Electric Power Components and Systems

ISSN: 1532-5008 (Print) 1532-5016 (Online) Journal homepage: https://www.tandfonline.com/loi/uemp20

Detection of Critical Conditions in Ceramic

Insulators Based on Harmonic Analysis of Leakage
Current

Mousalreza Faramarzi Palangar & Mohammad Mirzaie

To cite this article: Mousalreza Faramarzi Palangar & Mohammad Mirzaie (2016) Detection of
Critical Conditions in Ceramic Insulators Based on Harmonic Analysis of Leakage Current, Electric
Power Components and Systems, 44:16, 1854-1864, DOI: 10.1080/15325008.2016.1183723

To link to this article: https://doi.org/10.1080/15325008.2016.1183723

Published online: 09 Sep 2016.

Submit your article to this journal

Article views: 112

View related articles

View Crossmark data

Citing articles: 3 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=uemp20
Electric Power Components and Systems, 44(16):1854–1864, 2016
Copyright C Taylor & Francis Group, LLC

ISSN: 1532-5008 print / 1532-5016 online

DOI: 10.1080/15325008.2016.1183723

Detection of Critical Conditions in Ceramic Insulators

Based on Harmonic Analysis of Leakage Current
Mousalreza Faramarzi Palangar and Mohammad Mirzaie
Department of Electrical and Computer Engineering, Babol University of Technology, Babol, Iran

CONTENTS
Abstract—This article presents an online diagnostics method to pre-
1. Introduction dict critical conditions in ceramic insulators. For this purpose, a va-
2. Experimental Technique riety of experimental tests have been conducted on different kinds
of ceramic insulators under different environmental conditions in or-
3. Measurement Results der to analyze their leakage current. Fast Fourier transform analysis
4. Analysis of the Leakage Current Harmonic Components in of leakage currents shows a strong correlation between the ratio of
the Tested Insulators the third to fifth harmonic amplitude regarding occurrence of criti-
cal conditions and degree of insulator ageing. Therefore, the index
5. Discussion on the Function Range of the Index (R 3/5 ) and AOS
of the third to fifth harmonic amplitude is utilized as the detection
6. Conclusions indicator of critical conditions in ceramic insulators. An increase in
References the index of the third to fifth harmonic amplitude to more than one
represents a critical condition in the insulators and estimates a high
probability of flashover occurrence. The obtained results have also
shown the viability of the index in insulators monitoring procedures.
Training the proposed index to a microprocessor powered device
gives ability to the system for online diagnosis of ceramic insulator
qualifications, detecting critical conditions, and predicting flashover
occurrence.

Keywords: leakage current, harmonic analysis, critical condition, ceramice
insulators
Received 26 December 2014; accepted 7 February 2016
Address correspondence to Mousalreza Faramarzi Palangar, Department of
Electrical and Computer Engineering, Babol University of Technology,
Babol 25641, Iran. E-mail: r faramarzi@stu.nit.ac.ir
Color versions of one or more of the figures in the article can be found online
at www.tandfonline.com/uemp.
1. INTRODUCTION
Insulators are one of the main components of overhead power
transmission lines with the task of insulating the lines from
the tower body (earth) and also withstanding mechanical bur-
den of them. One of the main challenges of insulators is the
effect of pollution on their expected performance. Pollution
and humidity, during the time, cause changes in the electrical
behavior of insulators and also result in increase of leakage
current on the insulator surface. This also can provide elec-
trical discharge and finally flashover in the insulators surface
[1–8].
Several researchers have investigated the behavior of the
polluted insulators and presented various models for them. In
[1], a polluted insulator model has been proposed in which
the polluted insulator is modeled as two series of resistors
including polluted layer resistance and dry band resistance.
Then, a mathematical model was defined for the critical volt-
age with respect to the insulator profile and amount of pollu-
tion. A circuit model for the polluted insulator in [2] is as a
resistor in parallel with a capacitor. According to the circuit

1854
 Palangar and Mirzaie: Detection of Critical Conditions in Ceramic Insulators Based on Harmonic Analysis of Leakage Current 1855

model of the polluted insulator, a new mathematical model

has also been surveyed for critical parameters (e.g., distance
of the critical electrical arc, critical leakage current, and critical
voltage).
Moreover, there are variety methods in which detection
of the insulator conditions and prediction of flashover have
been considered [9–26]. In [9], peak amplitude changes of the
leakage current has been assumed as a criterion to detect the FIGURE 1. A schematic of the experimental setup.
critical conditions in insulators. So, using the leakage current
measurements in various conditions, the critical magnitude of
conditions of insulators, unlike the normal conditions, the ratio
the leakage current, in which the probability of flashover is
of the R3/5 is increased. Therefore, the index range in which the
high, has been selected as an identifier. The authors in [10],
probability of flashover is high, has been determined. Increas-
have recently offered a system that uses an antenna to capture
ing in the index R3/5 to more than one illustrates the abnor-
the electromagnetic radiation of partial discharges along with
mal function of the insulator. By analyzing the data obtained
leakage currents. Although this system has not been damaged
from the leakage current signal, which is recorded in critical
by flashover, it has not been tested on the field yet, where the
conditions, the critical range of the index resulting flashover is
electromagnetic interference generated by corona and other
when the third harmonic of the leakage current becomes larger
effects on the high-voltage conductors is expected to be large.
than the fifth harmonic. This indicator can be utilized to detect
Authors have proposed an online pollution leakage current
and predict critical conditions in ceramic insulators online by
monitoring system on insulators in transmission lines, to mea-
using an automated online system (AOS) before a flashover
sure the leakage current using current transformers [11]. In this
occurrence.
respect, many authors have examined how to predict flashover
in the composite insulator [12–14]. In [15], flashover has been
predicted by using the optical sampling of leakage current sig-
nal. In this paper, an optical monitoring system that stores rel- 2. EXPERIMENTAL TECHNIQUE
evant information of the leakage current have been developed. 2.1. Experimental Setup
Optical sensing of the leakage current and optical transmission
All experiments have been carried out in an artificial climatic
of the waveform signal to a processing module eliminates the
chamber (ACC) with dimensions of 1.5 m × 1.5 m × 1.8 m.
effects of electromagnetic interference. The leakage current of
This chamber has been made of a wooden structure with thick
polluted insulators is measured under different environmental
and transparent plastic-sheeted walls and Standard IEC60507
conditions and then curves of leakage current varying with
[25] standard design nozzles, which can be used to produce
time are plotted. Moreover, the effective factors of the wet-
clean fog. In order to perform experimental tests, measure
ting process on the polluted layer have been analyzed [16].
leakage current, and record the data, the experimental setup is
Artificial neural networks (ANNs) have been utilized in issues
used according to Figure 1.
requiring estimation, prediction, pattern recognition, and clas-
This setup, which is provided according to Standard
sification, etc. In the field of high-voltage insulators, ANNs are
IEC60507, includes a clean fog chamber, a high-voltage trans-
utilized to predict a flashover [18, 19], analyze surface tracking
former and a leakage current meter system. For measuring the
on polluted insulators [20], critical flashover voltage estima-
applied voltage to the insulator, a capacitor divider is utilized.
tion in polluted insulators [21], and forecast leakage current in
The leakage current meter system and oscilloscope are pro-
silicon rubber insulators [22].
tected against the increase of current and voltage. Sampling
In this article, a diagnostic index, as the ratio of the third to
of the leakage current signal is carried out by a digital oscil-
fifth harmonic amplitude (R3/5 ), of leakage currents has been
loscope and the data are stored as a comma separated values
presented for online detection of ceramic insulator conditions.
(CSV) file and transferred to the MATLAB (The MathWorks,
Experimental tests have been performed on ceramic insulators
Natick, Massachusetts, USA) software to analyze.
to evaluate the ability of the proposed index for recognizing
the operating condition of insulators. This indicator makes it
possible to predict the critical conditions of ceramic insulators 2.2. Tested Insulators
before a flashover occurrence. Through conducting several ex- Figure 2 shows tested insulators. The insulator samples, the
periment tests on the leakage current signals and analyzing the characteristics of which are reported in Table 1, are allocated
data by MATLAB software, it is demonstrated that in critical to be tested under different artificial pollutions.
1856 Electric Power Components and Systems, Vol. 44 (2016), No. 16

FIGURE 3. Image of insulator in polluted and clean state.

of light, medium, heavy, and very heavy. Figure 3 shows an

insulator after settling pollution on its surface.

2.4. Leakage Current Measurment

For each insulator, leakage current is measured in different lev-
els of pollution and dry and wet conditions under AC voltage.
Under dry conditions, tests have been carried out in environ-
mental humidity, and in wet conditions, the insulator is exposed
to a voltage in different humidities which is provided by the
fog generator. Figure 4 shows a view of the test objects used
in this research.
FIGURE 2. Tested insulators.

2.3. Artificial Pollution 3. MEASUREMENT RESULTS

Before testing, the insulator surfaces are cleaned by washing In the performed measurements, leakage current signals are
with isopropyl, alcohol, and rinsing with distilled water in or- stored as CSV file by the oscilloscope. Sampling includes
der to remove any trace of dirt and grease. To reproduce saline 4000 points of leakage current signal in the time domain. This
pollution typical of coastal areas, a contamination layer con- data is then transferred to MATLAB software and complete
sisting of NaCl and 40 g of kaolin mixed with 1 L of deionized features of the leakage current waveforms are extracted.
water is applied to the surface of the insulator. To pollute the
surface of the porcelain and glass insulators, the solid-layer
method is applied. This method, in accordance with Standard
IEC60507, is made to create uniform artificial pollution. The
standardized pollution levels are divided into four categories

Characteristics

H L D Electromechanical
Insulator (mm) (mm) (mm) failing load (KN)
A 146 305 255 120
B 146 320 255 160
C 178 546 321 210
D 146 320 255 70
E 176 530 280 210
F 146 360 280 290
FIGURE 4. Image of experimental setup (1—capacitive di-
vider, 2—HV resistance, 3—HV transformer, 4—fog maker,
TABLE 1. Characteristics of the tested insulators 5—humidity meter, 6—chamber, 7—suspension insulator).
 Palangar and Mirzaie: Detection of Critical Conditions in Ceramic Insulators Based on Harmonic Analysis of Leakage Current 1857

FIGURE 5. Leakage current waveforms and FFT of Insulator 1 in various pollutions at humidity of 77%: (a) clean, (b) light, (c) medium,
(d) heavy, and (e) very heavy.

Each insulator is exposed to different pollution and hu-
midity levels under the applied AC voltage of 50 Hz. As an
example, Figure 5 shows leakage current waveforms and their
fast Fourier transform (FFT) for Insulator 1 at a humidity of
77% and different pollution levels. In increasing in the amount
of pollution, which leads to enhance surface electrical con-
ductivity, the critical voltage level decreases. In other words,
because of the salt solution, the surface electrical conductivity
increases and more leakage current passes on the insulator sur-
face. An increase in the surface leakage current causes more
electrical discharges. Humidity also has a great effect on the
increase of these electrical discharges. In Figure 6, the leakage
1858 Electric Power Components and Systems, Vol. 44 (2016), No. 16

current waveform and their FFT in Insulator 1 are shown in

heavy pollution level and under different humidities.
The results show that the value of the fundamental harmonic
component of the leakage current increases considerably with
an increase in humidity. By applying fog and increasing hu-
midity, the amount of the third harmonic starts to increase and
becomes larger than the fifth harmonic. The effect of an in-
crease in humidity on the leakage current waveforms is shown
in Figure 7.

3.1. Leakage Current Waveforms During Flashover

Figure 8 presents typical changes in the leakage current in
the event of flashover occurrence against increasing voltage.
Flashover occurs as the following four steps.

Step 1: Leakage current waveform seems like sine waves at

the beginning of applying the voltage due to resistive
current flow (almost wet surface layer, no dry bands,
no discharge).
Step 2: The tips of triangular waveform become sharper and
lengthen when some visible point discharges occur
[26], with a continuous sound near the cap on the up-
per surface (dry band formation occurs during stages
1 to 3).
Step 3: The leakage current waveform becomes similar to the
symmetrical wave in the presence of linear filamen-
tary weak local arcs (streamer or partial discharge
[26]) in five or six spots near the cap on the upper
surface.
Step 4: In addition to the symmetrical waves at Step 3 in
the presence of intermittency, leakage current with a
large peak value appears with strong local arcs on the
bottom surface.

According to the experimental tests, the results of the leak-

age current harmonic measurements are shown in Tables 2–7
in different levels of humidity for the tested insulators. The re-
sults of the previously mentioned measurements demonstrate
that the under wet pollution, the value of the third harmonic
of the leakage current increases remarkably unlike the clean
state. With growth in wet pollution, the third harmonic in-
creases to the point which goes beyond the fifth harmonic, it at
this time the electrical discharge occurs. A growth in humidity
has a considerable effect on growth of the third harmonic of
the leakage current and behavior variations of the insulator.
It is seen that in high-humidity levels, due to complete wet-
ness, electrical conductivity reduction and liquid drops flow FIGURE 6. Leakage current waveforms and FFT of Insulator
on the surface, the leakage current is more stable. A rise in 1 in heavy pollution and in various levels of humidity: (a) 77%,
the amplitude and disorder of total harmonic distortion (THD) (b) 85%, and (c) 91%.
 Palangar and Mirzaie: Detection of Critical Conditions in Ceramic Insulators Based on Harmonic Analysis of Leakage Current 1859

FIGURE 7. Leakage current waveforms versus different levels

of humidity for Insulator 3.

of the leakage current waveform, in different pollutions com-

pared to clean mode, is an indication of pollution impact on
the electrical behavior of insulators.
After completing the leakage current measurement process
in the tested insulators, in order to determine their equiva-
lent salt deposit density (ESDD), the operation of sweeping
pollution off the surface is carefully conducted according to
Standard IES60507. The electrical conductivity of the solution
is measured. In Table 8, the results of electrical conductivity
and ESDD are shown for the tested insulators.

4. ANALYSIS OF THE LEAKAGE CURRENT

HARMONIC COMPONENTS IN THE TESTED
INSULATORS
In this section the analysis of results is discussed.
4.1. Clean Insulators
In a clean state, flashover is not seen in all tests. The results
indicate that the value of the fundamental harmonic component
of the leakage current increases significantly with respect to
the humidity augmentation. However, this increase is not so
significant in other harmonics. In clean conditions, the fifth
harmonic of the leakage current is always larger than the third
harmonic. Thus, it can be expressed that when the value of the
fifth harmonic is higher than the third, the insulators are in a
normal working state.
FIGURE 8. Shift of leakage current waveforms onto flashover
occurrence for Insulator 2 (medium pollution and humidity =
85%).
4.2. Polluted Insulators
In light pollution, the value of the fundamental harmonic of the
leakage current decreases primarily, but by applying of fog and
humidity, the value of the fundamental harmonic increases and
then decreases again. Because of the wetness of the insulator
surface and dissolution of soluble salts, surface conductivity
increases suddenly, and finally, due to flow of the leakage
current and creation of a dry band on the insulator surface, a
1860 Electric Power Components and Systems, Vol. 44 (2016), No. 16

Third Fifth Third Fifth

Severity harmonic harmonic R3/5 Severity harmonic harmonic R3/5
Humidity pollution (mA) (mA) (%) Humidity pollution (mA) (mA) (%)
Ambient, 64% Clean 0.0044 0.242 1.7 Ambient, 64% Clean 0.021 0.195 10.84
Light 0.295 0.387 76.28 Light 0.119 0.413 28.83
Medium 0.3 0.316 96.36 Medium 0.112 0.541 20.72
77% Heavy 0.295 0.247 119.3 77% Heavy 0.512 0.49 104.4
Very heavy 0.21 0.3 70.7 Very heavy 0.345 0.328 105.1
Light 0.204 0.298 68.42 Light 0.041 0.731 5.69
Medium 0.229 0.201 113.9 Medium 0.492 0.458 107.4
85% Heavy 0.259 0.194 133.5 85% Heavy 0.531 0.527 100.7
Very heavy 0.264 0.228 115.7 Very heavy 0.391 0.399 99.7
Light 0.176 0.3 58.58 Light 0.037 0.733 5.16
Medium 0.124 0.205 60.54 Medium 0.063 0.595 10.57
91% Heavy 0.239 0.339 70.7 91% Heavy 0.426 0.402 105.9
Very heavy 0.247 0.221 111.4 Very heavy 0.425 0.424 100.2

TABLE 2. Value of the third and fifth harmonics of leakage current TABLE 4. Value of the third and fifth harmonics of leakage current
and their ratio in Insulator 1 and their ratio in Insulator 3

slight discharge occurs on the surface. In some cases, electrical
discharges cause an electrical arc. After producing fog and
completing wetness of the insulator surface, the salt dissolves
in humidity. On the other hand, this humidity appears as small
water droplets on the surface dripping down on the edges of
the shed. This phenomenon causes the reduction of salt on the
surface and, consequently, decreases surface conductivity and
makes the insulator more stable.
In polluted insulators and in no-fog condition, the fifth har-
monic of the leakage current is always larger than the third
harmonic. This shows normal working state of the insulator. In
fact, the mentioned criterion for detection of normal working
condition in clean insulators is also applicable in polluted in-
sulators. By applying fog and increasing humidity, the amount
of the third harmonic starts to increase and gets larger than
the fifth harmonic. Although both the third and fifth harmon-

Third Fifth Third Fifth

Severity harmonic harmonic R3/5 Severity harmonic harmonic R3/5
Humidity pollution (mA) (mA) (%) Humidity pollution (mA) (mA) (%)
Ambient, 64% Clean 0.026 0.243 11.05% Ambient, 64% Clean 0.055 0.261 21.56
Light 0.216 0.4 53.95 Light 0.14 0.143 97.79
Medium 0.257 0.303 80 Medium 0.176 0.324 54.46
77% Heavy 0.313 0.296 105.7 77% Heavy 0.181 0.247 73.33
Very heavy 0.282 0.262 107.6 Very heavy 0.029 0.313 9.3
Light 0.101 0.365 27.69 Light 0.097 0.109 89.5
Medium 0.496 0.476 104.2 Medium 0.358 0.343 104.3
85% Heavy 0.371 0.344 107.8 85% Heavy 0.084 0.18 46.73
Very heavy 0.545 0.454 120 Very heavy 0.234 0.179 130.7
Light 0.062 0.237 26.21 Light 0.147 0.133 110.58
Medium 0.032 0.443 7.4 Medium 0.199 0.32 62.13
91% Heavy 0.382 0.341 112 91% Heavy 0.188 0.173 108.6
Very heavy 0.401 0.366 109.5 Very heavy 0.211 0.196 107.6

TABLE 3. Value of the third and fifth harmonics of leakage current TABLE 5. Value of the third and fifth harmonics of leakage current
and their ratio in Insulator 2 and their ratio in Insulator 4
 Palangar and Mirzaie: Detection of Critical Conditions in Ceramic Insulators Based on Harmonic Analysis of Leakage Current 1861

Third Fifth Kaolin ESDD Conductivity Pollution

Severity harmonic harmonic R3/5 Insulator (gr)/salt(gr) (mg/cm2) (µs/cm) severity
Humidity pollution (mA) (mA) (%)
1 40/10 0.0148 76 Light
Ambient 64% Clean 0.029 0.313 9.3 2 40/10 0.0214 116 Light
3 40/10 0.0086 102 Light
Light 0.081 0.34 23.96
4 40/10 0.0231 104 Light
Medium 0.035 0.488 7.36
5 40/10 0.01 109 Light
77% Heavy 0.298 0.275 108.3
Very heavy 0.561 0.546 102.7 6 40/10 0.022 122 Light
1 40/40 0.0673 330 Medium
Light 0.1 0.319 31.41
2 40/40 0.0668 350 Medium
Medium 0.469 0.463 101.2
3 40/40 0.0275 316 Medium
85% Heavy 0.281 0.239 117.5
4 40/40 0.078 340 Medium
Very heavy 0.44 0.416 105.7
5 40/40 0.03 298 Medium
Light 0.125 0.279 44.87
6 40/40 0.0798 412 Medium
Medium 0.084 0.465 18.16
1 40/80 0.0888 432 Heavy
91% Heavy 0.051 0.243 17.29
2 40/80 0.0953 494 Heavy
Very heavy 0.674 0.657 102.5
3 40/80 0.0446 505 Heavy
4 40/80 0.0986 425 Heavy
TABLE 6. Value of the third and fifth harmonics of leakage current 5 40/80 0.0516 503 Heavy
and their ratio in Insulator 5
6 40/80 0.114 583 Heavy
1 40/100 0.09 438 Very heavy
2 40/100 0.107 557 Very heavy
ics increase simultaneously, but the growth rate of the third
3 40/100 0.045 508 Very heavy
harmonic prevails over the fifth harmonic. In fact, as the third 4 40/100 0.107 463 Very heavy
harmonic becomes higher than the fifth, the insulator condi- 5 40/100 0.057 554 Very heavy
tions are critical and the probability of flashover increases. 6 40/100 0.128 655 Very heavy
As mentioned, changes of the third and fifth harmonics
of the leakage current can be the basis of the judgment TABLE 8. Electrical conductivity and ESDD value at 20◦ C
about the function of porcelain and glass insulators. Ac-
cording to the experimental tests, it can be stated that the
growth of the third harmonic is due to presence of pollu-
tion and humidity on the insulator. In this article, regard-
Third Fifth less of the amount of humidity and environmental pollution,
Severity harmonic harmonic R3/5 and only by measuring the leakage current and the value of
Humidity pollution (mA) (mA) (%) the third and fifth harmonics, a detection criterion for de-
Ambient, 64% Clean 0.028 0.328 8.62 scribing the status of porcelain and glass insulators has been
presented.
Light 0.049 0.195 25.32
Medium 0.347 0.367 94.55
77% Heavy 0.442 0.433 102.07
5. DISCUSSION ON THE FUNCTION RANGE OF
Very heavy 0.394 0.383 102.8
THE INDEX (R 3/5 ) AND AOS
Light 0.066 0.239 27.73
Medium 0.271 0.289 93.77 5.1. Index (R3/5 )
85% Heavy 0.474 0.428 110.7 As previously mentioned, by using leakage current analysis
Very heavy 0.395 0.39 101.2 and harmonic component measurement of clean insulators in
Light 0.086 0.28 30.82 normal working conditions, it is determined that in this condi-
Medium 0.084 0.288 29.23 tion for all insulators, the value of the fifth harmonic is more
91% Heavy 0.293 0.329 89.05 than the third harmonic of the leakage current. Therefore, this
Very heavy 0.343 0.329 104.2 criterion can be selected to detect the normal function con-
ditions of insulators. Whenever this condition changes, the
TABLE 7. Value of the third and fifth harmonics of leakage current insulator exits from its normal working state and enters into
and their ratio in Insulator 6 critical position.
1862 Electric Power Components and Systems, Vol. 44 (2016), No. 16

Voltage level (kV)

30 42 50
Insulator R3/5 (%)
1 82.8 113.6 110.2
2 101.6 105.7 103.7
3 104.3 112.4 106.3
4 92.6 73.33 118.4
5 107.3 108.3 102.5
6 103.7 121.07 121.3

TABLE 9. Values of indicator (R3/5 ) under different voltages at heavy

pollution and humidity of 77%

Through the presented results of the 78 tests performed un-

der different conditions of pollution and humidity, in 34 cases,
the value of the third harmonic has been larger than the fifth
harmonic. In some cases with heavy pollution, a flashover oc-
currence was expected, but it did not happen. The response
to this special event can be found in non-linear property of
pollution. In higher humidity, due to complete wetting of sur-
face which causes the solution to slow down the sheds and
reduce electrical conductivity on the surfaces, the probabil-
ity of electrical discharge becomes less compared with the
lower humilities and the insulators would be in stable condi-
tion. Among these 34 cases in which the third harmonic is
larger than the fifth harmonic, in 28 cases, a flashover has
frequently happened. Generally, in porcelain and glass insu-
lators in which flashover occurred, their third harmonic has
been higher than the fifth harmonic. Hence, it can be stated
that when a flashover occurs, the introduced index will defi-
nitely be more than one (R3/5 > 1 or 100%). Also, when the
value of the index is higher than one, flashover occurs with
probability of 82.3% (i.e., 28/34). Therefore, in all conditions
that the value of the third harmonic is higher than the fifth, a
flashover will not occur. Nevertheless, continuing work in con-
ditions will be critical in insulators. An increase in the applied
voltage, under pollution and humidity, has a great effect on the
change of normal working conditions. Table 9 shows the effect
of voltage increase on the indicator for the tested insulators,
in the constant pollution and humidity. The cases in which the
value of the third harmonic has been larger than the fifth har-
monic have been extracted, and in 15 cases this phenomenon
has occurred. Among these 15 cases, which have mostly been
in high voltages, there were 12 cases where a flashover has
been happened. In this test, the probability of a flashover for
18 test cases is 80% (i.e., 12/15). Therefore, in high volt-
ages, the probability of a flashover occurrence increases sig-
nificantly. This illustrates the effect of voltage growth on the
increase in particle discharges on the surface of insulators.
FIGURE 9. Flowchart of the AOS performance.
5.2. AOS for Detection of Critical Conditions
Training the proposed harmonic index to a microprocessor-
powered device gives ability to the system to log the insulator
qualifications, detect the faulty ones, and alarm in the criti-
cal conditions. The logger function in this system prepares a
periodic monitoring program with programmable time table
for suspicious and dusty power lines. With this effort, repair
and maintenance teams could take precautions such as repair-
ing, cleaning, or changing the faulty insulators. Following,
is a description of how the online automated system works:
first, a measurement device is easily mounted on the power
line insulator via an insulation reed. A remote control has
been prepared for the current clamp. Next, the system starts
the noise cancelation and external magnetic field suppression
steps. Precious analogue amplifiers and filters along with dig-
ital signal processor (DSP) techniques are the main engine in
this high-resolution measurement device. After clearly cancel-
ing external noises, the leakage current is read with a 10-kHz
sampling rate and current harmonics are calculated in the sys-
tem. Finally, the insulator condition is measured regarding to
 Palangar and Mirzaie: Detection of Critical Conditions in Ceramic Insulators Based on Harmonic Analysis of Leakage Current
the proposed index. Figure 9 shows flowchart of the AOS per-
formance utilized to detect insulators conditions.
6. CONCLUSIONS
This article aims to analyze functional status of porcelain and
glass insulators under pollution and humidity conditions and
finally presents a criterion to detect critical conditions in porce-
lain and glass insulators. Through the actions taken, the results
are listed as follows.
1) In a clean state, the value of the fifth harmonic is always
more than the third harmonic. In the other words, if
the value of the fifth harmonic is greater than the third
harmonic of the leakage current, the insulator will be in
a normal working state.
2) In dry pollution and normal working conditions, the
value of the fifth harmonic is still higher than the third.
3) In polluted insulators, particularly in medium and heavy
pollutions with fog and humidity, the insulator exits from
its normal working state. In this situation, the value of the
third harmonic becomes bigger than the fifth harmonic
and increase in the third harmonic is faster and higher
than the fifth harmonic. Accordingly, the insulator state
becomes critical and the probability of a flashover oc-
currence increases.
4) By analyzing leakage current harmonics, the index
(R3/5 ), and AOS, the functional conditions and flashover
occurrence in ceramic insulators can be predicted; while
this prediction is not affected by insulator profile, ESDD,
or the amount of humidity.
REFERENCES
[1] Obenaus, F., “Fremdschichtüberschlag und kriechweglänge
(Contamination flashover and creepage path length),” Deutsche
Electrotechnik, Vol. 4, pp. 135–136, 1958.
[2] Dhahbi-Megriche, N., Beroual, A., and Krähenbühl, L., “A new
proposal model for polluted insulators flashover,” J. Phys. D
Appl. Phys., Vol. 30, pp. 889–894, 1997.
[3] El-A. Slama, M., Beroual, A., and Hadi, H., “Analytical compu-
tation of discharge characteristic constants and critical param-
eters of flashover of polluted insulators,” IEEE Trans. Dielectr.
Electr. Insulat., Vol. 17, No. 6, pp. 1764–1771, 2010.
[4] Shu, L., Shang, Y., Jiang, X., Hu, Q., Yuan, Q., Hu, J., Zhang,
Z., Zhang, S., and Li, T., “Comparison between AC and DC
flashover performance and discharge process of ice-covered in-
sulators under the conditions of low air pressure and pollution,”
IET Gener. Transm. Distrib., Vol. 6, No. 9, pp. 884–892, 2012.
1863
[5] Topalis, F. V., Gonos, I. F., and Stathopulos, I. A., “Dielectric
behaviour of polluted insulators,” Proc. IEE Gener. Transm.
Distribut., Vol. 148, pp. 269–274, 2001.
[6] Dong, B., Jiang, X., Hu, J., Shu, L., and Sun, C., “Effects of ar-
tificial polluting methods on AC flashover voltage of composite
insulators,” IEEE Trans. Dielectr. Electr. Insulat., Vol. 19, No.
2, pp. 714–722, 2012.
[7] Zhang, Z., Liu, X., Jiang, X., Hu, J., and Gao, D. W., “Study
on AC flashover performance for different types of porcelain
and glass insulators with non-uniform pollution,” IEEE Trans.
Power Del., Vol. 28, No. 3, 2013.
[8] Douar, M. A., Mekhaldi, A., and Bouzidi, M. C., “Flashover
process and frequency analysis of the leakage current on in-
sulator model under non-uniform pollution conditions,” IEEE
Trans. Dielect. Electr. Insulat., Vol. 17, No. 4, pp. 1691–1698,
2010.
[9] Montoya, G., Ramirez, I., and Montoya, J. I., “Correlation
among ESDD, NSDD and leakage current in distribution in-
sulators,” IEE Proc. Gener. Transm. Distrib., Vol. 151, No. 3,
pp. 1284–1297, pp. 334–340, 2004.
[10] Youssef, M., and El-Alayly, A. A., “Remote monitoring of sur-
face current over insulators,” Proc. IEEE Transm. Distrib. Conf.,
Vol. 1, pp. 113–116, 2001.
[11] Shihab, S., Melik, V., Zhou, L., Melik, G., and Alame, N., “On-
line pollution leakage current monitoring system,” Proceedings
of IEEE Properties and Application of Dielectric Materials, pp.
538–541, 3–8 July 1994.
[12] Kordkheili, H. H., Abravesh, H., Tabasi, M., Dakhem, M., and
Abravesh, M. M., “Determining the probability of flashover
occurrence in composite insulators by using leakage current
harmonic components,” IEEE Trans. Dielectr. Electr. Insulat.,
Vol. 17, No. 2, pp. 502–512, 2010.
[13] Volat, C., Jabbari, M., and Farzaneh, M., “New method for in
live-line detection of small defects in composite insulator based
on electro-optic E-field sensor,” IEEE Trans. Dielectr. Electr.
Insulat., Vol. 20, pp.194–201, 2013.
[14] Bashir, N., and Ahmad, H., “Odd harmonics and third to fifth
harmonic ratios of leakage currents as diagnostic tools to study
the ageing of glass insulators,” IEEE Trans. Dielectr. Electr.
Insulat., Vol. 17, No. 3, pp. 819–832, 2010.
[15] Fontana, E., Campello Oliveira, S., do Monte de Melo Caval-
canti, F. J., Bandeira Lima, R., Martins-Filho, J. F., and Meneses-
Pacheco, E., “Novel sensor system for leakage current detection
on insulator strings of overhead transmission lines,” IEEE Trans.
Power Del., Vol. 21, No. 4, pp. 2064–2070, 2006.
[16] Zhang, Z., Jiang, X., Huang, H., Sun, C., Hu, J., and Gao, D.
W., “Study on the wetting process and its influencing factors
of pollution deposited on different insulators based on leakage
current,” IEEE Trans. Power Del., Vol. 28, No. 2, pp. 678–685,
2013.
[17] Suwarno, “Study on the waveform of leakage current on the
20kV postpin ceramic insulators under various conditions,”
IEEE International Symposium Electrical Insulating Materi-
als, pp. 387–390, Himeji, 2001.
[18] Samimi, M. H., Mostajabi, A. H., Ahmadi-Joneidi, I.,
Shayegani-Akmal, A. A., and Mohseni, H., “Performance evalu-
ation of insulators using flashover voltage and leakage current,”
Electr. Power Compon. Syst., Vol. 41, No. 2, pp. 221–233, 2013.
1864 Electric Power Components and Systems, Vol. 44 (2016), No. 16
[19] Clines, P., Lannes, W., and Richards, G., “Use of pollution
monitors with neural network to predict insulator flashover,”
Electr. Power Syst. Res., Vol. 42, pp. 27–33, 1997.
[20] Al Alawi, S., Salam Dr., M. A., Maqrashi, A. A., and Ahmad,
H., “Prediction of flashover voltage of contaminated insulator
using artificial neural networks,” Electr. Power Compon. Syst.,
Vol. 34, No. 8, pp. 831–840, 2006.
[21] Ugur, M., Auckland, D. W., Varlow, B. R., and Emin, Z., “Neural
networks to analyze surface tracking on solid insulators,” IEEE
Trans. Dielectr. Electr. Insulat., Vol. 4, No. 10, pp. 763–766,
1997.
[22] Kontargyri, V. T., Gialketsi, A. A., Tsekouras, G. J., Gonos, I. F.,
and Stathopulos, I. A., “Design of an artificial neural network
for the estimation of the flashover voltage on insulators,” Electr.
Power Syst.Res., Vol. 77, No. 12, pp. 1532–1540, 2007.
[23] Jahromi, N., El-Hag, A. H., Cherney, E. A., Jayaram, S.,
SanayePasand, M., Mohseni, H., “Prediction of leakage current
of composite insulators in salt fog test using neural network,”
IEEE Conference Electrical Insulation Dielectric. Phenomena
(CEIDP), pp. 309–312, 16–19 October 2005.
[24] Gouda, O. E., and El Dein, A. Z., “Laboratory simulation of
naturally polluted high-voltage transmission line insulators.”
IET Generat. Transm. Distrib., Vol. 8, No. 2, pp. 321–327,
2014.
[25] IEC, “Artificial pollution tests on high-voltage insulators to be
used on A.C. systems,” IEC Standard 60507, 1991.
[26] Nasser, E., “Some physical properties of electrical discharges
on contaminated surfaces,” IEEE Trans. Power Apparat. Syst.,
Vol. 87, No. 4, pp. 957–963, 1968.
BIOGRAPHIES
Mousalreza Faramarzi Palangar was born in Mashhad, Iran,
in 1989. He received his B.Sc. in electrical engineering from
Urmia University, Urmia, Iran, and his M.Sc. in electrical
power engineering and area of high-voltage engineering from
the Babol University of Technology, Babol, Iran, in 2015. His
research interests include condition monitoring of high-voltage
equipment, insulation performance, design and control detec-
tion system for protection of power system components, and
wireless power transfer.
Mohammad Mirzaie was born in Ghaem Shahr, Iran, in 1975.
He obtained his B.Sc. and M.Sc. in electrical engineering from
University of Shahid Chamran, Ahvaz, Iran, and Iran Univer-
sity of Science and Technology, Tehran, Iran, in 1997 and 2000,
respectively, and his Ph.D. in electrical engineering from the
Iran University of Science and Technology in 2007. He worked
as an assistant professor in the Electrical and Computer Engi-
neering Faculty of Babol University of Technology since 2007.
His research interests include life management of high-voltage
equipment; high-voltage engineering; intelligence networks
for internal faults assessment in equipment; and studying in-
sulation systems in transformers, cables, generators, breakers,
insulators, electrical motors, and overhead transmission lines.