A6/DO /y {>{•) 
DEVELOPMENT OF A NEURO-FUZZY SYSTEM 
FOR CONDITION MONITORING OF POWER 
TRANSFORMERS 
mwr.TTTV 0.'- SRI LAWK& 
i W O R A T U W A 
A dissertation submitted to the 
Department of Electrical Engineering, University of Moratuwa 
in partial fulfilment of the requirements for the 
Degree of Master of Science 
by 
JAGATH RENUKA SELLAHANNADI 
Supervised by: A/Prof. Lanka Udawatta 
Eng. W.D.Anura S Wijayapala 
Eng. K.P.Kusum Shanthi 
Department of Electrical Engineering 
University of Moratuwa, Sri Lanka 
January 2010 
University of Moratuwa 
1 
94540 
94540 
D E C L A R A T I O N 
The work submitted in this dissertation is the result of my own 
investigation, except where otherwise stated. 
It has not already been accepted for any degree, and is also not 
being concurrently submitted for any other degree. 
J.R.SEffaftlTnnacii 
Date (ci\^o)° 
We/I endorse the declaration by the candidate. 
A/Prof. Lanka Udawatta 
Eng. W.D.Anura S Wijayapala 
Eng. K.P.Kusum Shantlii 
Contents 
Declaration i 
Abstract iv 
Acknowledgement v 
* List of Figures vi 
List of Tables vii 
| Acronyms and Abbreviations viii 
Chapters 
1. Introduction 01 
1.1 General 01 
1.2 Problems Statement 02 
1.3 Objectives 02 
1.4 Methodology 02 
1.5 Achievement 02 
2. Power Transformer 03 
2.1 General 03 
2.2 Lifetime of a Power Transformer 03 
2.2.1 Economical Lifetime 03 
2.2.2 Technical Lifetime 04 
2.2.3 Ageing of Transformer 04 
3. Power Transformer Maintenance 07 
3.1 Why Maintain and Test 07 
3.2 Maintenance Strategies 07 
3.2.1 Run-to-Failure (RTF) Approach 07 
3.2.2 Inspect and Service as Necessary Approach 08 
3.2.3 Time-Based Maintenance (TBM) 08 
3.2.4 Condition-Based Maintenance (CBM) 08 
3.2.5 Reliability Centered Maintenance (RCM) 08 
3.3 Power Transformer Testing and Test Methods 09 
3.3.1 Dielectric Breakdown Voltage Test (Cup Tests) 11 
3.3.2 Colour Test 11 
3.3.3 Acidity Test 11 
3.3.4 Power Factor Test 11 
3.3.5 Water Content Test (Karl Fisher Method) 12 
3.3.6 Combustible Gas Analysis of Insulating Oil 12 
3.3.6.1 Faults and Indicator Gases 14 
3.3.6.2 Total Combustible Gas Analysis (TCGA) 15 
3.3.6.3 Dissolved Gas Analysis (DGA) 16 
3.3.6.3.1 Most commonly used DGA 16 
diagnosis methods 
ii 
,4 
4. Proposed Condition Monitoring System for Transformers 20 
4.1 Overview of the Proposed Monitoring Expert System 20 
4.2 The Condition Monitoring System 21 
4.2.1 Condition Monitoring System for DGA 22 
4.2.2 Condition Monitoring System for Oil Testing 27 
* 4.2.3 Condition Monitoring System for Insulation Resistance (IR) 
Testing 28 
| 4.3 Evaluation of DGA Test Results 30 
5. Adaptive Ncuro-Fuzzy Model for DGA 36 
5.1 Neural Networks 37 
5.2 Artificial Neural Networks 38 
5.3 Fuzzy Logic Analysis 38 
5.4 Adaptive Neuro-Fuzzy Inference System (ANFIS) 39 
5.5 ANFIS for DGA 39 
5.5.1 Checking the model for accuracy 43 
5.5.2 Analysis of ANFIS results 46 
6. Conclusion and Further Developments 49 
6.1 Conclusion 49 
* 6.2 Further development 50 
^ References 51 
Appendices 53 
Appendix A Program used for ANFIS 53 
in 
Abstract 
Well-being of Power Transformers is crucial to the reliable operation of a Power 
System. They represent a high capital investment in a Transmission Substation while 
being a key element determining the loading capability of the station within the 
network. With appropriate maintenance, including insulation reconditioning at the 
appropriate time, the technical life of a transformer can be extended. Assessment of 
Power Transformer condition is very important to maintenance engineers on the way 
to diagnose incipient faults and implementation of necessary maintenance plans to 
prolong their life span. Therefore different testing methods and diagnosis techniques 
are used for condition assessment of transformers; namely Dissolved Gas Analysis 
(DGA), moisture content measuring tests in oil/paper, Insulation Resistance (IR) 
measurement, acidity in oil etc. Accuracy of the final conclusion depends on the 
experience and knowledge of the maintenance engineer and the data which he referred 
to. Therefore, it is appropriate to have an Expert System, as a guide to maintenance 
engineers in the Ceylon Electricity Board (CEB), so as to address the above problems. 
This thesis describes the degradation of insulation, ageing process, faults and testing 
methods of transformers. Much attention was paid to DGA as a diagnosis tool. This 
thesis introduces two computer based expert systems to analyze the results from 
various diagnosis techniques and tests which are used in CEB at present. First 
program was written on Visual Basic environment and included essential tests 
including DGA which are carried out by CEB for its transformers. Knowledge base 
for this program was developed by using various standards, text books, transformer 
manufacturers' recommendations and the opinions of my supervisors and experienced 
engineers. Twenty Five (25) numbers of DGA test results of transformers were 
analyzed by using this program and such transformers were grouped according to the 
IEEE standards. Limitations of conventional DGA methods with frequent non-
decisions can be addressed by fuzzy-logic based diagnosis for power transformer 
incipient faults. Therefore, the second program was developed to meet the above 
demand by using Adaptive Neuro-Fuzzy Inference System (ANFIS) in MATLAB 
Environment. The developed ANFIS-based diagnosis system provides further 
improvement to fuzzy-logic based techniques by providing auto-learning capabilities. 
This program was tested for faults with 30 DGA test results and the outcome is within 
the satisfactory level. 
iv 
Acknowledgement 
I am heartily thankful to my supervisors, A/Professor Lanka Udawatta, Eng. W. D. 
Anura S Wijayapala, Eng. K. P. Kiisuin Shanthi, whose encouragement, guidance and 
support from the initial to the final level enabled me to develop an understanding of 
the subject. 1 would like to show my gratitude to the officers in Post Graduate Office, 
Faculty of Engineering, University of Moratuwa, Sri Lanka for helping in various 
ways to clarify the matters related to my academic work in time with excellent 
cooperation and guidance. I would also like to thank the people who serve in the 
Department of Electrical Engineering office. 
Lastly, I offer my regards to all of those who supported me in many respects during 
the completion of the project. 
v 
( List of Figures 
Figure Page 
v ( • 
vi 
2.1 Cellulose Molecule 04 
2.2 Diagram of Water Content in Oil and Paper 05 
2.3 Effect of Moisture on Dielectric Strength of Oil (Chart) 06 
3.1 Structure of RCM 09 
3.2 Combustible Gas Generation versus Temperature 13 
3.3 Halstead's Thermal Equilibrium Diagram 17 
4.1 Block Diagram of the Neuro-Fuzzy condition monitoring System 20 
4.2 Main Screen of the condition monitoring system 21 
4.3 DGA Results Entering Form 22 
4.4 IEEE DGA Test Result Analysis Outcome Screen 22 
4.5 IEC DGA Test Result Analysis Outcome Screen 24 
4.6 Flowchart for Extended IEC 599 25 
4.7 Age Compensated Outcome Screen 26 
4.8 Entering Form for Oil Testing Results 27 
4.9 Analyzed Results Screen for Oil Testing 27 
4.10 Allowable Values of Transformer Insulation Resistance 28 
4.11 Insulation Resistance Results Entering Form 29 
4.12 Program Outcome Screen for Insulation Resistance Analysis 29 
4.13 IEEE condition 01 Satisfied Transformers 32 
4.14 IEEE condition 02 satisfied transformers 33 
4.15 IEEE condition 03 satisfied transformers 34 
4.16 IEEE condition 04 satisfied transformers 35 
5.1 A Neuron 37 
5.2 Gaussian Bell Membership Function for C2H2/C2H6 41 
5.3 ANFIS Model Structure for DGA 41 
5.4 Developed Rule Base for IEC 599 42 
5.5 Fuzzy Rule Viewer for DGA 42 
5.6 Graphical Representation of Expected Outcome for Testing Data 45 
5.7 Graphical Representation of Expected Outcome and Calculated 45 
Outcome for Testing Data 
5.8 Duval Triangle Outcome for Testing Data 46 
List of Tables 
Table Page 
3.1 Transformer Inspection and Maintenance Checklist 10 
3.2 Acceptable values of the neutralization numbers for insulating oil 11 
3.3 Dornenburg Ratio-Fault Diagnosis Table 17 
3.4 Initial Rogers Ratio-Fault Diagnosis Table 18 
3.5 IEC 60599 evaluation code 19 
4.1 IEEE Dissolved gas concentration limits 23 
4.2 DGA analysis results for 25 transformers 31 
5.1 DGA evaluation code for Extended IEC 599 40 
5.2 Output (Fault) membership functions and assigned values 43 
5.3 Data set for 30 transformer faults 44 
5.4 Comparison of evaluated results 46 
5.5 Gas concentration data table for erroneous results 47 
5.6 Gas concentration comparison table for erroneous results 48 
vii 
Acronyms and Abbreviations 
°c Centigrade 
ANFIS Adaptive Neuro-Fuzzy Inference System 
A N N Artificial Neural Networks 
ASTM American Society for Testing and Materials 
C 2 H 2 Acetylene 
C 2 H 4 Ethylene 
C 2 H 6 Ethane 
C 3 H 6 Propylene 
CBM Condition-Based Maintenance 
CEB Ceylon Electricity Board 
CH4 Methane 
CO Carbon monoxide 
C 0 2 Carbon dioxide 
DGA Dissolved Gas Analysis 
DP Degree of Polymerization 
EPM Electrical Preventive Maintenance 
FIS Fuzzy Inference System 
FL Fuzzy Logic 
H 2 Hydrogen 
IEC International Electrotechnical Commission 
IEEE Institute of Electrical and Electronic Engineers 
IR Insulation Resistance 
KOH Potassium hydroxide 
kV Kilovolts 
kVA Kilovoltampere 
MTBF Mean Time Between Failures 
MVA Mega-volt-amps 
PD Partial Discharge 
PM Preventive Maintenance 
ppm Parts per million 
RCM Reliability-Centered Maintenance 
RTF Run-to-Failure 
TBM Time-Based Maintenance 
TCGA Total Combustible Gas Analysis 
TF Transformer 
viii