. It is found that the accuracy of the proposed approach
is relatively higher than the expressions provided by IEC 60060-1 and previously developed. The
proposed method is an alternative and useful approach for evaluating the waveform parameters of
the standard switching impulse voltage.
Keywords: evaluation of waveform parameters; high-voltage tests; insulation performance; switching
impulse voltage
1. Introduction
The main causes of insulation failures in high-voltage equipment are overvoltages
due to lightning and switching operations in high-voltage (HV) energy transmission and
distribution systems. Therefore, the international standards [1–5] suggest performing
impulse voltage tests on HV equipment to confirm its insulation performance. The stan-
dard lightning and switching impulse voltages can be generated by a simple resistor and
capacitor circuit [6]. Switching overvoltages are caused by switching operations of circuit
breakers and disconnecting switches in HV power systems, and the insulation performance
of HV equipment under switching overvoltages has still not been completely studied [7–9].
Switching overvoltages produce high electrical field stresses with quite a long time period
(several milliseconds) on the insulation of the HV equipment and can possibly cause in-
sulation failures. Thus, according to IEC standards [4], the insulation performance of HV
equipment operating at a system voltage of not less than 300 kV must be confirmed using a
switching impulse voltage test.
In impulse voltage tests, the software used for the evaluation of the impulse waveform
parameters must be assessed for accuracy using the waveform generated by the test data
generator (TDG) program attached to IEC 61083-2 [10]. The evaluation of the lightning
and switching impulse voltage waveform parameters has been widely studied [11–17]. In
the switching impulse voltage tests, the waveform parameters must be adjusted as per
the standard requirements [1–5]. As illustrated in Figure 1, the time to crest (Tp ), the time
to half (T2 ), and the peak voltage (Up ) shall be 250 µs ± 20%, 2500 µs ± 60%, and the
specified value depending on the system voltage with a tolerance of ±3%, respectively. T2
is defined as the duration time from the actual origin time (t0 ) to the time (t50% ) at 50% of
https://www.mdpi.com/journal/energies
Energies 2022, 15, 4760 2 of 10

the peak voltage at the tail part of the waveform. T2 can be determined by proper search
Energies 2022, 15, x FOR PEER REVIEW
and interpolation algorithms. Up can be determined by searching for the maximum2 value of 10
for the positive waveform or the minimum value for the negative waveform after the offset
voltage is removed from the considered waveform. However, there are some difficulties in
isdetermining t0 and
defined as the Tp duetime
duration to noise
fromsuperimposing on time
the actual origin the recorded waveform.
(t0) to the time (t50%)Inataddition,
50% of
T
the cannot be determined precisely due to noise and a long duration
p peak voltage at the tail part of the waveform. T2 can be determined by time around
properthe peak
search
voltage.
and According
interpolation to the standard
algorithms. be T
Up can[1], p can be calculated
determined by Equation
by searching K and
(1), where value
for the maximum
T AB are defined as Equations (2) and (3).
for the positive waveform or the minimum value for the negative waveform after the off-
set voltage is removed from the considered waveform. However, there are some difficul-
Tp = KTAB (1)
ties in determining t0 and Tp due to noise superimposing on the recorded waveform. In
addition, Tp cannot be determined
K = 2.42 −precisely
3.08 × 10due
−3 to noise and a −
TAB + 1.51 × 10 long
4
T2 duration time around
(2)
the peak voltage. According to the standard [1], Tp can be calculated by Equation (1),
TAB =(2)
where K and TAB are defined as Equations t90% − (3).
and t30% (3)

u(t)

1.0Up
0.9Up

0.5Up
0.3Up
TAB
0.1Up
t
t0 t10% t30% t90% t50%

T10-90
T10-50
Tp T2

Switchingimpulse
Figure1.1.Switching
Figure impulsevoltage
voltagewaveform
waveformand
andthe
thewaveform
waveformparameters.
parameters.

For determination of Tp , however, it was reported that Tp determined by Equation (1)

is accurate in only some cases where = KT
Tpthe AB to crest and the times to half are close
times (1)to
250 µs and 2500 µs, respectively. There is−3some research which presents approaches for
K = 2.42 − 3.08 10 TAB + 1.5110−4 T2 (2)
the determination of the accurate time to crest [16]. The analytical formulas for estimating
the waveform parameters Tp , T2 , Tand=U − t30% proposed in [16]. Tp is also determined
p were
t90% (3)
AB
from TAB and the estimated starting point of the considered waveform. The formulas can
For determination of Tp, however, it was reported that Tp determined by Equation (1)
estimate the waveforms derived from simulated two-exponential waveforms precisely, but
isdeviations
accurate in in only some casesare
the parameters where the times
increased in thetocases
crestofand the times to
experimental half are close
waveforms. to
In [17],
250 μs and 2500 μs, respectively. There is some research which
two-exponential fitting based on a non-linear regression on the waveform data around presents approaches for
the
thedetermination
crest value was ofemployed
the accurate time to crest
to estimate [16]. The analytical
the waveform parametersformulas for estimating
accurately. Since this
the waveform parameters T p, T2, and Up were proposed in [16]. Tp is also determined from
method requires software for estimation of the parameters, it is not useful for test engineers
Tfor
AB and the estimated starting point of the considered waveform. The formulas can esti-
determination of the waveform parameters when compared with the standard and
mate the waveforms
previously developedderived
formulas. from simulated two-exponential waveforms precisely, but
deviations in the parameters are
In this paper, a practical approach increased forinthe
thedetermination
cases of experimental waveforms.
of the waveform In [17],
parameters
two-exponential fitting based on a non-linear regression on the waveform
of the switching impulse voltage was developed. A simple, but relatively accurate approach data around
the
for crest value was employed
the determination of t0 is to estimate the
proposed. waveform parameters
Furthermore, the accurateaccurately.
formula usedSinceinthis
the
method
evaluation requires software
of Tp was for from
derived estimation of the parameters,
the analytically simulatedittwo-exponential
is not useful forwaveforms,
test engi-
neers
wherefor Tpdetermination
and T2 are in the of the waveform
ranges of 20 µs parameters
to 300 µs andwhen compared
1000 with
µs to 4000 µs,the standard
respectively.
and previously developed formulas.
The waveforms collected from the standard [10], simulations and experiments were utilized
In this paper,
for verification a practical
of the proposed approach
approach. for The
the determination
proposed approachof theprovides
waveform parameters
relatively high
of the switching impulse voltage was developed. A simple,
accuracy when compared with those calculated by the standard [1] and the previously but relatively accurate ap-
proach
developed for the determination
formula [16]. This of tpaper
0 is proposed. Furthermore,
is organized the accurate
as follows. Section 1formula
introducesusedthein
the evaluation
problem of Tp was
of evaluation derived
of the from the
waveform analytically
parameters of thesimulated
switchingtwo-exponential
impulse voltageswave-in HV
forms,
tests and where Tp andthe
performs T2 literature
are in the review.
ranges of 20 μs 2todescribes
Section 300 μs andan 1000 μs to for
approach 4000theμs, respec-
analytical
tively. The waveforms
generation of switching collected
impulse from the standard
voltages on the[10], simulations
basis and experiments
of the two-exponential were
function
utilized for verification of the proposed approach. The proposed approach provides rela-
employed for the development of the new expressions. The development of a new and
tively high accuracy when compared with those calculated by the standard [1] and the
previously developed formula [16]. This paper is organized as follows. Section I intro-
duces the problem of evaluation of the waveform parameters of the switching impulse
voltages in HV tests and performs the literature review. Section II describes an approach
Energies 2022, 15, x FOR PEER REVIEW 3 of 10

Energies 2022, 15, 4760 for the analytical generation of switching impulse voltages on the basis of the two-expo- 3 of 10
nential function employed for the development of the new expressions. The development
of a new and accurate expression for the determination of the waveform parameters is
also presented. The proposed expressions and their verification are presented in Section
accurate expression for the determination of the waveform parameters is also presented.
III. At the end of this paper, the conclusions are addressed.
The proposed expressions and their verification are presented in Section 3. At the end of
this paper, the conclusions are addressed.
2. Development of The Approach for Evaluation of The Switching Impulse Voltage
Parameters
2. Development of the Approach for Evaluation of the Switching Impulse
Voltage
The newParameters
accurate approach for evaluation of the switching impulse voltage parame-
ters is developed from the
The new accurate waveform
approach parametersofofthe
for evaluation theswitching
impulse voltage
impulsewaveforms in the
voltage parameters
form of a two-exponential
is developed function,parameters
from the waveform as given inofEquation
the impulse (4). voltage
It is quite straightforward
waveforms to
in the form
determine Tp and T2 from
of a two-exponential the known
function, as givenτ1 and inτEquation
2. From Equation
(4). It is(4), Tp can
quite be calculatedto
straightforward
analytically
determineby TpEquation (5), but
and T2 from the there
known is no analytical
τ 1 and expression
τ 2 . From Equation of (4),
T2; itTcan
p can bebe
calculated
calculated
byanalytically
Equation (6) byusing
Equation (5), but there
a numerical is no analytical
approach. A simpleexpression
Newton–Raphsonof T2 ; it can be calculated
algorithm [18]
is by
an Equation (6) usingused
effective method a numerical approach. A simple Newton–Raphson algorithm [18] is
in this paper.

( )
an effective method used in this paper. −t /
u (t ) = U 0 e 2 − e−t /1 (4)
 
−t/τ2 −t/τ1
u(t) = U0 e −e (4)
ln ( 2 / 1 )
Tp = (5)
1/ 1 −ln1/(τ2 /τ1 )
Tp = (5)
( )
1/τ1 − 1/τ2
−T / − T /
0.5U 0 = U 0 e 2 2 − e 2 1  (6)
0.5U0 = U0 e−T2 /τ2 − e−T2 /τ1 (6)
However, accurate τ1 and τ2 computed from Tp and T2 are required for the develop-
ment ofHowever, accurate
the proposed τ 1 and In
approach. computed
τ 2 this section,from Tp and T2approach
the effective are required for the develop-
for determination
of τ1 and τ2 from Tp and T2 will be presented in Section 2.1, and the analytical formula for of
ment of the proposed approach. In this section, the effective approach for determination
the and τ 2 from
τ 1required Tp andparameters
waveform T2 will be presented in Section
will be presented 2.1, and2.2.
in Section the analytical formula for
the required waveform parameters will be presented in Section 2.2.
2.1. Analytical Generation of Switching Impulse Waveforms
2.1. Analytical Generation of Switching Impulse Waveforms
For generation of the accurate waveform in the form of a two-exponential function
For generation of the accurate waveform in the form of a two-exponential function
described by T2 and Tp, it is necessary to know the accurate time constants of τ1 and τ2. In
described by T2 and Tp , it is necessary to know the accurate time constants of τ 1 and τ 2 . In
this paper, a time normalization technique was utilized to reduce the number of variables,
this paper, a time normalization technique was utilized to reduce the number of variables,
and
andthe secant
the secant method
method [18] was
[18] employed
was employed to to
determine
determine the solution
the oror
solution thethe
required time
required time
parameters. It is noticed that the ratio of T 2 to Tp (T2/Tp) has a monotonic relation with the
parameters. It is noticed that the ratio of T2 to Tp (T2 /Tp ) has a monotonic relation with
ratio
the of τ2 to
of ττ12 (τ
to2/τ
τ 11),(τas expressed in Figure 2, of which the range covers Tp of 20 μs to
ratio 2 /τ 1 ), as expressed in Figure 2, of which the range covers Tp of 20 µs
300 μs and T 2 of 1000 μs to 4000 μs. Therefore, the relation of T2/Tp and τ2/τ1 can be repre-
to 300 µs and T2 of 1000 µs to 4000 µs. Therefore, the relation of T2 /Tp and τ 2 /τ 1 can be
sented as functions
represented in Equations
as functions (7) and(7)
in Equations (8).and (8).
T2 / Tp = f ( 2 / 1) (7)
T2 /Tp = f (τ2 /τ1 ) (7)
−1
 2 / 1 = f (T2−1/ T p ) = g (T2 / T p ) (8)
τ2 /τ1 = f ( T2 /Tp ) = g( T2 /Tp ) (8)

Figure
Figure 2. 2. τ 21/τ
τ2/τ as function
as1the the function 2/TT
of Tof /Tp the
p 2and andfitting
the fitting curve.
curve.
Energies 2022,15,
Energies2022, 15,4760
x FOR PEER REVIEW 44 of 10
10

In
In this
this paper, the normalized
normalized base basewas
wasselected
selectedtotobebeτ1τ, and
1 , and
all all normalized
normalized time
time pa-
parameters
rameters (τ(τ1n,1n
τ2n, 2n2n, ,Tand
, τT Tpn) Tare
2n , and pn )given
are given by Equation
by Equation (9). (9).
1τn ==1 τ/ /τ1 =1 1= 1
1n 1
 τ22n
n == 2τ2/ /τ
11
(9)
(9)
T2n2n= T2 /21 1
T = T /τ
TTpnpn==T pT/p/τ1
1

The
Theprocedure
procedurefor
fordetermination
determinationofofττ1 1and
andττ22 from
fromthe
theinput variableofofTT2 2and
inputvariable andTTppisis
presented
presentedininthe
theflowchart
flowchartshown
shownininFigure
Figure3.3.

Start

Input data
(Tp and T2)

Set i = 1

Determine the initial values of

2n(i-1) and 2n(i) from (13) and (14)

Determine Tpn (2n(i-1)), T2n(2n(i-1)),

Tpn (2n(i)), and T2n(2n(i))

Calculate the deviation (e(i-1) and e(i)) updated 2n (i-1) and 2n (i)
of Tpn /T2n numerically by the secant method

Calculate 2n(i+1) and e(i+1) Set i = i+1

No
e (i+1) < 0.001%

Yes

Scaling 2n and 1n with the

calculated Tp/Tpn to obtain 2 and 1

End

Figure3.3.Algorithm
Figure Algorithmused
usedfor
forgeneration
generationof
ofthe
theimpulse
impulseswitching
switchingwaveform
waveformwith
withthe
thespecified
specifiedTT2
2
and Tp.
and Tp .

FromEquation
From Equation(10), (10),thethesecant
secantmethod
methodisisemployed
employedtotofindfindthe
thesolutions
solutionsofofττ2n2nfrom
from
T /T , which equals T /T . The goal of the secant method is to
Tpn /T2n , which equals Tp /T2 . The goal of the secant method is to determine τ 2n where
pn 2n p 2 determine τ2n where the
deviation (ε) of T /T is minimized, and the deviation goal was
the deviation (ε) of Tpn /T2n is minimized, and the deviation goal was set to be 0.001%.
pn 2n set to be 0.001%. Equation
(11) can be
Equation utilized
(11) can beto calculate
utilized the deviation
to calculate ε(i) of theε(i)
the deviation solution
of theof the ith iteration,
solution of the ithwhere
itera-
T2n(τwhere
tion, 2n(i)) is T
the normalized
2n (τ 2n (i)) is theparameter
normalized calculated
parameter by calculated
the Newton–Raphson algorithm [18]
by the Newton–Raphson
at τ2n(i), and
algorithm [18]Tat
pn(τ 2n(i))
τ2n (i), isand
calculated by is
Tpn (τ 2n (i)) thecalculated
analyticalbyformula expressed
the analytical by Equation
formula expressed(5).
From the calculated ε(i), (τ2n(i + 1)) is calculated by Equation (12). τ2n is calculated repeti-
tively until the deviation goal is reached.
Energies 2022, 15, 4760 5 of 10

by Equation (5). From the calculated ε(i), (τ 2n (i + 1)) is calculated by Equation (12). τ 2n is
calculated repetitively until the deviation goal is reached.

T2 /Tp = T2n /Tpn = f (τ2 /τ1 ) = f (τ2n ) (10)

T2n (τ2n (i )) T2
Tpn (τ2n (i ))
− Tp
ε (i ) = T2
(11)
Tp

ε(i )(τ2n (i ) − τ2n (i − 1))

τ2n (i + 1) = τ2n (i ) − (12)
ε ( i ) − ε ( i − 1)
The secant method requires the initial values of τ 2n (0) and τ 2n (1). From the relation of
T2 /Tp and τ 2 /τ 1 in Figure 3, the initial values are set as given by Equations (13) and (14).


τ2n (0) = 8 T2 /Tp − 24 (13)

τ2n (1) = 1.01τ2n (0) (14)

With the initial values in Equations (13) and (14), only two or three iterations are
required to reach the deviation goal. Finally, τ 1 and τ 2 can be determined from Equation (9)
with the known Tp and Tpn , which can be calculated by Equation (5). Using the proposed
method for the determination of τ 1 and τ 2 from the specified T2 and Tp , the switching
impulse voltage waveforms with Tp from 10 µs to 300 µs and T2 from 1000 µs to 4000 µs
can be precisely generated analytically.
From the calculated τ 2 /τ 1 determined by T2 /Tp using the proposed method, a nu-
merical curve fitting based on a Levenberg–Marquardt algorithm [19] was employed and
the formula of the fitting curve given in Equation (15) was obtained. It is noted that the
formula can be used for the initial values in the secant method. In fact, the formula is quite
accurate and the maximum deviation of the value from the formula and the actual value is
less than 0.5%. However, if more accurate results are required, it is recommended to utilize
the proposed method, as it is used in the paper with a deviation of less than 0.001%.
 1.25
T2
τ2n (0) = 3.389 − 9.89 (15)
Tp

2.2. The New Approach for Evaluation of the Switching Impulse Voltage Parameters
In the proposed method, analytical switching impulse voltage waveforms generated
by the approach presented in Section 2.1 were utilized for development of the formula used
for time to crest estimation. The time parameters, i.e., t10% , t90% , and t50% shown in Figure 1,
are calculated by the Newton–Raphson method due to no analytical solution, and Tp and
T2 are considered in the ranges of 20 µs to 300 µs and 1000 µs to 4000 µs, respectively, which
cover the waveform parameters used in the practical tests. t0 is zero. It is found that Tp /T2
has a monotonic relation with T10–50 /T10–90 , as shown in Figure 4, and parameters used for
calculation of such proportions can be calculated by Equations (16)–(18).

T2 = t50% − t0 (16)

T10−90 = t90% − t10% (17)

T10−50 = t50% − t10% (18)
 Energies 2022, 15, x FOR PEER REVIEW 6 of 10

Energies 2022, 15, x FOR PEER REVIEW 6 of 10

Energies 2022, 15, 4760 6 of 10

Figure 4. Real data and fitted values of Tp/T2 versus T10–50/T10–90.

Figure 4. Real data and fitted values of Tp /T2 versus T10–50 /T10–90 .
From
Figure 4. Real the
datadata of Tp/T
and fitted 2 andofTT
values p/T/T
10–50 10–90, the
2 versus fitting
T10–50 /T10–90.curve can be determined by a non-
linear regression based on a Levenberg–Marquardt algorithm [19]. The expression of the
From the data of Tp /T2 and T10–50 /T10–90 , the fitting curve can be determined by a
Fromcurve
fitted the data of Tp/T
is given and T10–50/T
as2 Equation (19). , the fitting
10–90From Equation curve (19)canand bethedetermined
required by
T 2, a
Tpnon-
canof be
non-linear regression based on a Levenberg–Marquardt algorithm [19]. The expression
linear regression
calculated by based on (20).
Equation a Levenberg–Marquardt
U p can also be determined algorithm by [19]. The for
searching expression
the crest ofvoltage
the
the fitted curve is given as Equation (19). From Equation (19) and the required T2 , Tp can be
fitted curve
of the
calculated is given
considered
by Equation as Equation
waveform.
(20). UpItcan (19).
is notedFrom
also bethatEquation
the approach
determined (19)
by andforthethe required
searching crest T2, Tdetermina-
voltage
for the crest p can be
voltage of
calculated
tion
the is thebysame
considered Equationas those
waveform. (20).of ItUthe
p can
is alsothat
standard
noted be the
determined
and previously
approach by
forsearching
developed for
the crest voltage the crest
methods voltage
[1,16].
determination
ofisthe
theconsidered
same as those waveform.
of It is noted that the approach 0.8506 for the methods
crest voltage determina-
T the standard
 T10  andand previously
 Tpreviously developed [1,16].
−4 methods [1,16].
tion is the same as thosep of = h
the − 90
standard = 1.389  10 − 90
  − 2.537  10
developed

(19)
TTp T2 T10T−1090 − 50   T10T−1050  0.8506
  0.8506
p = h  T10 − 90   T10 −90  − 90
= 1.389 − 2.537
−4 × 10−4 (19)
T2 = h  T10−50  = 1.389   T10− 50− 2.537  10 (19)
 10 −50  T
Tp = h10−50   T2
T2 T T 10 − 90
(20)
 T10T−1050−90 T
Tp =  T10 −90T
h 2 (20)
Comparisons of the times T = h  computed
top crest 10T−2 50 by the proposed, standard, and (20) previ-
ously developed of
Comparisons formulas
the times [1,16]
to crest presented
T10 −50 
arecomputed byinthe
Figures 5 and
proposed, 6. It is noticed
standard, that the
and previously
Comparisons
developed
proposed formulas
formula of the times
[1,16]
provides are to crest
presented
the bestcomputed
in Figures
accuracy by 5the
with andaproposed,
6.maximum standard,
It is noticed that the
deviation andof previ-
proposed
0.156%,
ously developed
formula
whereas provides formulas
the standard the best and [1,16]
accuracy arewith
previously presented
a maximum
developed in Figures 5 and
deviation
formulas of 6.
[1,16] It is noticed
0.156%,
provide whereas
maximum that
thethestan-
devi-
dard
proposed
ationsand previously
offormula
25.778%provides
and developed
1.147%, formulas
therespectively.
best accuracy[1,16]
From provide
with thearesultsmaximum
maximum of all test deviations
deviation
cases, the ofof 25.778%
0.156%,
previously
and
whereas 1.147%,
developed respectively.
the standard
and From
and previously
proposed formula the canresults
developed of allformulas
provide test
T cases,
p within theacceptable
[1,16]
the previously
provide maximum developed
limits devi-
defined andby
proposed
ations
the of
standard formula
25.778% and
[10]. can
The provide
1.147%,
standard Tp formula
within the
respectively. From acceptable
is accurate in limits
the results of all
cases defined
of test by the
the cases,
waveformthestandard
previously
with time [10].
to
The
developedstandard and formula
proposed is accurate
formula in
can cases
provide of the
T waveform
within
crest close to the standard time to peak (250 μs). The shorter the Tp is, the higher the devi-
p the with
acceptabletime to crest
limits close
defined tobythe
thestandard
standard
ation is. time[10].toThe
peakstandard
Additionally, (250 The
µs).formula
as shown inshorter the
is accurate
Figure 7 the Tpproposed
is, cases
in the higher thethe
ofapproach deviation
waveformfor thewithTis. Addition-
time to
2 determina-
ally,
crest
tion as
close shown
alsotoprovides in Figure
the standard promising7 the
time to proposed
peak (250
accuracy approach
when μs).itThe for the T
shorter the
is compared determination
2with Tp the
is, the also
higher the
previously provides
devi-
developed
promising
ation accuracy
is. Additionally,
formula [1]. The maximum when
as shown it is compared
in Figureof7with
deviations the previously
the proposed approach
and developed
previously formula
for thedeveloped [1].
T2 determina- The
formu-
maximum
tionlasalso deviations
areprovides
0.165% promising
and of therespectively.
0.98%, proposedwhen
accuracy and previously
it is compared developed
with theformulas
previously aredeveloped
0.165% and
0.98%, [1].
formula respectively.
The maximum deviations of the proposed and previously developed formu-
las are 0.165% and 0.98%, respectively.

Upper limit of the standard tolerance

+ Previously developed formula Proposed method o IEC method

UpperLower
limit of theof
standard tolerance
+ Previously developed formula Proposed method o IEC method
limit the standard tolerance
Deviation (%)

Lower limit of the standard tolerance

Deviation (%)

Tp = 20 mS Tp = 40 mS Tp = 80 mS Tp = 120 mS Tp = 160 mS Tp = 240 mS Tp = 300 mS

T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS

Tp = 20 mS Tp = 40 mS Tp = 80 mS Tp = 120 mS Tp = 160 mS Tp = 240 mS Tp = 300 mS

to 4000 mS5. TComparisons
T2 = 1000 to 4000 mS T2 = 1000 Figure 2 = 1000 to 4000 mS of
T2the to 4000 mS Ttimes
computed
= 1000 2 = 1000to crestmSbyT2the
to 4000 = 1000 to 4000 mS Tstandard,
proposed, 2 = 1000 to 4000
andmS previously

developed formulas [1,16] and the standard tolerance [10].

Energies 2022, 15, x FOR PEER REVIEW 7 of 10

Figure 5. Comparisons of the computed times to crest by the proposed, standard, and previously
developed formulas [1,16] and the standard tolerance [10].
Energies 2022, 15, 4760 Figure 5. Comparisons of the computed times to crest by the proposed, standard, and previously
7 of 10
developed formulas [1,16] and the standard tolerance [10].

+ Previously developed formula Proposed method o IEC method

Upper limit
+ Previously developed formula Proposed method o IEC method
Deviation (%)

Upper limit
Deviation (%)

Lower limit

Tp = 20 mS Tp = 40 mS Tp = 80 mS Tp = 120 mS Tp = 160 mS Tp = 240 mS Tp = 300 mS

Lower limit T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS
T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS

Tp = 20 mS Tp = 40 mS Tp = 80 mS Tp = 120 mS Tp = 160 mS Tp = 240 mS Tp = 300 mS

T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS

Figure 6. Zoom-in results of Figure 5.

Figure 6. Zoom-in results of Figure 5.
Figure 6. Zoom-in results of Figure 5.

+ Previously developed formula Proposed method

Upper limit
+ Previously developed formula Proposed method
Deviation (%)

Upper limit
Deviation (%)

Lower limit

Tp = 20 mS Tp = 40 mS Tp = 80 mS Tp = 120 mS Tp = 160 mS Tp = 240 mS Tp = 300 mS

Lower limit T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS
T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS

Tp = 20 mS Tp = 40 mS Tp = 80 mS Tp = 120 mS Tp = 160 mS Tp = 240 mS Tp = 300 mS

T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS T2 = 1000 to 4000 mS
Figure 7. Comparisons of the computed T2 by the proposed and previously developed formulas
[16]
and the standard tolerance [10].
Figure 7. Comparisons of the computed T2 by the proposed and previously developed formulas [16]
and the standard tolerance [10].
3. Verification
Figure of the of
7. Comparisons Proposed Method
the computed T2 by the proposed and previously developed formulas [16]
and the
By standard
using tolerance
the [10].
presented approach
3. Verification of The Proposed Method for determination of the waveform parameters, the
offset voltage
By usingisthe removed
presented andapproach
the time parameters
for determinationare determined. t0 can beparameters,
of the waveform estimated by the
3. Verification of The Proposed Method
searching for the time at which the voltage is less than 0.1% of the peak voltage, and the
offset voltage is removed and the time parameters are determined. t0 can be estimated by
approach for searching
By using startsapproach
from the for peak voltage in reversed order. Theparameters,
other time pa-
searching forthethepresented
time at which the voltagedetermination
is less than of theof
0.1% waveform
the peak voltage, andthe the
rameters,
offset i.e., tis
voltage , t90% , and
removed t50%
and shown
the time in Figure 1, are
parameters are calculated
determined. by tthe Newton–Raphson
0 can be estimated
10%
approach for searching starts from the peak voltage in reversed order. The other time by pa-
method
searching due tothe no analytical solutions. Then, using such time of parameters, T10–50and/T10–90
rameters,for i.e., t10%time
, t90%,atand
which the voltage
t50% shown is less
in Figure 1,than 0.1%
are calculated thebypeak voltage,
the Newton–Raphson the
is calculated
approach by Equations
for searching (17)
starts from and (18) and
the peak substituted
voltage in Equation (19) to obtain T /T 2.
method due to no analytical solutions. Then, usinginsuch
reversed
timeorder. The other
parameters, time
T10–50 p pa-
/T10–90 is
Eventually,
rameters, T
i.e., is calculated by Equation
2 t10%, t90%, and t50% shown in Figure (16), T is determined from
p 1, are calculated by the T /T with the
p Newton–Raphson
2 known
calculated by Equations (17) and (18) and substituted in Equation (19) to obtain Tp/T2.
Tmethod
2 , and Uduep is determined
to2 is
nocalculated by the
analytical crest value of the
solutions. considered waveform.
Eventually, T by EquationThen,
(16), Tusing such time parameters, T10–50/T10–90 is
p is determined from Tp/T2 with the known
The
calculated proposed
by Equationsmethod (17)is verified
and (18) by
andthe waveform
substituted inparameter
Equation evaluation
(19) to obtain of wave-
Tp/T2.
T2, and Up is determined by the crest value of the considered waveform.
forms composed
Eventually, T of four groups collected from (1) analytically simulated two-exponential
2 is calculated by Equation (16), Tp is determined from Tp/T2 with the known
The proposed method is verified by the waveform parameter evaluation of wave-
waveforms
Tforms used in the method development; (2) IEC 61083-2 [10], composed of SI-A1
2, and Up is determined by the crest value of the considered waveform.
composed of four groups collected from (1) analytically simulated two-exponential
to SI-A3 (simulation
The proposed waveforms) and SI-M1
by theand SI-M2 (waveforms measured in experi-
waveforms
ments); (3) used inmethod
simulations thewith is verified
methodand development;
without
waveform
(2) IECnoise
additional
parameter
61083-2 [10],
signal
evaluation
composed
(SI-X1 and ofofSI-A1
SI-X2),
wave-to
and
forms
SI-A3 composed
(simulation of four groups
waveforms) collected
and from
SI-M1 (1) analytically
and experimental
SI-M2 (waveforms simulated two-exponential
measured in experi-
(4) experiments
waveforms (SI-X3 and SI-X4). The example waveform is presented in
ments);
Figure 8. (3)used
Additionally,
in the method
simulations the with
accuracy
development;
and without
of the proposed
(2) IEC 61083-2
additional [10], (SI-X1
noiseissignal
method compared
composed of
andthose
with
SI-A1and
SI-X2),of
to
the
SI-A3 (simulation
(4) experiments waveforms)
(SI-X3 and and SI-M1
SI-X4). methods and
The example SI-M2 (waveforms measured in experi-
standard
ments); and
(3) previously
simulations developed
with and without [1,16]experimental
additional asnoise
expressed
signal
waveform
in(SI-X1
Tablesand 1isand
presented toin
2. It isand
SI-X2),
beFigure 8. Additionally, the accuracy of the
the proposed method is compared with those of
(4)mentioned
experiments that the reference
(SI-X3 and SI-X4). values
Theinexample casesexperimental
of SI-X3 and waveform
SI-X4 haveisbeen collected
presented in
from commercial software [20]. The approved measuring system, composed of a 1500 kV
Figure 8. Additionally, the accuracy of the proposed method is compared with those of
voltage divider, a measuring cable, and a 12-bit digital recorder, was utilized to measure
the switching impulse voltage in the experiments.
 is to be mentioned that the reference values in the cases of SI-X3 an
collected from commercial software [20]. The approved measuring sy
a 1500 kV voltage divider, a measuring cable, and a 12-bit digital recor
Energies 2022, 15, 4760
measure the switching impulse voltage in the experiments. 8 of 10

Figure 8. Switching impulse voltage waveform in the Case SI-X3.

Figure 8. Switching impulse voltage waveform in the Case SI-X3.
Table 1. Reference and calculated values, acceptable limits, and deviations of Tp .
From the test results ofValue
Reference
the from
evaluation
the
of the waveform
Value from
parameter
Value from the
accuracy
Case of the proposed method is relatively
Values Standard
Formula in higher
[16] than that of the s
Proposed
(Acceptable Formula Method
ously developed methods.
Limits) All waveform parameters determined
(Deviations)
(Deviations)
(Deviations) by th
are within
SI-A1
the standard
250.7 µs tolerances
248.703 µs [1,10].251.630
The µs deviations
251.621ofµs Tp in the
(±2%) (−0.797%) (+0.371%) (+0.367%)
forms, simulated19.89
by µstwo-exponential
15.300 µs function, provided
20.232 µs byµsthe standa
20.051
SI-A2
ulation with and(±without
2%) noise signal, and
(−23.076%) collected(+0.809%)
(+1.720%) from the expe
43.08 µs 35.999 µs 43.214 µs 43.252 µs
SI-A3
0.156%, 1.34%, 1.45%,
(±2%) and 2.00%, respectively.
(−16.437%) The deviations
(+0.310%) (+0.398%) of T2 of th
186.6 µs 177.619 µs 170.227 µs 189.085 µs
simulation
SI-M1 and measured
(±5%)
experimentally
(−4.81%)
are within (+1.332%)
(−8.775%)
0.06% and 1.60
casesSI-M2
of the waveforms
218.0 µs from221.943
simulation
µs and
225.264measured
µs experimentally
220.781 µs
(±5%) (+1.809%) (+3.332%) (+1.276%)
Up are within 0.005% and 0.48%,
250.00 µs respectively.
246.908 µs 249.862 µs 249.842 µs
SI-X1
(±2%) (−1.237%) (−0.056%) (−0.063%)
250.00 µs 251.779 µs 254.712 µs 253.626 µs
TableSI-X2
1. Reference and
(±5%)calculated values, acceptable
(+0.711%) (+1.885%)limits, and deviations of
(+1.450%)
247.35 µs 250.087 µs 253.221 µs 252.291 µs
SI-X3
(±5%) (+1.105%) (+2.372%) (+1.996%)
Reference Value from the
SI-X4
269.44 µs 273.007 µs 276.028 µs Value
271.913 µs from the
(±5%) Values
(+1.324%) Standard
(+2.445%) (+1.289%)
Case Formula in [16
(Acceptable Formula
From the test results of the evaluation of the waveform parameters, it is(Deviations)
clear that
Limits)
the accuracy of the proposed method (Deviations)
is relatively higher than that of the standard and
previously developed methods. All waveform parameters determined by the proposed
250.7 μs 248.703 μs 251.630 μs
SI-A1
method are within the standard tolerances [1,10]. The deviations of Tp in the cases of the
(±2%) function, provided
waveforms, simulated by two-exponential (−0.797%) (+0.371%)
by the standard, additional
simulation with and without noise signal, and collected from the experiments are within
19.89 μs 15.300 μs 20.232 μs
SI-A2
0.156%, 1.34%, 1.45%, and 2.00%, respectively. The deviations of T2 of the waveforms from
(±2%)are within 0.06%
simulation and measured experimentally (−23.076%)
and 1.60%, respectively.(+1.720%)
In cases
of the waveforms from simulation and measured experimentally, the deviations of Up are
43.08 μs 35.999 μs 43.214 μs
SI-A3
within 0.005% and 0.48%, respectively.
(±2%) (−16.437%) (+0.310%)
186.6 μs 177.619 μs 170.227 μs
SI-M1
(±5%) (−4.81%) (−8.775%)
218.0 μs 221.943 μs 225.264 μs
SI-M2
(±5%) (+1.809%) (+3.332%)
250.00 μs 246.908 μs 249.862 μs
SI-X1
(±2%) (−1.237%) (−0.056%)
250.00 μs 251.779 μs 254.712 μs
Energies 2022, 15, 4760 9 of 10

Table 2. Reference and calculated values, acceptable limits, and deviations of T2 and Up .

Reference Values Value from Proposed Method

Case (Acceptable Limits) (Deviations)
T2 Up T2 Up
2512.00 µs 950.28 kV 2511.50 µs 950.23 kV
SI-A1
(±2%) (±0.1%) (−0.0199%) (−0.005%)
1321.00 µs 987.67 V 1320.75 µs 987.67 V
SI-A2
(±2%) (±0.1%) (−0.0189%) (+0.001%)
3987.00 µs 99.219 kV 3987.05 µs 99.219 kV
SI-A3
(±2%) (±0.1%) (+0.0013%) (0.0000%)
655.00 µs −590.70 V 655.335 µs −590.70 V
SI-M1
(±2%) (±0.5%) (+0.0512%) (+0.0799%)
2407.00 µs 3.680 kV 2396.20 µs 3.680 kV
SI-M2
(±2%) (±0.5%) (−0.4487%) (−0.0027%)
2500.00 µs 521.582 kV 2499.90 µs 521.582 kV
SI-X1
(±2%) (±0.1%) (−0.0039%) (0.0000%)
2500.00 µs 521.988 kV 2465.06 µs 521.988 kV
SI-X2
(±2%) (±0.5%) (−1.3978%) (+0.4754%)
2293.37 µs −569.982 kV 2256.80 µs −569.982 kV
SI-X3
(±2%) (±0.5%) (−1.5948%) (+0.4023%)
3775.51 µs 453.788 kV 3784.21 µs 449.321 kV
SI-X4
(±2%) (±0.5%) (−0.9843%) (+0.2304%)

4. Conclusions
This paper presents an effective approach for the analytical generation of the switch-
ing impulse waveforms with the required Tp and T2 . From the generated waveforms, a
new, simplified, and accurate approach for the waveform parameters determination of the
switching impulse voltages has been developed and verified successfully by simulated and
experimental waveforms. Using the developed analytical formula, the proportion of T10–50
and T10–90 is employed to determine the time to crest (Tp ) precisely. The deviation of Tp
calculated by the proposed formula is very small and within the standard tolerance for both
simulated and experimental waveforms. The proposed technique has the largest deviations
of +1.45% and +2.00% for the simulated and experimental waveforms, respectively, whereas
the maximum error according to the standard formula [1] is −26.4% for the simulated
waveforms and −4.81% for the experimental waveforms, and the maximum deviations
determined by the previously developed formula [16] are +1.89% for the simulated wave-
forms and −8.78% for the experimental waveforms. Furthermore, the proposed approach
provides the small deviations of T2 and Up and also falls within the standard tolerances. It
can be concluded that the presented approach is superior to the approaches proposed by
the IEC standard and a previously developed formula for the determination of waveform
parameters of switching impulse voltages from simulations and experiments.

Funding: This research was funded by National Research Council of Thailand.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: The authors would like to give special acknowledgement to the School of
Engineering, King Mongkut’s Institute of Technology Ladkrabang, for providing the facilities for this
research work, and the National Research Council of Thailand for financial support.
Conflicts of Interest: The author declare no conflict of interest.

References
1. IEC 60060-1. High-Voltage Test Techniques–Part 1: General Definitions and Test Requirements, 3rd ed.; IEC: Geneva, Switzerland, 2010.
2. IEC 60060-2. High-Voltage Test Techniques–Part 2: Measuring Systems, 3rd ed.; IEC: Geneva, Switzerland, 2010.
Energies 2022, 15, 4760 10 of 10

3. IEC 60060-3. High-Voltage Test Techniques–Part 3: Definitions and Requirements for On-Site Testing, 1st ed.; IEC: Geneva,
Switzerland, 2006.
4. IEC 60071-1. Insulation Co-Ordination–Part 1: Definitions, Principles and Rules, 8th ed.; IEC: Geneva, Switzerland, 2011.
5. IEEE Standard 4TM-2013. IEEE Standard for High-Voltage Testing Techniques; Institute of Electrical and Electronics Engineers (IEEE):
Piscataway, NJ, USA, 2013.
6. Kuffel, E.; Zaengl, W.S.; Kuffel, J. High Voltage Engineering: Fundamentals, 2nd ed.; Newnes: Oxford, UK, 2000.
7. Zhimin, Y.; Kai, Y.; Shihang, W.; Yajun, Z.; Jianying, L.; Fang, Y. Electrical tree in silicone rubber: Roles of silicone grease and
switching impulses. IEEE Trans. Dielectr. Electr. Insul. 2020, 27, 829–836.
8. Pradhan, A.K.; Tenbohlen, S. Estimation of moisture content in oil-impregnated pressboard through analyzing dielectric response
current under switching impulse. IEEE Trans. Dielectr. Electr. Insul. 2021, 28, 938–945. [CrossRef]
9. Pradhan, A.K.; Tenbohlen, S. A new approach to estimate activation energy of oil-impregnated pressboard stressed under
switching impulse at different temperatures. IEEE Trans. Dielectr. Electr. Insul. 2021, 28, 1162–1170. [CrossRef]
10. IEC 61083-2. Instruments and Software Used for Measurement in High-Voltage and High Current Tests–Part 2: Requirements for Software
for Tests with Impulse Voltages and Currents, 2nd ed.; IEC: Geneva, Switzerland, 2013.
11. Gockenbach, E. Impact of new lightning and switching impulse definitions on the test results for insulation systems. In Proceed-
ings of the 2005 International Symposium on Electrical Insulating Materials, Kitakyushu, Japan, 5–9 June 2005; pp. 792–795.
12. Okabe, S.; Tsuboi, T.; Takami, J. Evaluation of K-factor based on insulation characteristics under non-standard lightning impulse
waveforms. IEEE Trans. Dielectr. Electr. Insul. 2009, 16, 1124–1126. [CrossRef]
13. Pattanadech, N.; Yutthagowith, P. Fast curve fitting algorithm for parameter evaluation in lightning impulse test technique. IEEE
Trans. Dielectr. Electr. Insul. 2015, 22, 2931–2936. [CrossRef]
14. Yutthagowith, P.; Pattanadech, N. Improved Least-Square Prony Analysis Technique for Parameter Evaluation of Lightning
ImpulseVoltage and Current. IEEE Trans. Power Deliv. 2016, 31, 271–277. [CrossRef]
15. Yutthagowith, P.; Kitwattana, K.; Kunakorn, A. Fast and effective technique in evaluation of lightning impulse voltage parameters.
J. Electr. Eng. Technol. 2021, 16, 459–467. [CrossRef]
16. Božidar, F.; Dalibor, F. A new method for estimation of time parameters of standard and non-standard switching impulse voltages.
Int. J. Electr. Power Energy Syst. 2018, 96, 126–131.
17. Andreas, N.; Anders, B.; Jari, H. An improved method for switching-impulse evaluation. In Proceedings of the 2012 Conference
on Precision electromagnetic Measurements, Washington, DC, USA, 1–6 July 2012; pp. 20–21.
18. Chapra, S.C.; Canale, R.P. Numerical Methods for Engineers, 5th ed.; McGraw-Hill: Singapore, 2006.
19. Marquardt, D. An algorithm for least-squares estimation of nonlinear parameters. Siam J. Appl. Math. 1963, 11, 431–441. [CrossRef]
20. User Manual for Digital Impulse Measuring Systems, Transient Recorders and Evaluating Systems, Downloaded in January 2022.
Available online: http://www.strauss-mess.de/dokumente/WinTRAS-KAL-E.pdf (accessed on 8 January 2022).