Transformer fault Diagnosis using Gas Analysis

1. CONDITION MONITORING OF TRANSFORMER
Guided By:-
Prof. Ankit Shahpatel
Electrical Engineering Department
G.C.E.T-V.V.NAGAR
Prepared by:-
Mehul Makwana
ME (Power system )
Enroll.No 130110737020
G.C.E.T-V.V.NAGAR
Co-Guided By:-
Prof. Sameer B Patel
Electrical Engineering Department
G.C.E.T-V.V.NAGAR

2. CONTENTS
• Introduction
• Proposed Methods
• Dissolved Gas Analysis
• Case study on DGA
• Fuzzy Inference System
• References

3. INTRODUCTION
Transformer is static electrical device
• From the day of This equipment in service, Different stresses like electrical ,
mechanical, chemical and environmental factor effect the condition of the
transformer .
• At the initial stage , Degradation of insulation quality occurs slowly. But this
deterioration multiplies in due course of time and leads to final failure of the
transformer.
• So, to overcome this situation, continuous monitoring of the condition and
preventive measures is required for correct maintenance of the transformer
• The faults like partial discharge , electrical arcs , or hot spots generally deteriorate
the condition of transformer in quick progression.
• Hence Early detection of faults is very important for saving transformer from any
catastrophic failure.

4. PROPOSED METHODS
• 1. Dissolve gas analysis
• 2.Fuzzy Inference System

5. Dissolve Gas Analysis
• Analysis Dissolved gases-in-oil analysis (DGA) is a common practice in
transformer fault diagnosis. Electrical insulation such as mineral oils and
cellulosic materials degrade under excessive thermal and electrical stresses,
forming by product gases which can serve as indicators of the type of stress and its
severity. Dissolved gas-in-oil concentrations, relative proportion of gases, and gas
generation rates (gassing rates) are used to estimate the condition of a
transformer.
• Commonly used gases include hydrogen (H2), methane (CH4), acetylene (C2H2)
ethylene (C2H4), ethane (C2H6), carbon monoxide (CO), and carbon dioxide
(CO2). These gases are extracted from the oil under high vacuum and analyzed by
Gas Chromatograph to get each gas concentration separately.

6. CONT..
• DGA process
Gas
Chromatogra
phy
Extraction
of all the
gases from
Xmer oil
Oil Sampling
Interpretation of
gas data

7. 1. OIL SAMPLING
6/29/2016 7
2. GAS CHROMATOGRAPHY APPARATUS

8. METHODS OF INTERPRETATION(IEC, IEEE)
• The standards IEC [4] and IEEE [5] provide guidelines for interpretation of
DGA. In these standards one can for example find graphical interpretations
of gas ratios and numerical tables for typical gas concentrations within oil
immersed transformer in operation.
6/29/2016 8

9. PRINCIPAL CAUSES OF GAS FORMATION IN TRANSFORMER
• Fault gases are caused by corona (partial discharge), thermal heating
(pyrolysis) and arcing.
PARTIAL DISCHARGE is a fault of low level energy which usually
occurs in gas-filled voids surrounded by oil impregnated material.
• The main cause of decomposition in partial discharges is ionic
bombardment of the oil molecules.
• The major gas produced is Hydrogen. The minor gas produced is
Methane.
6/29/2016 9

10. CONT..
THERMAL FAULTS
• A small amount of decomposition occurs at normal operating temperatures.
• As the fault temperature rises, the formation of the degradation gases change from
Methane (CH4) to Ethane (C2H6) to Ethylene (C2H4).
• A thermal fault at low temperature (<300deg/C) produces mainly Methane and
Ethane and some Ethylene.
• A thermal fault at higher temperatures (>300deg/C) produces Ethylene.
• The higher the temperature becomes the greater the production of Ethylene.
• ARCING is a fault caused by high energy discharge.
6/29/2016 10

11. Cont..
• The CO And CO2 gases indicate the condition of cellulose and paper
insulation.
• If The ratio CO/CO2 > 0.1 It indicate the Over heating of the cellulose
& insulation over heating.

12. IEC ratio codes & Fault classification
Gas
Ratios
Range of
gas ratio
code
Range of
code /
IEC
code
𝐶2 𝐻2
𝐶2 𝐻4
<0.1 0
0.1–1.0 1
1.0–3.0 1
>3.0 2
𝐶𝐻4
𝐻2
<0.1 1
0.1–1.0 0
1.0–3.0 2
>3.0 2
𝐶2 𝐻4
𝐶2 𝐻6
<0.1 0
0.1–1.0 0
1.0–3.0 1
>3.0 2
𝑪 𝟐 𝑯 𝟐
𝑪 𝟐 𝑯 𝟒
𝑪𝑯 𝟒
𝑯 𝟐
𝑪 𝟐 𝑯 𝟒
𝑪 𝟐 𝑯 𝟔
FAULT TYPE
0 0 2 Partial discharge of low energy
0 1 1 Thermal Fault of low temperature 150 − 300 .
0
𝐶
0 1 2 Thermal Fault of low temperature < 150 .
0
𝐶
1 0 0 Flashover, Intermittent sparking
1 1 1 Thermal Fault of low temperature 150 − 300 .
0
𝐶
1 1 2 Thermal Fault of high temperatures > 700 .
0
𝐶
1 2 0 Core and tank circulating currents.
1 2 1 Winding Circulating currents.
1 2 2 Core and tank circulating currents.
2 0 0 Partial discharge of high energy density, Corona
2 0 1 Discharge of high energy, Arcing.
2 0 2 Discharges of low energy, Continuous sparking
2 1 0 Partial discharge of high energy density, Corona
2 1 1 Discharge of high energy, Arcing.
2 1 2 Discharges of low energy, Continuous sparking
2 2 0
Severe arcing, Overheating of oil. > 1000 .
0
𝐶2 2 1
2 2 2

13. MATLAB Programming
• From The Analysis of Key gases concentration in the transformer oil
and its ratio according to rogers method & IEC code I have prepped
MATLAB program for the identify , which type of fault in the
transformer May occurs in future.
• The MATLAB programming method is simple for users as its not
required much knowledge and data as other method like ANN , Fuzzy
Logic , Expert system .
• By the Entering the Value of concentration of gages in oil in Program
we get the resultant fault according to IEC standard .
6/29/2016 13

14. Flowchart

15. CASE STUDY ON DGA
• Transformer detail
RATING : 15 MVA 66/11.55KV
MAKE : ATLANTAS ELECT PVT LTD
Sample take : Bottom point
DGA indicates Hydrogen & Hydrocarbons So need to Investigate the fault
& Resample after investigation.
Also the CO/CO > 0.1 Indicate the Insulation overheating.
2
3635
12513
4 17 159 29 4 0 560
2000
4000
6000
8000
10000
12000
14000
H2 O2 N2 CH4 CO CO2 C2H4 C2H6 C2H2 TDCG
VALUEINPPM
GASES CONCENTRATION
DGA
BEFOR INVESTIGATION

16. CONT..
• Enter the value in MATLAB Program:
Result: Thermal fault of low temperature
H2 N2 CH4 C2H2 C2H4 C2H6 CO2 CO
2 123513 4 0 29 4 159 17

17. CONT..

18. CONT..
LV-U phase bus bar Paper insulation burned

19. CONT..
LV-U phase bus bar
bolt affected
Paper and Bolt remove
for investigation
Rectification and bolting
after cleaning

20. CONT..
• TEST SAMPLE AFTER INVESTIGATION AND MAINTAINING
0
3185
10357
0 6 150 0 0 0 60
2000
4000
6000
8000
10000
12000
H2 O2 N2 CH4 CO CO2 C2H4 C2H6 C2H2 TDCG
VALUEINPPM
GASES CONCENTRATION
DGA
AFTER INVESTIGATION
Paper covering and joint again

21. Fuzzy Inference system
• There are lots of indeterminate factors in process of transformer fault
diagnosis whose influence to the transformer operation status is
usually fuzzy and uncertain.
• Ratio codes are quantized to define the crisp boundaries of 0,1 and 2.
In practice these boundaries are non crisp (Fuzzy) especially under
multiple faults condition.
• These codes could lead to errors in diagnosis moving across the crisp
boundaries from one fault to another. To overcome these limitations,
Fuzzy System for diagnosis of multiple faults is developed.

22. Fuzzy set description
• An ordinary set can be characterized as a binary function. Elements in
the set can be assigned to1 and remaining elements of the universe can
be assigned to 0. The function is generalized so that value assigned to
the elements of the universal set located within a specified range
which indicates membership grades of these elements within the sets,
such function is called membership function and the corresponding set
is a fuzzy set.

23. Fuzzy Control Algorithm
C2H2
C2H4
CH4
C2H6
H2
C2H4/C2H6
CH4/H2
C2H2/C2H4
FUZZY
CONTROL
LER
OUTP
UT

24. Proposed fuzzy control algorithm
• The proposed FIS editor prepared using MATLAB Fuzzy Logic
Toolbox is shown in Fig.
FIS Editor

25. Cont..
• This fuzzy system consists of 3 ratios C2H2/C2H4 ,CH4/H2 and C2H4/C2H6 as
inputs. Each ratio is fuzzyfied as Very Low, Low, Medium, High and Very High
according to membership intervals as defined in Following table ,
Ratio CH4/H2 C2H2/C2H4 C2H4/C2H6
Very Low X<0.09 X<0.09 X<0.9
Low 0.09<=X<=0.11 0.09<=X<=0.11 0.9<=X<=0.11
Medium 0.11<=X<=0.9 0.11<=X<=2.9 0.11<=X<=2.9
High 0.9<=X<=1.1 2.9<=X<=3.1 2.9<=X<=3.1
Very High X>1.1 X>3.1 X>3.1

26. Cont..
• The membership boundaries of Low and High fuzzy are fuzzyfied by
using triangular function.
• The membership boundaries of other fuzzy ratios are fuzzyfied by
using trapezoidal function.

27. Fuzzy Sets for Gases Ratios
Fuzzy set for C2H4/C2H6

28. CONT..
Fuzzy set for CH4/H2

29. Rules Editing window :

30. Result Window

31. REFERENCES
[1] Er.S.Sehgal and Er.D.singh,“Dissolve gas analysis” , IJERT Vol. 1
Issue 6, August - 2012 .
[2]R.pandey,MT.Deshpande,“Dissolved Gas Analysis of Mineral Oil
Used in Transformer” IJAIEM Volume 1, Issue 2, October2012.
[3]SY.Jasmin and J.Shreevastava,“Dissolve gas Analysis of Power
transformer”, IJEEER Vol. 3, Issue 5, Dec 2013, 1-10.
[4]IEC Standards 60599, “Mineral oil-impregnated electrical equipment
in service” - Guide to the interpretation of dissolved and free gases
analysis, second edition, Mars 1999.
[5]IEEE Standards C57.104-1991. IEEE Guide for the Interpretation of
gases generated in Oil-Immersed Transformers, 1991.

32. CONT..
[6]Deepika Bhalla, RK Bansal, and Hari Om Gupta , “Application of Artificial
Intelligence Technique for DGA of Transformers-A Review ,World Academy of
Science, Engineering and Technology Vol;4 2010-02-27.
[7] Ali A. Albakry, “Modified Fault Diagnosis Method For Power Transformers
Using Fuzzy Logic Technique”,Journal of Babylon university / pure and applied
science.
6/29/2016 32