Received: 16 April 2020

DOI: 10.1049/elp2.12031

REVIEW
- -Revised: 6 November 2020 Accepted: 16 November 2020

- IET Electric Power Applications

An overview of various faults detection methods in synchronous

generators

Mohsen Mostafaei | Jawad Faiz

Centre of Excellence on Applied Electromagnetic Abstract

Systems, School of Electrical and Computer
Engineering, College of Engineering, University of
Synchronous generator (SG) plays a vital and critical role in the power system by sup-
Tehran, Tehran, Iran plying electric power to consumers. Various faults in SGs can cause some catastrophic
events such as power disruption or blackout. These faults can be classified into two
Correspondence electrical and mechanical faults. Short circuit in stator windings and field winding are
Jawad Faiz, Centre of Excellence on Applied electrical fault while bearing, static/dynamic eccentricity, and broken damper bars faults
Electromagnetic Systems, School of Electrical and
Computer Engineering, College of Engineering,
are mechanical one. Unlike the induction machines, there are no much researches in SGs
University of Tehran, Tehran 1439957131, Iran. condition monitoring owing to its complex behaviour against the faults. Herein, the SG
Email: jfaiz@ut.ac.ir modelling approaches are presented briefly to elaborate shortcoming and challenging
issues in the modelling, and then a comprehensive review of various electrical and me-
chanical fault detection methods is presented.

1 | INTRODUCTION and the mechanical faults include:

Synchronous generator (SG) is one of the most important � Bearing failure.

apparatus in the power systems, which supply the electrical en- � Broken damper bars.
ergy to consumers. The SGs are highly reliable equipment but � Rotor mechanical integrity failure.
the faults are unavoidable and can interrupt the electrical power � Stator mechanical integrity failure.
supply. Since the SG reliability is a critical issue in the power
systems, various protection techniques have been so far pro- Creating a real fault on the SG and analysing its behaviour
posed to enhance the power system reliability. Although reli- is mostly destructive and costly, because the SG cannot operate
ability enhancement improves by redundancy (adding another properly after such experiment. Moreover, a real fault can be
SG as the backup unit), it increases the cost, volume and weight dangerous and cause some problems. That is why modelling
of the power generation system. Hence, an appropriate approach plays a major role in the SGs analysis by reducing costs and
uses a sensitive and accurate fault diagnosis method to detect the dangers.
faults in the incipient level. In large SGs, the stator‐winding fault Herein, various SGs modelling firstly are presented. Then,
is the most prevalent one, and the rotor‐winding fault has second rotor/stator winding faults, as the most probable faults, are
priority. Figure 1 shows the proportion of the faults which occur reviewed. Rotor eccentricity and broken damper bars faults are
in large electrical machines [1]. Although, there are a lot of work discussed in details. Finally, the last section provides the
on the fault diagnosis of induction machines, less attention has conclusions.
been paid to SGs faults detections. The reason includes
complicated structure and difficult starting process of the
generator. The failure modes can be classified into electrical and 2 | SYNCHRONOUS GENERATOR
mechanical ones. The electrical faults include [2]: MODELLING
� Core insulation failure. Model‐based approaches are useful for developing a condition
� Stator winding or its insulation failure. monitoring method in electrical machines [3]. In fact, electrical
� Rotor winding or its insulation failure machines modelling can help to understand the faulty machine

-
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is
properly cited.
© 2021 The Authors. IET Electric Power Applications published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

IET Electr. Power Appl. 2021;15:391–404. wileyonlinelibrary.com/journal/elp2 391

392
- MOSTAFAEI AND FAIZ

space harmonics of the winding [7]. To overcome this

shortcoming winding function approach (WFA) has been
proposed for modelling.
Although, the WFA is more complicated than the dq0
model, it is applicable to the asymmetrical systems. This model is
able to estimate the inductances in various healthy and faulty
conditions accurately. The WFA shows the air‐gap magneto‐
motive force (MMF) distribution in the air‐gap due to the
winding [8], which takes into account the space harmonics,
slotting, air‐gap permeance and slot skewing effects [9,10]. The
WFA is capable to model various electrical machines in different
fault conditions and it only needs to consider the effect of fault
in the estimated inductance matrix. The internal fault has been
modelled in Ref. [11] based on the WFA, in which a new
inductance estimation procedure is given, and it also integrate all
F I G U R E 1 Proportion of faults in large electrical machines [1]
space harmonics of the fault modelling. To apply this method
the following assumptions must be considered:
behaviour. Then, it will be possible to develop a fault detection
method or increase the sensitivity and accuracy of the available 1. Neglecting stator slots effects.
techniques before doing any test. 2. Ignoring magnetic hysteresis.
Speed and accuracy are two important factors in the 3. Neglecting iron saturation effects.
model‐based condition monitoring methods. Simple models
are mostly applicable at high speed and low accuracy. However, In the WFA, the general expression of the mutual induc-
the complicated ones have adequate sensitivity and a large tance between winding “i” and winding “j” of electrical ma-
computational burden. Hence, a compromise must be made chine is given by:
between these two factors to achieve a reasonable model.
Modelling can predict the SG behaviour in various con- 2π
ditions such as over‐loading and failures modes, before the Lij ðθÞ ¼ μ0 rl ∫ N i ðφ; θÞ:N j ðφ; θÞ:g 1 ðφ; θÞdφ ð1Þ
examination; this decreases the experiment costs significantly. 0
Generally, the SGs are modelled by the following four main
methods: where μ0 is the free space permeability, r is the air gap
average radius, l is the axial stack length of the machine, and
1. dq0 frame. g 1(φ,θ) is the inverse air‐gap length function which is constant
2. Winding function approach (WFA). in the case of uniform air gap. Ni(φ,θ) and Nj(φ,θ) are the
3. Phase domain model (PDM). winding functions of the windings i and j, φ is the angle along
4. Finite element method (FEM). the inner surface of the stator, and θ is the angular position of
the rotor with respect to the stator reference axis.
Each model has its own advantages and disadvantages. Referring to machine parameters, (1) is modified as
The simplest modelling method of the SG is the use of. follows:
the dq0 frame. Although, this modelling technique is
applicable in symmetrical and balanced conditions, it cannot � �
Lmd þ Lmq φB0 2 Lmd þ Lmq
be recommended for asymmetrical systems [4]. Owing to the LBA ¼ � ∫ nB ðφ; θÞ:N A ðφ; θÞdφ − � �
3πw2s φB 3πw2s
simplicity of this approach, its computations time is relatively
short, and the parameters of the SG can be calculated quickly. φB 0
The dq0 model is a useful tool for prediction of healthy SG ∫ nB ðφ; θÞ:N A ðφ; θÞdφ:cos 2 pðφ − θÞdθ
φB
behaviour. At this end, the dq0 parameters are calculated for
stator/rotor windings and damper bars [5]. However, the ð2Þ
dynamic model of the brushless SG has been presented in
Ref. [6] which models the SGs with the turn‐to‐turn fault where LBA is the mutual inductance between circuit A and
(TTF) using the dq0 frame. Simulation and experimental B, and Lmd and Lmq are the d‐axis and q‐axis magnetising
results show that this model is appropriate for a minor fault, inductances, respectively. Moreover, ws is the amplitude of the
but it is not preferred for severe faults such as phase‐to‐- fundamental harmonic of the stator phase winding function of
ground or phase‐to‐phase short‐circuits. The reason is that a healthy machine, nB(φ,θ) is the total turns function of circuit
the high severity faults can lead to asymmetrical and unbal- B, NA(φ,θ) is the total winding function produced by circuit A
anced system. It means that the dq0 model is not accurate for and finally p is the number of pole pairs.
severe faults detection in the SGs. Moreover, the dq0 models By using (2), the stator inductance, rotor inductance, and
used in the simulation of the SGs cannot account for the mutual inductance between the stator and rotor of the healthy
MOSTAFAEI AND FAIZ
- 393

machine are easily calculated. However, in faulty machine, the Few modelling methods consider the saturation effect in
following equation must be solved numerically. the faulty SGs. In Ref. [23], the saturation effect has been taken
into account in the induction motor with uniform air‐gap. A
� � �� robust non‐linear model of the SGs has been introduced in
dI dL
¼ L 1 Rþ I þ L 1V ð3Þ Ref. [24] to overcome some non‐linear phenomena such as the
dt dt
SGs saturation.
The MWFA model of the mixed eccentricity (ME) fault has
This is a vector form of the voltage equation. The been reported in Ref. [25] in which the self‐ and mutual‐in-
mentioned method of the inductance matrix calculation in the ductances are calculated precisely.
SG improves the accuracy of the SG model in an asymmetrical Last but not least, the finite element method (FEM) is the
system such as phase‐to‐ground fault and phase‐to‐phase fault. most accurate modelling approach that is able to model various
The results reported in Ref. [11] illustrate that experimental electrical machines by considering different conditions such as
and simulation waveforms are reasonably close. saturation, core loss, and rotor and stator faults. In fact, FEM
Apart from severe short circuit faults, rotor misalignment is able to validate other models when a real machine is not
is another failure that leads to an asymmetrical system and accessible, or an experiment is costly and challenging. More-
needs a more accurate model; however, it is slower than the over, the magnetic flux pattern can be visualised in FEM
dq0 frame model. The dynamic eccentricity fault has been software, which gives a clear viewpoint. Despite of the FEM
modelled using WFA in Ref. [12], in which the saturation effect modelling accuracy, it needs a powerful computer particularly
has been included. for complex calculations. Hence, using this model for fault
To shorten the computation time of WFA and to improve detection is difficult and time‐consuming process, and it is only
the accuracy of the dq0 frame method, a hybrid modelling applicable for validation or optimisation of electrical machines.
method has been proposed by merging the two modelling FEM has been applied to validate the MWFA modelling
approaches. It means that the hybrid model [13] can work in method in the saturation condition and eccentricity fault [26].
both symmetrical and asymmetrical systems. As shown in Table 1 shows a brief review of various SG modelling methods.
Figure 2, for low severity fault (SC of 5% to 10% of stator
winding turns), dq0 and hybrid method are accurate, but in
saturation condition with more severe faults, the dq0 model 3 | STATOR WINDING FAULTS
cannot follow the real machine waveform and this error in-
creases in a higher saturation level and more severe faults. In According to Figure 1, 60% of the faults are dedicated to the
low severity fault, the faults such as turn‐to‐turn fault (TTF), stator winding. The most probable causes for winding fault are
stator core does not saturate, but in severe ones, saturation is high temperature, short circuit, electrical discharge, mechanical
common, and the dq0 model cannot work properly. In fact, the stress, magnetic force and insulation deterioration [27].
proposed hybrid method makes a compromise between the Generally, stator winding fault diagnosis methods use magnetic
accuracy and speed, which can cover the entire fault region flux, current and voltage or a combination of them.
from low to high severity. Although the phase‐to‐phase and phase‐to‐ground faults
The phase domain model (PDM) is more complicated than are more severe than the inter‐turn winding fault, they can be
the dq0 model, but it does not require the sinusoidal winding detected easier by protection relays in a fraction of second.
distribution [8]. Besides, the dq0 model has constant induc- Detection of the inter‐turn fault is difficult at the incipient
tance matrices [13]. The PDM uses voltage and flux‐linkage stage because it does not significantly influence the terminal
equations in a reference phase to derive the fault model in the currents. Therefore, condition monitoring is essential for inter‐
SGs [14–16]. turn faults to diagnose the fault at starting point and prevent
The PDM can deal with the saturation conditions, as the following problems. There is a widespread belief that many
described in Ref. [17]. Some have presented a modified phase‐to‐ground or phase‐to‐phase faults started as undetected
winding function approach (MWFA), which combines the TTF, which grew and propagated until disaster finally occurs
WFA and PDM model for detecting the phase to ground faults [28]. Here, the stator winding fault detection methods are
[9,18,19]. The MWFA, as an extended version of the WFA, is briefly reviewed in transformers and induction motors, and
used in a non‐uniform air‐gap machine [20]. The dynamic then the methods are extended to the SGs. In the incipient
eccentricity (DE) fault causes the asymmetrical air‐gap; there- stage, the insulation failure is not dangerous, but if it is not
fore, the SGs must be modelled by MWFA [20–22]. detected earlier, it may lead to a severe fault. In fact, the TTF in
The mutual inductance of two arbitrary windings x and y in the winding induces a high‐induced current, which flows in the
respect to the winding distribution nx and ny is as follows [23]: shorted loops and causes the winding failure [29]. Detecting
the fault at an early stage decreases the machine damage, and
� � the machine can be put back into service by rewinding the
〈Pnx 〉 〈Pny 〉
Lyx ¼ μ0 rl 2π〈nx ny 〉 2π ð4Þ stator [30].
〈P〉 Differential relays cannot detect the TTF due to instru-
mental transformers saturation, which reduces its sensitivity.
where P is the permeance distribution of the air‐gap. Hence, other methods such as negative or zero sequence‐based
394
- MOSTAFAEI AND FAIZ

F I G U R E 2 TTSC fault signature of a real

machine, developed WRSG model using hybrid
approach and dq0 model at (a) 5% (b) 10% (c) 15%
(d) 20% of stator winding short circuit [13]

methods should be used for the TTF diagnosis, beside the pattern. As shown in Figure 3, the stator‐winding fault reduces
differential relay. To detect the stator‐winding fault in the SGs, the magnetic flux near the fault region, and the stray flux in-
various quantities such as current, voltage, and magnetic flux creases adversely. Therefore, flux monitoring can be an effec-
can be utilised. The following techniques can be applied for tive choice for fault diagnosis. Generally, sensors through
stator fault detection. measuring the leakage or air‐gap flux in different parts monitor
the magnetic flux. For flux monitoring in electrical machines,
some sensor such as fluxgate magnetometer, Hall Effect sen-
3.1 | Magnetic flux sors, and the search coil (SC) magnetometer can be used [31].
Each sensor has its frequency range, and the search coil is the
Most electrical and mechanical faults in electrical machines simplest sensors for magnetic flux measurement with high‐
including stator‐winding fault change the magnetic flux frequency range.
MOSTAFAEI AND FAIZ
- 395

TA B L E 1 A brief review of various SG modelling methods

Method Applications Advantages Disadvantages

dq0 Symmetrical systems Simple Unable in the fault condition.
Steady‐state conditions Short computational time Sinusoidal winding distribution.
Minor stator winding faults Unable to take into account the effect of all space
harmonics.
Unable to model the saturation.

WFA Asymmetrical systems Accurate Long computational time.

Severe stator winding faults Sensitive
Eccentricity faults For uniform air‐gap

Hybrid Symmetrical and asymmetrical systems More accurate than WFA Moderate computational time.
Minor and severe stator winding faults Simpler than WFA

PDM Internal phase and ground fault in stator Accuracy More complex than dq0.
Transient conditions Independency of winding distribution High computational time.
Considering saturation effect

MWFA Extended WAF Suitable for non‐uniform air‐ gap Complexity.

Phase‐to‐ground fault Predict transient and steady‐ state performance. High computational time.
Rotor dynamic eccentricity Independent of winding distribution.
Mixed eccentricity Considering the saturation
All kinds of faults Considering the effect of all spatial and time
harmonics.

FEM Electrical machine optimisation High accuracy Time consuming.

Validation of modelling methods Modelling of all conditions Need powerful computers.
Checking electrical parameters in various Complexity.
conditions

F I G U R E 3 Magnetic field density distribution in a synchronous generator: (a) healthy condition, (b) inter‐turn short circuit in stator winding [46]

The magnetic flux sensors are used for TTF detection in space and time, but in faulty case, air‐gap MMF is distorted as
transformers by measuring the leakage flux [32,33], and flux‐ shown in Figure 3 [46].
linkage [34–37] with high sensitivity. Besides, the air‐gap and The search coil can measure total harmonic distortion
stray magnetic flux monitoring are enthusiastic methods in (THD) of the air‐gap magnetic flux; the THD of the flux in-
induction motors [38–44], which are used for stator TTF creases the fault in the stator winding; the saturation and
detection. Four different search coils mounted in various parts loading conditions can be also taken into account. In addition,
of induction motor have been introduced in Ref. [39] to the fault location is identified using the ANN and wavelet
measure the magnetic flux. transform [46].
The MMF near the shorted turns has been decreased in the Although the magnetic flux sensor is mostly inexpensive,
TTF case [45]. In the healthy case, the MMF is sinusoidal in reliable, and sensitive, its main drawback is practical limitations
396
- MOSTAFAEI AND FAIZ

during installation [46–48]. In fact, the stray flux sensors are In faulty case ZSV ¼ 0, but due to the intrinsic asymmetry
non‐invasive and inside/in the vicinity of the SG stator of the electrical machine, the ZSV is generally non‐zero even in
housing, but the stray flux is so small even in the faulty case the healthy condition. Hence, a threshold should be consid-
and decreases the condition monitoring sensitivity. ered. When the ZSV is higher than the threshold (k), a TTF
In the large SGs, the stator winding fault is mostly severe, has been occurred in the winding.
and the TTF is more probable in small ones. Therefore, con- Time‐harmonic of the stator current and rotor current can
ventional protection devices can detect the stator fault in a be used for stator winding fault detection. The interaction
large SG without any problems, but the TTF in a small SG can between the electrical quantities at the supply frequency and
be diagnosed by novel methods like air‐gap flux or stray flux different space harmonics produce additional time‐harmonic
monitoring. Although the mentioned methods are sensitive components in the stator and rotor currents [39]. The TTF of
and accurate, the current/voltage‐based methods are more the stator winding produces some even harmonics in the rotor
popular because they do not need additional sensors. field current, which is noticeable to define a fault detection
method. In this case, some even harmonics increase in the
faulty case [52] and this validates the simulation results [53,54].
3.2 | Current/voltage signature For instance, when the TTF occurs in the stator winding, the
eighth‐order harmonic is generated in the rotor of a 4‐pole
Current or voltage is the most common quantity for fault SG; however, it is noted that even harmonics are also pro-
diagnosis, because they are accessible through a current duced in the unbalanced supply conditions in synchronous
transformer (CT), potential transformer (PT), or probes. Un- motors [55]. The third harmonic of the rotor current can be
like the magnetic flux‐based methods, the current/voltage‐ used for fault diagnosis, but it has no acceptable sensitivity.
based ones do not need additional sensors, and they are more Therefore, by a search coil in the rotor and measuring its
reliable. For the SGs with multi‐branch winding, comparison 90 Hz component (for fundamental frequency of 60 Hz) can
of the different branches currents is common, which is known significantly improve the protection without any correction
as differential TTF protection. In. factor [56].
this method, currents in different branches of stator In transient and steady state modes, for stator TTF in a
winding are compared. In healthy condition, the difference of brushless SG, the third harmonic of the positive sequence
the currents is almost zero, but for TTF in the SG, the dif- stator voltage is selected as an indicator [57]. Most stator TTF
ference increases, and differential current circulates between detection methods in induction motors can be extendedto the
the windings [49]. SGs, because of the similarity of the stator in both machines.
Another approach is the residual winding voltage appli- Therefore, the extended Park's vector approach (EPVA) has
cable in all SGs. To measure the residual voltage, an additional been presented for the SGs and induction motors in Ref. [58],
open delta PT is used. In the healthy generator, phase voltage which uses the spectral analysis of the decomposed current by
of the winding is almost the same, therefore, the sum of Park's transform.
fundamental voltage harmonics in secondary is zero, while in The second harmonic in the EPVA spectrum is sensitive to
the TTF, the faulty winding voltage drops, and the residual the stator TTF and it is a good indication for fault diagnosis.
voltage is no longer zero. In this method, the sum of other However, the mentioned method of performance is ambitious
harmonics such as third harmonics is non‐zero, and all must be in the unbalanced supply. Negative sequence impedance is also
removed to prevent unwanted tripping [50]. In the above‐ another indicator to detect the TTF in the stator winding of
mentioned detection method, an additional PT is necessary, it electrical machines. The proposed method has an appropriate
may cause some installation problems, and in some cases, it is performance against the TTF, but it is not so in minor faults
not economical. To solve this problem (required additional and unbalanced supply [59,60]. A modified negative sequence
PT), a novel protection scheme shown in Figure 4 has been impedance method uses a voltage mismatch detector to solve
proposed which calculates zero‐sequence voltage (ZSV) the mentioned problems [61]. Negative sequence voltage
instead of measuring it [51]. In this method, the ZSV is (NSV) can be also used for this purpose as introduced in the
calculated as follows. large SGs [62].
Recently, slope measurement for condition monitoring
has become popular. The slope of a Lissajous curve has
V AN þ V BN þ V CN ¼ 3V 0N ð5Þ been used in Ref. [63] for short‐circuit fault diagnosis in
turbo‐generators. In Figure 5, a turbo‐generator with a
where VON is the ZSV and VAN, VBN and VCN are as short‐circuit fault in phase has been connected to the power
follows: grid A. According to the figure, the fault current is written
as follows:
8
< V AN ¼ V AG V NG
VA grid − V A Gen ðnÞ
V ¼ V BG V NG ð6Þ IF A ¼
: BN ð7Þ
V CN ¼ V CG V NG Zgrid þ ZGen ðnÞ − jX AB2 ðnÞ − jX AC2 ðnÞ − j|γ|X A−AF
MOSTAFAEI AND FAIZ
- 397

F I G U R E 4 Proposed protection scheme for

TTF detection using zero‐sequence voltage [51]

�
xðtÞ ¼ Ax sinðωx tÞ �
ð8Þ
yðtÞ ¼ Ay sin ωy t þ φ

Therefore, by solving (8):

�� � � � �
ωy xðtÞ
y t ¼ Ay sin � sin 1 þφ ð9Þ
ωx Ax

where x(t) is the current sample in a specific time and y(t) is

F I G U R E 5 Schematic model of a generator with an internal fault in the current sample in the previous cycle. Since the frequency
phase A change is negligible over a short time, by assuming ωx¼ωy, (9)
is simplified as follows:
where γ is the phase difference between VA_grid, VA_Gen
and β is the phase difference between the healthy and faulty �� � � � �
xðtÞ
current. y t ¼ Ay sin sin 1 þφ ð10Þ
The fault currents in three phases can be classified into two Ax
categories according to β. When β < 90°, the fault current is
called class I and else is called class II. In this case, the current The slope of the Lissajous curve is an appropriate factor to
in the faulty phase (IFA) has a reverse direction, and when determine the current class, where the positive slope means the
β > 90°, it is the class II. class I and negative slope means the class II. The sign of Δx/
To determine the current class, the Lissajous curve is used Δy determines the sign of the large axis slope of the ellipse in
for the phase difference between the two time‐dependent this curve.
signals. In this curve, two time‐dependent signals are plotted To distinguish the internal fault from other disturbances and
versus each other. For two sinusoidal signals with amplitude A, faults, the difference between the SG phase currents is estimated.
their equations can be written as follows: In the internal fault, the difference between the phase currents of
398
- MOSTAFAEI AND FAIZ

the two healthy phases is negligible, but the difference between 4 | ROTOR FAULTS
the currents of the faulty phase and two healthy phases is large. If
the difference of instantaneous current of phase A and phase B Owing to the rotor rotation, both mechanical and electrical
in the faulty case is ΔFAB and in healthy case, just in the previous faults can take place in the rotors. Detection of the electrical
cycle is ΔHAB, the ratio ΔFAB/ΔHAB can help to distinguish failure in the rotor, that is the SG short‐circuit.
the internal fault from other disturbances and faults. Figure 6 winding is quite different from the stator, because the DC
shows the flow‐chart of fault diagnosis methods which can current flows in the rotor winding and the faults cannot be
detect the fault in different phases. detected by sinusoidal waveform‐based approaches. To
Table 2 shows a brief review of the various SG winding discriminate rotor electrical and mechanical faults, symmetrical
fault detection methods. component monitoring can be considered as an appropriate
Frequency response analysis (FRA) is another well‐known method [66]. In the proposed method, the current spectrum
technique in detecting the various defects such as winding short and voltage have been analysed to discriminate against the
circuit, core deformation, and winding movement in power rotor faults. The results show that the rotor TTF increases the
transformers and it is now a standard method. It uses the equiv- positive, negative, and zero sequences of the rotational fre-
alent impedance for fault diagnosis in the frequency domain [64]. quency of the first right sideband, while the mechanical fault
Figure 7 presents the equivalent scheme for the rotor winding. increases only the positive sequence of the rotational frequency
A novel approach uses the FRA in a rotating machine, [66]. Furthermore, vibration can be used to diagnose both
especially SGs with static excitation [65]. It is able to detect winding and eccentricity faults.
both ground fault and TTF. Moreover, FRA can be applied
while the rotor is turning at full speed with no excitation. [65].
The disadvantages of the FRA are the difficulties in inter- 4.1 | Rotor winding faults
preting the results and unavailability of the healthy results for
on‐service machines. The TTF in the rotor winding is caused by the insulation
Despite the transformer, the rotor winding of the SG failure due to ageing, thermal, and mechanical stresses [67].
(similar to the secondary winding of transformer) is rotational, As a result of short‐circuit turns, the total ampere‐turn is
and its position changes the FRA results. However, Figure 8 reduced in the affected pole, and consequently, the air‐gap
shows that the differences between the various rotor positions magnetic flux density distribution becomes asymmetrical [68].
are negligible. Magnetic flux and vibration are two major quantities,
measured by sensors, which can be used for the rotor
winding fault diagnosis. For this purpose, magnetic flux
probes were used in the 1970s for the first time and devel-
oped up to now [69]. Some methods focus on the stray flux
or leakage flux measured by stator wedge‐mounted sensors;
however, its magnitude is small and not recommended for
condition monitoring [70,71]. A new flux probe introduced in
Ref. [72] has been mounted on the stator core tooth to
measure the main magnetic flux passing through the core
tooth. These probes can be removed easier than the con-
ventional flux probes [71].
A non‐invasive method has been proposed in Refs. [73,74],
which uses the external search coil to sense the stray magnetic
flux in the SGs. Analysing both the stray flux and frame vi-
brations can detect the rotor TTF in SGs.
The search coil near the frame is firstly applied to in-
duction motors for fault detection [75]. Similar search coil
can be also used in the SGs. The frequencies in the
magnetic flux spectrum and signal spectrum of vibration
define the healthy and faulty SGs. In the faulty condition,
the magnitude of some frequencies increases compared with
the healthy case. It is noted that although there are
amplitude changes every 25 Hz in the vibration spectrum,
the most significant modification appears at low frequencies,
especially at 225 Hz, but the healthy frequencies remain
unchanged [73].
As shown in Figure 9, to diagnose the rotor‐winding fault in
turbo‐ and hydro‐SGs, the flux probe is installed in the stator slot.
F I G U R E 6 Fault diagnosis flow‐chart The flux probe is sensitive to the of the air‐gap radial flux change.
MOSTAFAEI AND FAIZ
- 399

TA B L E 2 Various SG stator and stator fault detection methods

Method Advantages Disadvantages

Air‐gap flux High sensitivity and accuracy Invasive
High reliability Installed inside the stator slots
Wide range measurement May change the machine design
Useful for various kinds of faults

Stray flux Non‐invasive Low sensitivity and accuracy.

Installed in or in vicinity of the machine housing
High range measurement
Useful for various kinds of faults.

Current/Voltage Accessibility and simple measuring Low sensitivity and accuracy

Using current/voltage transformers which is installed on SG Instruments errors
Non‐invasive
Reliable
Using various transform function i.e. Park's vector
transform or symmetrical component transform

FRA Detect both electrical and mechanical faults a Costly test.

Needs the healthy condition data.
Needs expert's opinion.

F I G U R E 7 Quadrupole representation of field winding of a F I G U R E 9 Position of flux probe around stator tooth near air‐gap [76]
synchronous machine during the FRA test

associated with the two opposite slots containing the faulted coil,
thus, the presence of shorted turns could be detected [76].
Discrete wavelet transform (DWT) is alternative way for
the rotor winding fault detection, which provides a set of
decomposed signals in independent frequency bands, including
the independent dynamic information due to the orthogonality
of wavelet function [77]. Herein, the.
rotor current in cylindrical SG is used as DWT input, and
the results show that the fifth detail level gives useful infor-
mation about the rotor winding fault.

4.2 | Rotor eccentricity fault

Rotor eccentricity is a mechanical fault in rotational ma-

chines related to the rotor. Generally, the air gap is
F I G U R E 8 Standstill FRA of stator winding at different rotor positions
(0–360 electrical degrees) [65]
distributed homogeneously, but the rotor eccentricity is
defined as asymmetric air‐gap that exists between the stator
and rotor [78].
As each rotor slot passes the flux probe, a difference in the Figure 10 shows the static eccentricities (SE) and dynamic
induced voltage waveform in a search coil caused by magnetic eccentricity (DE) faults. In the case of the SE fault, the posi-
poles is detectable. An inter‐turn fault in a coil reduces the peaks tion of the minimal radial air‐gap length is fixed in space. The
400
- MOSTAFAEI AND FAIZ

DE occurs when the centre of the rotor is not at the centre of the damper bars damp the transient power and torque oscil-
the rotation, and the position of minimum air‐gap rotates with lations. The rotor speed oscillates around the synchronous
the rotor [79]. These faults are critical for electrical machines speed in transition, and this oscillation should be damped by
and must be detected in the early stage. Different methods the damper bars [91].
have been presented for the eccentricity fault detection in in- The damper bars in a steady‐state mode are quite different
duction motors but less in SGs. Therefore, the detection from the fault condition. Hence, the damper bar fault detection
methods for induction motors are presented firstly, and then methods in SGs is proposed here. The correlation of the
some proposed approaches in SGs are discussed. broken bars fault in induction motors and the damper bars in
The eccentricity fault can be detected in large induction SGs has been considered in Ref. [92].
motors using line current and motor frame vibration [80]. Although the current and thermal stress are considered for
Besides, based on a new theoretical analysis [81] has presented designing the damper bars, failure is possible due to the defi-
to diagnose SE and DE faults simultaneously. Search coil cient construction of damper cage [93] or frequent and hard
installation in stator slots is another approach for detecting the duty cycle [94,95]. This kind of fault is not common in SGs,
SE fault [82], which is not practical because the sensors must but it occurs in some SGs and leads to severe damages [94].
be fixed in the machine during the manufacturing process [21]. Due to the similarity of the damper bars and induction
The current harmonic analysis has been presented in Refs. motors cage bars and the probability of the broken damper
[22,23] for eccentricity fault diagnosis. The volunteer harmonic bars fault, proposing the novel detection methods were not
components are the 17th and 19th. However, the mentioned enthusiastic for researchers, and a few methods has been
harmonics depend on the structure and geometry of the SGs. presented for fault detection. An online diagnosis method for
Moreover, these harmonics are similar to power network damper bars breakage presented in Ref. [94] uses a flux probe
harmonics [83]. Rotor current double‐frequency ripple can be to measure the air‐gap flux from starting to the rated speed.
used for the SE fault detection [84], where the impact of the Another method for diagnosing the broken damper bars in
regulator has been ignored. Ref. [96] is the Hilbert–Huang Transform (HHT) based on
The mixed eccentricity (ME) fault may occur in electrical Empirical Mode Decomposition (EMD) method reported in
machines, where both symmetrical and rotor rotation axes are Refs. [97,98].
displaced with respect to the stator rotation axis. In fact, all Temperature is a key parameter for broken damper bars
stator, rotor, and rotational symmetrical axes are displaced with fault detection, and temperature sensors are mounted on the
respect to each other [26]. The SE, DE, and ME faults in SGs rotor to measure the temperature rise during the fault. Resis-
are discussed in Ref. [26]. The ME fault diagnosis method has tance temperature detector (RTD) is one of the sensors used to
been also introduced for permanent magnet synchronous sense the temperature [99]. Besides, the temperature sensors
motors in Refs. [85,86]. have been reported in Ref. [100]. Therefore, temperature
monitoring can be an easy and economical method for broken
damper bar fault detection, but some believe that the mounted
5 | BROKEN DAMPER BARS FAULTS RTD is an invasive method.

Damper bars are used in large SGs. These bars are located
axially into the pole face slots, and two ends are connected. At 6 | OTHER WELL‐KNOWN FAULT
starting, the rotor speed and synchronous speed differ, and DETECTION METHODS
currents are induced in the damper bars and developed the
torque, which called the asynchronous operation of the SG Some techniques such as artificial intelligence (AI) approach
[78,87]. Damper bars in SG are used to counteract an asyn- help to improve diagnosis methods and their sensitivities. The
chronous air‐gap flux causing by electrical and mechanical AI application in electrical machines and drives are proposed in
transients [88,89]. Moreover, they are used in direct‐online Refs. [101,102] focusing on the stator winding fault diagnosis
applications to bring the SG to synchronous speed [90]. In fact, of induction motors [102].
In the AI‐based systems, several quantities such as stator
currents and voltages, magnetic fluxes, and frame vibration are
utilised as input signals [39]. Generally, an expert system, an
artificial neural network (ANN), a fuzzy neural network [103],
and their combinations are the well‐known methods in AI.
ANN has been widely studied during the last 2 decades and
successfully applied to dynamic system modelling [104,105] as
well as fault diagnosis [106–110].
Vibration monitoring has a long history in condition
monitoring and used in various electrical equipment. Vibration
is used in Ref. [111] for stator winding fault in.
induction motors. For SGs, the vibration is employed for
F I G U R E 1 0 Various kinds of eccentricity in electrical machines various mechanical faults detection [112], especially bearing
MOSTAFAEI AND FAIZ
- 401

fault [113]. Generally, vibration monitoring is an expensive There are two key advantages of the presented algorithm
method because it requires some costly vibration sensors [113]. for realising the monitoring system. The first advantage is the
Recently, many have focused on thermal condition moni- integration of the two methods for the measurement of
toring using infrared thermography. This apparatus identifies magnetic flux. The applied measuring procedure includes a
temperatures and hotspots of different locations. Thermog- comparative analysis of the results of magnetic flux measure-
raphy is used in induction motors [114] and SGs [115] for ments with the results obtained by the system for measuring
condition monitoring. The high cost of an infrared device is mechanical vibrations to approach the fault detection
the main disadvantage of the thermal monitoring methods. comprehensively. The second advantage is the mobility of the
The above‐mentioned methods can be combined to pre- monitoring system and its application without interrupting the
sent a comprehensive approach which has the advantages of generators operation (the case when the leakage stator flux is
the both for fault diagnosis. One of these methods uses air‐gap measured or when the system is connected with the previously
flux, stray flux, and vibration to detect the rotor‐winding fault built‐in sensors in the air gap of the generator) [115].
in hydro‐generator precisely [116]. Capacitive sensors are installed on the surface of the stator
[117], while the inductive sensors are fixed in the ventilation
ducts of the stator [48]. Figure 11 shows the proposed algo-
rithm, which can detect the rotor‐winding fault of hydro‐
generator. As shown, mounted sensors must measure the air‐
gap and stray flux firstly, and then the signal processing method
is applied to determine the THD of stray flux. Any change in
magnetic flux may be the sign of a fault in the SG. Therefore,
vibration measurement can complete the fault detection pro-
cess. The unbalanced magnetic flux and increased vibration
show the rotor‐winding fault, and the SG must be shut down
immediately.
Table 3 presents appropriate comparisons between the
measurement methods.

7 | CONCLUSION

The current‐ and voltage‐based methods are generally more

useful than the flux based ones, because these parameters can be
measured as easy as possible by instrumental transformers. For
this purpose, analysing harmonics in various conditions and
using signal‐processing techniques such as Wavelet and Fourier
transform. Nowadays, vibration analysis is used for various fault
detection in electrical machine and this processed vibration data
can present much information about the SG condition especially
under mechanical faults. Apart from the mentioned parameters,
frequency spectrum of SG offers invaluable data about condi-
tion monitoring. In transformers, frequency based methods are
popular and it can be used in SGs.
All proposed methods can be combined by AI techniques
(as a hybrid method) to increase their sensitivities. Moreover,
AI can help the researchers to reduce the errors and prevent
some catastrophic problems that can be occurred because of
FIGURE 11 Fault detection algorithm flow chart CT, VT or other sensors mal‐operation.

TA B L E 3 Comparative analysis of measurement methods [116]

Measurement methods Mounting of sensors Shorted turns detection Air‐gap asymmetry detection
Main magnetic flux in air‐gap Time‐consuming Yes (from one to more shorted turns) Yes

Leakage magnetic flux of the stator Not time‐consuming Yes (from one to more shorted turns) ‐

Vibrations of upper and lower guide bearing Time‐consuming Yes (if the number of shorted turns is Yes
significant/low number of rotor poles)
402
- MOSTAFAEI AND FAIZ

OR CID 21. Toliyat, H.A., Al‐Nuaim, N.A.: Simulation and detection of dynamic
Mohsen Mostafaei https://orcid.org/0000-0001-9976-4956 airgap eccentricity in salient‐pole synchronous machines. IEEE Trans.
Jawad Faiz https://orcid.org/0000-0003-0844-8523 Ind. Appl. 35(1), 86–93 (1999)
22. Tabatabaei, I., et al.: Modeling and simulation of a salient‐pole syn-
chronous generator with dynamic eccentricity using modified winding
REFE RE NCE S function theory. IEEE Trans. Magn. 40(3), 1550–1555 (2004)
1. Seinsch, H.O.: Monitoring und diagnose elektrischermaschinen und. In: 23. Faiz, J., Tabatabaei, I.: Extension of winding function theory for non‐
Stand der forschung, entwicklungstendenzen, Frankfuft am Main. uniform airgap of electric machinery. IEEE Trans. Magn. 38(6),
VDE‐Haus (2001) 3654–3657 (2002)
2. Tavner, P.J.: Review of condition monitoring of rotating electrical 24. Nandi, S.: A detailed model of induction machines with saturation
machines. IET Electr. Power Appl. 2(4), 215–247 (2008) extendable for fault analysis. IEEE Trans. Ind. Appl. 40(5), 1302–1309
3. Bellini, A., et al.: Advanced in diagnosis techniques for induc- (2004)
tion machines. IEEE Trans. Ind. Electron. 55(12), 4109–4126 25. Faiz, J., Ebrahimi, B.M., Valavi, M.: Mixed eccentricity fault diagnosis in
(2008) salient‐pole synchronous generator using Modified winding function
4. Aroquiadassou, G., et al.: A simple circuit‐oriented model for predicting method. Prog. Electromagn. Res. 11, 155–172 (2009)
Six‐Phase induction machine performances. In: 32nd Annual Confer- 26. Sadeghi, I., Ehya, H., Faiz, J.: Analytic method for eccentricity fault
ence on IEEE Industrial Electronics (IECON), pp. 1441–1446. Paris diagnosis in salient‐pole synchronous generators, Intl Aegean Confer-
(2006) ence on Electrical Machines and Power Electronics (ACEMP). In:
5. Krause, P., et al.: Analysis of electric machinery and drive systems, 3rd Optimisation of electrical and Electronic equipment, pp. 261–267.
ed. John Wiley & Sons, Hoboken, New Jersey (2013). https://doi.org/ OPTIM (2017)
10.1002/9781118524336 27. Tavner, P.J., Penman, J.: Condition monitoring of electrical machines.
6. Sivakumar, N., et al.: Dynamic model of brushless synchronous John Wiley & Sons, Hoboken, New Jersey (1987)
generator with turn‐to‐turn short circuit fault for condition monitoring. 28. Kliman, G.B., et al.: A new approach to online turn fault detection in
Int. Conf. ECCE Asia, 258–263 (2013) AC motors. In:35th IAS Annual Meeting Industry Applications
7. Dehkordi, A.B., et al.: Development and validation of a comprehensive Conference. 1, 687–693 (1996)
synchronous machine model for a real‐time environment. IEEE Trans. 29. Crawford, D.E.: A mechanism of motor failures. In: 12th IEEE elec-
Energy Conv. 25(1), 34–48 (2010) trical/Electronics and insulation Conference (EIC), pp. 126–129 (1975)
8. Wang, L., Jatskevich, J., Dommel, H.W.: Re‐examination of synchro- 30. Tallam, R.M., et al.: A survey of methods for detection of stator‐related
nous machine modeling techniques for electromagnetic transient sim- faults in induction machines. IEEE Trans. Ind. Appl. 43(4), 920–933
ulations. IEEE Trans. Power Syst. 22(3), 1221–1230 (2007) (2007)
9. Kumar, P.S., et al.: Modified winding function approach to stator fault 31. Lenz, J., Edelstein, A.S.: Magnetic sensors and their applications. IEEE
modelling of synchronous generator. In: 12th IEEE International Sens. J. 6(3), 631–649 (2006)
Conference of Control and Automation (ICCA), 161–166 (2016) 32. Cabanas, M.F., et al.: A new online method based on leakage flux
10. Queval, L., Ohsaki, H.: Non‐linear abc‐model for electrical machines analysis for the early detection and location of insulating failures in
using N‐D lookup tables. IEEE Trans. Energy Conv. 30(1), 316–322 power transformers: application to remote condition monitoring. IEEE
(2015) Trans. Power Deliv. 22(3), 1591–1602 (2007)
11. Tu, X., Dessaint, L.A., Kahel, M.: A new model of synchronous 33. Haghjoo, F., Mostafaei, M., Mohammadi, H.: A new leakage flux‐based
machine internal faults based on winding distribution. IEEE Trans. Ind. technique for turn‐to‐turn fault protection and faulty region identifi-
Electron. 53(6), 1818–1828 (2008) cation in transformers. IEEE Trans. Power Deliv. 33(2), 671–679
12. Babaei, M., et. al.: A detailed analytical model of a salient‐pole syn- (2018)
chronous generator under dynamic eccentricity fault. IEEE Trans. 34. Mostafaei, M., Haghjoo, F.: Flux‐based turn‐to‐turn fault protection for
Magn. 47(4), 764–771 (2011) power transformers. IET Gener. Transm. Distrib. 10(5), 1154–63 (2016)
13. Nadarajan, S., et al.: Hybrid model for wound‐rotor synchronous 35. Venikar, P.A., et al.: Search coil based online diagnostics of transformer
generator to detect and diagnose turn‐to‐turn short‐circuit fault in internal faults. IEEE Trans. Power Deliv. 32(6), 2520–2529 (2017)
stator windings. IEEE Trans. Ind. Electron. 62(3), 1888–1900 (2015) 36. Kang, Y.C., et al.: Protection, faulted phase and winding identification
14. Reichmeider, P., et al.: Internal faults in synchronous machines—Part I: for the three‐winding transformer using the increments of flux linkages.
the machine model. IEEE Trans. Energy Conv. 15(4), 376–379 (2000) IET Gener. Transm. Distrib. 4(9), 1060–1068 (2010)
15. Subramaniam, P., Malik, O.P.: Digital simulation of a synchronous 37. Mostafaei, et al.: Turn‐to‐turn fault monitoring methods in electrical
generator in direct‐phase quantities. Proc. Inst. Electr. Eng. 118(1), power transformers—state of the art. Int. Trans. Electr. Energy Syst.
153–160 (1971) 28(12), 1–22 (2018). https://doi.org/10.1002/etep.2644
16. Muthumuni, D.D.S., Mclaren, P.G., Dirks, E.: Internal fault simulation 38. Dehghan, H., Haghjoo, F., Cruz, S.M.: A flux‐based differential tech-
in synchronous machines. In: Proc. Canadian conference electrical and nique for turn‐to‐turn fault detection and defective region identification
computer engineering, pp. 1202–1206.Canada, Halifax (2000) in line‐connected and inverter‐fed induction motors. IEEE Trans.
17. Marti, J.R., Louie, K.W.: A phase‐domain synchronous generator model Energy Conv. 33(4), 1876–1885 (2018)
including saturation effects. IEEE Trans. Power Syst. 12(1), 222–229 39. Negrea, M.D.: Electromagnetic flux monitoring for detecting faults in
(1997) electrical machines. Dissertation, Helsinki University of Technology,
18. Hemmati, S., et al.: Modelling and experimental validation of internal Espoo, Finland (2007)
short‐circuit fault in salient‐pole synchronous machines using numerical 40. Penman, J., et al.: Detection and location of inter‐turn short circuits in
gap function including stator and rotor core saturation. IET Electr. the stator windings of operating motors. IEEE Trans. Energy Conv.
Power Appl. 7(5), 391–399 (2013) 9(4), 652–658 (1994)
19. Jiang, H., et al.: A new approach to synchronous generator internal fault 41. Toni, K., Slobodan, M., Aleksandar, B.: Detection of turn to turn faults
simulation using combined winding function theory and direct phase in stator winding with axial magnetic flux in induction motors. In:
quantities. In: 9th International Conference of Electrical Machines and IEEE International Electric Machines & Drives Conference (IEMDC).
Drives, pp. 105–111 (1999) Vol. 1, pp. 826–829 (2007)
20. Al‐Nuaim, N.A., Toliyat, H.A.: A novel method for modeling dynamic 42. Mirzaeva, G., Saad, K.I.: Advanced diagnosis of stator turn‐to‐turn
airgap eccentricity in synchronous machines based on modified winding faults and static eccentricity in induction motors based on internal flux
function theory. IEEE Trans. Energy Conv. 13(2), 156–162 (1998) measurement. IEEE Trans. Ind. Appl. 54(4), 3961–3970 (2018)
MOSTAFAEI AND FAIZ
- 403

43. Bui, V.P., et al.: Non‐invasive fault monitoring of electrical machines by 65. Blánquez, F.R., et al.: Field‐winding fault detection in synchronous
solving the steady‐state magnetic inverse problem. IEEE Trans. Magn. machines with static excitation through frequency response analysis. Int.
44(6), 1050–1053 (2008) J. Electr. Power Energy Syst. 1(73), 229–239 (2015)
44. Henao, H., Demian, C., Capolino, G.A.: A frequency‐domain detection 66. Salomon, C.P., et al.: Discrimination of synchronous machines rotor
of stator winding faults in induction machines using an external flux faults in electrical signature analysis based on symmetrical components.
sensor. IEEE Trans. Ind. Appl. 39(5), 1272–1279 (2003) IEEE Trans. Ind. Appl. 53(3), 3146–3155 (2017)
45. Luo, X., et al.: Multiple coupled circuit modeling of induction machines. 67. Stone, G.C., et al.: Using magnetic flux monitoring to detect synchro-
IEEE Trans. Ind. Appl. 31(2), 311–318 (1995) nous machine rotor winding shorts. Proc. EEE PCIC, 1–7 (2011)
46. Yaghobi, H., Ansari, K., Rajabi‐Mashhadi, H.: Stator turn‐to‐turn fault 68. Valavi, M., Jørstad, K.G., Nysveen, A.: Electromagnetic analysis and
detection of synchronous generator using total harmonic distortion electrical signature‐based detection of rotor inter‐turn faults in salient‐
(THD) analysing of magnetic flux linkage. Iran. J. Sci. Technol. Trans. pole synchronous machine. IEEE Trans. Magn. 54(9), 1–9 (2018)
Electr. Eng. 37(2), 161–182 (2014) 69. Albright, D.R.: Inter turn short‐circuit detector for turbine‐generator
47. Yaghobi, H.: Impact of static synchronous compensator on flux‐based rotor windings. IEEE Trans. Power Appl. Syst, PAS‐90(2), 478–483
synchronous generator loss of excitation protection. IET Gener, (1971)
Transm. Distrib. 9(9), 874–83 (2015) 70. Jenkins, M.P., Wallis, D.J.: Rotor shorted turns: description and utility
48. Xuan, M.T., et al.: A novel airgap monitoring system for large low speed evaluation of a continuous online monitor. The EPRI Predictive
hydro‐generators. In: IEEE Power Engineering Society General Maintenance Refurbishment Conf, San Francisco (1993)
Meeting, pp. 1–8 (2006) 71. Whittle, A.: Continuous rotor flux monitoring. EPRI Gener. Workshop,
49. Reimert, D.: Protective relaying for power generation systems. Taylor & Phoenix (2007)
Francis Group, Milton Park, Oxfordshire (2006) 72. Šašić, M., Lloyd, B., Elez, A.: Finite element analysis of turbine
50. IEEE Guide for AC Generator Protection: In: IEEE Std C37.102‐2006 generator rotor winding shorted turns. IEEE Trans. Energy Conv.
(Revision of IEEE Std C37.102‐1995), pp. 1–177. IEEE, New York 27(4), 930–937 (2012)
City, US (2006). https://doi.org/10.1109/IEEESTD.2006.320495 73. Cuevas, M., et al.: Non‐invasive detection of rotor short‐circuit fault in
51. Platero CA, et al., Scheme for synchronous machine stator turn‐to‐turn synchronous machines by analysis of stray magnetic field and frame
protection. IET Electr. Power Appl. 14(4), 716–22 (2019) vibrations. IEEE Trans. Magn. (7), 1–4 (2016)
52. Toliyat, H.A., et. al.: Electric machines: modelling, condition moni- 74. Cuevas, M., et al.: Non‐invasive detection of winding short‐circuit faults
toring, and fault diagnosis, 1st ed. CRC Press, Boca Raton, New York in salient‐pole synchronous machine. In: International Electric
(2012) Machines and Drives Conference (IEMDC), pp. 1–7 (2017)
53. Rabbi, A.: Detection of stator inter‐turn fault of synchronous machine 75. Frosini, L., et al.: Development of a leakage flux measurement system
by rotor current analysis: a simulation approach. Master thesis. KTH for condition monitoring of electrical drives. In: Symposium on
Royal Institute of Technology, Stockholm, Sweden (2016) Diagnostics for Electric Machines Power Electronics & Drives
54. Rodriguez, et al.: Field current signature analysis for fault detection in (SDEMPED). 2, 356–363 (2011)
synchronous motors. In: IEEE 10th International Symposium on 76. Fišer, R., et al.: Modeling, analysis and detection of rotor field winding
Diagnostics for Electrical Machines, Power Electronics and Drives, pp. faults in synchronous generators. In: XIX International Conference on
246–252.SDEMPED (2015) Electrical Machines, pp. 1–6.ICEM (2010)
55. Neti, P., Nandi, S.: Stator inter‐turn fault detection of synchronous 77. Kim, Y.J., et al.: Detection of shorted‐turns in the rotor winding of
machines using field current signature analysis. In: IEEE 41st IAS cylindrical synchronous generators using discrete Wavelet Transform.
Annual Meeting Conference Record of the Industry Applications In: Twenty‐First Annual IEEE Applied Power Electronics Conference
Conference. 5, pp. 2360–2367 (2006) and Exposition (2006)
56. Neti, P., Nandi, S.: Stator inter‐turn fault detection of synchronous 78. Vas, P.: Parameter estimation, condition monitoring, and diagnosis of
machines using field current and rotor search‐coil voltage signature electrical machines. In: Monographs in electrical and Electronic Engi-
analysis. IEEE Trans. Ind. Appl. 45(3), 911–920 (2009) neering. Oxford University Press, London (1993)
57. Sottile, J., Trutt, F.C., Leedy, A.W.: Condition monitoring of brushless 79. Nandi, S., Bharadwaj, R. M., Toliyat, H. A.: Performance analysis of a
three‐phase synchronous generators with stator winding or rotor circuit three‐phase induction motor under mixed eccentricity condition. IEEE
deterioration. In: Conf. Rec. IEEE IAS Annu. Meeting. Vol. 3, pp. 1587– Trans. Energy Conv. 17(3), 392–399 (2002)
1594 (2001) 80. Cameron, J.R., Thomson, W.T., Dow, A.B.: Vibration and current
58. Cruz, S.M., Cardoso, A.M.: Stator winding fault diagnosis in three‐phase monitoring for detecting airgap eccentricity in large induction motors.
synchronous and asynchronous motors, by the Extended Park's vector IEE Proc. 133(3), 155–163 (1986)
approach. IEEE Trans. Ind. Appl. 37(5), 1227–1233 (2001) 81. Dorrell, D.G., Thomson, W.T., Roach, S.: Analysis of airgap flux, cur-
59. Kohler, J.L., Sottile, J., Trutt, F.C.: Alternatives for assessing the electrical rent and vibration signals as a function of combination of static and
integrity of induction motors. IEEE Trans. Ind. Appl. 28(5), 1109–1117 dynamic airgap eccentricity in 3‐phase induction motors. IEEE Ind.
(1992) Appl. Conf. Rec., vol. 1, pp. 563–570.Orlando (1995)
60. Kohler, J.L., Sottile, J., Trutt, F.C.: Condition monitoring of stator 82. Verma, S.P., Natarajan, R.: Effect of eccentricity in induction motors.
windings in induction motors. I. Experimental investigation of the In: Proc. of International Conference on Electrical Machines, pp. 930–
effective negative‐sequence impedance detector. IEEE Trans. Ind. 933.Hungry, Budapest (1982)
Appl. 38(5), 1447–1453 (2002) 83. Liao, H.: Power system harmonic state estimation and observability
61. Sottile, J., Trutt, F.C., Kohler, J.L.: Condition monitoring of stator analysis via sparsity maximisation. IEEE Trans. Power Syst. 22(1),
windings in induction motors. II. Experimental investigation of voltage 15–23 (2007)
mismatch detectors. IEEE Trans. Ind. Appl. 38(5), 1454–1459 (2002) 84. Bruzzese, C., Joksimovic, G.: Harmonic signature of static eccentricities
62. Karthik, T., Raju, G.S.: New Criterion for stator inter turn fault in the stator voltages and in the rotor current of no load salient pole
detection of synchronous generator. In: 17th National Power Systems synchronous generators. IEEE Trans. Ind. Elec. 58(5), 1606–1624
Conference (2012) (2011)
63. Faiz, J., et al.: Diagnosis of inter‐turn fault in stator winding of turbo‐ 85. Ebrahimi, B.M., Faiz, J., Roshtkhari, M.J.: Static‐, dynamic‐, and mixed‐
generator. Int. Trans. Electr. Energy Syst. 29(12), e12132 (2019) eccentricity fault diagnoses in permanent‐magnet synchronous motors.
64. IEEE Guide for the application and interpretation of frequency IEEE Trans. Ind. Elec. 56(11), 4727–4739 (2009)
response analysis for oil‐immersed transformers, IEEE Std C57.149‐ 86. Ebrahimi, B.M., Faiz, J.: Diagnosis and performance analysis of three‐
2012, pp. 1–72. IEEE, New York City, US (2013). https://doi.org/ phase permanent magnet synchronous motors with static, dynamic and
10.1109/IEEESTD.2013.6475950 mixed eccentricity. IET Electr. Power Appl. 4(1), 53–66 (2010)
404
- MOSTAFAEI AND FAIZ

87. Mc‐Pherson, G., Laramore, R.D.: Electrical machines and transformers. 104. Haykin, S.: Neural networks: A comprehensive foundation, 2nd ed.
John Wiley & Sons, Hoboken, New Jersey (1990) Prentice‐Hall, New Jersey, US (1999)
88. Toliyat, H.A., Kliman, G.B.: Handbook of electric motor, 2nd ed. CRC 105. Gupta, M.M., Jin, L., Homma, N.: Static and dynamic neural networks
Press, Boca Raton, New York (2004) from fundamentals to advanced theory. John Wiley & Sons, New
89. Wallin, M.: Measurement and modeling of unbalanced magnetic pull in Jersey, USA (2003)
hydropower generators. Acta Universitatis Upsaliensis, Uppsala, Swe- 106. Kowalski, C.T.: Monitoring and fault diagnosis of induction motors
den (2013) using neural networks. Wroclaw University of Technology Academic
90. Holmgren, F.: Damper winding fault detection in synchronous ma- Press, Wroclaw (2005)
chines. Uppsala University, Uppsala (2016) 107. Tun, P.P., et al.: Brushless synchronous generator turn‐to‐turn short cir-
91. Vukosavić, S.N.: Electrical machines. Springer, New York (2013). cuit fault detection using multilayer neural network. Asian Conference on
https://doi.org/10.1007/978-1-4614-0400-2 Energy, Power and Transportation Electrification (ACEPT), 1–8 (2018)
92. Rahimian, M.M.: Broken bar detection in synchronous machines based 108. Korbicz, J., et. al.: Fault diagnosis. models, artificial intelligence, applica-
wind energy conversion system. Dissertation, Texas A&M University, tions. Springer. Berlin (2004). https://doi.org/10.1007/978-3-642-
College Station, US (2012) 18615-8
93. Bacher, J.: Detection of broken damper bars of a turbo generator by the 109. Patan, K.: Artificial neural networks for the modeling and fault diag-
field winding. Renewable Energy Power Quality J. 1(2), 35–41 (2004) nosis of technical processes. Springer, Berlin (2008). https://doi.org/
94. Karmaker, H.C.: Broken damper bar detection studies using flux probe 10.1007/978-3-540-79872-9
measurements and time‐stepping finite element analysis for salient‐pole 110. Yaghobi, H., Mashhadi, H. R., Ansari, K.: Artificial neural network
synchronous machines. Proc. Symp. Diagnostics Electr. Mach., Power approach for locating internal faults in salient pole synchronous
Electron. Drives, 193–197 (2003) generator. In: Elsevier Expert System with Applications, vol. 38, pp.
95. Neti, P., Dehkordi, A.B., Gole, A.M.: A new robust method to detect 13328–13341
rotor faults in salient‐pole synchronous machines using structural 111. Trutt, F. C., Sottile, J., Kohler, J. L.: Condition monitoring of induction
asymmetries. Proc. IEEE Ind. Appl. Soc. Annu. Meeting, 1–8 (2008). motor stator windings using electrically excited vibrations. In: IEEE
5–9 October Industry Applications Conference: 37th IAS Annual Meeting
96. Antonino‐Daviu, J. A., et al.: Towards condition monitoring of damper Conference, 4, pp. 2301–2305 (2002)
windings in synchronous motors via EMD analysis. IEEE Trans. on 112. Ediriweera, S. H.: Fault diagnosis by vibration analysis of synchronous
Energy Conversion. 27(2), 432–439 (2012). generators. Dissertation, University of Moratuwa, Moratuwa (2011)
97. Huang, N.E., Shen, S.S.P.: Hilbert‐Huang transform and its applica- 113. Choi, J. W.: Analysis of electrical signatures in synchronous generators
tions, vol.5. World Scientific, Singapore (2005). https://doi.org/ characterised by bearing faults. Dissertation, Texas A&M University,
10.1142/5862 College Station, US (2010)
98. Huang, N. E., et al.: The empirical mode decomposition and the Hilbert 114. López‐Pérez, D., Antonino‐Daviu, J.: Application of infrared ther-
spectrum for non‐linear and non‐stationary time series analysis. In: mography to failure detection in industrial induction motors: case
Proceedings of the Royal Society of London. Series A: Mathematical, stories. IEEE Trans. Ind. Appl. 53(3), 1901–1908 (2017)
Physical and Engineering Sciences, 454, (1971), pp. 903–995 Royal 115. Stipetic, S., et al.: Measurement of excitation winding temperature on
Society (1998). http://dx.doi.org/10.1098/rspa.1998.0193 synchronous generator in rotation using infrared thermography. IEEE
99. Berry, P. J., Essam, H. S.: An investigation into damper winding failure Trans. Ind. Electron. 59(5), 2288–2298 (2012)
in a large synchronous motor. In: 50th International Universities power 116. Babić, B.M., Milić, S.D., Rakić, A.Ž.: Fault detection algorithm used in a
Engineering Conference (UPEC), pp. 1–4 (2015) magnetic monitoring system of the hydrogenerator. IET Electr. Power
100. Overview of temperature sensors. National Instruments, National Appl. 11(1), 63–71 (2017)
Instruments. http://www.ni.com/white-paper/4218/en/ (2020). 117. Pollock, G.B., Lyles, J.F.: Vertical hydraulic generators experience with
Accessed 22 Jun 2020 dynamic air gap monitoring. IEEE Trans. Energy Conv. 7(4), 660–668
101. Vas, P.: Artificial‐intelligence based electrical machines and drives: (1992)
application of Fuzzy, neural, Fuzzy‐neural and genetic‐algorithm based
techniques. Oxford University Press, Oxford, UK (1999)
102. Filippetti, F., et al.: Recent developments of induction motor drives fault
diagnosis using AI techniques. IEEE Trans. Ind. Electron. 47(5), How to cite this article: Mostafaei M, Faiz J. An
994–1004 (2000) overview of various faults detection methods in
103. Fang, H., Xia, C.: A Fuzzy neural network based fault detection scheme
synchronous generators. IET Electr. Power Appl.
for synchronous generator with internal fault. In: IEEE 6th Interna-
tional Conference on Fuzzy Systems and Knowledge Discovery. 4, 2021;15:391–404. https://doi.org/10.1049/elp2.12031
433–437 (2009)