Professional Documents
Culture Documents
A R T I C L E I N F O A B S T R A C T
Keywords: Owing to their vital function and high cost, oil-immersed power transformers represent key links in electricity
Power transformers grids. While extensive effort has been invested by industry in developing various condition monitoring and fault
Asset management diagnosis apparatuses, it is not economically feasible to apply rigorous inspection and extensive testing to all
Fault diagnosis
power transformer fleets in the network. The fundamental purpose of asset management is to balance cost and
Health Index
Remaining life
reliability of the system. Maintaining the reliability of power transformers involves regular condition monitoring,
fault diagnosis and overall health assessment. In addition, estimating the remnant life of power transformers can
help asset managers plan capital investments. By reviewing the methods used for power transformer asset
management, assets integrity team can decide the most suitable maintenance strategy for each transformer in the
network. This paper is aimed to present state-of-the-art review for power transformer asset management, health
assessment and remnant life methodologies. While many review papers on transformer fault diagnosis have been
published in the literature, not much attention was given to present a state-of-the-art review on the three
modules of the transformer asset managements that include fault diagnosis, reliability assessment and life
management modules. These three modules are comprehensively presented in this paper that also presents future
challenges and improvement in this field of research.
* Corresponding author.
E-mail address: a.abusiada@curtin.edu.au (A. Abu-Siada).
https://doi.org/10.1016/j.epsr.2023.109194
Received 9 November 2022; Received in revised form 23 January 2023; Accepted 5 February 2023
Available online 11 February 2023
0378-7796/© 2023 Elsevier B.V. All rights reserved.
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
techniques published in the literatures and currently used by industry replacements for asset managers.
practice. While many review papers on transformer fault diagnosis have
been published in the literature, not much attention was given to present 2. Fault diagnosis
a comprehensive review paper on the three modules of the transformer
asset managements. To the best of authors’ knowledge, no such review An oil-immersed power transformer is a complicated piece of
paper is currently available in the literature. equipment that comprises many components including bushing, tap
The basic structure of the transformer condition-based assessment is changer, insulation, winding and magnetic core. Through condition
shown in Fig. 2. As shown in the figure, asset management consists of monitoring methods, transformer condition data can be analyzed ac
three modules: fault diagnosis, reliability assessment and life manage cording to different failure modes. Due to the wide variety of sizes and
ment. A fault diagnosis module is utilized to assess the health condition types of power transformers, there are no standardized failure modes for
of individual transformer components. Different fault diagnostic all power transformers. Failure modes can be generally classified based
methods (presented in Section 0) feed into failure modes, which are the on its location and cause [5–7]. This section presents commonly used
underlying conditions that can lead to failures. In Section 3, reliability condition monitoring methods and their diagnostic techniques.
assessment methods will be reviewed. Different from fault diagnosis,
reliability assessment is utilized to determine the overall health condi
tion of the transformer. The Reliability Index or Health Index (HI) is the 2.1. Condition monitoring methods
overall health score given by the reliability assessment. HI is designed to
identify the transformer most in need of the asset manager’s attention Dissolved Gas Analysis (DGA): DGA is one of the most popular fault
and is calculated based on condition monitoring parameters or fault detection techniques for power transformers [8]. Under harsh operation
diagnosis data. In Section 4, life management approaches utilized to conditions, both oil and paper insulation decompose and release certain
estimate the transformer aging are presented. The estimated remaining gasses that dissolved into the oil and decrease its dielectric strength. The
life of power transformer is conduced based on the history and current fault gasses, namely, hydrogen (H2), methane (CH4), ethane (C2H6),
condition of the transformer. It highlights the priority of the need for ethylene (C2H4), acetylene (C2H2), carbon monoxide (CO) and carbon
dioxide (CO2) can be measured using online or offline methods in
2
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
3
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Table 3
Conventional fault diagnostic methods.
Condition Monitoring Methods Refs Threshold/Ranges/Boundaries Signal Figures/Plot Graph
temperature. 10 min, has the advantage of eliminating the effect of such parameters
Dielectric Response Analysis - Frequency Domain Spectroscopy on the IR reading [28]. A sharp decline in PI requires urgent attention
(FDS): FDS technique is a frequency domain analysis for insulation and further investigation.
condition assessment. The method is based on assessing the dielectric Degree of Polymerization (DP): DP is one of the key properties of
parameters such as relative complex permittivity, complex capacitance cellulosic insulation material [29]. The mechanical strength of paper is
and dissipation factor, which can reflect the insulation polarization strongly related to the length of the cellulose molecules. The cellulose
characteristics. PDS is commonly used for checking the moisture content polymer molecules are connected through long chains of glucose rings
in different parts of solid insulation [17]. FDS measurement is also [30]. Under thermal stresses, the glucose chains break, shorten the
sensitive to temperature [17,26]. length of the cellulose molecules and reduce the paper’s mechanical
Insulation Resistance (IR): IR test is utilized to assess the insulation strength. The average length of these chains is measured in terms of the
conditions affected by contaminations, which can be measured using the DP which should be in the range of 1000 – 1200 for new material.
leaking current [27]. IR values are measured by a Megger or Megohm Dropping the DP value to 250 is considered the end of paper life [15,
meter at different time intervals. As the magnitude of IR value varies due 31–33]. The conventional method to measure the DP value is through
to environmental conditions such as moisture and temperature, Polari laboratory tests [34]. This requires paper samples to be taken from the
zation Index (PI), which uses the IR measurements taken in 1 min and transformer, which is impractical for the transformers in operation.
4
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Furan Analysis: Furan analysis is used to assess the paper insulation marked spot shows the fault could be D1 or D2. This issue also
condition. Cellulose degradation due to high temperature, oxygen, happens in the Duval Pentagon method. In Duval Pentagon, the
moisture and acid contents produce chemical furan compounds. Furan marked spot shown in Fig. 3(b) could indicate possible faults be T1,
compounds include 2-furaldehyde (2-FAL), 5-methyl-2-furaldehyde (5- T3 or D2. These faults are detailed in Table 1.
M2F), 5-hydroxymethyl-2-furaldehyde (5-H2F), 2-acetyl furan (2-ACF) • Data cannot be interpreted unless measured values pass minimum
and 2-furfurol (2-FOL) [35]. The most stable and abundant furan com level [47,48]. As an example, according to IEEE 57.104, to apply the
pound is the 2-FAL [36,37]. The concentration of the furan compounds DGA Doernenburg Ratio method, at least one of the critical gasses
in ppm can be measured using high-performance liquid chromatography (H2, CH4, C2H2 and C2H6) concentration used in the ratio must
or gas chromatograph-mass spectrometry. There is a strong correlation exceed a certain level.
between DP value and 2-FAL concentration. More detailed calculations • Different interpretation methods may lead to different conclusions,
to predict DP value using 2-FAL measurements are explained in Section which can be observed in DGA interpretation [47,48]. Also,
Life Management4. graphical-based interpretation such as frequency response analysis,
is not always consistent as it relies on the personnel expertise.
2.2. Fault analyzing methods Conventional methods depend heavily on visual inspection, expert
knowledge and judgement. Therefore, other diagnostic methods based
Data analyzing for oil-immersed power transformers has always been on mathematical statistics, equivalent electric models, and artificial
challenging. False diagnosis could cause unnecessary additional main intelligence, have been proposed as presented in the following sections.
tenance, and even wrong component replacement. Depending on the
characteristics of measuring data, feature extraction in some diagnostic 2.2.2. Mathematical statistics method
methods is based on numerical threshold limits, and others by recog In this method, statistical indicators are employed to extract features
nizing the variations on their patterns, as shown in Table 3. with reference data. The chosen indicator(s) should only be sensitive to
Data analyzing methods can be categorized into conventional the differences in fault conditions [51]. The statistical assessment is used
manual diagnosis and advanced computational techniques such as using for data comparison, such as FRA data interpretation [43]. In FRA,
artificial intelligence. The difference is that manual diagnosis requires winding deformation can be identified by statistical indicators such as
experts with knowledge and experience to manually select data features the Correlation Coefficient (CC) and the Standard Deviation (SD) [17].
and determine the possibility of abnormality. Advanced computational CC is used to calculate the variation between data variables while SD is
techniques utilize computer science to perform automatic data feature used to assess the deviation of one data set from its mean value. The
selection and fault diagnosis. formulas of CC and SD are given by (1) and (2), respectively. Other
commonly used indicators are Mean Square Error (MSE), Absolute Sum
2.2.1. Challenges to conventional interpretation methods of Logarithmic Error (ASLE) and Maximum Minimum (MM), which have
Conventional interpretation methods are still widely adopted by been presented in [40,52–54].
current industry practice to identify and quantify various faults within ∑N
power transformers based on diagnostic methods listed in Table 3. Such i=1 Xi Yi
CC(X,Y) = √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ (1)
methods are facing several challenges that include: ∑N 2 ∑N 2̅
i=1 X i i=1 Y i
• The exact value of ratios and ranges could lead to misdiagnosed re √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
∑N 2
sults [47,48]. For instance, according to Rogers Ratio method used SD(X,Y) = i=1 [Yi − Xi ]
(2)
for DGA interpretation, if the ratios for both C2H2/C2H4 and N− 1
C2H4/C2H6 are less than 0.1, then the ratio value of 0.1 for CH4/H2 Xi is the ith elements of FRA fingerprint.
becomes the critical boundary between normal and PD fault. Yi is the ith elements of measured FRA trace.
• The defined ranges vary between different research papers. For N is the number of samples.
example, some authors suggested DP value less than 200 as end-of- No general rules define the fault level using statistics indicators [43].
life instead of 250 that is suggested in other papers [30,32,49,50]. While mathematical statistics method has improved the accuracy of
• When measurements fall onto graphical-based methods’ boundaries, graphical-based interpretation, they fail in quantifying actual fault level.
it is difficult to draw a conclusion [47,48]. An example of using the
DGA Duval Triangle method is shown in Fig. 3(a). In this case, the
5
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
6
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Fig. 6. Principal diagram of SVM with two classes. [72]. The output of this model [Y i ∈ {1, 2, 3, 4}], if there are four
possible classified states – normal, thermal fault, low energy discharge
node and leaf node. In the example of DGA diagnosis, the root node can and high energy discharge. The inputs can also be the sets of gas ratios
be used to separate data into thermal-fault and electrical-fault, which is and the outputs can be six fault types [71]. Logistic regression can be
the most relevant attribute/feature of the data sets. Decision nodes trained on small data sets; therefore, it is simpler and more efficient. The
represent a possible choice available at that point in the tree structure. test results show that logistic regression has an accuracy rate of over
Finally, each leaf shows an outcome (T1, T2, T3, D1, D2 and PD are the 90% and has much better performance than conventional methods such
symbols of DGA fault types, as shown in Table 1). as Rogers Ratios (60%), IEC Ratios (60%) and Duval methods (80%)
Random Forest: Random Forest is an ensemble learning algorithm [72].
that can be considered an extension of Decision Tree [111]. Its model Support Vector Machine (SVM): SVM is a binary classifier that
uses several individual decision trees, and each tree learns from a constructs an optimal separating decision surface, called a hyperplane,
random sample of the training data sets [68,112]. After all Decision to maximize the margins between the hyperplane and the data set as
Trees give their predictions, the prediction of Random Forest is calcu shown in Fig. 6 [69]. The support vectors are the data used to find the
lated by aggregating over the ensemble. Compared to a Decision Tree, optimal location of the hyperplane.
Random Forest is able to process multiple classification trees simulta SVM can map non-linear data into higher dimensional feature space
neously; therefore, it is a fast classifier. It also improves the predictive and turn it into a linear regression problem [115]. This non-linear
accuracy and controls over-fitting issues by taking the average of various relationship can be expressed as a decision function given by (3)
Decision Trees [8,113]. [116]. The training data set is {Xi , Yi }m i=1 , where Yi is the label of the
K-nearest Neighbour (KNN): KNN is a non-parametric classification input vector Xi . Xi ∈ RN , RN is the N dimension of input vectors. K(Xi , X)
method. It stores and allocates training data to certain classes, which “K” is the Kernel function, which maps low-dimensional space into
means the number of classes. The new data will then be classified into high-dimension. The popular kernel functions are Linear kernel, Poly
one of these classes where its nearest neighbor is located [114]. The idea nomial kernel, Gaussian kernel, Sigmoid and Radial Basic Function
of KNN is very similar to decision trees but instead of using a tree (RBF) [69]. The hyperplane can be found when f(x) = 0. The two groups
structure, it calculates the distance on the graph. Running KNN is nor of data that are separated by boundaries are negative f(x) = − 1 and
mally faster than a decision tree [113]. There are many different positive f(x) = + 1. SVM is a popular technique used in transformer
methods to calculate the distance in KNN, such as Euclidean, Chebyshev, fault diagnosis in which the function f(x) is used to classify the condi
Minkowski, city block, Mahalanobis, Spearman correlation, Hamming, tions such as discharge fault (f(x) = + 1) and thermal fault (f(x) = −
Jaccard and cosine [114]. 1).
Logistic Regression: Logistic Regression is a machine-learning tech [ ]
nique for analyzing discrete data [71]. Each set of input Xi ∈ RN , RN is an ∑
m
f (x) = sign αi Yi K(Xi , X) + b (3)
N dimensional feature vector. As an example, if the DGA set comprises 5 i=1
gasses, then [ X1i , X2i , X3i , X4i , X5i ] = [H2 , CH4 , C2 H6 , C2 H4 , C2 H2 ]
K(Xi , X) is Kernel function.
7
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
8
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
HI is a powerful tool which has been used to assess the technical and
operating condition of the transformer [132]. HI integrates data from
inspection, operating conditions and condition monitoring into a
quantitative index to provide an overall health information to asset
managers [131]. HI applied in reliability assessment is used to assess the
long-term degradation level, prioritize critical equipment, and propose a
maintenance plan through condition parameters data.
9
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Table 5
AI methods used for HI algorithm.
AI methods Input Variables References Remarks
Logistic Regression DGA, DF, Moisture, Acidity, BDV, 2-FAL • Provides explicit probability of HI giving feature
[146]
vector.
General Regression DGA, DF, 2-FAL, Dielectric strength, Moisture, Acidity [132] • Multi-dimensional measurements combine through an
Neural Network optimal scoring-weighting system.
(GRNN) DGA, 2-FAL, Acidity, IFT, Moisture, BVD, Insulation Resistance, PF, Turn [136] • Allow using small training sets.
Ratio, winding resistance, short circuit impedance, excitation current, • Simpler algorithm to implement than SVM.
Bushing – age, DDF, tap changer type and operations per month, loading and • Condition of transformer subsystems estimated using
maintenance history GRNN.
• Condition scores for each subsystem are delivered by
using GRNN.
• Overall HI is non-linear combination of condition
scores for each subsystem by using additive and mul
tiplicative means.
Fuzzy Logic DGA, Furan (2-FAL), moisture, acidity, BDV, DF [147] • Results highly closer to experts’ diagnosis.
DGA, Furan, DF, Moisture, Acidity [148] • Hard to determine the membership functions.
Fuzzy C-Means Inaccurate and uncertain data [133] • Help with database construction, assigning oil
characteristic record with weight to a particular HI
level.
Decision Tree DGA, BDV, IFT, Acidity, Moisture, Color, DF, 2-FAL [114] • In general, it has lower accuracy than SVM and KNN.
KNN DGA, BDV, IFT, Acidity, Moisture, Color, DF, 2-FAL [114,149] • Easy to implement, but poor performance if the
predictor variables increase or the number of relevant
attributes is low.
• Non-parametric method, its performance doesn’t vary
significantly depending on distribution pattern of data.
• Robust to noise.
• Can be time-consuming during the classification phase.
SVM DGA, BDV, IFT, Moisture, DF, 2-FAL [150,151] • Non-linear classifier for large dimension of
independent variable.
• Possible deterioration if dimension is too large relative
to the size of training data.
• SVM works better with optimization techniques.
ANN DGA, IFT, Moisture, Acidity, BDV, DDF, Colour, Furan [135] • Simplify the process of data training.
• Fast diagnosing speed.
• Performance depends on the completeness of the
training sample.
• The process is not visible.
• Oscillation easily occurs in the identification.
ANFIS DGA, BDV, DF, Acidity, IFT, Moisture in oil and paper, 2-FAL [152] • ANFIS provides more accurate results than ANN.
10
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
dependent on the data availability. estimate the degree of deterioration of power transformers. When the
deterioration reaches a level where the transformer is too risky to be in
3.2.2. Probability of failure operation, it is defined as technical EoL.
The probability of failure method is worth mentioning here due to its The degradation of components such as bushings, tap-changers,
simplicity. If the probability of each failure mode is known from the fault tank, cooling systems, and active parts (magnetic core, windings, and
diagnostic module and historical data, this method can provide insulation) can cause gradual reduction in the dielectric, mechanical and
approximate failure probability using (8) [5,145]. thermal strength. Even under normal operating conditions, losing
strength may eventually end the transformer’s life. Without economic
HI =1 − ((1 − est.PoFFM1 ) × (1 − est.PoFFM2 ) × (1 − est.PoFFM3 )
(8) considerations, replacing or refurbishing the above-mentioned compo
× …(1 − est.PoFFMn )) nents, apart from the active parts, is possible. Therefore, in previous
n is the total number of failure modes. studies, calculating the remnant life of oil-immersed power transformers
est.PoFFMi (i = 1 ∼ n) is the estimated probability of each failure is mainly focused on winding solid insulation. The solid insulations
mode. including paper, pressboard and wood are made of cellulosic materials
Assuming that DGA and Furan tests have been used for failure modes [158]. The chemical decomposition of cellulose is caused by oxidation,
1 and 2, respectively, then est.PoFFM1 is the estimated probability of DGA hydrolysis and pyrolysis mechanism [31,159]. Hence, the degradation
failure mode and est.PoFFM2 is the estimated probability of Furan failure of cellulose is strongly influenced by the level of oxygen, water content
mode. If est.PoFFM1 is approximately 10% and est.PoFFM2 is approxi and thermal conditions. The oxygen and water contents are considered
mately 20%, then according to (8), HI is 0.28 or 28%. If DGA result to be controllable by using modern oil preservation systems with
shows an increased failure rate from 10% to 60%, then HI value also improved sealing systems [160]. The thermal stress is considered to
increases to 68%. Although this method provides only approximate dominate the degradation process [161]. The hotspot temperature in
probability of HI, it has the advantage of not masking any failure mode side the transformer can cause the most significant aging. Therefore, the
over the scoring-weighting method. winding hot-spot temperature (HST) is the main factor determining the
insulation lifespan.
3.2.3. Artificial intelligence methods
AI-based HI algorithm does not require a pre-defined formula. It uses
analytic data techniques to find a new correlation between transformer 4.1. Thermal-based method
condition indicators and HI classes: Very poor, Poor, Fair, Good, and
Very Good. Many scholars have proposed several AI-based methods to The thermal methods introduced in IEEE Std C57.91 and IEC
enhance the HI algorithm. Table 5 lists some of the proposed AI methods 60,076.7 can be used to estimate the percentage loss of life (% Loss of
used to build HI algorithms. Life) of the insulation through calculating the aging acceleration factor
The advantage of using AI algorithms for HI classification can avoid FAA [162,163]. The calculation of the aging acceleration factor is based
the direct calculation of HI scores based on empirical weighting factors. on the winding HST and the Arrhenius reaction rate theory [160].
Therefore, this can increase the accuracy by not applying crisply defined Eq. (10), which is presented in IEEE Std C57.91 is practically applied
ranges to determine HI classes. for thermally upgraded paper. It commonly uses a reference HST of
110 ◦ C for 65 ◦ C average winding rises [162,164]. It means the value of
3.2.4. Markov model the accelerated aging factor is greater than 1 for a winding reference
The above HI methodologies focus on calculating HI score/classifi HST over 110 ◦ C; and less than 1 for a temperature below 110 ◦ C.
cation based on the transformers measured conditions. In some research Eq. (11) that is presented in IEC 60,076.7 (AS/NZS 60,076.7) is
papers [140,153,154], the Markov model has been proposed to predict practically applied for non-thermally upgraded paper (Kraft paper). The
the future HI state based on the current HI state. The future state of HI reference HST is 98 ◦ C when the aging acceleration factor FAA is equal to
can be expressed by (9). The transition probability P denotes the prob 1 according to Table 2 in the standard. The IEC standard also defines that
ability of transitions to the next state. the relative aging rate is doubled for every 6 ◦ C increase in HST.
[ ]
HI(t + 1) = HI(t) × P(t) (9) 15000 − 15000
(10)
110+273 θHST (t) + 273
For thermally upgraded paper FAA (t) = e
HI(t) is the current state HI at time t.
P(t) is the transition probability matrix at time t.
(11)
θHST (t) − 98
For non − thermally upgraded paperFAA (t) = 2 6
The equivalent aging factor of the transformer for the total time is
The designed lifespan for oil-immersed power transformer is gener calculated by (12) according to IEEE Std C57.91. Usually, it uses a total
∑
ally 25 - 40 years [155]. Due to the harsh operating conditions of period of 24-hour cycle, then Nn=1 Δtn is 24.
in-service power transformers, the aging degradation rate increases, ∑N
causing earlier retirement. Life management, as one of asset manage FEQA = n=1 FAA,n Δtn
∑N (12)
ment’s significant tasks, is designed to recognize the level of degradation n=1 Δtn
and estimate the remaining lifespan of the transformer under current
FEQA is the equivalent aging factor for the entire period.
conditions. The remaining lifespan calculation predicts time intervals
n is the index of the time interval Δt.
between fleets’ service life, allowing businesses to replace them gradu
N is the total number of time intervals.
ally [156].
Δtn is the time interval.
From the business point of view, the definition of End of Life (EoL)
FAA,n is aging acceleration factor for the temperature that exists
can be divided into three categories: economical EoL, strategic EoL and
during the time interval Δtn .
technical EoL [157]. An example of economical EoL is when mainte
% Loss of Life is the lifetime of cellulose lost over time, which is
nance cost is too high, i.e., economically, it is no longer worth main
calculated using the equivalent aging factor given by (13). The% Loss of
taining the transformer. Examples of strategic EoL could be power
Life is to be subtracted from the normal insulation life to estimate the
network upgrades or the obsolescence of critical components. This sec
remaining life of cellulosic paper. According to both IEEE (Table I.2 of
tion will focus on technical assessment, and the methods used to
IEEE C57.91) and IEC (IEC 60,076.7, Table 3) standards, the minimum
11
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Normal Insulation Life is 180,000 h (20.5 years) for a thermally Filter (UKF) have been developed. The drawback of this recursive al
upgraded insulation system at reference temperature of 110 ◦ C. gorithm is the problem of handling long-term predictions by accumu
lating prediction errors.
FEQA × t × 100 FEQA × 24 × 100
% Loss of Life = = (13) Although computational methods can improve the accuracy of the
Normal Insulation Life 180 000
thermal model, the function of temperature and time from IEEE or IEC is
FEQA is the equivalent aging factor for the total time-period. based on a well-dried and oxygen-free insulation system [31]. Alterna
The remaining useful life (RUL) at time t can be determined by using tively, the transformer’s water and oxygen content remain constant, and
the previous state of the insulation paper and its current condition along in reality, most in-service transformers are not operating under such
with the process noise as given by (14) [49,165,166]. perfect condition [165]. The following section introduces another
thermal method considering moisture and oxygen contents in oil.
RUL(t) = RUL(t − 1) − FAA (t) + w(t) (14)
w(t) is the process noise at time t. 4.2. DP-value method
The thermal models from the standards are simple techniques to
measure the aging of insulation. However, using sensors to measure the 4.2.1. Kinetic model to predict DP
winding HST directly is not cost-effective. It is normally estimated from The thermal aging process of paper can also be assessed by
other measurements and factors. In IEEE Std C57.91, the HST θHST can measuring the Degree of Polymerization (DP). As mentioned, it is
be calculated using top-oil temperature (TOT) θtop measurement. The difficult to measure DP value directly. Therefore, the Kinetic model
TOT can also be estimated through the ambient temperature as per Eq. method assesses the degradation of cellulose using estimated DP value
(15). with consideration of the effects of moisture and oxygen contents [165].
θHST = θtop + ΔθHST = θamb + Δθtop + ΔθHST (15) The kinetics of cellulose degradation is calculated using (16) based on
the pseudo-zero-order kinetic equation and Arrhenius reaction rate
θtop is the top-oil temperature ( ◦ C). theory [36,170]. The value “A” in the equation is decided by environ
θHST is the winding hotspot temperature ( ◦ C). mental factors like moisture content of the paper and dissolved oxygen
θamb is the average ambient temperature during the load cycle to be in oil. While the value of A can be experimentally measured, it is not
studied ( ◦ C). constant through the transformer life [50,171]. Δt is the aging period.
Δθtop is the top-oil rise over ambient temperature ( ◦ C). The value of “Ea” depends on the type of cellulose and chemical reaction
ΔθHST is the winding hotspot rise over top-oil temperature ( ◦ C). [50]. From the International System of Units (SI), the gas constant R is
In practice, the calculation of HST can be much more complex as the 8.314 J.mol− 1K− 1. The temperature of the paper typically refers to the
thermal conditions depend on the load, environmental and transformer HST of winding as the hottest temperature causes the most degradation.
conditions [49,167]. All these uncertainties will result in errors in the
1 1
TOT estimation, HST calculation and RUL prediction. It is worth (16)
− Ea
− = A(t− 1) × Δt × eR×T(t− 1)
mentioning that even slight error like 2 ◦ C in HST calculation can make DP(t) DP(t − 1)
about 20% difference in acceleration aging factor. Dynamic thermal DP(t − 1) and DP(t) are the DP values at the start and the end of the
models have been used to increase the accuracy of monitoring contin period.
uous temperature changes. A is the environmental factor, which can depend on the moisture and
Kalman Filter: some research suggests a state estimation tool called oxygen content in oil at (t-1) where (t-1) is the previous state and t can be
Kalman filter for thermal modeling [167–169]. Kalman filter is an hours, days or years.
optimal recursive data-processing algorithm, which can optimally esti Δt is the aging period.
mate the temperature. The principle of this mathematical model is Ea is the activation energy of the aging reaction (J.mol− 1).
shown in Fig. 12. The thermal dynamics present the actual system and R is the gas constant (J.mol− 1.K − 1).
ambient temperature, and TOT can be measured in the real system. The T is the temperature of the paper (K).
thermal model aims to have the estimated HST value as close as possible Rearranging (16), the expected life of the paper can be calculated as
to the actual HST value by comparing the estimated and measured TOT below Eq. (17) [161].
values. The controller gain K is to control the decay rate of the error
1 1
function, which means offering a faster elimination of the error [169]. Expected Life Δt =
DP(t)
− DP(t− 1) Ea
× eR×T(t− 1) (17)
The standard Kalman filter is only designed to model a linear rela A(t− 1)
tionship with Gaussian noise, which does not change with time [169].
Compared to the thermal models from IEEE/IEC standards, this
However, the correlation of the TOT with its relevant parameters, such
model has the advantage of considering moisture and oxygen involve
as temperature and load is non-linear [49]. To deal with non-linear
ment. However, both the values of A and HST are complicated to be
conditions, the Extended Kalman Filter (EKF) and Unscented Kalman
determined.
12
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Table 6
various DP-Furan correlational models.
Chendong model: For non-thermally upgraded paper
( )
log10 2 ∼ FALppm − 1.51
DP = (19)
− 0.0035
Pahlavanpour model: consider the different degradation rate for winding paper at different locations
800
DP = ( ) (22)
0.186 × 2 ∼ FALppm + 1
13
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
Table 7 algorithms into a single model, which can bring benefits from these al
Examples of hybrid model used in condition assessment. gorithms and avoid their downsides. Some examples of the hybrid model
References Hybrid models Remarks are listed in Table 7. Although researchers have claimed, their new
models improved the accuracy, such models do not seem to be a stan
[155] Bayesian ordered regression & • Classify HI into 5 classes (very
ANN bad, bad, moderate, good, very dard model that can work for all different datasets.
good) without calculating HI
score.
6. Conclusion
• Easily adapted.
• Improved accuracy compared to
some single AI methods. Cost-effective and reliable transformer asset management techniques
• Full quantized uncertainty of its are essential to design strategic maintenance and replacement plans.
parameters and predictions. Continuously improved technologies in condition monitoring and
• More efficient use of
computational methods increase the opportunity to use condition-based
unsymmetrical datasets.
[187] Unsupervised Clustering & • Unsupervised ML Clustering maintenance. This paper reviewed various methods used in condition
supervised classification used for feature extraction. assessment major modules: fault diagnostic module, reliability assess
• Reduce the size of data and ment module and life management module. In the fault diagnostic
processing time.
module, fault diagnostic methods according to individual transformer
[80] Fuzzy Logic & decision tree • Avoid crisp decision tree rules
[188,189] Principal Component Analysis • PCA is used to reduce the
components have been introduced. However, the transformer condition
(PCA) & Particle Swarm dimension of data sets. could often be between normal and faulty conditions, which means that
Optimization (PSO) & SVM • PSO is used to obtain the optimal the measurements are not in the normal range but still acceptable in
parameters for SVM algorithm. practice. Therefore, reliability assessment using HI is utilized to evaluate
[190] Relevance vector machine • RVM performs binary
the overall health condition of the transformer. Calculating HI is
(RVM) & ANFIS separation.
• ANFIS performs further fault commonly used in scoring-weighting and enhanced through Artificial
diagnosis. Intelligence methods. In the life management module, the degradation
• Higher accuracy than single of the insulation system has been a major indicator of End-of-Life for
ANFIS, SVM and ANN.
power transformers. The remaining service life of a transformer can be
estimated through the degradation factors and other indicators.
5. Future improvement areas
Current industry practice in the field of asset condition monitoring Declaration of Competing Interest
and fault diagnosis still relies on personnel expertise, which may result
in inconsistent interpretation and recommended asset management ac The authors declare no conflict of interest regarding this paper.
tions. AI methods are the future tools to automate and standardize asset
management methods of oil-immersed power transformers. However, Data availability
there are two areas that call for further development. One is to develop a
suitable failure mode analysis method for the power transformers, and No data was used for the research described in the article.
the second is to develop a proper hybrid AI-based mode that can provide
proven accuracy regardless the size and operating conditions of the
transformer. These two areas are briefly elaborated on below. Acknowledgments
5.1. Failure modes definition The authors sincerely acknowledge the support from BHP for
providing a scholarship to the first author to conduct her research on
As mentioned above, no standardized failure modes are used for high voltage assets condition monitoring methodologies. The authors
power transformer condition assessment. The failure modes can be also acknowledge the support from Curtin University, Australia, for
constructed based on the source of fault, as shown in Fig. 13 [180,181]. facilitating research resources.
However, implementation for such failure mode structure could be
confusing. For example, thermal and electrical faults are closely related. References
Hence, it is hard to determine whether the fault is thermal or electrical
originated. The failure mode structure can also be built based on loca [1] M. Dong, et al., A novel maintenance decision making model of power
transformers based on reliability and economy assessment, IEEE Access 7 (2019)
tion – winding, oil insulation, paper insulation, core, bushing, tap 28778–28790, https://doi.org/10.1109/ACCESS.2019.2897606.
changer, cooling system and tank [5–7,120,182,183]. It seems clear to [2] H. Ma, T.K. Saha, C. Ekanayake, D. Martin, Smart transformer for smart
choose the condition monitoring method for individual components and grid—intelligent framework and techniques for power transformer asset
management, IEEE Trans. Smart Grid 6 (2) (2015) 1026–1034, https://doi.org/
locate the fault. However, this calls extensive number of sensors to be 10.1109/TSG.2014.2384501.
installed. In most cases, this is not a cost-benefit for businesses. Some [3] S. Koziel, P. Hilber, P. Westerlund, E. Shayesteh, Investments in data quality:
research efforts have been conducted to identify various faults in the evaluating impacts of faulty data on asset management in power systems, Appl.
Energy 281 (2021), 116057, https://doi.org/10.1016/j.apenergy.2020.116057.
transformer oil and paper insulation using fuzzy logic, ANFIS, and gene [4] A. Jahromi, R. Piercy, S. Cress, J. Service, W. Fan, An approach to power
expression programming [184–186]. However, fesibility of these models transformer asset management using health index, IEEE Electric. Insulation
has not been assessed in field applications yet. Magazine 25 (2) (2009) 20–34, https://doi.org/10.1109/MEI.2009.4802595.
[5] Condition assessment of power transformers " Power transformers and reactors,
CIGRE Tech. Brochure A2-761 (2019).
5.2. Hybrid models [6] Asset health indices for equipment in existing substations, Substations and Electr.
Installations, CIGRE Tech. Brochure (2021) B3–858.
[7] R. Murugan, R. Ramasamy, Understanding the power transformer component
In the above sections, only individual AI algorithms have been failures for health index-based maintenance planning in electric utilities, Eng.
summarized. Due to significant measurements and history, using AI and Fail Anal. 96 (2019) 274–288, https://doi.org/10.1016/j.
minimizing human intervention seem to be the future for big data engfailanal.2018.10.011.
[8] G. Odongo, R. Musabe, D. Hanyurwimfura, A multinomial dga classifier for
analysis. Researchers are continuously exploring the optimal AI incipient fault detection in oil-impregnated power transformers, Algorithms 14
methods, including hybrid models. Hybrid models integrate different AI (4) (2021) 128, https://doi.org/10.3390/a14040128.
14
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
[9] IEEE guide for the interpretation of gases generated in mineral oil-immersed [38] M. Wu, H. Cao, J. Cao, H.L. Nguyen, J.B. Gomes, S.P. Krishnaswamy, An overview
transformers, IEEE Std C57.104-2019 (Revision of IEEE Std C57.104-2008) of state-of-the-art partial discharge analysis techniques for condition monitoring,
(2019) 1–98, https://doi.org/10.1109/IEEESTD.2019.8890040. IEEE Electric. Insulation Magazine 31 (6) (2015) 22–35, https://doi.org/
[10] M. Duval, J. Buchacz, Gas formation from arcing faults in transformers—part II, 10.1109/MEI.2015.7303259.
IEEE Electric. Insulation Magazine 38 (6) (2022) 12–15, https://doi.org/ [39] J. Seo, H. Ma, T. Saha, A novel signal extraction technique for online partial
10.1109/MEI.2022.9916225. discharge (PD) measurement of transformers, Int. Trans. Electric. Energy Syst. 26
[11] L. Cheim, M. Duval, S. Haider, Combined duval pentagons: a simplified approach, (5) (2016) 1032–1048.
Energies (Basel) 13 (11) (2020) 2859, https://doi.org/10.3390/en13112859. [40] J. Ni, Z. Zhao, S. Tan, Y. Chen, C. Yao, C. Tang, The actual measurement and
[12] L. Jin, D. Kim, A. Abu-Siada, S. Kumar, Oil-immersed power transformer analysis of transformer winding deformation fault degrees by FRA using
condition monitoring methodologies: a review, Energies (Basel) 15 (9) (2022) mathematical indicators, Electr. Power Syst. Res. 184 (2020), 106324, https://
3379, https://doi.org/10.3390/en15093379. doi.org/10.1016/j.epsr.2020.106324.
[13] H. Chai, B.T. Phung, S. Mitchell, Application of UHF sensors in power system [41] X. Liu, Y. He, L. Wang, Adaptive transfer learning based on a two-stream densely
equipment for partial discharge detection: a review, Sensors (Basel) 19 (5) (2019) connected residual shrinkage network for transformer fault diagnosis over
1029, https://doi.org/10.3390/s19051029. vibration signals, Electronics (Basel) 10 (17) (2021) 2130, https://doi.org/
[14] D. Kim, A. Abu-Siada, Partial discharge analysis on-site in various high voltage 10.3390/electronics10172130.
apparatus. 2018 Condition Monitoring and Diagnosis (CMD), 2018, pp. 1–6, [42] A. Secic, M. Krpan, I. Kuzle, Vibro-acoustic methods in the condition assessment
https://doi.org/10.1109/CMD.2018.8535987, 23-26 Sept. of power transformers: a survey, IEEE Access 7 (2019) 83915–83931, https://doi.
[15] K.R.M. Nair, Power and Distribution Transformers: Practical Design Guide, Taylor org/10.1109/ACCESS.2019.2923809.
& Francis Group, Milton, 2021. Milton. [43] M. Žarković, Z. Stojković, Analysis of artificial intelligence expert systems for
[16] T.-.D. Do, V.-.N. Tuyet-Doan, Y.-.S. Cho, J.-.H. Sun, Y.-.H. Kim, Convolutional- power transformer condition monitoring and diagnostics, Electr. Power Syst. Res.
neural-network-based partial discharge diagnosis for power transformer using 149 (2017) 125–136, https://doi.org/10.1016/j.epsr.2017.04.025.
UHF sensor, IEEE access 8 (2020) 207377–207388, https://doi.org/10.1109/ [44] IEEE guide for acceptance and maintenance of insulating mineral oil in electrical
ACCESS.2020.3038386. equipment, IEEE Std C57.106-2015 (Revision of IEEE Std C57.106-2006) (2016)
[17] A.-.S. Ahmed, Power transformer condition monitoring and diagnosis. (Energy 1–38, https://doi.org/10.1109/IEEESTD.2016.7442048.
Engineering), The Institution of Engineering and Technology, Stevenage, 2018. [45] S. Gubanski, et al., Dielectric response methods for diagnostics of power
[18] Y. Shi, S. Ji, F. Zhang, Y. Dang, L. Zhu, Application of operating deflection shapes transformers, Electra 202 (2002) 23–34.
to the vibration-based mechanical condition monitoring of power transformer [46] M.M. Islam, G. Lee, S.N. Hettiwatte, A review of condition monitoring techniques
windings, IEEE Trans. Power Delivery 36 (4) (2021) 2164–2173, https://doi.org/ and diagnostic tests for lifetime estimation of power transformers, Electric. Eng.
10.1109/TPWRD.2020.3021909. 100 (2) (2017) 581–605, https://doi.org/10.1007/s00202-017-0532-4.
[19] J. Seo, H. Ma, T.K. Saha, A joint vibration and arcing measurement system for [47] J. Faiz, M. Soleimani, Dissolved gas analysis evaluation in electric power
online condition monitoring of onload tap changer of the power transformer, transformers using conventional methods a review, IEEE Trans. Dielectr. Electric.
IEEE Trans. Power Delivery 32 (2) (2017) 1031–1038, https://doi.org/10.1109/ Insulation 24 (2) (2017) 1239–1248, https://doi.org/10.1109/
TPWRD.2016.2531186. TDEI.2017.005959.
[20] Y. Hu, J. Zheng, H. Huang, Experimental research on power transformer vibration [48] C. Lefeng, Y. Tao, Dissolved gas analysis principle-based intelligent approaches to
distribution under different winding defect conditions, Electronics (Basel) 8 (8) fault diagnosis and decision making for large oil-immersed power transformers: a
(2019) 842 [Online]. Available: https://www.mdpi.com/2079-9292/8/8/842. survey, Energies (Basel) 11 (4) (2018) 913.
[21] T. Mariprasath, V. Kirubakaran, A real time study on condition monitoring of [49] J.I. Aizpurua, S.D.J. McArthur, B.G. Stewart, B. Lambert, J.G. Cross, V.
distribution transformer using thermal imager, Infrared Phys. Technol. 90 (2018) M. Catterson, Adaptive power transformer lifetime predictions through machine
78–86, https://doi.org/10.1016/j.infrared.2018.02.009. learning and uncertainty modeling in nuclear power plants, IEEE Trans. Ind.
[22] X. Zhang, E. Gockenbach, Asset-management of transformers based on condition Electron. 66 (6) (2019) 4726–4737, https://doi.org/10.1109/TIE.2018.2860532.
monitoring and standard diagnosis [Feature article], IEEE Electric. Insulation [50] N. Lelekakis, D. Martin, J. Wijaya, Ageing rate of paper insulation used in power
Magazine 24 (4) (2008) 26–40, https://doi.org/10.1109/MEI.2008.4581371. transformers part 1: oil/paper system with low oxygen concentration, Dielectr.
[23] S. Degeratu, et al., Condition monitoring of transformer oil using thermal analysis Electric. Insulation, IEEE Trans. on 19 (2012) 1999–2008, https://doi.org/
and other techniques, J. Therm. Anal. Calorim. 119 (3) (2015) 1679–1692, 10.1109/TDEI.2012.6396958, 12/01.
https://doi.org/10.1007/s10973-014-4276-3. [51] D. Bormann and P. Picher, Mechanical condition assessment of transformer windings
[24] M. Wang, A.J. Vandermaar, K.D. Srivastava, Review of condition assessment of using frequency response analysis (FRA). 2008.
power transformers in service, IEEE Electric. Insulation Magazine 18 (6) (2002) [52] A.S.A. Razzaq, M.F.M. Yousof, S.M. Al-Ameri, M.A. Talib, and A.J. Mshkil,
12–25, https://doi.org/10.1109/MEI.2002.1161455. "Interpretation of inter-turn fault in transformer winding using turns ratio test
[25] I. Fofana, A. Bouaicha, M. Farzaneh, Aging characterization of transformer oil- and frequency response analysis," vol. 11, pp. 218–225, 2022, doi: 10.18178/
pressboard insulation using modern diagnostic techniques, Eur. Trans. Electric. ijeetc.11.3.218-225.
Power 11 (2011) 1110–1127, https://doi.org/10.1002/etep.499, 01/01. [53] K.H. Ibrahim, N.R. Korany, S.M. Saleh, Effects of power transformer high-
[26] D. Linhjell, L. Lundgaard, U. Gafvert, Dielectric response of mineral oil frequency equivalent circuit parameters non-uniformity on fault diagnosis using
impregnated cellulose and the impact of aging, IEEE Trans. Dielectr. Electric. SFRA test, Ain Shams Eng. J. 13 (4) (2022), 101674, https://doi.org/10.1016/j.
Insulation 14 (1) (2007) 156–169, https://doi.org/10.1109/TDEI.2007.302884. asej.2021.101674.
[27] H. Torkaman, F. Karimi, Measurement variations of insulation resistance/ [54] S.M.A.N. Al-Ameri, et al., Understanding the influence of power transformer
polarization index during utilizing time in HV electrical machines – A survey, faults on the frequency response signature using simulation analysis and
Measurement 59 (2015) 21–29, https://doi.org/10.1016/j. statistical indicators, IEEE Access 9 (2021) 70935–70947, https://doi.org/
measurement.2014.09.034. 10.1109/ACCESS.2021.3076984.
[28] IEEE guide for diagnostic field testing of fluid-filled power transformers, [55] M.A. Tsili, E.I. Amoiralis, A.G. Kladas, A.T. Souflaris, Power transformer thermal
regulators, and reactors, IEEE Std C57.152-2013 (2013) 1–121, https://doi.org/ analysis by using an advanced coupled 3D heat transfer and fluid flow FEM
10.1109/IEEESTD.2013.6544533. model, Int. J. Therm. Sci. 53 (2012) 188–201, https://doi.org/10.1016/j.
[29] J. Lukic, et al., Changes of new unused insulating kraft paper properties during ijthermalsci.2011.10.010.
drying—impact on degree of polymerization, Cigrè Sci. Eng. 20 (2021) 161–170. [56] H. Hongsheng, Q. Suxiang, C. Jian, Monitoring and fault diagnosing system
[30] R.D. Medina, A.A. Romero, E.E. Mombello, G. Rattá, Assessing degradation of design for power transformer based on temperature field model and DGA feature
power transformer solid insulation considering thermal stress and moisture extraction, in: 2008 7th World Congress on Intelligent Control and Automation,
variation, Electr. Power Syst. Res. 151 (2017) 1–11, https://doi.org/10.1016/j. 2008, pp. 1800–1805, https://doi.org/10.1109/WCICA.2008.4593195, 25-27
epsr.2017.04.006. June.
[31] IEEE guide for loading mineral-oil-immersed transformers and step-voltage regulators, [57] P. Lu, et al., Real-time monitoring of temperature rises of energized transformer
0-7381-7195-6, 2012. cores with distributed optical fiber sensors, IEEE Trans. Power Delivery 34 (4)
[32] S.S.M. Ghoneim, The degree of polymerization in a prediction model of insulating (2019) 1588–1598, https://doi.org/10.1109/TPWRD.2019.2912866.
paper and the remaining life of power transformers, Energies 14 (3) (2021) 670 [58] Y. Cui, H. Ma, T. Saha, C. Ekanayake, D. Martin, Moisture-dependent thermal
[Online]. Available: https://www.mdpi.com/1996-1073/14/3/670. modelling of power transformer, IEEE Trans. Power Delivery 31 (5) (2016)
[33] S. Forouhari, A. Abu-Siada, Application of adaptive neuro fuzzy inference system 2140–2150, https://doi.org/10.1109/TPWRD.2016.2569123.
to support power transformer life estimation and asset management decision, [59] A. Santisteban, F. Delgado, A. Ortiz, I. Fernández, C.J. Renedo, F. Ortiz,
IEEE Trans. Dielectr. Electric. Insulation 25 (3) (2018) 845–852, https://doi.org/ Numerical analysis of the hot-spot temperature of a power transformer with
10.1109/TDEI.2018.006392. alternative dielectric liquids, IEEE Trans. Dielectr. Electric. Insulation 24 (5)
[34] J. Foros, M. Istad, Health index, risk and remaining lifetime estimation of power (2017) 3226–3235, https://doi.org/10.1109/TDEI.2017.006228.
transformers, IEEE Trans. Power Delivery 35 (6) (2020) 2612–2620, https://doi. [60] M. Mikhak-Beyranvand, J. Faiz, B. Rezaeealam, Thermal analysis of power
org/10.1109/TPWRD.2020.2972976. transformer using an improved dynamic thermal equivalent circuit model, Electr.
[35] Mineral insulating oils - methods for the determination of 2-furfural and related Power Components and Syst. 47 (18) (2019) 1598–1609, https://doi.org/
compounds, 1993. 10.1080/15325008.2019.1689452.
[36] Ageing of liquid impregnated cellulose for power transformers, CIGRE Tech. [61] L. Li, S. Niu, S.L. Ho, W.N. Fu, Y. Li, A novel approach to investigate the hot-spot
Brochure 738 (2018). D1 Materials and emerging test techniques. temperature rise in power transformers, IEEE Trans. Magn. 51 (3) (2015) 1–4,
[37] T.V. Oommen, T.A. Prevost, Cellulose insulation in oil-filled power transformers: https://doi.org/10.1109/TMAG.2014.2359956.
part II maintaining insulation integrity and life, IEEE Electric. Insulation
Magazine 22 (2) (2006) 5–14, https://doi.org/10.1109/MEI.2006.1618996.
15
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
[62] W.H. Tang, Q.H. Wu, Condition Monitoring and Assessment of Power [86] Z. Zhao, C. Tang, Q. Zhou, L. Xu, Y. Gui, C. Yao, Identification of power
Transformers Using Computational Intelligence, Springer London, London, 2011. transformer winding mechanical fault types based on online IFRA by support
London:. vector machine, Energies (Basel) 10 (12) (2017) 2022, https://doi.org/10.3390/
[63] H. Wu, X. Wang, J. Zhang, Online monitoring and diagnosis method of en10122022.
transformer winding deformation, IEEJ Trans. Electric. Electron. Eng. 14 (12) [87] G. Bucci, F. Ciancetta, E. Fiorucci, S. Mari, A. Fioravanti, Deep learning applied to
(2019) 1747–1753, https://doi.org/10.1002/tee.23000. SFRA results: a preliminary study, in: presented at the 2021 7th International
[64] A. Abu-Siada, O. Aljohani, Detecting incipient radial deformations of power Conference on Computing and Artificial Intelligence, Tianjin, China, 2021,
transformer windings using polar plot and digital image processing, IET Sci., https://doi.org/10.1145/3467707.3467753 [Online]. Available:.
Measurement & Technol. 12 (4) (2018) 492–499, https://doi.org/10.1049/iet- [88] A. Tavakoli, L.D. Maria, B. Valecillos, D. Bartalesi, S. Garatti, S. Bittanti,
smt.2017.0412. A machine Learning approach to fault detection in transformers by using
[65] O. Aljohani, A. Abu-Siada, Application of digital image processing to detect short- vibration data, IFAC PapersOnLine 53 (2) (2020) 13656–13661, https://doi.org/
circuit turns in power transformers using frequency response analysis, IEEE 10.1016/j.ifacol.2020.12.866.
Trans. Ind. Inf. 12 (6) (2016) 2062–2073, https://doi.org/10.1109/ [89] Z. Wang, A. Sharma, Research on transformer vibration monitoring and diagnosis
TII.2016.2594773. based on internet of things, J. Intelligent Syst. 30 (1) (2021) 677–688, https://
[66] A.S. Masoum, N. Hashemnia, A. Abu-Siada, M.A.S. Masoum, S.M. Islam, Online doi.org/10.1515/jisys-2020-0111.
transformer internal fault detection based on instantaneous voltage and current [90] K. Hong, M. Jin, H. Huang, Transformer winding fault diagnosis using vibration
measurements considering impact of harmonics, IEEE Trans. Power Delivery 32 image and deep learning, IEEE Trans. Power Delivery 36 (2) (2021) 676–685,
(2) (2017) 587–598, https://doi.org/10.1109/TPWRD.2014.2358072. https://doi.org/10.1109/TPWRD.2020.2988820.
[67] A.G.C. Menezes, M.M. Araujo, O.M. Almeida, F.R. Barbosa, A.P.S. Braga, [91] A. Doolgindachbaporn, G. Callender, P. Lewin, E.A. Simonson, G. Wilson, Data
Induction of Decision Trees to Diagnose Incipient Faults in Power Transformers, driven transformer thermal model for condition monitoring, IEEE Trans. Power
IEEE Trans. Dielectr. Electric. Insulation 29 (1) (2022) 279–286, https://doi.org/ Delivery (2021) 1, https://doi.org/10.1109/TPWRD.2021.3123957. -1.
10.1109/TDEI.2022.3148453. [92] M. Kunicki, S. Borucki, A. Cichoń, J. Frymus, Modeling of the winding hot-spot
[68] N. Haque, A. Jamshed, K. Chatterjee, S. Chatterjee, Accurate sensing of power temperature in power transformers: case study of the low-loaded fleet, Energies
transformer faults from dissolved gas data using random forest classifier aided by 12 (18) (2019), https://doi.org/10.3390/en12183561.
data clustering method, IEEE Sens. J. 22 (6) (2022) 5902–5910, https://doi.org/ [93] G.M. dos Santos, R.R.B. de Aquino, M.M.S. Lira, Thermography and artificial
10.1109/JSEN.2022.3149409. intelligence in transformer fault detection, Electric. Eng. 100 (3) (2017)
[69] Y. Kim, T. Park, S. Kim, N. Kwak, D. Kweon, Artificial intelligent fault diagnostic 1317–1325, https://doi.org/10.1007/s00202-017-0595-2.
method for power transformers using a new classification system of faults, [94] A. Setayeshmehr, A. Akbari, H. Borsi, E. Gockenbach, On-line monitoring and
J. Electric. Eng. Technol. 14 (2) (2019) 825–831, https://doi.org/10.1007/ diagnoses of power transformer bushings, IEEE Trans. Dielectr. Electric.
s42835-019-00105-0. Insulation 13 (3) (2006) 608–615, https://doi.org/10.1109/TDEI.2006.1657975.
[70] O. Kherif, Y. Benmahamed, M. Teguar, A. Boubakeur, S.S.M. Ghoneim, Accuracy [95] T.K.Saha H.Ma, C. Ekanayake, Machine learning techniques for power
improvement of power transformer faults diagnostic using KNN classifier with transformer insulation diagnosis. AUPEC 2011, 2011, pp. 1–6. Brisbane, QLD,
decision tree principle, IEEE Access 9 (2021) 81693–81701, https://doi.org/ Australia.
10.1109/ACCESS.2021.3086135. [96] M. Hui, T.K. Saha, C. Ekanayake, Statistical learning techniques and their
[71] Y.D. Almoallem, I.B.M. Taha, M.I. Mosaad, L. Nahma, A. Abu-Siada, Application applications for condition assessment of power transformer, IEEE Trans. Dielectr.
of logistic regression algorithm in the interpretation of dissolved gas analysis for Electric. Insulation 19 (2) (2012) 481–489, https://doi.org/10.1109/
power transformers, Electronics (Basel) 10 (10) (2021) 1206, https://doi.org/ TDEI.2012.6180241.
10.3390/electronics10101206. [97] S. Amidedin Mousavi, A. Hekmati, M. Sedighizadeh, M. Bigdeli, A. Bazargan,
[72] M. Božić, M. Stojanović, Z. Stajić, Đ. Vukić, Power transformer fault diagnosis ANN based temperature compensation for variations in polarization and
based on dissolved gas analysis with logistic regression, Przegląd depolarization current measurements in transformer, Therm. Sci. Eng. Progress
Elektrotechniczny 6 (2013) 83–87. 20 (2020), 100671, https://doi.org/10.1016/j.tsep.2020.100671, 2020/12/01/.
[73] H.A. Illias, W.Z. Liang, Identification of transformer fault based on dissolved gas [98] Y. Zhang, S. Li, X. Fan, J. Liu, J. Li, Prediction of moisture and aging conditions of
analysis using hybrid support vector machine-modified evolutionary particle oil-immersed cellulose insulation based on fingerprints database of dielectric
swarm optimisation, PLoS ONE 13 (1) (2018), https://doi.org/10.1371/journal. modulus (in English), Polymers (Basel) 12 (8) (2020) 1722, https://doi.org/
pone.0191366 pp. e0191366-e0191366. 10.3390/polym12081722.
[74] K. Bacha, S. Souahlia, and M. Gossa, "Power transformer fault diagnosis based on [99] J. Liu, et al., A BPNN model based adaboost algorithm for estimating inside
dissolved gas analysis by support vector machine," Electric power systems research, moisture of oil-paper insulation of power transformer, IEEE Trans. Dielectr.
vol. 83, no. 1, pp. 73–79, 2012, doi: 10.1016/j.epsr.2011.09.012. Electric. Insulation 29 (2) (2022) 1, https://doi.org/10.1109/
[75] J.I. Aizpurua, et al., Power transformer dissolved gas analysis through Bayesian TDEI.2022.3157909. -1.
networks and hypothesis testing, IEEE Trans. Dielectr. Electric. Insulation 25 (2) [100] B.A. Thango, P.N. Bokoro, Prediction of the degree of polymerization in
(2018) 494–506, https://doi.org/10.1109/TDEI.2018.006766. transformer cellulose insulation using the feedforward backpropagation artificial
[76] R.A. Hooshmand, M. Parastegari, Z. Forghani, Adaptive neuro-fuzzy inference neural network, Energies 15 (12) (2022) 4209 [Online]. Available: https://www.
system approach for simultaneous diagnosis of the type and location of faults in mdpi.com/1996-1073/15/12/4209.
power transformers, IEEE Electric. Insulation Magazine 28 (5) (2012) 32–42, [101] M. Nezami, M. Equbal, S.A. Khan, S. Sohail, An ANFIS based comprehensive
https://doi.org/10.1109/MEI.2012.6268440. correlation between diagnostic and destructive parameters of transformer’s paper
[77] J. Faiz, M. Soleimani, Assessment of computational intelligence and conventional insulation, Arabian J. Sci. Eng. 46 (2) (2021) 1541–1547.
dissolved gas analysis methods for transformer fault diagnosis, IEEE Trans. [102] R.A. Prasojo, K. Diwyacitta, S. Suwarno, H. Gumilang, Transformer paper
Dielectr. Electric. Insulation 25 (5) (2018) 1798–1806, https://doi.org/10.1109/ expected life estimation using ANFIS based on oil characteristics and dissolved
TDEI.2018.007191. gases (Case study: indonesian transformers), Energies (Basel) 10 (8) (2017) 1135,
[78] H.A. Illias, X.R. Chai, A.H. Abu Bakar, Hybrid modified evolutionary particle https://doi.org/10.3390/en10081135.
swarm optimisation-time varying acceleration coefficient-artificial neural [103] A. Abu-Siada, S.P. Lai, S.M. Islam, A novel fuzzy-logic approach for furan
network for power transformer fault diagnosis, Measurement 90 (2016) 94–102, estimation in transformer oil, IEEE Trans. Power Delivery 27 (2) (2012) 469–474,
https://doi.org/10.1016/j.measurement.2016.04.052. https://doi.org/10.1109/TPWRD.2012.2186986.
[79] T. Kari, et al., An integrated method of ANFIS and Dempster-Shafer theory for [104] P.K. Pandey, Z. Husain, R.K. Jarial, ANFIS based approach to estimate remnant
fault diagnosis of power transformer, IEEE Trans. Dielectr. Electric. Insulation 25 life of power transformer by predicting furan contents, Int. J. Electric. Comput.
(1) (2018) 360–371, https://doi.org/10.1109/TDEI.2018.006746. Eng. (Malacca, Malacca) 4 (4) (2014) 463, https://doi.org/10.11591/ijece.
[80] T.K. Abdel-Galil, R.M. Sharkawy, M.M.A. Salama, R. Bartnikas, Partial discharge v4i4.5567.
pattern classification using the fuzzy decision tree approach, IEEE Trans. Instrum. [105] L. Ruijin, Z. Hanbo, S. Grzybowski, Y. Lijun, Z. Yiyi, L. Yuxiang, An integrated
Meas. 54 (6) (2005) 2258–2263, https://doi.org/10.1109/TIM.2005.858143. decision-making model for condition assessment of power transformers using
[81] Y. Sun, et al., Partial discharge pattern recognition of transformers based on fuzzy approach and evidential reasoning, IEEE Trans. Power Delivery 26 (2)
mobilenets convolutional neural network, Appl. Sci. 11 (15) (2021) 6984, (2011) 1111–1118, https://doi.org/10.1109/TPWRD.2010.2096482.
https://doi.org/10.3390/app11156984. [106] N.A. Bakar, A. Abu-Siada, A novel method of measuring transformer oil interfacial
[82] X. Zhou, et al., Research on transformer partial discharge uhf pattern recognition tension using UV–Vis spectroscopy, IEEE Electric. Insulation Magazine 32 (1)
based on cnn-lstm, Energies (Basel) 13 (1) (2019) 61, https://doi.org/10.3390/ (2016) 7–13, https://doi.org/10.1109/MEI.2016.7361098.
en13010061. [107] M. Arshad, S.M. Islam, A. Khaliq, Fuzzy logic approach in power transformers
[83] Y. Xi, L. Yu, B. Chen, G. Chen, Y. Chen, Research on pattern recognition method of management and decision making, IEEE Trans. Dielectr. Electric. Insulation 21
transformer partial discharge based on artificial neural network, Security and (5) (2014) 2343–2354, https://doi.org/10.1109/TDEI.2014.003859.
Commun. Networks (2022), https://doi.org/10.1155/2022/5154649. [108] R.M. Arias Velásquez, Support vector machine and tree models for oil and Kraft
[84] J. Wang, et al., Evaluation on partial discharge intensity of electrical equipment degradation in power transformers, Eng. Fail Anal. 127 (2021), 105488, https://
based on improved ANFIS and ultraviolet pulse detection technology, IEEE Access doi.org/10.1016/j.engfailanal.2021.105488.
7 (2019) 126561–126570, https://doi.org/10.1109/ACCESS.2019.2938784. [109] Machine learning tools in support of transformer diagnostics, CIGRE Tech.
[85] J.R. Secue, E. Mombello, Sweep frequency response analysis (SFRA) for the Brochure A2-206 (2018).
assessment of winding displacements and deformation in power transformers, [110] S. obulareddy, S.K. Munagala, Application of machine learning algorithms for
Electric Power Syst. Res. 78 (6) (2008) 1119–1128, https://doi.org/10.1016/j. power transformer internal faults identification, IEEE 1 (2021) 331–335, https://
epsr.2007.08.005. doi.org/10.1109/ICACCS51430.2021.9441719.
16
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
[111] Z. Zhao, J. Xu, Y. Zang, R. Hu, Adaptive abnormal oil temperature diagnosis [135] A. Alqudsi, A. El-Hag, Assessing the power transformer insulation health
method of transformer based on concept drift, Appl. Sci. 11 (14) (2021) 6322, condition using a feature-reduced predictor mode, IEEE Trans. Dielectr. Electric.
https://doi.org/10.3390/app11146322. Insulation 25 (3) (2018) 853–862, https://doi.org/10.1109/TDEI.2018.006630.
[112] M.E.A. Senoussaoui, M. Brahami, I. Fofana, Transformer oil quality assessment [136] M. Islam, G. Lee, S.N. Hettiwatte, K. Williams, Calculating a health index for
using random forest with feature engineering, Energies (Basel) 14 (7) (2021) power transformers using a subsystem-based GRNN approach, IEEE Trans. Power
1809, https://doi.org/10.3390/en14071809. Delivery 33 (4) (2018) 1903–1912, https://doi.org/10.1109/
[113] A. Alqudsi, A. El-Hag, Application of machine learning in transformer health TPWRD.2017.2770166.
index prediction, Energies (Basel) 12 (14) (2019) 2694, https://doi.org/10.3390/ [137] D.G.T. Da Silva, H.J. Braga Da Silva, F.P. Marafão, H.K.M. Paredes, F.A.
en12142694. S. Gonçalves, Enhanced health index for power transformers diagnosis, Eng. Fail.
[114] S.S.M. Ghoneim, I.B.M. Taha, Comparative study of full and reduced feature Anal. 126 (2021), 105427, https://doi.org/10.1016/j.engfailanal.2021.105427.
scenarios for health index computation of power transformers, IEEE Access 8 [138] D. Rediansyah, R.A. Prasojo, Suwarno, A. Abu-Siada, Artificial intelligence-based
(2020) 181326–181339, https://doi.org/10.1109/ACCESS.2020.3028689. power transformer health index for handling data uncertainty, IEEE Access 9
[115] S.-.W. Fei, Y. Sun, Forecasting dissolved gases content in power transformer oil (2021) 150637–150648, https://doi.org/10.1109/ACCESS.2021.3125379.
based on support vector machine with genetic algorithm, Electr. Power Syst. Res. [139] R.M. Arias Velásquez, J.V. Mejía Lara, A. Melgar, Converting data into knowledge
78 (3) (2008) 507–514, https://doi.org/10.1016/j.epsr.2007.04.006. for preventing failures in power transformers, Eng. Fail. Anal. 101 (2019)
[116] C.H. Wei, W.H. Tang, Q.H. Wu, A hybrid least-square support vector machine 215–229, https://doi.org/10.1016/j.engfailanal.2019.03.027.
approach to incipient fault detection for oil-immersed power transformer, Electr. [140] M.S. Yahaya, N. Azis, M.Z.A. Ab Kadir, J. Jasni, M.H. Hairi, M.A. Talib,
Power Components and Syst. 42 (5) (2014) 453–463, https://doi.org/10.1080/ Estimation of transformers health index based on the markov chain, Energies 10
15325008.2013.857180. (11) (2017) 1824 [Online]. Available: https://www.mdpi.com/1996-1073/10/11
[117] S.A. Wani, A.S. Rana, S. Sohail, O. Rahman, S. Parveen, S.A. Khan, Advances in /1824.
DGA based condition monitoring of transformers: a review, Renew. Sustain. [141] R. Murugan and R. Raju, "Evaluation of in-service power transformer health
Energy Rev. 149 (2021), 111347, https://doi.org/10.1016/j.rser.2021.111347. condition for Inspection, Repair, and Replacement (IRR) maintenance planning in
[118] T. Kari, W. Gao, A. Tuluhong, Y. Yaermaimaiti, Z. Zhang, Mixed kernel function electric utilities," International J. Syst. Assurance Eng. Manag., vol. 12, no. 2, pp.
support vector regression with genetic algorithm for forecasting dissolved gas 318–336, 2021, doi: 10.1007/s13198-021-01083-1.
content in power transformers, Energies (Basel) 11 (9) (2018) 2437, https://doi. [142] P. Bohatyrewicz, J. Płowucha, J. Subocz, Condition assessment of power
org/10.3390/en11092437. transformers based on health index value, Appl. Sci. 9 (22) (2019) 4877, https://
[119] Z. Zhe-wen, W. Yong, Y. Ding, T. lei, Z. Ying-jian, A transformer fault diagnosis doi.org/10.3390/app9224877.
method based on bayesian network, J. Phys.: Conf. Ser 1213 (5) (2019) 52002, [143] M. Pompili, F. Scatiggio, Classification in iso-attention classes of hv transformer
https://doi.org/10.1088/1742-6596/1213/5/052002. fleets, IEEE Trans. Dielectr. Electric. Insulation 22 (5) (2015) 2676–2683, https://
[120] L. Cheim, L. Lin, A. Dagnino, Probabilistic transformer fault tree analysis using doi.org/10.1109/TDEI.2015.005252.
Bayesian networks, in: 2014 IEEE PES T&D Conference and Exposition, 2014, [144] A.M. Selva, et al., Application of statistical distribution models to predict health
pp. 1–5, https://doi.org/10.1109/TDC.2014.6863177, 14-17 April. index for condition-based management of transformers, Appl. Sci. 11 (6) (2021)
[121] C. Hu, C. Zhang, Z. Zhang, S. Xie, Comparative study on defects and faults 2728, https://doi.org/10.3390/app11062728.
detection of main transformer based on logistic regression and naive bayes [145] J.H. Jürgensen, L. Nordström, P. Hilber, Estimation of individual failure rates for
algorithm, J. Phys.: Conf. Ser 1732 (1) (2021) 12075, https://doi.org/10.1088/ power system components based on risk functions, IEEE Trans. Power Delivery 34
1742-6596/1732/1/012075. (4) (2019) 1599–1607, https://doi.org/10.1109/TPWRD.2019.2913777.
[122] A. Li, X. Yang, H. Dong, Z. Xie, C. Yang, Machine learning-based sensor data [146] W. Zuo, H. Yuan, Y. Shang, Y. Liu, T. Chen, Calculation of a health index of oil-
modeling methods for power transformer PHM, Sensors (Basel) 18 (12) (2018) paper transformers insulation with binary logistic regression, Math. Problems in
4430, https://doi.org/10.3390/s18124430. Eng. 2016 (2016), 6069784, https://doi.org/10.1155/2016/6069784.
[123] C. Yan, M. Li, and W. Liu, "Transformer fault diagnosis based on BP-adaboost and [147] A.E.B. Abu-Elanien, M.M.A. Salama, M. Ibrahim, Calculation of a health index for
PNN series connection," Math. Problems in Eng., vol. 2019, pp. 1–10, 2019, doi: oil-immersed transformers rated under 69kV using fuzzy logic, IEEE Trans. Power
10.1155/2019/1019845. Delivery 27 (4) (2012) 2029–2036, https://doi.org/10.1109/
[124] S. Chen, H. Ge, H. Li, Y. Sun, X. Qian, Hierarchical deep convolution neural TPWRD.2012.2205165.
networks based on transfer learning for transformer rectifier unit fault diagnosis, [148] R.D. Medina, D.A. Zaldivar, A.A. Romero, J. Zuñiga, E.E. Mombello, A fuzzy
Measurement: J. Int. Measurement Confederation 167 (2021), 108257, https:// inference-based approach for estimating power transformers risk index, Electr.
doi.org/10.1016/j.measurement.2020.108257. Power Syst. Res. 209 (2022), 108004, https://doi.org/10.1016/j.
[125] J. Lin, J. Ma, J. Zhu, Hierarchical federated learning for power transformer fault epsr.2022.108004.
diagnosis, IEEE Trans. Instrum. Meas. 71 (2022) 1–11, https://doi.org/10.1109/ [149] A. Mohmad, M. Ibrahim Shapiai, M. Solehin Shamsudin, M.A. Abu, A.A. Hamid,
TIM.2022.3196736. Investigating performance of transformer health index in machine learning
[126] S.A. Khan, M.D. Equbal, T. Islam, ANFIS based identification and location of application using dominant features, J. Phys.: Conf. Ser 2128 (1) (2021) 12025,
paper insulation faults of an oil immersed transformer, in: 2014 6th IEEE Power https://doi.org/10.1088/1742-6596/2128/1/012025.
India International Conference (PIICON), 2014, pp. 1–6, https://doi.org/ [150] K. Ibrahim, R.M. Sharkawy, H.K. Temraz, M.M.A. Salama, Selection criteria for oil
10.1109/POWERI.2014.7117715, 5-7 Dec. transformer measurements to calculate the Health Index, IEEE Trans. Dielectr.
[127] F. Cheng, L. Qu, W. Qiao, A case-based data-driven prediction framework for Electric. Insulation 23 (6) (2016) 3397–3404, https://doi.org/10.1109/
machine fault prognostics, in: 2015 IEEE Energy Conversion Congress and TDEI.2016.006058.
Exposition (ECCE), 2015, pp. 3957–3963, https://doi.org/10.1109/ [151] A.D. Ashkezari, H. Ma, T.K. Saha, C. Ekanayake, Application of fuzzy support
ECCE.2015.7310219, 20-24 Sept. vector machine for determining the health index of the insulation system of in-
[128] H. Metin Ertunc, M. Hosoz, Comparative analysis of an evaporative condenser service power transformers, IEEE Trans. Dielectr. Electric. Insulation 20 (3)
using artificial neural network and adaptive neuro-fuzzy inference system, Int. J. (2013) 965–973, https://doi.org/10.1109/TDEI.2013.6518966.
Refrig. 31 (8) (2008) 1426–1436, https://doi.org/10.1016/j.ijrefrig.2008.03.007. [152] H. Zeinoddini-Meymand, B. Vahidi, Health index calculation for power
[129] S.A. Khan, M.D. Equbal, T. Islam, A comprehensive comparative study of DGA transformers using technical and economical parameters, IET Sci. Measurement &
based transformer fault diagnosis using fuzzy logic and ANFIS models, IEEE Technol. 10 (7) (2016) 823–830, https://doi.org/10.1049/iet-smt.2016.0184.
Trans. Dielectr. Electric. Insulation 22 (1) (2015) 590–596, https://doi.org/ [153] S.K. Abeygunawardane, P. Jirutitijaroen, H. Xu, Adaptive maintenance policies
10.1109/TDEI.2014.004478. for aging devices using a markov decision process, IEEE Trans. Power Syst. 28 (3)
[130] M.S. Katooli, A. Koochaki, Detection and classification of incipient faults in three- (2013) 3194–3203, https://doi.org/10.1109/TPWRS.2012.2237042.
phase power transformer using DGA information and rule-based machine learning [154] S. Milosavljevic, A. Janjic, Integrated transformer health estimation methodology
method, J. Control, Automation & Electric. Syst. 31 (5) (2020) 1251–1266, based on markov chains and evidential reasoning, Math. Problems in Eng. 2020
https://doi.org/10.1007/s40313-020-00625-5. (2020), 7291749, https://doi.org/10.1155/2020/7291749.
[131] T.A. Aman, A.S. Mokhtar, M. AizamTalib, Assessment the overall health condition [155] P. Sarajcev, D. Jakus, M. Nikolic, Wide & deep machine learning model for
of transformer using health index and critical index approach: TNB grid case transformer health analysis, in: 2019 4th International Conference on Smart and
study, in: 2020 IEEE International Conference on Power and Energy (PECon), Sustainable Technologies (SpliTech), 2019, pp. 1–6, https://doi.org/10.23919/
2020, pp. 235–239, https://doi.org/10.1109/PECon48942.2020.9314483, 7-8 SpliTech.2019.8783122, 18-21 June.
Dec. [156] M.M.S NAKAI, M. YOSHIDA, K. AOKI, Rationalization and high precision of
[132] M.M. Islam, G. Lee, S.N. Hettiwatte, Application of a general regression neural transformer lifetime evaluation method, CIGRE Tech. Brochure A2-315 (2020).
network for health index calculation of power transformers, Int. J. Electric. Power [157] A.d. C.P. Krause, F. Devaux, H. Ding, V. Katshuna, J. Lukic, L. Melzer, S. Miyazaki,
& Energy Syst. 93 (2017) 308–315, https://doi.org/10.1016/j. M. Munro, A. Peixoto, M. Scala, D. Walker, The condition of solid transformer
ijepes.2017.06.008. insulation at end-of-life CIGRE Reference Paper (2022). vol. A2-321
[133] Pattern recognition techniques for determining the health index of oil-paper [158] R.A. Prasojo, S. Suwarno, Power transformer paper insulation assessment based
insulation of in-service power transformers, CIGRE Tech. Brochure (2014) vol. on oil measurement data using SVM-classifier, Int. J. Electric. Eng. Inf. 10 (4)
A2_105_2014. (2018) 661–673, https://doi.org/10.15676/ijeei.2018.10.4.4.
[134] R.A. Prasojo, N.U. Maulidevi, B.A. Soedjarno, S. Suwarno, Health index analysis [159] B. Gorgan, et al., Calculation of the remaining lifetime of power transformers
of power transformer with incomplete paper condition data, in: 2019 IEEE 4th paper insulation, in: 2012 13th International Conference on Optimization of
International Conference on Condition Assessment Techniques in Electrical Electrical and Electronic Equipment (OPTIM), 2012, pp. 293–300, https://doi.
Systems (CATCON), 2019, pp. 1–4, https://doi.org/10.1109/ org/10.1109/OPTIM.2012.6231792, 24-26 May.
CATCON47128.2019.CN0073, 21-23 Nov. 2019. [160] L. Kong, W. Mo, X. Ge, T. Liu, F. Zhou, H. Chen, Application of insulation aging
evaluation method for distribution transformers in practice, in: 2020 IEEE
17
L. Jin et al. Electric Power Systems Research 218 (2023) 109194
International Conference on High Voltage Engineering and Application (ICHVE), [176] Furanic compounds for diagnosis, CIGRE Working Group D1.01(TF13) 494
2020, pp. 1–4, https://doi.org/10.1109/ICHVE49031.2020.9279932, 6-10 Sept. (2012).
[161] Ageing of cellulose in mineral-oil insulated transformers, CIGRE TASK FORCE [177] B.A. Thango, A.O. Akumu, L.S. Sikhosana, A.F. Nnachi, J.A. Jordaan, Study of the
D1.01.10 323 (2007). impact of degree of polymerization on the remnant life of transformer insulation.
[162] IEEE guide for loading mineral-oil-immersed transformers and step-voltage 2021 IEEE PES/IAS PowerAfrica, 2021, pp. 1–5, https://doi.org/10.1109/
regulators, IEEE Std C57.91-2011 (Revision of IEEE Std C57.91-1995) (2012) PowerAfrica52236.2021.9543450, 23-27 Aug. 2021.
1–123, https://doi.org/10.1109/IEEESTD.2012.6166928. [178] P. Sen, S. Pansuwan, Overloading and loss-of-life assessment guidelines of oil-
[163] AS/NZS 60076.7 Power Transformers Part 7: loading guide for oil-immersed power cooled transformers, in: 2001 Rural Electric Power Conference. Papers Presented
transformers (IEC 60076-7, Ed. 1.0 (2005) MOD), 2013. at the 45th Annual Conference (Cat. No. 01CH37214), IEEE, 2001 pp. B4/1-B4/8.
[164] P.K. Sen, P. Sarunpong, Overloading and loss-of-life assessment guidelines of oil- [179] Z. Tian, An artificial neural network method for remaining useful life prediction
cooled transformers, in: 2001 Rural Electric Power Conference. Papers Presented of equipment subject to condition monitoring, J. Intell. Manuf. 23 (2) (2012)
at the 45th Annual Conference (Cat. No.01CH37214), 2001, https://doi.org/ 227–237, https://doi.org/10.1007/s10845-009-0356-9.
10.1109/REPCON.2001.949516, 29 April-1 Maypp. B4/1-B4/8. [180] A.R. Abbasi, Fault detection and diagnosis in power transformers: a
[165] H. Gorginpour, H. Ghimatgar, M.S. Toulabi, Lifetime estimation and optimal comprehensive review and classification of publications and methods, Electr.
maintenance scheduling of urban oil-immersed distribution-transformers Power Syst. Res. 209 (2022), 107990, https://doi.org/10.1016/j.
considering weather-dependent intelligent load model and unbalanced loading. epsr.2022.107990.
IEEE Transactions On Power Delivery, 2022, p. 1, https://doi.org/10.1109/ [181] M. Arshad, S.M. Islam, A novel fuzzy logic technique for power transformer asset
TPWRD.2022.3146154. -1. management, in: Conference Record of the 2006 IEEE Industry Applications
[166] V.M. Catterson, Prognostic modeling of transformer aging using Bayesian particle Conference Forty-First IAS Annual Meeting 1, 2006, pp. 276–286, https://doi.
filtering, in: 2014 IEEE Conference on Electrical Insulation and Dielectric org/10.1109/IAS.2006.256536, 8-12 Oct.
Phenomena (CEIDP), 2014, pp. 413–416, https://doi.org/10.1109/ [182] R.M. Arias Velasquez, J.V. Mejia Lara, Root cause analysis improved with
CEIDP.2014.6995874, 19-22 Oct. machine learning for failure analysis in power transformers, Eng. Fail. Anal. 115
[167] A. Bagheri, M. Allahbakhshi, M.M. Arefi, N. Najafi, M.S. Javadi, A new approach (2020), 104684, https://doi.org/10.1016/j.engfailanal.2020.104684.
for top-oil thermal modelling of power transformers using Unscented Kalman [183] S. Li, G. Wu, H. Dong, L. Yang, X. Zhen, Probabilistic health index-based apparent
filter considering IEEE C57.91 standard, IET Electr. Power Appl. 16 (5) (2022) age estimation for power transformers, IEEE Access 8 (2020) 9692–9701, https://
536–547, https://doi.org/10.1049/elp2.12174. doi.org/10.1109/ACCESS.2020.2963963.
[168] W. Chen, X. Su, Application of Kalman filter to hot-spot temperature monitoring [184] A. Abu-Siada, M. Arshad, S. Islam, Fuzzy logic approach to identify transformer
in oil-immersed power transformer, IEEJ Trans. Electric. Electron. Eng. 8 (4) criticality using dissolved gas analysis, in: IEEE PES General Meeting,
(2013) 322–327, https://doi.org/10.1002/tee.21862. Minneapolis, MN, USA, 2010, pp. 1–5, https://doi.org/10.1109/
[169] W. Lai, H. Luo, W. Li, Y. Cao, L. Ye, Y. Wang, Prediction of top oil temperature for PES.2010.5589789.
oil-immersed transformer based on Kalman filter algorithm, in: 2017 2nd [185] A.B. Norazhar, A. Abu-Siada, S. Islam, A review on chemical diagnosis techniques
International Conference on Power and Renewable Energy (ICPRE), 2017, for transformer paper insulation degradation, in: 2013 Australasian Universities
pp. 132–136, https://doi.org/10.1109/ICPRE.2017.8390514, 20-23 Sept. Power Engineering Conference (AUPEC), Hobart, TAS, Australia, 2013, pp. 1–6,
[170] S. Li, H. Ma, T.K. Saha, Y. Yang, G. Wu, On particle filtering for power transformer https://doi.org/10.1109/AUPEC.2013.6725476.
remaining useful life estimation, IEEE Trans. Power Delivery 33 (6) (2018) [186] A. Abu-Siada, “Improved Consistent interpretation approach of fault type within
2643–2653, https://doi.org/10.1109/TPWRD.2018.2807386. power transformers using dissolved gas analysis and gene expression
[171] D. Martin, C. Yi, C. Ekanayake, M. Hui, and T. Saha, "An updated model to programming”, Energies, vol 12, no. 4, p. 730, February 2019.
determine the life remaining of transformer insulation," IEEE Trans. Power [187] L. Dias, et al., An unsupervised approach for fault diagnosis of power
Delivery, vol. 30, no. 1, pp. 395–402, 2015, doi: 10.1109/TPWRD.2014.2345775. transformers, Quality and Reliability Eng. Int. 37 (6) (2021) 2834–2852, https://
[172] E. Zhang, et al., Aging state assessment of transformer cellulosic paper insulation doi.org/10.1002/qre.2892.
using multivariate chemical indicators, Cellulose 28 (4) (2021) 2445–2460. [188] J. Wu, K. Li, J. Sun, L. Xie, A novel integrated method to diagnose faults in power
[173] V. Vasovic, et al., Aging of transformer insulation — Experimental transformers transformers, Energies (Basel) 11 (11) (2018) 3041, https://doi.org/10.3390/
and laboratory models with different moisture contents: part I — DP and furans en11113041.
aging profiles, IEEE Trans. Dielectr. Electric. Insulation 26 (6) (2019) 1840–1846, [189] Y. Benmahamed, M. Teguar, A. Boubakeur, Application of SVM and KNN to Duval
https://doi.org/10.1109/TDEI.2019.008183. Pentagon 1 for transformer oil diagnosis, IEEE Trans. Dielectr. Electric. Insulation
[174] C.P. Krause, Alfonso de, Francois Devaux, Hongzi Ding, Victor Katshuna, 24 (6) (2017) 3443–3451, https://doi.org/10.1109/TDEI.2017.006841.
Jelena Lukic, Lena Melzer, Satoru Miyazaki, Mike Munro, Anabela Peixoto, [190] J. Fan, F. Wang, Q. Sun, F. Bin, F. Liang, X. Xiao, Hybrid RVM–ANFIS algorithm
Mario Scala, David Walker, The condition of solid transformer insulation at end- for transformer fault diagnosis, IET Generation, Transm. Distribution 11 (14)
of-life, CIGRE Reference Paper 321 (2022) vol. Electra. (2017) 3637–3643, https://doi.org/10.1049/iet-gtd.2017.0547.
[175] A.d. Pablo, Furfural and ageing: how are they related. IEE Colloquium on
Insulating Liquids (Ref. No. 1999/119), 1999, https://doi.org/10.1049/ic:
19990667, 27-27 May 1999pp. 5/1-5/4.
18