IRJET- Combined Dissolved Gas Analysis: A Prescient Methodology for Recognizing Faults in Transformer by MATLAB GUI

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 11 | Nov 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 981
Combined Dissolved Gas Analysis: A Prescient Methodology for
Recognizing Faults in Transformer by MATLAB GUI
Simon Babukutty1, S.S. Khule2
1PG Student, Electrical Engineering Dept., Matoshri College of Engineering and Research Centre Nashik, India
2HOD, Electrical Engineering Dept, Matoshri College of Engineering and Research Centre Nashik, India
----------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The transformer is one of the most important
equipment playing a significant role in Power transmission
network. For utilities which require a reliable power supply in
the long run, utmost attention on transformers is required.
The mineral oil & insulation inside the transformer is subject
to high thermal and electrical stresses and thus gasses are
formed due to the decomposition of the mineral oil&cellulose.
Transformer failure will cause huge loss to Industries & thus
as a proactive approach transformers diagnosis of mineral oil
must be carried out. Combined Dissolved gas analysis (DGA)is
used for assessing oil for dissolved gases formed due to faults
in the transformer. Five Classical methods in DGA are Key Gas
Method, IEC Ratio method, RogersRatioMethod, Doernenburg
Ratio Method and Duval triangle Method. This paper presents
Combined Dissolved Gas Analysis withfiveclassicalmethodsof
100% accuracy when compared to a reliable individual
method of 90% accuracy.
Key Words: DGA, mineral oil, transformer, Fault Analysis,
Gas, MATLAB GUI.
1. INTRODUCTION
Power Transformers are one of the most expensive
equipment for Utilities. They are operated24*7byutilities&
industries. For reliablepowersupplycontinuousmonitoring,
routine maintenance & testing is essential. During
continuous operation and aging, the transformer insulating
materials are subjected to electrical and thermal stresses.
These stresses lead to breakdown of insulation and several
gases are released. These gases help in identifying the type
and severity of fault. Total Combustible Gas analysis, Gas
Blanket analysis and dissolved gas analysis are the three
different methods for detecting gases. Out of these three
methods Dissolved Gas Analysis is proved as most accurate
method for condition assessment of transformers.
Most of the power utilities carry out DGA analysis by only
one method preferably Key Gas Method.Thispaperpresents
computer based MATLAB GUI, developed to combine five
DGA interpretation techniques thereby improving the
accuracy whencomparedto mostreliableindividual method.
2. FORMATION OF GASES IN TRANSFORMER OIL
Gas is formed in mineral oil due to decomposition of
cellulose and oil. Decomposition of cellulose and oil takes
place due to thermal fault and electrical fault. Electrical
faults occur due to Corona (Partial discharge) or Arcing
(High energy discharge). Gases formed are CH4, H2, C2H6,
C2H4, C2H2, CO2, CO etc. The percentage amount of gases
formed in transformer depends on type of fault and its
source of formation. Table 1 [4] [3] gives a dissolved gas
concentration permissiblelimitwhichdeterminesthehealth
of the transformer.
Table -1: Dissolved Gas concentration limit
Gases (in
PPM)
Normal Caution Abnormal Danger
Hydrogen 100 101-700 701-1800 >1800
Methane 120 121-400 401-1000 >1000
Acetylene 35 36-50 51-80 >80
Ethylene 50 51-100 101-200 >200
Ethane 65 66-100 101-150 >150
Carbon
Monoxide
350 351-570 571-1400 >1400
Carbon Dioxide 2500 2501-4000 4001-10000 >10000
TDCG 720 721-1920 1921-4630 >4630
3. TRANSFORMER FAULT DIAGNOSIS BY
CLASSSICAL METHODS
One of the main advantages of Dissolved Gas Analysis over
other gas identification methods is it gives early
identification of incipient fault. Fig1 shows the steps
involved in DGA [7]-[11].
Fig -1: DGA Techniques Flow Chart

The oil sample is collected from the transformer unit. From
this oil sample, the dissolved gas is extracted. The
concentration of different gasesinppmisevaluatedfrom the
extraction. The concentration values of different gases are
analyzed for incipient fault. The five classical methods for
detection of incipient faults are Key Gas Method, IEC Ratio
Method, Rogers Ratio Method, Dornenburg Method, Duval
Triangle.
3.1 Key Gas Method
Key Gas method depends on the percentage of key gas
liberated at the time of fault. Fig 2 Shows different incipient
faults based on quantity of gases released inside the
transformer [5][7]. When the gas crosses the limits,
subsequent fault is the root cause.
a) Key Gas: Ethylene b) Key Gas: Carbon Monoxide
c) Key Gas: Hydrogen d) Key Gas: Acetylene
Fig -2: Key Gas Method with Four Faults
3.2 Rogers Ratio Method
As the namesuggeststhismethoddiagnosesthefaultwith
the help of gas ratios generated at the time of fault. The four
gas ratios are CH4/H2, C2H6/CH4, C2H4/C2H6 and
C2H2/C2H4 [8][9][3]. One of the main advantages of this
method is irrespective of the volume of oil it depends onlyon
ratios of gas released. Table IIshows codes forgasratios.The
value of each ratio is calculated, based on these four ratios
range, code (0-5) is provided in Table 2. Table 3 shows fault
identified with reference to the ratio of gases with code.
There are 12 possible causes of incipient faults as shown in
Table 3.
Table -2: Gas Ratio Codes [2][3]
Sr.No Gas Ratios Ratio Codes Range Code
1 CH4/H2 i
<=0.1 5
0.1-1.0 0
>=1.0,<3.0 1
>=3.0 2
2 C2H6/CH4 j
<1.0 0
>=1.0 1
3 C2H4/C2H6 k
<=1.0 0
>=1.0,<3.0 1
>=3.0 2
4 C2H2/C2H4 l
<0.5 0
>=0.5,3.0 1
>=3.0 2
Table -3: Different faults based on Rogers Ratio [5]
Sr.No i j k l Diagnosis
1 0 0 0 0 No fault. Normal deterioration
2 5 0 0 0 Partial discharge
3 1-2 0 0 0 Slight overheating(<150°C)
4 1-2 1 0 0 Overheating(150C-200°C)
5 0 1 0 0 Overheating(200C-300°C)
6 0 0 1 0
General conductor over
heating
7 1 0 1 0 Winding circulating currents
8 1 0 2 0
Core & circulating currents
(300C-700°C)
9 0 0 0 1
Flash over without power
follow through
10 0 0 1-2 1-2
Arc with power follow
through
11 0 0 2 2
Continuous sparking to
floating potential
12 5 0 0 1-2
Partial discharge with
tracking
3.3 IEC Ratio Method
IEC ratio method is derived from the Roger’s Ratio
method. In this method, ratio C2H6/CH4 was dropped as it
revealed only partial temperature range at the time of
decomposition inside the transformer. Following Table 4
shows the IEC Ratio codes with gas ratio range and its code
(0-2). [2][3][11]

Table -4: IEC Ratio Codes [3]
Sr.No Gas Ratios Ratio Codes Range Code
1 CH4/H2 i
<0.1 1
0.1-1.0 0
>1.0 2
3 C2H4/C2H6 k
<1.0 0
1.0-3.0 1
>3.0 2
4 C2H2/C2H4 l
<0.1 0
0.1-3.0 1
>3.0 2
Table 5 shows fault identified with reference to the ratio
of gases with code. There areninepossiblecausesofincipient
faults as shown below.
Table -5: Different faults based on IEC Ratio [3]
Sr.No i k l Diagnosis
1 0 1 0 Normal ageing
2 1 0 >0
Partial discharge of low
energy density
3 1 0 1
Partial discharge of high
energy density
4 0 1-2 1-2 Discharge of low energy
5 0 2 1
Discharge of high
energy
6 0 1 0
Thermal fault(<150°C)
heating
7 2 0 0
Thermal fault(150°C -
300°C)
8 2 1 0
Thermal fault(300°C -
700°C)
9 2 2 0
Thermal fault(>700°C)
follow through
3.4 Dornenburg Ratio Method
This method is based on ratios of four gases CH4 / H2,
C2H2 /C2H4, C2H2 /CH4 and C2H6 / C2H2 or six gases H2,
CH4, C2H4, C2H2, C2H6 and CO. Through this method three
types of fault can be analyzed i.e. thermal decomposition,
corona and arcing [7][14][15]. This method is validwhenthe
concentration of at least one of the gas H2, CH4, C2H4, or
C2H2 must exceed twice the concentration limit. However
gases viz. C2H6 and CO must exceed the concentration limit.
Table 6 shows the concentration limit of key gases. [4][3]
Table -6: Concentration limit for Dornenburg Ratio
Method [4][3]
Sr.No Key gas
Concentration
(PPM)
1 Hydrogen (H2) 100
2 Methane(CH4) 120
3 Acetylene (C2H2) 35
4 Ethylene (C2H4) 50
5 Ethane (C2H6) 65
6
Carbon monoxide
(CO)
350
Table 7 gives the diagnosis for different gas ratio range
and its subsequent fault
Table -7: Diagnosis by Dornenburg Ratio Method [5]
Fault
CH4 /
H2)
C2H2
/C2H4)
C2H2
/CH4
C2H6
/
C2H2
Thermal
decomposition
>1.0 <0.75 <0.3 >0.4
Partial Discharge
(PD)
<0.1
Not
significant
<0.3 >0.4
Arcing fault
>0.1 to
<1.0
>0.01 to <
0.1
>0.3 <0.4
3.5 Duval Triangle Method
Duval Triangle method is one of the most accurate
method when compared to other fourmethods.Thismethod
is based on three gases CH4, C2H4 and C2H2 [9][16]. The
problem exists only if at least one of the hydrocarbon gases
or hydrogen is greater or equal to L1 level. The gas
generated must at the rate of G1 & G2 as shown in Table 8.
[10][11]
Table -8: L1 limit and gas generation rates for Duval
Triangle Method [3]
Sr.N
o
Gas
L1
limits
G1 limits
(ppm per
month)
G2 limits
(ppm per
month)
1 H2 100 10 50
2 CH4 75 8 38
3 C2H2 3 3 3
4 C2H4 75 8 38
5 C2H6 75 8 38
6 CO 700 70 350
7 CO2 7000 700 3500
If the above conditions as mentioned in table VIII are
satisfied then the problem exists. The percentageofeachgas
with total gas (CH4 + C2H4 + C2H2) is calculated. This
percentage of each gas is plotted on the triangular chart
which is subdivided into six fault zones. The fault zone
where the point is located indicates the type of fault
produced in the transformer as shown in Fig3. [10][11]

Fig -3: Duval Triangle.
4. COMBINED DGA ANALYSIS
Fig 4. Shows the process for combined dissolved gas
analysis by all the five methods which we have discussed,
based on the percentage of fault diagnosed by each method
the fault is concluded.
Fig. 4. Block diagram of Combined Dissolved Gas
Analysis (DGA) [1].
Table 9 gives the codes for seven different faults
Table -9: Fault classification with codes [6][12]
Sr.No Fault Code
1 Normal Condition F1
2 Thermal faults<300°C F2
3 Thermal faults 300°C-700°C F3
4 Thermal faults >700°C F4
5 Arcing F5
6 Partial Discharge F6
7 No prediction F7
Table -10: Mapping process of each fault with different
diagnostic methods
The result of the combined analysis is displayedonMATLAB
GUI. The above codes are incorporated to provide the
incipient fault based on the percentage of the fault30data of
gases was analysed for DGA by comparing the results with
individual methods and combined DGA analysis
5. RESULT AND CONCLUSION
Using MATLAB GUI results for 30 data of gasesfromdifferent
papers were calculated [1]. The result of the tested DGA is as
shown in Table XI. It was observed that the most accurate
method is duval triangle method with 90% accuracy.
However with combine DGA analysis, the percentage of
accuracy is 100%. Thus it is preferred to carry out combined
DGA analysis, instead of the most accurate individual Duval
triangle method for fault analysis in case of internal fault in
the transformer.
Table -11: Comparing percentage prediction of individual
and combined DGA method
Sr.
No
Method
No. of
Data
tested
Computation
Results
obtained
No
Prediction
%
Accuracy
1 Key Gas 30 2 28 7%
2 Dornenburg 30 14 16 47%
3 Rogers 30 11 19 37%
4 IEC 30 25 5 83%
5 Duval 30 27 3 90%
6
Combined
DGA
30 30 0 100%

REFERENCES
[1] S. H. E.-H. Osama E. Gouda, Saber M. Saleh, “Power
Transformer Incipient Faults Diagnosis Based on
Dissolved Gas Analysis," TELKOMNIKA Indonesian
Journal of Electrical Engineering, vol. 17, pp. 10-16,
January 2016.
[2] A. S. A. AbubakarA. Suleiman, “Improving accuracy of
dga interpretation of oil filled power transformers
needed for effective condition monitoring," IEEE
International Conference on Condition Monitoring and
Diagnosis, September 2012.
[3] T. B. K. L. N.A. Muhamad, B.T. Phung, “Comparative
Study and Analysis of DGA Methods for Transformer
Mineral Oil," IEEE, vol. 8, April 2007.
[4] IEEE, IEEE Guide for the Interpretation of Gases
Generated in Oil-Immersed Transformers," IEEE Std
C57.104-2008, 2008.
[5] T. S. Ekkarach Wannapring, Cattareeya
Suwanasri,”Dissolved Gas Analysis Methods for
Distribution Transformers”, IEEE, July 2016.
[6] H. B. E. G. A. Akbari, A. Setayeshmehr, “A Software
Implementation of the Duval Triangle Method,” IEEE
International Symposium on Electrical Insulation,vol.6,
April 2008.
[7] J. Golarz, “Understanding Dissolved Gas Analysis (DGA)
Techniques and Interpretations”, IEEE, May 2016.
[8] M. S. Jawad Faiz, “Dissolved gas analysis evaluation in
electric power transformers using conventional
methods a review,” IEEETransactionsonDielectricsand
Electrical Insulation, vol. 24, pp. 1239-1248, April2017.
[9] E. A.-S. Norazhar Abu Bakar and S. Islam, “A review of
dissolved gas analysis measurement and interpretation
techniques," IEEE Electrical InsulationMagazine,vol.30,
pp. 39-49, May/June 2014
[10] L. Hamrick, “Dissolved gas analysis for transformers,"
Neta World, pp. 01-04, May/June 2009- 2010.
[11] M. L. M. J. Mr.Rahul Soni, Mr.Siddharth Joshi, “Condition
monitoring of power transformer using dissolved gas
analysis of mineral oil: A review," International Journal
of Advance Engineering and Research Development
(IJAERD), 2015.
[12] P. S. B. P. Deepak, Munisha Bage, “Transformer Fault
Diagnosis Based on DGA using Classical Methods,"
International Journal of Engineering Research
Technology (IJERT) CMRAES, vol. 4, April 2016.
[13] V. N.Ravichandran, Investigations on Power
Transformer Faults Based on Dissolved Gas Analysis ,"
International Journal of Innovative Technology and
Exploring Engineering (IJITEE), vol. 8, April 2019.
[14] H. G. Z. S. S. M. G. Ibrahim B. M. Taha, “Comparative
Study between Dorneneburg and Rogers Methods for
Transformer Fault Diagnosis Based on Dissolved Gas
Analysis Using Matlab Simulink Tools ," IEEE, Feb 2016.
[15] H. H. E.-T. Osama E. Gouda, Salah H. El-Hoshy,“Proposed
three ratios technique for the interpretation of mineral
oil transformers based dissolved gas analysis ," IET
Generation,TransmissionDistribution,vol.12,pp.2650-
2661, April 2018.
[16] D. J. Sukhbir Sngh and M. Bandyopadhyay, “Software
Implementation of Duval Triangle Technique forDGAin
Power Transformers," International Journal ofElectrical
Engineering, vol. 4, pp. 529{540, 2011.
[17] S. Singh and M. Bandyopadhyay, “Duval Triangle: A
Noble Technique for DGA in Power Transformers,"
International Journal of Electrical and Power
Engineering, vol. 4, pp. 193-197, 2010.
[18] A. A. S. . N. A. M. . N. B. . A. S. Alghamdi, “Introducing the
Hybrid-DGA Interpretation software as an effective
power transformer management tool," IEEE, October
2013.
[19] W. G. W. J. Lelekakis. N, Martin. D, “Comparison of
dissolved gas-in-oil analysis methods using a dissolved
gas-in-oil standard,"IEEEElectrical InsulationMagazine,
vol. 27, pp. 29-35, September-October 2011.

IRJET- Combined Dissolved Gas Analysis: A Prescient Methodology for Recognizing Faults in Transformer by MATLAB GUI

More Related Content

Similar to IRJET- Combined Dissolved Gas Analysis: A Prescient Methodology for Recognizing Faults in Transformer by MATLAB GUI (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Combined Dissolved Gas Analysis: A Prescient Methodology for Recognizing Faults in Transformer by MATLAB GUI