A Novel Application of Gene Expression Programming in Transformer Diagnostics

jinksimaginaryAI and Robotics

Nov 7, 2013 (3 years and 10 months ago)

172 views

A Novel Application of Gene Expression
Programming in Transformer Diagnostics

A.Abu-Siada, Lai Sin Pin and Syed Islam
Curtin University of Technology, Australia


Abstract-Furans are the major degradation of insulating
paper in transformer oil. Hence the concentration of furans in oil
can be used as a good indicator of paper deterioration. Furan
concentration in transformer oil is currently measured using
High-Performance Liquid Chromatography (HPLC) or Gas
Chromatography-Mass Spectrometry (GC/MS). Both methods
provide accurate and reliable results in detecting furan
concentration. However, the two methods need very expensive
equipments and take long time to get the result for one sample.
Moreover it requires a trained expert to perform and interpret
the results. This paper introduces a novel approach for detecting
furan concentration in transformer oil through measuring its
spectral response. The Ultraviolet-to-Visible (UV-Vis) 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. Also, the
paper introduces a novel application for Gene Expression
Programming (GEP) to estimate the relationship between furan
concentration and spectral response of transformer oil.
I. INTRODUCTION
Power transformers are a vital link in a power system.
Monitoring and diagnostic techniques are essential to decrease
maintenance and improve reliability of the equipment.
Currently there are several of chemical and electrical
diagnostic techniques applied for power transformers[1]. As
the entire energized and high temperature transformer
components are immersed in the transformer oil, the
transformer oil is a key source to detect incipient faults, fast
developing faults, insulation trending and generally reflects the
health condition of the transformer. The electrical windings in
a power transformer consist of paper insulation immersed in
insulating oil. Paper insulation is composed of approximately
90% of cellulose, 6-7% hemi-cellulose and 3-4% of lignin.
Due to electrical and thermal stresses, oil and cellulose
decomposition occurs evolving gases that will decrease the
heat dissipation capability and the dielectric strength of the oil.
When degradation of paper insulation occurs, the cellulose
molecular chains get shorter and chemical products such as
furanic derivatives are produced and dissolve in the oil. Furans
are the major degradation of paper in oil and the 2-furaldehyde
in oil is the most prominent component of paper decomposition.
The increase in furans concentration in the oil corresponds to
the decrease in the tensile strength and the degree of
polymerization (DP) of the paper. De Pablo reported the
following relation between furfural and Degree of
polymerization based on viscosity DP
v
[2] :
FAL
DP
V
2
88
.
8
7100
+
= (1)
where 2FAL is the furfural concentration in mg/kg of oil.
Then the concentration of 2-furfuraldehyde in the oil can be
used as a good indicator of paper deterioration. It has been
estimated that new paper, under normal running conditions will
generate furfural at the rate of 1.7 ng/g of paper/h. The rate of
production increases with increasing degree of degradation and
a total yield of 0.5 mg of furfural/g of paper is expected in
100000 h of running (15-20 years)[3-5]. Furan level in a
transformer can be correlated with paper DP, and therefore an
in-service assessment of the mechanical strength of the paper
insulation can be made. Furans concentration is measured by
High-Performance Liquid Chromatography (HPLC) or Gas
Chromatography-Mass Spectrometry (GC/MS) based on
American Society for Testing and Materials (ASTM D5837,
Standard Specifications for Mineral Insulating Oil in Electrical
Apparatus)[6]. Both analysis methods provide accurate and
reliable results in detecting the concentration of furan
derivatives; 2- furfural (2-FAL), 2-Furfurol (2-FOL), 5-Hyroxy
methyl-2-furfural (5-HMF), 5-Methyl-2-furfural (5-MEF), and
2-Acetylfuran (2-ACF). However, these two methods need
very expensive equipments and take long time to get the results
in addition it requires for an expert person to perform and
interpret the results. The technique for furan testing is
described in IEC, but there is no guideline for interpretations.
This paper presents a novel approach for determination of
furan concentration in transformer oil through measuring its
spectral response. The Ultraviolet-to-Visible (UV-Vis) 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.
II. LABORATORY AGED OIL
The study has been performed on in-service as well as
laboratory aged transformer oil. Laboratory aged insulating oil
is prepared by utilizing the heating process available in IEC
61125[7]. Section of new craft paper (20mmx280mm) was cut
and wrapped around copper strips (3mmx10mm) then it was
impregnated in 25ml of new transformer oil (shell Diala B).
All samples were heated up to 100°C in a thermostatically-
controlled aluminum alloy block heater for 7 days. Oxygen
2008 Australasian Universities Power Engineering Conference (AUPEC'08)
Paper P-054 page 1
flow at a rate of 1 l/hr was supplied into each dry tube to
further accelerate the aging process.
III. FURAN ANALYSIS
All samples were prepared in accordance to standard ASTM
D 5837 and tested using GC/MS system for furan derivatives
identification and quantification. Table I shows furan
derivative concentration in particle per million (ppm) for
different oil samples using GC/MS. Results show that the 2-
furaldehyde (2-FAL) is the most prominent component of
paper decomposition. Therefore, the level of 2-furaldehyde in
transformer oil can be used as an indicator for paper
deterioration.
Table I
Furan concentration result by GC/MS
Test Sample 2-FAL 2-FOL 2-ACF 5-MEF 5-HMF
New Oil <0.01 <0.01 <0.01 <0.01 <0.01
Sample 1 3.1 <0.01 0.01 0.01 <0.01
Sample 2 5.1 <0.01 0.02 0.01 <0.01
Sample 3 10.0 <0.01 0.03 0.03 <0.01
Sample 4 15.0 0.01 0.05 0.05 <0.01
The correlation between 2-furaldehyde and DP with respect
to the solid insulation extent of damage is given in table II
depicting insulation paper dielectric and mechanical properties.
When DP test reveals a value of 250 or less, the paper is
considered to have lost all its mechanical strength, and the
transformer has reached its end of life [8].
Table II
DP and 2-Furfuraldehyde (2-FAL) Correlation
2-FAL (ppm) DP Value Significance
0-0.1 1200-700 Healthy Insulation
0.1-1.0 700-450 Moderate Deterioration
1-10 450-250 Extensive Deterioration
> 10 < 250 End of Life Criteria

IV. UV-VIS SPECTRAL RESPONSE
UV-Spectrophotometry is a non-intrusive test used to
determine the transformers integrity. UV-Spectrophotometry
is an accurate and sensitive method to analyze impurities in the
transformer oil using light absorbing properties of a sample.
Light transmitted through the oil sample containing various
contaminations is decreased by that fraction being absorbed
and is detected as a function of wavelength[9]. A
spectrophotometer measures the transmission, absorption or
reflection of the light spectrum for a given wavelength.
Absorption spectroscopy provides a measure of how much
light is absorbed by the oil sample which can be calculated
as[2]:










=



DR
DS
A log
(2)
Where

A is the absorbance, S is the sample intensity at
wavelength

, D is the dark intensity at wavelength

, R is
the reference intensity at wavelength

.
Same oil samples used for furan concentration measurement
using GC/MS were tested using a laboratory grade
spectrophotometer for absorbance spectrophotometry. The
experiment procedure was set up in reference to ASTM
E275[10]. Figure 1 shows the lab set up for measuring spectral
response for one oil sample. Figure 2 shows the spectral
response (absorbance) for different oil samples with different
furan concentration.



Figure 1. Lab set up for measuring the spectral response of transformer oil


Figure 2. UV/Vis Spectrum (Absorbance) for different oil samples with
different furan concentration

It can be shown from Fig. 2 that the new oil exhibits its
characteristics between 200 and 350nm uv-spectrum with
maximum absorbance at 250nm wavelength. However, in-
service and laboratory aged oil samples exhibit their respective
characteristics in the range of 200 and 470nm wavelength uv-
spectrum. Results show that absorbance as well as bandwidth
for maximum absorbance increases by a significant and easily
observable margin with oil deterioration and contaminations
which are reflected by the furan concentration level in the oil.
UV spectrum shows considerable noise for contaminated oil
which can be attributed to the variety of contaminations
including very high carbon and water content in the oil. Figure
2 shows a good correlation between the furan concentration in
transformer oil and its spectral response. To prove this
correlation in more details, Figure 3 shows the relationship
between spectral response parameters (maximum absorbance
2008 Australasian Universities Power Engineering Conference (AUPEC'08)
Paper P-054 page 2
and bandwidth wavelength) of transformer oil and its furan
concentration level; the more furan concentration the more
absorbance and more wavelength. Then the spectral response
parameters can be used as an alternative method to GC/MS to
determine the furan concentration in transformer oil. Spectral
response of transformer oil can be measured instantly with
relatively cheap equipment and does not need an expert person
to conduct the test. In the next section, Gene Expression
Programming (GEP) software tool is used to find a
mathematical correlation between furan concentration in
transformer oil and its spectral response.


Figure 3. Correlation between furan concentration in transformer oil and its
spectrum response parameters (wavelength and peak absorbance)

V. GENE EXPRESSION PROGRAMMING
Gene Expression Programming (GEP) is a new technique for
data analysis was first invented by Candida Ferreira in 1999[11,
12]. GEP is a learning algorithm that can find relationships
between variables in sets of data and builds models to explain
these relationships. GEP is similar to genetic algorithm (GA)
and genetic programming (GP) as it uses populations of
individuals, selects them according to fitness, and introduces
genetic variation using one or more genetic operators. The
nature of individuals is the fundamental difference between the
three algorithms. In GA the individuals are linear strings of
fixed length (chromosomes); in GP the individuals are
nonlinear entities of different sizes and shapes (parse tree); and
in GEP the individuals are encoded as linear strings of fixed
length expressed as nonlinear entities of different sizes and
shapes so that GEP combines the advantage of both GA and
GP while overcomes their some shortcomings[13]. One
important application of GEP is symbolic regression or
function finding, where the goal is to find expression that
performs well for all fitness cases within a certain error of the
correct value. In GEP individuals are selected according to its
fitness by roulette-wheel sampling with elitism and the best
individual is preserved. Mathematically, the fitness
i
f
of an
individual program i is expressed by the following
equation[14]:

=
=
t
C
j
jjii
TCMf
1
,
)(
(3)
Where M is the range of selection, C
i,j
is the value returned by
the individual chromosome i for fitness case j (out of C
t
fitness
cases), and T
j
is the target value for fitness case j. Figure 4
illustrates the flowchart of the typical GEP algorithm[12].


Figure 4. The flowchart of GEP algorithm
GEP genes are composed of head and tail. The head contains
symbols that represent both functions (elements from the set
function F) and terminals (elements from the terminal set T),
whereas the tail contains only terminals. Therefore two
different alphabets occur at different regions within a gene. For
each problem, the length of the head h is chosen, whereas the
length of the tail t is a function of h and the number of
arguments of the function with the most arguments n, and is
evaluated by the equation:
T=h(n-1)+1 (4)
TABLE III
GEP Model results
Target Model Residual
0.5 0.539963 4.00E-02
1 0.886286 0.113714
2 1.963771 3.62E-02
3 3.139409 0.139409
4 4.014475 1.45E-02
5 4.912101 8.79E-02
7 7.12213 0.12213
11 11.13847 0.138472
12 11.85625 0.143752
13 12.98926 1.07E-02
15 14.94548 5.45E-02
2008 Australasian Universities Power Engineering Conference (AUPEC'08)
Paper P-054 page 3
To find the correlation between furan concentration in
transformer oil and its spectral response parameters, the
bandwidth and the maximum absorbance for each oil sample
are provided as the input (independent) parameters to the GEP
while the corresponding furan concentration is provided as the
target. Some other parameters should be specified to GEP such
as number of Genes, number of chromosomes, length of head
and the set functions. Table III shows the GEP model results
compared to the actual experimental measurements. The
absolute error between GEP model and the actual
measurements is given in the third column of table III. The
maximum error between the GEP model and the actual
measurements is 0.143752. .


Figure 5. GEP model and actual correlation between Furan and spectral
response parameters


Figure 6. Expression tree for the GEP model

Figure 5 shows the relationship between furan concentration
and spectral response parameters for experimental
measurements and GEP model. In Fig. 5, the vertical axis
represents the output (furan concentration in ppm), However, it
is difficult to represent the horizontal axis by real values as it
represents the two independent input variables (maximum
absorbance and bandwidth). From table III and Fig. 5, it can be
concluded that the GEP model is very accurate and efficient in
modeling the correlation between furan concentration in
transformer oil and its spectral response. The mean square
error between the model and target is only 0.065 and the
correlation coefficient is 0.9987. The model comes with an
Expression Tree (ET) (shown in Figure 6) that representing the
input/output relationship. The ET is translated to a
mathematical expression using the same program which
provides the model in many computer languages including
Matlab, FORTRAN, C++, Visual basic etc.

To check the capability of the model to estimate the furan
concentration in the transformer oil using its spectrum response
parameters; different random spectrum parameters have been
provided to the model as shown in table IV. The spectrum
parameters have been chosen to be inside and outside the
spectrum response range shown in Figure 2. The estimated
furan concentration corresponding to each parameter coincides
with the practical results shown in Figure 2. This proves the
capability of the model to interpolate/extrapolate the given
parameters to estimate a precise furan concentration in the
transformer oil.

Table IV
Capability of GEP model to interpolate/extrapolate results
BW (nm) Max. Absorbance Furan (ppm)
400 1.6 1.98714443
417 1.75 3.75018108
440 1.82 9.20123901
450 1.85 11.0897414
460 1.9 13.8267452
475 1.95 17.4311133

VI. CONCLUSIONS
Results show that there is a good correlation between oil
spectral response and its furan contents. Gene Expression
Programming has been used to find this correlation
mathematically. The great insight of GEP consisted in the
invention of chromosomes capable of representing any
expression tree. The structural and functional organisation of
GEP genes and their interplay with expression trees always
guarantee the production of valid programs. The paper has
introduced a novel application for GEP in transformer
diagnostics. Results show that GEP model can estimate
precisely the furan concentration of transformer oil using the
spectral response parameters (bandwidth and maximum
absorbance). The UV-Vis spectral response of transformer oil
can be measured instantly with relatively cheap equipment and
does not need an expert person to conduct the test or to
interpret the results or to interpret the results. Results show that
the model is very efficient and can replace the current
expensive and time consuming techniques to detect furan
concentration in transformer oil.

2008 Australasian Universities Power Engineering Conference (AUPEC'08)
Paper P-054 page 4
REFERENCES
[1] T. K. Saha, "Review of Modern Diagnostic techniques for
Assessing Insulation Condition Aged Transformers," IEEE
Transaction on Dielectrics and Electrical Insulation, vol. 10, pp.
903-917, 2003.
[2] M. Arshad, "Remnant Life Estimation Model Using Fuzzy Logic
for Power Transformer Asset Management," PhD thesis, Curtin
University of Technology, 2005.
[3] Y. Shang, L. Yang, Z. J. Guo, and Z. Yan, "Assessing aging of
large transformers by furfural investigation," in Solid Dielectrics,
2001. ICSD '01. Proceedings of the 2001 IEEE 7th International
Conference on, 2001, pp. 272-274.
[4] C. H. Zhang and J. M. K. Macalpine, "Furfural Concentration in
Transformer Oil as an Indicator of Paper Ageing, Part 1: A
Review," in Power Systems Conference and Exposition, 2006.
PSCE '06. 2006 IEEE PES, 2006, pp. 1088-1091.
[5] J. M. K. MacAlpine and C. H. Zhang, "Observations from
measurements of the furfural content of oil samples from
transmission transformers," in Advances in Power System Control,
Operation and Management, 2000. APSCOM-00. 2000
International Conference on, 2000, pp. 317-321 vol.2.
[6] ASTM, "Standard Test Method for Furanic Compounds in
Electrical Insulating Liquids by High-Performance Liquid
Chromatography (HPLC)," D5837-05, 2005.
[7] ITC, "Unused Hydrocarbon-Based Insulating Liquids-Test Methods
for Evaluating The Oxidation Stability " IEC61125, 1992.















































[8] M. Wang, A. J. Vandermaar, and K. D. Srivastava, "Review of
condition assessment of power transformers in service," Electrical
Insulation Magazine, IEEE, vol. 18, pp. 12-25, 2002.
[9] J. A. Palmer, W. Xianghui, R. A. Shoureshi, A. Mander, D.
Torgerson, and C. Rich, "Effect of aging on the spectral response of
transformer oil," in Electrical Insulation, 2000. Conference Record
of the 2000 IEEE International Symposium on, 2000, pp. 460-464.
[10] ASTM, "Standard Practice for Describing and Measuring
Performance of Ultraviolet, Visible, and Near-Infrared
Spectrophotometers,," ASTM E275-01, vol. 03.06, pp. 72-81, 2001.
[11] W. Chuan-Sheng, H. Li, and K. Li-Shan, "The automatic modeling
of complex functions based on gene expression programming," in
Machine Learning and Cybernetics, 2005. Proceedings of 2005
International Conference on, 2005, pp. 2870-2873 Vol. 5.
[12] G. Zhaohui, L. Gaobin, Y. Zhenkun, and J. Min, "Automatic
Modeling of Complex Functions with Clonal Selection-Based Gene
Expression Programming," in Natural Computation, 2007. ICNC
2007. Third International Conference on, 2007, pp. 228-232.
[13] C. Ferreira, "Gene Expression Programming: Mathematical
Modeling by an Artificial Intelligence," Springer, 2006.
[14] C. Ferreira, "Gene Expression Programming: a New Adaptive
Algorithm for solving Problems," Complex Systems, vol. 13, pp. 87-
129, 2001.


2008 Australasian Universities Power Engineering Conference (AUPEC'08)
Paper P-054 page 5