Journal of Physics: Conference Series

PAPER • OPEN ACCESS You may also like

- High-Frequency Transformer Design with
Prediction of Transformer Health Index Using Hollow Core for Solid State Transformer
Haitham A Obaid and Yasir M Y Ameen
Condition Situation Monitoring (CSM) Diagnostic - The Illustrated Wavelet Transform
Handbook: Introductory Theory and
Techniques Applications in Science, Engineering,
Medicine and Finance
J Ng and N G Kingsbury
To cite this article: AM Abdullah et al 2021 J. Phys.: Conf. Ser. 1878 012007
- Correct use of Lorentz - Einstein
transformation equations for
electromagnetic fields
Oleg D Jefimenko

View the article online for updates and enhancements.

This content was downloaded from IP address 203.202.253.178 on 09/10/2023 at 10:36

ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

Prediction of Transformer Health Index Using Condition

Situation Monitoring (CSM) Diagnostic Techniques

AM Abdullah1,2, R Ali1, SB Yaacob1, TMNT Mansur1, NH Baharudin1

1
Universiti Malaysia Perlis (UniMAP),
Faculty of Electrical Engineering Technology, Perlis, Malaysia
Email : [email protected]
2
Public Works Department Malaysia (JKR), Kuala Lumpur, Malaysia
Email : [email protected]

Abstract. This paper presents newly developed condition situation monitoring (CSM) diagnostic
techniques for predicting the health index and health condition of practical cases in any working
transformer. Investigation on measurement and testing can produce results on the insulation
resistance between phase and ground test, polarization index with different time duration test,
winding resistance of transformer while winding tap changer rating transformer can be
determined by the performance health index of the transformer. Real data was collected from 8
units of transformers from Company Transformer Manufacturing and Servicing (CTMS) in
Malaysia. The data was applied to calculate the healthy index to provide the transformer
performance indicator condition. The main purpose for the use of transformer health index is to
indicate the condition of the transformer and attend to failures that can be prevented, with
consequences on safety and cost.

1 Introduction
Currently, many industries monitoring condition transformers specially in insulation. It also diagnosis
and data interpretation for asset management. Condition Transformer is base on the electrical,
mechanical and also environmental stresses.
Previously, the transformer insulation systems of the oil paper transformer normally using the oil
characteristic tests for evaluating the quality. Some also using the dissolved gas analysis (DGA) to
indicate health condition. It may detect the fault by using the dissolved gas analysis technique which
happens in the transformers. Besides that, the ageing condition transformer insulation system using oil
characteristics test is used to evaluate. Therefore, it was very costly by used the oil characteristics test
and also time consuming.
Previously several researchers were proposed various methods to determine the transformer
conditions. A.E.B Abu Elainien et al. [1] recommended technique transformer asset management. A.
Jahromi et al. [2] proposed health index transformer asset management. D. Hughes et al. [3] proposed
condition base risk management (CBRM) to corporate decision making. J. Young et al. [4] proposed the
equipment condition assessment index. Liu, L. X et al [5] proposed feedback artificial learning algorithm
for the state. M. Koch et. al [6] proposed transformer diagnostic moisture equilibrium.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

M. Martins [7] proposed transformer an alternative to mineral oil. Q. Liu et al [8] proposed the
characteristic and breakdown transformer under lightning impulse. S. Besner et. al [9] proposed on
unusual ethylene service transformer at low temperature. S. Miletic et al [10] proposed health index
benchmarking asset assessment and Z. Yang et. al [11] proposed transformer on rule mining based for
fault diagnosis.
The entire health transformer represents by index forms according performance incidence
replacement, explosions and also system failure. Besides that, it also to manage control maintenance.
Abu Elanien et al., [12]. Wang et.al proposed improve transformer performance due to demand, and
asset maintenance cost. Wang et al., 2004 [15]. The transformer design base on the industrial. Hughes,
2010 [3].
Naderian proposed the asset health index to identify investment needs, minimize operation and
maintenance cost. Previously, researchers proposed like reliability life, transformer life span and
consumption. Naderian et al., 2008 [14]. Proposed by calculate transformer health to analysis complex
asset and also not reflect repair nor inspection for the transformer. Jahromi et al., 2009 [13]. The health
index may indicate transformer life span also the risk. It also can define capital cost like replacement,
repairing and prevent increase maintenance.
Purpose the Health Index (HI) to monitor transformer and also to improve quality transformer
insulation system. The factor increasing age insulation may to system failure. It is because life span
insulation base on the strength and reliability of the electrical and mechanical.
The paper insulation may indicate the transformer life span and ageing. Normally, the information
its test on windings faults and problems of dielectric. It is also included measure temperature and
dissolved gas analysis. These tests may detect fault such as ageing, moisture and electrical & mechanical
faults.
This paper presents a developed condition situation monitoring (CSM) technique to predict
transformer health using methods calculate health index and performance conditions. It also provides
about knowing transformer life span condition.
Among all the electrical equipment from a power grid, a case study in the power transformer was
chosen because it is the most important equipment due to its operational role and due economic reasons.
Monitoring system status through this, knowing the transformer health index may help to plan and
operation more reliable and also effective.

2 Condition Situation Monitoring (CSM) Techniques

This paper proposed the new condition situation monitoring (CSM) diagnostic technique for prediction
of power transformers health index and monitoring a condition of a transformer. The new condition
situation monitoring (CSM) technique was developed for the purpose of detecting faults in condition
transformer. Besides that, it can be monitored online by several parameters without sent oil sample to
test.

The new condition situation monitoring diagnostic technique measures and tests the insulation
resistance from high voltage to ground (HV-E), low voltage to ground (LV-E), high voltage to low
voltage (HV-LV). It also includes winding resistance primary winding and secondary winding. The
method that is used to predict the trends of the monitored transformer to calculate the transformer’s
indicator of whether it is in good, fair or poor conditions. The data collections were tested and measured
using 8 units of transformer from different locations. Besides that, the health index result was provided

2
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

by CTMS for comparison. The new condition situation monitoring (CSM) diagnostic technique can
determines the health index for the transformer’s life span. It saves cost without the need to send the
sample oil for testing and maintenance.

3 Measurement Method for Data Collection Transformer

The initial contact resistance has very strong influence on life contact resistance. If the
connection contact resistance is known, it may to define remaining life contacts. This can make by the
resistance test of temperature oil transformer and electrical connection design in this research.

3.1 Winding Resistance Measurement

Winding resistance measurement is performed with a low resistance ohmmeter. For a three- phase
wye-connected transformer, the resistance is measured for each phase to neutral winding. If delta is
connected, the resistance is measured for each phase to phase winding. For delta connected transformers,
the measured resistances for each phase compose of a parallel combination of the winding under test
and the series combination of the remaining windings. It is therefore recommended to make three
measurements for each phase to phase winding in order to obtain the most accurate results. It is also
recommended to allow the transformer to be de-energized until the temperature is equalized before
making resistance measurements.
In order to reduce measurement errors due to changes in temperature, some precautions should
be taken before the measurement is made. For oil immersed transformer, the transformer should be
under the oil and without excitation for at least 3 hours. In addition, it is important to ensure that the
average oil temperature is approximately the same as the winding temperature. To avoid an inadmissible
winding temperature rise during the measurement, it is also recommended that the measuring current
should be limited to no more than 10 percent of rated current of the winding.
In order to diagnose the possible problems, the measured results are compared to the factory
values, value of other phases of the same transformer. Consistency in measurements and record keeping
are the keys to making the proper analysis using this test.

3
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

Start

Stage 1 Tests Figure 1. This shows the illustration

(Insulation Resistance) workflow practiced for the development of
CSM for transformers. The newly developed
online condition situation monitoring (CSM)
Health Good / Fair
technique gave fast and immediate result after
Index
data analysis and interpretation.
Poor

Stage 2 Tests
(Winding Resistance)
Test)

Good / Fair Maintain /

Health Revise
Index Frequency / Oil
Treatment
Poor

Repair / Upgrade
/Replacement of
Transformer

End

3.2 Insulation Resistance Measurement

The insulation resistance test or Megger test is used to indicate the insulation leakage current
resistance. The resistance is a moisture function and insulation impurity content and also the insulation
temperature. The resistance depends on the electric field strength across the insulation. Primarily, this
measurement provides information regarding the insulation condition and make sure that the leakage
current is small.
Transformer Insulation resistance is measured by resistance meter using DC voltage. In
measuring resistance, it has to make sure that the core and tank are grounded. Each transformer winding
is short circuited at terminals. Resistance measurements in between of each winding and also other
winding grounded. Windings never left floating during insulation resistance measurements. When
winding is installed with solid ground, the ground must be removed to measure winding insulation
resistance to another winding grounded. If the ground is not removed, the winding insulation resistance
could not be measured. It is a part of the circuit grounded section. Insulation resistance is determined
into mega ohms (MΩ).

4
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

Two winding transformer the following measurement configurations are used:

i. Measure from the high voltage winding to ground (HV-E)
ii. Measure from the low voltage winding to ground (LV-E)
iii. Measure form high voltage winding to low voltage winding (HV-LV)
This test is easily performed and many manufacturers required to emerging transformer, to
preclude start up failure caused by entry moisture into the transformer during shipment or storage. The
test can detect other ground circuits in the transformer that have been caused by shipping damage, the
test checks complete circuit bushings and coils.
The measurement for the duration is in minutes. Normally, the resistance readings are taken 15
seconds and 60 seconds. It is very important to record all the tests in report because these readings may
use to compare for future likes measurements, the mater used, measuring voltage and the measured
resistance. Reading or any comparisons measurements must in the same voltage because vary of the
insulation resistance with applied voltage.
Precautions should be taken when doing the insulation resistance test as following:
i. If it is found out the current in begins to increase, the test must be discontinued immediately
without stabilizing.
ii. Under no conditions should the test be made while the transformer is under the vacuum.
iii. All terminals should be grounded after the test has been completed. It allows trapped charges to
decay until negligible value in certain period.

4 Calculation of HI Transformer Method

The transformer health index is determined by condition indicator applying data in MS Excel
spreadsheet and also given the total health index.
Through CSM methodology, these data were provided by various diagnostic techniques.
Currently, the problem facing of the transformer is difficulties to evaluate for condition transformer and
may also have high risks. Therefore, a new technique for health index is formulated for calculation,
analysis and prediction to indicate condition of the transformer.
The formula health index was derived to evaluate, determine and calculate the condition
transformer according performance transformer, like example testing insulation resistance high voltage
to ground (HV-E), low voltage to ground (LV-E), high voltage to low voltage (HV-LV). It also includes
winding resistance primary winding and secondary winding. The weighting factor is according the
previously or any historical data condition transformer from CTMS.
5 Result and Discussion
There are three stages involved in the diagnostic techniques. The first stage is using high voltage
to ground (HV-E), low voltage to ground (LV-E), and high voltage to low voltage (HV-LV) to determine
insulation resistance. In the second stage, it tests on the winding resistance of 11kV transformer in
primary winding and secondary winding. The result of the health index test is based on 8 unit of
transformers in CTMS.

5
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

The result of the health index will be compared with the result from the CTMS health index from
industrial facilities to compare the transformer conditions measured in accuracy.
From the result, the transformer health index more than 0.85 is considered good condition. The
transformer health index more 0.6 until 0.85 health index is considered fair condition. For the
transformer health index below 0.6, it is considered poor condition. The transformer in poor condition,
it may have high risk and the risk will increase. Therefore, transformer in poor condition needs for repair
or replacement transformer.

5.1 Result Transformer Health Index between Calculate and Transformer Health Index Given by
CTMS
The result of the health index using the proposed method is compared with the health index
which given by the Company Transformer Manufacturing and Servicing in Malaysia are shown in
Figure 2 and Table 1. From Figure 2 and Table 1, it is found out that the error percentage of the health
index result is between range 7% to 10%.

5.2 Performance of the calculated Health Index and CTMS Data

It can be concluded that the calculated health index which is based on the condition indicator can
classify the condition of the transformers with an acceptable result. The result also indicates that the
calculated health index from the proposed model show an acceptable value compare to the data given
by CTMS.

The results of the calculated health index are based on the condition indicator for all of 10
transformers. From the result, the transformer which has health index value more than 0.85 is
considering good. The health index value between 0.6 until 0.85 is considers Fair condition where an
enhance maintenance and service is expected. For the transformer health index below 0.6, it is classified
as poor condition. In other word the transformer is in risk condition and need to be repair or replaced
immediately.

6
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

Figure 2. Comparison of transformer health index between calculated and CTMS

Table 1. Transformer health index between calculated and given by CTMS

Transformer Transformer Health Index

Number Rating kVA CMTS Calculated

1 1000/11/0.433 0.0897 0.12

2 1000/11/0.433 0.8417 0.87

3 1000/11/0.433 0.7887 0.68

4 1000/11/0.433 0.5130 0.65

5 1000/11/0.433 0.8187 0.86

6 1000/11/0.433 0.0870 0.08

7 1000/11/0.433 0.8397 0.85

8 1000/11/0.433 0.8287 0.86

9 1000/11/0.433 0.2177 0.23

10 1000/11/0.433 0.9140 0.93

7
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

6 Conclusion
This paper presents new approach to condition situation monitoring (CSM) diagnostic technique,
a health index to determine health condition of the transformer. This method is based on the transformer
condition to evaluate the health index. The condition situation monitoring (CSM) diagnostic technique
models may able to determine the health index for identify of the transformer. The results of the health
index showed by (CSM) diagnostic technique models. It provides results faster compared to method
using oil testing which need analysis for few days.

7 References
[1] A.E.B.Abu Elanien and M.M.A.Salama, (2010). Asset management for transformer, Electric.
Power System., April 2010, pp. 456–464.
[2] A.Jahromi, R.Piercy, S.Cress, J.Service, and W.Fan, (2009). Approach transformer asset
management using health index, IEEE., Mar. 2009, pp. 20–34.
[3] D.Hughes, (2010). Condition based risk management (CBRM) asset condition information
central corporate decision making, Electrical Distrib., Turin, Italy.
[4] J.Young, Ma Lin, W.Liyuan, T.Jian, and H.Lin, (2010). Determine index equipment condition
assessment, in Proc. Prognost. Health Conference., pp. 1–6.
[5] Liu, L.X., Han, B.X. & Ren, X.B., (2010). Learning Algorithm Feedback Artificial Sixth
International Natural Computation., pp. 357-361
[6] M. Koch, S. Tenbohlen, and T. Stirl, (2010). Diagnostic moisture equilibrium transformers,
IEEE Trans. Power., pp. 2574–2581.
[7] M.Martins, (2010). An alternative mineral oil for transformers, IEEE Electrical Insulation.
Magazine., pp. 7-13.
[8] Q.Liu, Z.D.Wang, (2011). Streamer characteristic synthetic and transformer liquid under
lightning impulse, IEEE Transaction Dielectric and Electrical Insulation., pp. 285-294.
[9] S.Besner, J.Jalbert and B.Noirhomme, (2012). Unusual Ethylene Transformer Low
Temperature, IEEE Transactions on Dielectric and Electrical Insulation.
[10] S. Miletic, (2009). Asset assessment index benchmarking, Int. Conference. Elect. Distrib.,
Prague, Czech Republic.
[11] Z.Yang, W.H.Tang, A.Shintemirov, and Q.H.Wu, (2009). Association Rule Dissolved Gas
Analysis Transformers Fault Diagnosis, IEEE Transactions on Systems, Man and Cybernetics.
[12] A.E.B.Abu Elanien and M.M.A.Salama (2011) Approach Calculate Thermal Lifetime
Transformer Insulation, Electrical Power Systems Research Journal.
[13] A.Jahromi, R.Piercy (2009) Approach Transformer Asset Management Health Index. Electrical
Insulation Magazine., IEEE, p.p 20-34.

8
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012007 doi:10.1088/1742-6596/1878/1/012007

[14] A.Naderian, S.Cress, and R.Peircy (2008) Approach Determine Transformer health index, IEEE
Int. Symp. Electrical Insulation, Canada., p.p 192–196.
[15] N.Wang, F.C.Lu and H.M.Li (2004) Analytical On line Dissipation Factor Base Morphological
Filter, IEEE Transaction Dielectric and Electrical. Insulation., p.p 840-846.

Acknowledgements
The authors would like to acknowledge the support from Faculty of Electrical Engineering Technology,
Universiti Malaysia Perlis for the funding of this work.