FUZZY BASED HEALTH

INDEXING OF TRNSFORMERS

A PROJECTREPORT

Submitted by
Abhishek Khemka (11501613001)
Arindam Pal (11501613011)
Koustav Dutta (11501613020)
Nitin Gupta (11501613026)
Samanway Bera (11501613035)
Shivam Sinha (11501613042)

In partial fulfillment for the award of the

degree of

Bachelor of Technology
IN

ELECTRICAL ENGINEERING

B. P. PODDAR INSTITIUTE OF MANAGEMENT &

TECHNOLOGY
4TH YEAR
 B.P.Poddar Institute of Management & Technology

BONAFIDE CERTIFICATE

Certified that this project report FUZZY BASED HEALTH INDEXING OF

TRNSFORMER is the bonafide work of
Abhishek Khemka (11501613001)
Arindam Pal (11501613011)
Koustav Dutta (11501613020)
Nitin Gupta (11501613026)
Samanway Bera (11501613035)
Shivam Sinha (11501613042)

Who carried out the project work under my supervision.

SIGNATURE SIGNATURE

Dr.Nandita Sanyal Dr. Nandita Sanyal

HEAD OF THE DEPARTMENT SUPERVISOR
(Associate Professor)
Electrical Engineering Electrical Engineering
B.P.P.I.M.T Kolkata. B.P.P.I.M.T Kolkata.
 ABSTRACT
To ensure satisfactory operation of this large power distribution network,
satisfactory operation of the most valuable assets that is the Power transformer
is necessary. Information about the overall health of these assets is very much
necessary for optimized maintenance plan. This can be done by Health Indexing
of power transformers.

In this thesis, a Fuzzy based health index model for condition evaluation of
power transformers is proposed. A health index is a tool that processes service
and condition data into a score which describes the overall health of an asset.
The motivation behind this is to objectively and condently assess the condition
of power transformers so that reinvestment and maintenance decisions might be
justied. This way, the technical lifetime of healthy assets might be safely
increased, while risky assets can be identied and taken care of before they fail.
Health indexing is particularly useful for evaluation of large transformer eets,
since it makes it easy to identify the assets most in need of additional attention.
In order to identify which assessment methods that are best suited for use health
indexing , fuzzy logic is identified as it is based on the notion of relative graded
membership, where human perception and cognition. Some health indexing
parameters have been reviewed. Based on those reviews and general data
collected, a health index model has been proposed. Its main inputs for
assessment are: Service Life, Furan Concentration, Moisture Content, acid
content, acid number and maximum operating load. This information is
processed through assessment modules that each evaluate dierent subsystems
of the transformer. For these assessment modules to reect the degradation of
transformers in a best possible way, emphasis has been put on identifying the
most important failure modes and aging mechanisms.

In order to test the performance of the proposed model, it has been applied to 47
dierent transformers. The results from this analysis show that the model is
capable of dierentiating between assets in dierent conditions and that the
health index score appears to be a rough, but reliable, indication of the actual
condition of a transformer.

iii
 CONTENTS

ABSTRACT
Chapter 1 Introduction

1.1 Introduction To Fuzzy Logic 1

1.2 Why Fuzzy Logic 1
1.3 Membership Function 2
1.4 Advantage of Fuzzy Logic 4
Chapter 2 Project Idea
2.1 Project Objective 6
2.2 Health Deterioration Indexing(HDI) 7
2.2.1 HDI Parameters 7
Chapter 3 Service Life 8
Chapter 4 Acid Number of The Transformer Oil 9
Chapter 5 Furan Concentration 10
Chapter 6 Dissolved Gas Analysis 13
Chapter 7 Furfural Test Of Transformer Oil 14
Chapter 8 Thermal Ageing 16
Chapter 9 Moisture Content 17
9.1 Types Of Water In Transformer Oil
9.2 Solubility Of Water In Oil
Chapter 10 Insulating Oil Condition 19
10.1 Types Of Insulating Oil
10.2 Test Of Insulating Oil
Chapter 11 Present Work 21
Chapter 12 Conclusion 27
References 28

iv
 LIST OF FIGURES

1. MEMBERSHIP FUNCTION EXAMPLE(1.3.1) 4

2. FUZZY LOGIC SYSTEM ARCHITECTURE(1.4.1) 5
3. DEVELOPMENT OF DP WITH TIME AT DIFFERENT 11
MOISTURE CONTENT AT 95C
4. APPROXIMATE OIL DECOMPOSITION 13
TEMPERATURE ABOVE 150C
5. HOT SPOT TEMPERATURE 16
6. PPM MOISTURE IN OIL 17
7. BDV TESTING TOOLKIT 20
8. FIS1 21
9. FIS2 21
10. Input membership functions, Output s of FIS 2 22
11. Input membership functions, Output s of FIS 2 23
12. Block diagram of HDI 27

v
vi
 1
1. INTRODUCTION

1.1. INTRODUCTION TO FUZZY LOGIC

Fuzzy Logic (FL) is a method of reasoning, that resembles human reasoning. The
approach of FL imitates the way of decision making in humans that involves all
intermediate possibilities between digital values YES and NO.

The conventional logic block that a computer can understand takes precise input and
produces a definite output as TRUE or FALSE, which is equivalent to humans YES
or NO. The inventor of fuzzy logic, Lotfi Zadeh in 1965, observed that unlike
computers, the human decision making includes a range of possibilities between YES
and NO, such as
CERTAINLY YES
POSSIBLY YES
CANNOT SAY
POSSIBLY NO
CERTAINLY NO
The fuzzy logic works on the levels of possibilities of input to achieve the definite output.

Fuzzy logic can deal with information arising from computational perception and
cognition, that is, uncertain, imprecise, vague, partially true, or without sharp
boundaries. Fuzzy logic allows for the inclusion of vague human assessments in
computing problems. Also, it provides an effective means for conflict resolution of
multiple criteria and better assessment of options. New computing methods based on
fuzzy logic can be used in the development of intelligent systems for decision making,
identification, pattern recognition, optimization, and control.

1
 Implementation:-
It can be implemented in systems with various sizes and capabilities ranging from
small micro-controllers to large, networked, workstation-based control systems.

It can be implemented in hardware, software, or a combination of both.

1.2. Why Fuzzy Logic?

Fuzzy logic is useful for commercial and practical purposes.

It can control machines and consumer products.
It may not give accurate reasoning, but acceptable reasoning

Fuzzy logic helps to deal with the uncertainty in engineering.

1.3. Membership Function:-

A collection of things that belong to some definition is known as a Set in mathematics. Any
item either belongs to that set or does not belong to that set.

The function shown above describes the membership of the 'tall' set, you are either in it or
you are not in it. The membership function makes no distinction between somebody who
is 6'5" and someone who is 7'4", they are both simply tall againthere is very small
difference between a 4'11" and 5' man. This is only a difference of one inch, however this
membership function just says one is tall and the other is not tall.

2
The following fuzzy set approach to the set of tall men provides a much better representation
of the tallness of a person.

So for the two people shown above the first person having low membership of 0.3 and so
is not very tall. The second person having membership of 0.95 he is definitely tall.
Defining Fuzzy Sets Mathematically

Fuzzy sets are defined as

Let X be a space of points, with a generic element of X denoted by x. Thus X = {x}.

A fuzzy set A in X is characterized by a membership function fA(x) which associates

with each point in X a real number in the interval [0,1], with the values
of fA(x) at x representing the "grade of membership" of x in A. Thus, the nearer the
value of fA(x) to unity, the higher the grade of membership of x in A.

Membership functions for fuzzy sets can be defined in any number of ways as long
as they follow the rules of the definition of a fuzzy set. The most common membership
functions are shown below:

3
 Membership functions thus allow to quantify linguistic term and represent a fuzzy
set graphically. A membership function for a fuzzy set A on the universe of
discourse X is defined as A:X [0,1].

Here, each element of X is mapped to a value between 0 and 1. It is called membership

value or degree of membership. It quantifies the degree of membership of the element
in X to the fuzzy set A.

x axis represents the universe of discourse.

y axis represents the degrees of membership in the [0, 1] interval.
There can be multiple membership functions applicable to fuzzify a numerical value.
Simple membership functions are used as use of complex functions does not add more
precision in the output.

All membership functions for LP, MP, S, MN, and LN are shown as below

Fig. 1.3.1:- Membership Function Example

The triangular membership function shapes are most common among various other
membership function shapes such as trapezoidal, singleton, and Gaussian.

Here, the input to 5-level fuzzifier varies from -10 volts to +10 volts. Hence the
corresponding output also changes.

1.4. Advantages of FLS s:-

Mathematical concepts within fuzzy reasoning are very simple.
Modification of a FLS by just adding or deleting rules due to flexibility of fuzzy logic is possible.
Fuzzy logic Systems can take imprecise, distorted, noisy input information.
FLSs are easy to construct and understand.

4
Fuzzy Logic Systems Architecture:-

Fig. 1.4.1:- Fuzzy Logic System Architecture

5
 2
2. Project Idea:-
As a result of the deregulations in the power system networks, power utilities have been
competing to optimize their operational costs and enhance the reliability of their electrical
infrastructure. So, the importance of implementing effective asset management plans which
improve the life cycle management of the electrical equipment is highlighted. From an asset
management point of view, commonly practiced maintenance strategies are considered to
have large portions of redundant costs. Therefore, applying cost-effective, reliable and
condition-based maintenance policy is a priority.-

2.1. Project Objective: -

The main objectives of this project are:-
a) To know about the overall health conditions of the existing Power transformers along with
its accessories & to track the rate of deterioration with respect to service life of those
transformers.
b) To make a proper replacement strategy for the older transformers.
c) To revise the routine maintenance process with the RCM i.e. Risk / Reliability centered
maintenance. This scoring helps to prepare a proper maintenance plan by which spending of
money for the maintenance can be done in an optimized way.
d) The individual parameter score gives the information about the specific parameters for
which the transformers health is deteriorating, so improvement of that individual parameter
can elongate a transformers life.

2.2. Health Deterioration Indexing or HDI?

It is a process of measuring the condition of health of a Power transformer by calculating a
score based on certain parameters. As it is a deterioration index, it indicates how much
a transformer is prone to damage or fault. The more the score of HDI of a distribution
transformer, the more its health condition is bad i.e. it is more prone to fault.

6
 2.2.1.HDI assessment parameters:-

Like human health, the health condition of a Power transformer is also assessed by some
condition assessment tests. Unlike the usual procedure, the health assessment method is not
only restricted to the condition of the insulation system of the transformer, but in this
indexing process we also have considered the condition of the peripherals of the transformer
like the condition of the PRV valve, Oil sealants, Bushing, OLTC etc., because they
directly/indirectly affect the life of a transformer & any defects in those peripherals can
cause severe damage to the Power transformer, although the main weightage is given on the
insulation system.

The parameters considered for HDI assessment are given below,

HDI Assessment parameters

1. SERVICE LIFE

2. MOISTURE CONTENT
3. ACID CONTENT

4. FURAN CONCENTRATION

5.THERMAL AGEING

6.ACID NUMBER

The objective of the present work is to develop an expert system to determine the health of a
power transformer. Expert system has shown potential in industrial and commercial
applications by Artificial Intellegence .The expert system in the present thesis embodies
knowledge about HDI of power transformers. With six Health Deterioration Indexing
parameters chosen fuzzy diagnostic method is prepared using the MATLAB Fuzzy Logic Toolbox [13].
Sugeno type FIS [14-16] as a fuzzy inference method. Intelligent algorithms, for example, expert system [1], FIS
[2, 3], AI networks [4-6], PNN [7],AN FIS [8-11], wavelet networks [12] have been used to interpret DGA
results in previous works. These algorithms are found not suitable for multiple faults, only dominant fault is
indicated by these methods.

The present data set obtained from CESC limited the Kolkata based Electricity distribution company.

7
 3
3. Service Life Normal Ageing of Power Transformer:-

Normal service life of a Power transformers mainly depends on the normal ageing of the
insulation system which consists of mineral oil, paper, and cellulosic material which have a
finite life even under normal/ideal operating condition. The normal ageing process depends
on the temperature (Loading condition) and time, whereas moisture and presence of oxygen
accelerates the ageing. According to IEEE C57.91-1995 normal life expectancy of the power
transformers are about 25 to 30 years (approximately 1, 80,000hrs), when it loaded below the
name plate ratings but under low ambient temperature and by cyclic loading process a
transformer can be loaded above nameplate rating without affecting the normal ageing. In
our scoring process safe service life of a transformer is considered as 25 years & maximum
service life is considered as 31 years. The age of the transformers are calculated based on the
commissioning year of the respective transformers. The more the age of the transformer, the
high will be its HDI score.
To help achieving reliable and maximum uninterrupted service life of your transformer,
Hartford Steam Boiler offers comments and recommendations. The advice is intended to
complement the equipment manufacturers recommendations not replace them.
The guidelines for transformers include a substation overview and information about:
Installation
Oil Testing
Gauges
Connections
Cooling
Safety
Cost

8
 4
4. ACID NUMBER:-
Acid number (acidity) is the amount of potassium hydroxide (KOH) in milligrams (mg) that
it takes to neutralize the acid in 1 gram (gm) of transformer oil. The higher the acid number,
the more acid is in the oil. New transformer oils contain practically no acid. Oxidation of the
insulation and oils forms acids as the transformer ages. The oxidation products form sludge
and precipitate out inside the transformer. The acids attack metals inside the tank and form
soaps (more sludge).

Acid also attacks cellulose and accelerates insulation degradation. Slugging begins when the
acid number reaches 0.40; it is obvious that the oil should be reclaimed before it reaches
0.40. It is recommended that the oil be reclaimed when it reaches 0.20 mg potassium
hydroxide per gram (KOH/gm) [21]. As with all others, this decision must not be based on
one DGA test, but watch for a rising trend in the acid number each year. Plan ahead and
begin budget planning before the acid number reaches 0.20.

9
 5
5. Furan Concentration:-
Furan is a heterocyclic organic compound, consisting of a five-membered aromatic ring
with four carbon atoms and one oxygen. Chemical compounds containing such rings are also
referred to as furans. Furan is a colorless, flammable, highly volatile liquid with a boiling
point close to room temperature. It is soluble in common organic solvents,
including alcohol, ether, and acetone, and is slightly soluble in water. It is toxic and may
be carcinogenic in humans. Furan is used as a starting point to other specialty chemicals.
Furans are the major degradation of insulating study in transformer oil. Hence, the
concentration of furans in oil can be used as a good indicator of study deterioration. Furan
concentration in transformer oil is currently measured using high-performance liquid
chromatography or gas chromatography-mass spectrometry. Both methods provide accurate
and reliable results in detecting furan concentration. However, the two methods need very
expensive equipment and take a long time to perform the test on oil samples that should be
chemically treated before conducting the test. Moreover, both methods require an expert to
perform the test and to interpret its results. This study introduces a novel technique for
detecting furan concentration in transformer oil through measuring oil spectral response. The
ultraviolet-to-visible spectral response of transformer oil can be measured instantly with
relatively cheap equipment and does not need an expert person to conduct the test. Results
show that there is a good correlation between oil spectral response and its furan contents.
The study introduces two approaches using expert systems to estimate the relationship
between furan concentration and spectral response of transformer oil; they are gene
expression programming and fuzzy logic.
The solid insulation (cellulose based products) in transformers degrades with time at rates
which depends on the temperature, moisture content, oxygen and acids in the insulation
system. When degradation of the solid insulation occurs, the cellulose molecular chains get
shorter and chemical products such as furanic derivatives (5 hydroxymethyl-2-furfural, 2
furfural alcohol, 2 furfural, 2 acetyl furan, 5-methyl-2furfural), CO and CO2 are produced
and dissolved in the transformer oil. Of the furanic compounds, the 2-furaldehyde is the most
abundant & as it is a polar compound, it cannot be removed by normal filtering. Its
concentration in oil has been related in many cases to the degree of polymerization (DP) and
consequently to the mechanical strength of the solid insulation.
DP o the Degree of Polarization is an average easiest of the use of glucose units per
molecular chain. New Kraft paper has an average cellulose polymer chain that is 1,000
glucose molecules to 1,200 glucose molecules long. The manufacturing and transformer
drying process breaks down the cellulose. New paper in a new transformer, therefore, has
shorter polymer chainsfrom 800 glucose molecules to 1,000 glucose molecules long. The
10
end of life of a transformer is when the paper reaches DP of 200 or less when the paper
becomes very brittle and the short circuit capability is much reduced. The more the value of
the furan compounds in oil, the less the value of DP, the less the remaining life of the
transformer.

Fig. 5.1:- Development of DP with time at different moisture content at 95C

11
 6
6. DGA Indicator of Health of Power Transformer :-
Insulating materials within transformers and electrical equipment break down to liberate
gases within the unit. The distribution of these gases can be related to the type of electrical
fault, and the rate of gas generation can indicate the severity of the fault. The identity of the
gases being generated by a particular unit can be very useful information in any preventative
maintenance program.
The collection and analysis of gases in an oil-insulated transformer was discussed as early as
1928. Many years of empirical and theoretical study have gone into the analysis of
transformer fault gases.
The Dissolved Gas Analysis is the most important tool for determining the condition of a
Power Transformer. The operation of power transformer under abnormal thermal
environment and electrical stresses has impact to the insulation deterioration until
breakdown. Both the oil and cellulose insulation materials are the origin of combustible
gases inside power transformer. The rate and amount of gas generation are important as the
normal aging of power transformer produces gases with extremely slow rate. Therefore, the
generating rate of gases more than normal indicates the problem occurred with power
transformer. The method has been used for several purposes: to detect incipient faults, to
supervise suspect transformer, to test a hypothesis or explanation for the probable cause of
failures or disturbances which have already occurred and to ensure that new transformers are
healthy.
Whenever electrical power transformer goes under abnormal thermal and electrical stresses,
certain gases are produced due to decomposition of transformer insulating oil, when the fault
is major, the production of decomposed gases are more and they get collected in Buchholz
relay. But when abnormal thermal and electrical stresses are not significantly high the gasses
due to decomposition of transformer insulating oil will get enough time to dissolve in the oil.
Hence by only monitoring the Buchholz relay it is not possible to predict the condition of the
total internal healthiness of electrical power transformer. That is why it becomes necessary to
analyses the quantity of different gasses dissolved in transformer oil in service.
From dissolved gas analysis of transformer Oil or DGA of transformer oil, one can
predict the actual condition of internal health of a transformer.
It is preferable to conduct the DGA test of transformer oil in routine manner to get prior
information about the trend of deterioration of transformer health and life. Actually
in dissolved gas analysis of transformer oil or DGA of transformer oil test, the gases in
oil are extracted from oil and analyse the quantity of gasses in a specific amount of oil. By
observing percentages of different gasses present in the oil, one can predict the internal
condition of transformer.
12
Generally the gasses found in the oil in service are hydrogen (H2), methane (CH4), Ethane
(C2H6), ethylene (C2H4), acetylene (C2H3), carbon monoxide (CO), carbon dioxide (CO2),
nitrogen (N2) and oxygen (O2).
Most commonly used method of determining the content of these gases in oil, is using a
Vacuum Gas Extraction Apparatus and Gas Chronographs. By this apparatus first gasses are
extracted from oil by stirring it under vacuum. These extracted gasses are then introduced in
gas Chronographs for measurement of each component.
Generally it is found that hydrogen and methane are produced in large quantity if internal
temperature of power transformer rises up to 150oC to 300oC due to abnormal thermal
stresses. If temperature goes above 300oC, ethylene (C2H4) is produced in large quantity. At
the temperature is higher than 700oC large amount of hydrogen (H2) and ethylene (C2H4) are
produced.
Ethylene (C2H4) is indication of very high temperature hot spot inside electrical transformer.
If during DGA test of transformer oil, CO and CO2 are found in large quantity it is
predicted that there is decomposition of proper insulation.

Fig 6.1:- Approximate Oil Decomposition Temp Above 150C

13
 7
7. Furfural Test or Furfur aldehyde Analysis of Transformer Oil:-

Transformer core and winding have mainly paper insulation. Base of paper is cellulose.
The Cellulose has a structure of long chain of molecules. As the paper becomes aged,
these long chains are broken into number of shorter parts. This phenomenon we often
observe in our home. The pages of very old books become very much brittle.
In transformer, the aging effect of paper insulation is accelerated due to oxidation occurs
in oil. When insulating paper becomes mechanically weak, it cannot withstand the
mechanical stresses applied during electrical short circuit and leads to electrical
breakdown.
It is therefore necessary to monitor the condition of paper insulation inside a power
transformer.
It is not possible to bring out a piece of paper insulation from a transformer in service for
testing purpose. But we are lucky enough, that there is a testing technique developed,
where we can examine the condition of paper insulation without touching it. The method
is called Furfuradehyde analysis of in short Furfural test.
Although by dissolved gas analysis one can predict the condition of the paper insulation
primarily, but it is not very sensitive method. There is a guide line in IEC-599, where it is
told that if the ratio of CO2 and CO in DGA results is more than 11, it is predicted that the
condition of paper insulation inside the transformer is not good. Healthy cellulose
insulation gives that ratio in a range of 4 to 11. But still it is not a very sensitive way of
monitoring the condition of paper insulation. Because CO2 and CO gases also produced
during oil breakdown and sometimes the ratio may misleads the prediction.
When oil is soaked into paper, it is damaged by heat and some unique oil soluble
compounds are realized and dissolved in the oil along with CO 2 and CO. These
compounds belong to the Furfur aldehyde group. These are sometimes
called Furfural in short. Among all Furfurals compounds 2- Furfural is the most
predominant. This Furfural family compound can only be released from destructive
heating of cellulose or paper. Furfur aldehyde analysis is very sensitive as because
damage of few grams of paper is noticeable in the oil even of a very large size
transformer. It is a very significant diagnostic test. It is best test for assessing life of
transformer. The rate of rise of percentage of Furfurals products in oil, with respect to
time, is used for assessing the condition and remaining life of paper insulation in power
transformer.

14
Benefits of DGA:-
Assesses the internal condition of the transformer.
Helps calculate probability of failure and end of life.
Identifies degradation before it leads to failure.
Essential for effective maintenance and replacement strategies.
Low cost test process.

15
 8. Thermal Ageing:-
8
The service life of a transformer is equal to the service life of its insulation (paper
insulation and mineral oil) & as the ageing of insulation is a function of temperature along
with other parameters, therefore the information about the thermal operation i.e.
temperature rise of the windings or loading condition of the transformer, can give us the
condition of its health.
The ageing of the insulation system is normally affected by pyrolysis, hydrolysis and
oxidation. Among these three processes, pyrolysis involves heating & degradation of
paper insulation. As a result of it, the tensile strength of insulation reduces & oil
decomposes to various gasses (like methane, hydrogen, ethane etc.).The information about
the hot spot temperature gives some idea about the severity of degradation & the
remaining life of the insulation system. The IEC 60076-7:2005 loading guide proposes, in
accordance to Mot siges equation that the ageing of a transformer solid insulation is
doubled for every increase of 6C of the hot-spot temperature in the range 80C - 140C,
for normal Kraft paper, whereas the IEEE C57.91:1995 loading guide proposes another
model for thermally upgraded Kraft paper .

Fig. 8.1:- Hotspot Temperature

In the IEC guide (normal Kraft paper) the paper is expected to rate at a nominal rate at a
constant temperature of 98C. In the IEEE guide (upgraded insulation) the nominal aging
rate is achieved at a constant operating temperature of 110C.
We have considered thermal operation of the power transformer as an indicator of insulation
degradation & therefore take it as a parameter to determine the previous condition of health
of a power transformer. As it is difficult to collect the hot spot temperature for all the
operating years of a transformer therefore scoring is based on the thermal loading chart
.Assuming that if a transformer operates with an average of 90% loading throughout its
service life its insulation will last for 25 years & it gets a score of 10.

16
 9
9. Moisture Content:-
The state of the solid insulation determines the life expectance of a transformer. Moisture,
oxygen and temperature lead to an inevitable decrease of the mechanical strength of the solid
insulation. Insulation grade paper is made by the delignification of wood pulp by the Kraft
process. It contains about 90% Cellulose, 6-7% lignin and the balance is hemicelluloses the
natural humidity of paper is 4-5% by weight and the insulation is dried after winding to less
than 0.5%.Too much moisture in the solid insulation has a detrimental effect on the integrity
of the insulation but high dryness is not also good because to bind the glucose molecules,
water is needed. Therefore assessment of water content in solid insulation is rightfully an
essential part of any comprehensive condition assessment program.

Fig 9.1.1:- PPM Moisture In Oil

9.1. Types of Water in Transformer Oil:-

Water can exist in several different states within the transformer. There are three basic
types of water found associated with transformer oil.
Dissolved water is hydrogen bonded to the hydrocarbon molecules of which oil is
composed.
Emulsified water is supersaturated in solution but has not yet totally separated from
the oil. It usually gives oil a milky appearance.
Free water is also supersaturated in solution but in a high enough concentration to
form water droplets and separate from the oil.

In most cases, when one is analyzing or discussing the amount of water in oil, dissolved
water is being referred to as emulsified, and free water is visually apparent.

17
9.2. Solubility of water in Oil:-
Solubility is defined as the total amount of water than can be dissolved in the oil at a
specific temperature. The solubility of water is not constant in oil but changes due to
temperature. As the temperature increases, the amount of water that can be dissolved in
oil also increases. The increase is not linear but exponential in function. For example, at
10C only 36 ppm of water can be dissolved in the oil, whereas when the temperature
increases to 90C, the amount of water that can be dissolved in the oil increases
tremendously to almost 600 ppm. The table shown lists the calculated solubility limits for
oil at various temperatures.

Oil Temperature Water Content in Oil, PPM

0 22
10 36
20 55
30 83
40 121
50 173
60 242
70 331
80 446
90 592
100 772

Table 9.1.1:- Water in Oil Solubility as a function of Temperature

In the former version of IEEESTD 62 1995 moisture content in solid insulation was
defined as follows:
Dry Insulation:- 0 2%
Wet Insulation:- 2 4%
Very Wet Insulation:- 4.5% +
In our scoring process the moisture content in oil is considered as a deciding factor for
solid insulation moisture content and limiting value for WCO is 30 ppm for 132 KV
transformers and 40 ppm for 33 KV transformers. Scoring is done based on the recent
report of oil testing of each transformer.

18
 10
10. Insulating Oil Condition:-
Insulating oil in an electrical power transformer is commonly known as transformer oil. It is
normally obtained by fractional distillation and subsequent treatment of crude petroleum.
That is why this is also known as mineral insulating oil. Insulating oil, in a transformer,
serves two major purposes i.e. insulation & cooling. Major portion of the insulation system
of a transformer consists of transformer oil; therefore good condition of this oil also ensures
good heath of the transformer.
10.1. Types of Insulating Oil:-
Generally there are two types of Transformer Oil used in transformer,
1. Paraffin based transformer Oil
2. Naphtha based transformer Oil

Naphtha oil is more easily oxidized than Paraffin oil. But oxidation product i.e. sludge in the
naphtha oil is more soluble than paraffin oil. Thus sludge of Naphtha based oil is not
precipitated in bottom of the transformer. Hence it does not obstruct convection circulation
of the oil, means it does not disturb the transformer cooling system. But in the case of
Paraffin oil although oxidation rate is lower than that of Naphtha oil but the oxidation
product or sludge is insoluble and precipitated at bottom of the tank and obstruct the
transformer cooling system.

10.2. Test of Insulating Oil:-

Many characteristics test is done to know about the degradation of the insulating oil, out of
which 4 major tests i.e. Acidity test, BDV test, Resistivity & moisture content have been
chosen as determining factor insulation strength of oil. Among these 4 tests the importance
of acidity and moisture content test has been discussed in the Acid operation & Moisture
operation section earlier.
Break Down Voltage:-
The breakdown voltage of an insulator refers to the minimum voltage which results in part of
an insulator to become an electric conductor. It measures the dielectric strength of the
insulating oil. Greater the value of Break Down Voltage, more the dielectric strength, &
more the ability to withstand at high voltage.

19
 Fig 10.2.1:- BDV Testing Toolkit

Resistivity:-
It represents the opposition in passage of electric current of 1cc of insulating oil, usually
expressed in -cm. For good insulating property, the value of resistivity should be as high
as possible. The actual minimum standard for specific resistance of transformer oil at 90C is
35X1012 cm (maximum) and at the lower boundary is estimated at 27C is 1500*10^12
cm.
In the scoring process for determining insulating oil condition, total score of 10 is divided
into 4 categories. Maximum weightage is put on the Acid No as it causes more harm to the
insulation system.

20
 11
11.Present work
In this present work two FIS system is developed. Six HDI paarameters are grouped into three in two groups.
First group contain service life, moisture content and acid content and the second FIS deals with furan
concentration, maximum operating load and acid number.
The zero-order Mamdani model is given below for both the FIS. Trapezoidal membership function was chosen
by trial and error.

Fig. 11.1 FIS1

Fig. 11.1 FIS 2

21
Fig. 11.3 Input membership functions, Output s of FIS 1

22
Fig. 11.4 Input membership functions, Output s of FIS 2

23
 These two FIS operated on transformer data . Those transformers are located at different location of the city
Kolkata. Two FIS systems are Joined by AND operation.
The simulation results thus obtained

Simulation Results

0 - 2 => Very Good; 2 - 5 => Good; 5 - 8=> Bad, 8 - 10 => Very Bad

Acid
Transfor Thermal Output
Sl. Servic Content / Output 2 Overall Remarks
mer PPM 2FAL Loading 1(S.L,PPM,
No. e Life Acid (2FAL,TL) Output
Name (%) AC)
number
B.T.ROAD
1 5 11 0.025 0 60 2.23
S/S T-1 0.976 3.5 Good
B.T.ROAD
2 4 12 0.025 0 60 2.23
S/S T-2 0.976 3.5 Good
BBD BAG
3 20 12 0.075 10 45 2.23
S/S T-1 3.5 0.976 Good
BBD BAG
4 19 12 0.1 10 45 0.976 2.23
S/S T-2 3.5 Good
BELUR
5 52 39 0.075 2740 70 9.02
R/S T-1 9.02 9.02 Very Bad
BELUR
6 48 34 0.1 4060 70 5.3 7
R/S T-2 9.02 Bad
BELUR
7 52 28 0.075 1010 70 7
R/S T-3 6.5 6.5 Bad
BOTANIC
AL
8 33 10 0.025 340 80 2.18 5.23
GARDEN
S/S T-1 6.5 Bad
BOTANIC
AL
9 52 11 0.025 0 55 3.72
GARDEN
S/S T-2 6.5 0.976 Good
CHAKMIR
10 8 28 0.025 410 45 6.5 3.72
S/S T-1 0.976 Good
CHAKMIR
11 52 60 0.05 3760 45 5.23
S/S T-2 9.02 2.18 Good
DUM
12 DUM S/S 4 11 0.025 0 55 2.18 2.23
T-1 0.976 Good
DUM
13 DUM S/S 3 15 0.025 20 55 2.23
T-2 0.976 3.5 Good
EAST
14 CALCUTT 20 11 0.05 10 65 3.5 2.23
A S/S T-1 3.5 Good

24
 EAST
15 CALCUTT 12 7 0.1 0 60 2.23
A S/S T-2 2.99 3.5 Good
EAST
16 CALCUTT 15 7 0.025 20 60 3.5 2.23
A S/S T-3 3.5 Good

17 EMSS T-1 8 10 0.025 0 40 0.7

0.976 0.976 Very Good

18 EMSS T-2 8 10 0.025 0 40 0.976 0.7

0.976 Very Good

19 EMSS T-3 8 19 0.025 0 45 0.7

0.976 0.976 Very Good

20 EMSS T-4 8 10 0.025 0 40 0.976 0.7

0.976 Very Good

21 EMSS T-5 5 8 0.025 0 40 0.7

0.976 0.976 Very Good
JADAVPO
22 RE S/S T- 3 10 0.075 0 50 0.976 0.7
1 0.976 Very Good
JADAVPO
23 RE S/S T- 5 12 0.025 0 50 0.7
2 0.976 0.976 Very Good
JADAVPO
24 RE S/S T- 9 12 0.025 0 60 3.5 2.23
3 0.976 Good
MAJERHA
25 11 11 0.1 0 65 2.23
T S/S T-1 2.25 3.5 Good
MAJERHA
26 6 11 0.05 0 65 3.5 2.23
T S/S T-2 0.976 Good
MAJERHA
27 33 15 0.025 40 70 5.23
T S/S T-3 6.5 3.5 Good
NCSS 160
28 3 12 0.025 30 70 3.5 2.23
T-1 0.976 Good
NCSS 160
29 3 9 0.025 0 70 2.23
T-2 0.976 3.5 Good
NEW
30 COSSIPOR 25 41 0.4 4660 50 6.5 7
E S/S T-2 6.5 Bad
NEW
31 COSSIPOR 17 14 0.05 540 40 2.23
E S/S T-3 3.5 3.32 Good
PARK
32 LANE S/S 25 26 0.05 10 60 3.5 5.23
T-1 5.61 Good
PARK
33 LANE S/S 20 37 0.2 5930 65 7
T-2 3.5 9.02 Bad
PATULI
34 3 9 0.025 0 35 0.7
S/S T-2 0.976 0.976 Very Good

25
 PATULI
35 3 9 0.025 0 35 0.7
S/S T-1 0.976 0.976 Very Good
PRINSEP
36 40 21 0.075 210 50 3.72
R/S T-1 6.5 0.976 Good
PRINSEP
37 STREET 40 9 0.025 0 50 3,72
R/S T-2 6.5 0.976 Good
PRINSEP
38 STREET 4 12 0.025 0 50 0.7
R/S T-3 0.976 0.976 Very Good
PRINSEP
39 STREET 6 22 0.075 10 75 3.72
R/S T-4 0.976 6.5 Good
SOUTHER
40 6 35 0.288 910 80 3.72
N R/S T-1 3.5 6.5 Good
SOUTHER
41 6 47 0.05 1840 70 3.72
N R/S T-2 3.5 6.5 Good
SOUTHER
42 21 52 0.05 3480 70 7
N R/S T-3 3.5 9.02 Bad
TITAGAR
43 34 27 0.05 470 70 5.23
H R/S T-1 6.5 3.5 Good
TITAGAR
44 21 33 0.05 480 70 2.23
H R/S T-2 3.5 3.5 Good
TITAGAR
45 32 23 0.075 1340 65 7
H R/S T-3 6.5 6.5 Bad
TITAGAR
46 23 29 0.05 870 65 2.23
H R/S T-4 4.07 3.5 Good
TITAGAR
47 23 28 0.05 220 60 2.23
H R/S T-5 3.85 3.5 Good
WEST
48 BANK R/S 6 17 0.05 0 70 2.23
T-1 0.976 3.5 Good
WEST
49 BANK R/S 5 11 0.025 10 70 2.23
T-2 0.976 3.5 Good

26
BLOCK DIAGRAM:-

Service Life

Moisture Fuzzy Inference

Content Output
System

Acid Content

Health
Deterioration
Indexing

Furan
Concentration
Fuzzy
Inference Output
System
Thermal
Ageing

Fig. 11.5- Block Diagram Of HDI Method

27
 12
12.Conclusion:-
Now a days health of transformer should be checked up regularly to avoid damages of
transformers. This project will help to observe the health condition of a transformer and we
can take action as required. We have used Fuzzy Logic System to do it more perfectly.
Available result shows slight difference with the test result. The Expert system needs fine
tuning in future work.
The Fuzzy models can also be tried with different Configuration to give a Robust HDI
system.
This work will mark one of the few attempts and contributions in the world in the field of
transformer health indexing using Fuzzy logic, where such projects could be implemented
extensively. With the increasing improvements in Fuzzy technologies and Health
deterioration indexing, such projects would have more value added and should receive more
attention and support.

28
 References
1. C. E. Lin, J. M. Ling, and C. L. Huang, Expert system for transformer fault diagnosis using dissolved gas
analysis, IEEE Transactions on Power Delivery, vol. 8, no. 1, pp. 231238, 1993. View at Publisher View at
Google Scholar View at Scopus
2. Y.-C. Huang, H.-T. Yang, and C.-L. Huang, Developing a new transformer fault diagnosis system through
evolutionary fuzzy logic, IEEE Transactions on Power Delivery, vol. 12, no. 2, pp. 761767, 1997.View at
Publisher View at Google Scholar View at Scopus
3. H.-T. Yang and C.-C. Liao, Adaptive fuzzy diagnosis system for dissolved gas analysis of power
transformers, IEEE Transactions on Power Delivery, vol. 14, no. 4, pp. 13421350, 1999. View at
Publisher View at Google Scholar View at Scopus
4. Y. Zhang, X. Ding, Y. Liu, and P. J. Griffin, An artificial neural network approach to transformer fault
diagnosis, IEEE Transactions on Power Delivery, vol. 11, no. 4, pp. 18361841, 1996. View at Publisher View
at Google Scholar View at Scopus
5. J. L. Guardado, J. L. Naredo, P. Moreno, and C. R. Fuerte, A comparative study of neural network efficiency in
power transformers diagnosis using dissolved gas analysis, IEEE Transactions on Power Delivery, vol. 16, no.
4, pp. 643647, 2001. View at Publisher View at Google Scholar View at Scopus
6. D. V. S. S. Siva Sarma and G. N. S. Kalyani, ANN approach for condition monitoring of power transformers
using DGA, in Proceedings of the IEEE Region 10 Conference (TENCON '04), vol. 3, pp. C444C447,
November 2004. View at Scopus
7. W. M. Lin, C. H. Lin, and M. X. Tasy, Transformer-fault diagnosis by integrating field data and standard codes
with training enhancible adaptive probabilistic network, IEE Generation Transmission Distribution, vol. 152,
no. 3, pp. 335341, 2005. View at Google Scholar
8. Y.-C. Huang, Evolving neural nets for fault diagnosis of power transformers, IEEE Transactions on Power
Delivery, vol. 18, no. 3, pp. 843848, 2003. View at Publisher View at Google Scholar View at Scopus
9. R. Naresh, V. Sharma, and M. Vashisth, An integrated neural fuzzy approach for fault diagnosis of
transformers, IEEE Transactions on Power Delivery, vol. 23, no. 4, pp. 20172024, 2008. View at
Publisher View at Google Scholar View at Scopus
10. D. Bhalla, R. K. Bansal, and H. O. Gupta, Application of artificial intelligence techniques for Dissolved Gas
Analysis of transformersa review, World Academy of Science, Engineering and Technology, vol. 62, pp.
221229, 2010. View at Google Scholar View at Scopus
11. D. R. Morais and J. G. Rolim, A hybrid tool for detection of incipient faults in transformers based on the
dissolved gas analysis of insulating oil, IEEE Transactions on Power Delivery, vol. 21, no. 2, pp. 673680,
2006. View at Publisher View at Google Scholar View at Scopus
12. R. A. Hooshmand, M. Parastegari, and Z. Forghani, Adaptive neuro-fuzzy inference system approach for
simaltaneous diagnosis of the type and location of faults in power transformers, IEEE Electrical Insulation
Magzine, vol. 28, no. 5, pp. 3242, 2012. View at Google Scholar
13. Z. Wang, Y. Liu, and P. J. Griffin, A combined ANN and expert system tool for transformer fault
diagnosis, IEEE Transactions on Power Delivery, vol. 13, no. 4, pp. 12241229, 1998. View at Publisher View
at Google Scholar View at Scopus
14. Mathworks, Fuzzy Logic Toolbox, User's Guide.
15. D. Driankov, H. Hellendoorn, and M. Reinfrank, An Introduction to Fuzzy Control, Narosa Publishing House,
New Delhi, India.
16. G. J. Klir and T. A. Folger, Fuzzy Sets, Uncertainity and Information, PHI, New Delhi, India.
17. M. McNeill and E. Thro, Fuzzy Logic, A Practical Approach.
18. IEEE guide for dissolved gas analysis in transformer load tap changers, IEEE Standard C57.139-2010.

29