621-1779-1-PB

7/28/2019 621-1779-1-PB

1/5

Advances in Electrical Engineering Systems (AEES) 152Vol. 1, No. 3, 2012, ISSN 2167-633XCopyright World Science Publisher, United Stateswww.worldsciencepublisher.org

Dissolved Gas Analysis as a Diagnostic Tools for Early

Detection of Transformer Faults

Sherif S. M. Ghoneim1, 2

, IEEE Member, Sayed A. Ward3

1Canal Suez University, Suez, Egypt

2Taif University-Taif- KSA

3Benha University- Benha- Egypt

E-mail:[email protected]

Abstract- Transformers is a device on which cost effective supply of electricity mostly depends. Hence, to manage the

life of transformers, to reduce failures and to extend the life of transformer, some measures are being adopted. Dissolved

gas-in-oil analysis (DGA) is a common practice in transformer fault diagnosis. Some classical methods that depend on

gases concentration in transformers oils are used to interpret transformer faults such as Dornenberg, Rogers, Duvaltriangle and key gases methods. These methods in some cases did not give the same results; therefore, an expertise

method is developed to give the fault type according to the dissolved gases concentration in oil. A software code isdesigned using logic functions to get the type of the faults in transformers. Comparing the results from this software with

the real laboratory cases as well as some cases in literatures is developed. The results explain the validation of software

to detect the fault in transformer.

Key words: Dissolved gases analysis; Transformer oil; Interpretation of transformer faults.

1. Introduction

The cost of unplanned outages can be reduced by theearly detection of such internal faults in transformer. The

interpretation of transformer faults using dissolved gases

analysis is produced using some techniques that assumedby Dornenberg, Rogers, Duval triangle and key gases

methods [1-5].Dissolved gas analysis is the most important test in

determining the condition of a transformer. It is the first

indicator of a problem and can identify deterioratinginsulation and oil, over heating hot spots, partial discharge

and arcing. Dissolved gas analysis is made on the basis of

the standard IEC60599 [6] and IEEEC 57-104TM [7]

standards. A four condition DGA guide to classify risks totransformers with no previous problems has been published

in the standard IEEE C57- 104TM.Several artificial intelligence methods such as Fuzzy

Logic and Artificial Neural Network (ANN) were

developed as a technique to interpret the faults in

transformer [8-15].

In this paper an expertise method based upon the

dissolved gases analysis (DGA) techniques is developed.

Based on the interpretation of the classical techniques to

the cause of transformer faults according to the gases

concentration in oil transformer, an expertise system is

suggested to give the cause of the transformer fault with

the aid of logic functions that is used as in Fuzzy andNeural Network. A lot of real cases of analyzing the

dissolved gases were collected and used to illustrate the

validity of the proposed expertise method. In addition,

some cases from previous literatures were used to

compare their results with the proposed methods results.

2. Classical methods to diagnose transformer

faults

When thermal or electrical stresses, which affect theinsulating oil and cellulose material in transformers, are

higher than the normal permissible value, then certain

combustible gases, referred as fault gases, started to be

produced inside the transformer. The most significant

fault gases produced by oil decomposition are H2

(Hydrogen), C2H6 (Ethane), C2H4 (Ethylene) and C2H2(Acetylene) as well as Carbon monoxide (CO) and carbon

dioxide (CO2) which produce from decomposition of

insulated paper (Cellulose). The type of the faults[corona, arcing discharge (both electrical faults) and

overheating (thermal fault)] as well as their severity, play

an important role in producing different combustible

gases.

Based on DGA, many interpretative methods have

been introduced to diagnose the nature of the incipient

deterioration occurred in transformer.

Over the years, several techniques have been

developed to facilitate the diagnoses of fault gases such as

Dornenberg method [5], Roger's ratio method [3], Key

gases method [5, 16], and Duval Triangle method [4] as

well as the recently developed techniques such as neural

network and fuzzy logic.
http://www.worldsciencepublisher.org/http://www.worldsciencepublisher.org/mailto:[email protected]:[email protected]:[email protected]:[email protected]://www.worldsciencepublisher.org/

7/28/2019 621-1779-1-PB

2/5

Sherif S. M. Ghoneim & Sayed A. Ward, AEES, Vol. 1, No. 3, pp. 152-156, 2012 153

2.1. Key gases method

The key gas method interprets the incipient faults in

transformer according to some significant gases to assign

four typical fault types. These gases are called key

gases [5] which are shown in Fig. 1.

Overheated Cellulose

0

20

40

60

80

100

CO H2 C H4 C2H6 C2H2 C2H2

Gases

%

(a)

Overheated Oil

0

20

40

60

80

100

C O H2 C H4 C 2H6 C2H2 C2H2

Gases

%

(b)

Arc ing in Oil

0

20

40

60

80

100

CO H2 CH4 C2H6 C2H2 C2H2

Gases

%

(c)

Corona in Oil

0

20

40

60

80

100

CO H2 CH4 C2H6 C2H2 C2H2

Gase s

%

(d)

Fig. 1: Key gas method and four typical faults

2.2. Dornenburg ratio method

The Dornenburg method utilizes four calculated gas

ratios to indicate a single fault type from three general

fault types. This procedure requires significant levels of

the gases to the present in order for the diagnosis to be

valid. The four ratios and their diagnosis values are given[17]. Dornenburg method uses five individual gases or

four-key gas ratios, which are:-

R1=CH4/H2,

R2=C2H2/C2H4,

R3=C2H2/CH4,

R4=C2H6/C2H2.

A flow chart that describes step by step procedure to

identify the reason behind transformer faults, is found in

[5].

2.3. Roger's ratio method

This method was further modified into an IEC standard

[3], [18]. The original Rogers ratio method uses four gas

ratios which are CH4/H2, C2H6/CH4, C2H4/C2H6 and

C2H2/C2H4 for diagnosis. The refined Rogers method

uses two tables: one defined the code of the ratio, and the

other defined the diagnosis rule. The ratio C2H6/CH4only indicated a limited temperature range of

decomposition, but did not assist in further identification

of fault. Therefore, in IEC standard 599, the further

development of Roger's ratio method, was deleted.

Roger's ratio method and IEC 599 have gained popularity

in industrial practices. However, it may give no

conclusion in some cases. This is the "no decision"problem.

2.4. Duval triangle method

The Duval Triangle was first developed in 1974[19].

Three hydrocarbon gases only (CH4, C

2H

4and C

2H

2) are

only used. These three gases are generated as a result of

increasing the level of energy necessary to generate gasesin transformers in service. Figure 2 indicates the Triangle

method. In addition to the 6 zones of individual faults

(PD, D1, D2, T1, T2 or T3), an intermediate zone DT has

been attributed to mixtures of electrical and thermal faultsin the transformer.

D1 D2 DT

T3

T2

T1

PD

%CH

4

%C2H2

%C2H

4

Fig.2: Duval triangle as a diagnostic tool to detect the incipient faults in

transformer.

(T1 the zone of low thermal fault

7/28/2019 621-1779-1-PB

3/5


tree is developed which contains the information between

different faults types. Every fault type takes a number to

help us to get the main cause of the transformer fault.

This is shown in Figure 3.

Fault type

No Fault0

Fault1

No Fault Identify2

Discharge7

Thermal3

Partial8

Arcing11

Low9

High10

Low12

High13

300 and

700oC6

Thermal Cellulose14

Fig. 3: Decision fault tree

A software code in excel sheet is developed using the

logic function to get transformer fault from the four

classical method that mentioned before, the resultsdepend on the combustible gases that arise when fault

occurs in transformer. After determining the fault type

from these methods, the program decides the incipientfault type.

Figure 4 explain the form of the excel sheet that is used to

explain the main fault in transformer and Figure 5

illustrates the final report.

Fig. 4: The excel sheet that use to give the main cause of transform fault

Fig. 5: The final report

Figure 6 illustrates the flow chart that used to determine

the main cause of the fault using if statement and logicfunctions. It depends on the code from the decision tree

fault.

Fig. 6: Flow chart to determine the main cause of the transformer fault

4. Fault Type in Transformers According to

Gas Concentration

Some oil samples are taken from real transformers which

are in operation to carry out the study. In table 1, the

results by the expertise method with that take from the

chemical lab is explained. Figure 7 shows the results ofcase 1 as in table 1.

7/28/2019 621-1779-1-PB

4/5


Table 1: Comparison between the expertise method results and lab

results

CASE 1 2 3 4 5

H2 191 154 36 86 10

CH4 47 11 245 187 24

C2H2 0.0001 3 0.0001 0.0001 0.0001

C2H4 15 8 332 363 24

C2H6 43 14 144 136 372

CO 634 487 538 26 343

CO2 6623 3395 2326 198 2583

LAB.

RES.

Therm.

300-700oC

Disch. of

highenergy

Therm.

>700oC

Therm.

>700oC

Thermal

300-700oC

EXP.

METH.

Med.

thermal

fault

High

arcing

discharge

High

thermal

fault

High

thermal

fault

Med.

thermal

fault

Fig. 7: Software results for case 1

It is seen from the table that, when the C 2H2 is

increased above 2, the expected fault in transformer is

arcing discharge as in case 2. In cases 1 and 3, the

dominant gases are CH4 and C2H6, hence the expected

fault is medium thermal fault. The high thermal fault is

expected when CH4 and C2H4 are the dominant gases.

The results explain that the reliability of the expertise

method to detect the actual fault in transformer as thelaboratory does.

5. Comparison between the Expertise

Method and other Methods in Literatures

Table 2 shows that the comparison between the results

from the software code based on expertise method and the

results from other methods in literatures.

Figure 8 shows the result from the software code forsample 2 in [20].

Table 2 explains also the reliability and validation of

the proposed expertise method for detecting the inceptionfaults in transformer based on DGA.

Table 2: Comparison between the expertise method results and results

in literatures

CASE Sample 2 in

[20]

Example 2

in [21]

Case in [11] Case III in

[22]

H2 59 206 769 127

CH4 93 42 999 24

C2H2 1 221 31 81

C2H4 6 82 1599 32

C2H6 89 16 234 0.0001

CO 736 334 - 0.0001

CO2 1519 3432 - 2024

LAB.RES.

Therm.decomp.

ArcingDisch.

Thermal faultwith

temp.>700oC

Arcing notinvolve

cellulose

EXP.

METH.

Low therm.

fault

High

arcingdisch.

High thermal

fault

High

arcingdisch.

Fig. 8: Software results for sample 2 in [3]

6. Conclusions

The results from different cases under study reveal that

the proposed technique (expertise method) is reliable to

use as a diagnostic tools to detect the fault in transformer

in its early stage. The conclusions from the real cases

explain that the nature of the insulating materials involved

in the fault and the nature of the fault itself affect on

distribution of dissolved gases. Based on The results from

the software code and the lab results, we see that the

software code is reliable to diagnosis the transformer fault

based on the gas concentrations. The results from thesoftware code have a good agreement with the previous

conventional methods for the fault detection in

transformers.

References

[1] E. Dornenburg, and W. Strittmater, Monitoring Oil CoolingTransformers by Gas Analysis, Brown Boveri Review, vol. 61, pp.

238-274, 1974.

[2] IEC Publication 599, Interpretation of the Analysis of Gases inTransformers and Other Oil-Filled Electrical Equipment in Service, First

Edition, 1978.

7/28/2019 621-1779-1-PB

5/5


[3] R. R. Rogers, IEEE and IEC Codes to Interpret Incipient Faults in

Transformers, Using Gas in Oil Analysis, IEEE Trans. on ElectricalInsulation, vol. 13, no. 5, pp. 349-354, 1978.

[4] M. Duval, Dissolved Gas Analysis: It Can Save Your Transformer,

IEEE Electrical Insulation Magazine, vol. 5, no. 6, pp. 22-27, 1989.[5] ANSI/IEEE Std C57.104-1991, IEEE Guide for The Interpretation of

Gases Generated in Oil-Immersed Transformers, IEEE Power

Engineering Society, 1992.

[6] Mukund Hazeeb, S. K. Chand, B. N. De Bhowmick, M. M.Goswami, - Dissolved Gas Analysis Some case studies, Power Grid

Corporation of India Limited.[7] CIGRE WG A2.18 Guide for Life Management Techniques for

Power Transformers, 20th January, 2003.

[8] Fbio R. Barbosa, Otaclio M. Almeida, Arthur P. S. Braga, CceroM. Tavares, Mrcio A. B. Amora, Francisco A. P., Artificial Neural

Network Application In Estimation Of Dissolved Gases In Insulating

Mineral Oil From Physical-Chemical Datas For Incipient FaultDiagnosis, IEEE Trans. Power Del., vol. 6, no. 2, pp. 601-607, Apr.

1991.

[9] CS Chang,CW Lim,Q Su, Fuzzy-Neural Approach for DissolvedGas Analysis of Power Transformer Incipient Faults", Australasian

Universities Power Engineering Conference (AUPEC 2004) 26-29

September 2004, Brisbane, Australia.[10] Diego Roberto Morais and Jacqueline Gisle Rolim,"A Hybrid

Tool for Detection of Incipient Faults in Transformers Based on theDissolved Gas Analysis of Insulating Oil", IEEE Transactions on PowerDelivery, Vol. 21, No. 2, April 2006.

[11] Rahmat-Allah Hooshmand, Mahdi Banejad, Fuzzy LogicApplication in Fault Diagnosis of Transformers Using Dissolved Gases",

Journal of Electrical Engineering & Technology, Vol. 3, No. 3, pp.

293~299, 2008.[12] Rahmatollah Hooshmand, Mahdi Banejad," Application of Fuzzy

Logic in Fault Diagnosis in Transformers using Dissolved Gas based on

Different Standards", World Academy of Science, Engineering and

Technology 17 2006.[13] J. P. Lee, D. J. Lee, S. S. Kim, P. S. Ji and J.Y. Lim," Dissolved

Gas Analysis of Power Transformer Using Fuzzy Clustering and Radial

Basis Function Neural Network", Journal of Electrical Engineering &Technology, Vol. 2, No. 2, pp. 157~164, 2007.

[14] N.A. Muhamad, B.T. Phung, T.R. Blackburn, K.X Lai,"

Comparative Study and Analysis of DGA Methods for Transformer

Mineral Oil", Journal of Electrical Engineering & Technology, Vol. 2,No. 2, pp. 157~164, 2007.

[15] Mang-Hui Wang," A Novel Extension Method for TransformerFault Diagnosis", IEEE Transactions on Power Delivery, Vol. 18, No. 1,

January 2003.

[16] J.P. Gibeault, On-line monitoring of key fault gases in transformeroil. Operational experience accumulated over the years, SYPROTEC

INC., August 1997.

[17] R.D.Stebbins, J.J.Kelly. S.D.Myers,''Power Hansformer FaultDiagnosis', 1997 IEEE PESWM, Panel Session, New York, Feb.6, 1997.

[18] IEC Publication 60599, Interpretation of the analysis of gases intransformer and other oil med electrical equipment in &, Geneva,

Switzerland, 1999.[19] M. Duval, Fault Gases Formed in Oil-Filled Breathing EHV

Power Transformers- The Interpretation Of Gas Analysis Data, IEEE

PAS Conf. Paper No C 74 476-8, 1974.

[20] Sayed A. Ward Evaluating Transformer Condition Using DGA OilAnalysis, 2003 Annual Report Conference on Electrical Insulation and

Dielectric Phenomena.[21] J. BILBAO, et. al. "Expertise Method to Diagnose Transformer

Conditions", WSEAS/IASME Conferences, Corfu, Greece, August 17-

19, 2004.[22] Joseph B. DiGiorgio, "Dissolved Gas Analysis of Mineral Oil

Insulating Fluids" , Northern Technology and testing,

http://www.nttworldwide.com.
http://www.nttworldwide.com/http://www.nttworldwide.com/http://www.nttworldwide.com/

Documents

621-1779-1-PB