BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.

156

A Review Study on Mathematical Methods for

Fault Detection Problems in Induction Motors
E. Ayaz

AbstractInduction motors are frequently used in industrial

processes. Failure of these machines may cause economic, quality
and safety losses. In this paper, the mathematical methods used in
detection of mechanical and electrical faults of these motors are
reviewed together with theory and application examples on the
current and vibration data which is acquired during performance
tests of the motors followed by accelerated aging.
Index TermsInduction motor, fault detection, aging, signal
processing

2.1. FREQUENCY DOMAIN METHODS

Since different fault types generate different frequency
spectrum distributions, the diagnosis results are based on
frequency features of signals. Frequency content of a signal at
frequency mf can be found by Fourier transform given as
below (21-22)
1

() =

() exp (

=0

1. INTRODUCTION

NDUCTION motors are widely used electrical drives in

industrial processes due to their wide power range, simple
and rugged structure. When the motor ages or a fault occurs in
the motors, these may cause losses in their efficiency, economic
and safe operation of industrial processes. The faults can be
categorized into as being mechanical or electrical faults.
According to the statistical surveys performed on motors
revealed that 41% of failures are resulted from bearing faults,
37% of them are from stator faults, 10% is from rotor faults,
and 12% are from other faults such as unbalanced phase supply,
soft foot, asymmetries in the magnetic circuits etc. (1-7)
This paper addresses fault detection methods in induction
motors together with theory and applications on experimental
data acquired during performance test of the motors subjected
to accelerated aging (6-7). Detection of eccentricity fault (7-9),
bearing fault (10-18), and stator insulation fault (19-20) is
considered. Applications of statistical methods, power spectral
density analysis, coherence analysis, continuous and discrete
wavelet transform, autoregressive modeling method, adaptive
neuro-fuzzy inference system, artificial neural network is
presented by means of the experimental data.
2. MATHEMATICAL METHODS
This section includes mathematical methods used in signal
analysis for fault detection and diagnosis studies. These are
frequency domain methods, time-frequency/scale domain
methods, stochastic methods, and soft computing methods.

2
)

(1)

where N is the number of samples, f is the frequency

resolution, m is integer number and t is the data-sampling
interval. The auto-power spectral density (APSD) of x(t) is
estimated as
() =

1
|()|2 ,

(2)

The cross power spectral density (CPSD) between x(t) and

y(t) is similarly estimated. The statistical accuracy of the
estimate in Equation (2) increases as the number of data points
or the number of blocks of data increases.
The cause and effect relationship between two signals or the
commonality between them is generally estimated using the
coherence function. The coherence function is given by
() =

| ()|
() ()

0 < < 1

(3)

where Sxx and Syy are the APSDs of x(t) and y(t), respectively,
and Sxy is the CPSD between x(t) and y(t). A value of coherence
close to unity indicates highly linear and close relationship
between the two signals.
2.2. TIME-FREQUENCY/SCALE DOMAIN METHODS
To find time localization of the frequency content of the
signal short-time Fourier transform (STFT) can be used which
is defined as (23-24)

(, ) = () ( ) exp[2] .

(4)

E. AYAZ is with the Faculty of Electrical and Electronic Engineering,

Istanbul Technical University, Istanbul, Turkey (e-mail: [email protected]).

Copyright BAJECE

ISSN: 2147-284X

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

In STFT, the signal x(t) is rst windowed using dierent type
of window functions g(t) such as triangular window,
rectangular window, Gaussian window etc. centered at time
location . Then Fourier Transform of resulting windowed
signal is taken. This gives the STFT of signal for particular
time. As window slides along time axis, so basically STFT
maps input signal x(t) into two dimensional function in a timefrequency plane with fixed resolution.
Wavelet transform can be used to separate the signal into
frequency bands and to get variable resolution which is good
time resolution for rapidly changing signals and good frequency
resolution for slowly changing signals. The continuous wavelet
transform is defined as

157

Fig.1. Signal decomposition at the second stage.

2.3. STOCHASTIC METHODS

This section gives statistical methods and time series
modeling methods.
2.3. 1. STATISTICAL METHODS
Several statistical parameters, calculated in the time domain,
are generally used to define average properties of machinery
data. The two basic parameters are the mean value and the
standard deviation . For a given data set {xi} these are defined
as follows (25):

1
=

(, ) = () , ()

(9)

=1

(5)

1
= ( )2

where

(10)

=1

, () =
(
) ;

||

, ; 0

(6)

is called the mother wavelet. The dilation parameter a

controls the scale or frequency of wavelet, and the translation
parameter b controls the position of the wavelet in time. The

parameters a and b are defined as = 0 , = 0 0 where

, , 0 > 1, and 0 > 0, the Discrete Wavelet
Transformation (DWT) is given as
[, ] =

0 0
[] [
]

(7)

S. Mallat introduced an efficient algorithm to perform the

DWT known as the Multi-Resolution Analysis (MRA). The
MRA is similar to a two-channel sub-band coder used in highpass and low-pass filters, from which the original signal can be
reconstructed.
The frequency decomposition of the signal is shown
schematically (Fig.1). The low-frequency sub-band is referred
to as approximation ai and the high-frequency sub-band by
detail di. Thus, at the second stage the signal may be
reconstructed as
(8)
= 2 + 1 + 2
S

a1

a2

where N is the number of the data points. For the Gaussian

(normal) probability distribution, two parameters that reflect
the departure from the normal distribution are skewness (c) and
kurtosis (k). These are calculated as follows.

d1

1
3
[
=1( ) ]

=
3

(11)

1
4
[
=1( ) ]

=
4

(12)

For a perfect normal distribution, c is equal to zero. A

negative value is due to skewness towards lower values while a
positive value indicates non-symmetry towards higher values.
For small data sets, one often gets values that differ from zero.
The kurtosis or flatness k, is very close to unity for a normal
distribution. These statistical parameters may be used to
perform a quick check of the changes in the statistical behaviour
of a signal.
2.3. 2. TIME SERIES MODELING METHODS
In time series analysis the autoregressive (AR) modeling
method is used commonly due its simplicity and ability to show
sharp peaks in the frequency domain. The AR method establish
the mathematical model for regression and forecast, hence the
AR coefcients of the model represent the signal features, and
can be used for fault detection purposes (16,26).
AR modeling is a parametric method and represents the
signal as a linear combination of its previous values plus an
error term. AR model of order p for the signal x(n) is given as

d2

() +
=1

Copyright BAJECE

ISSN: 2147-284X

September 2014

( ) = ()

Vol:2 No:3

(13)

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

where ak for k = 0,1,, p are AR coefficients and v(n) is the

white noise with variance 2 . The filter coefficients are
determined using the autocorrelation sequence of the AR
process which satisfies the Yule-Walker equations given by

() +

=1

( ) = 2 () ;

(14)

where (k) is the unit sample sequence. Thus given the

autocorrelations rx(k) for k = 0,1,, p these equations can be
solved for the AR coefficients ak and . In most applications
rx(k) is unknown and estimated from a sample realization of the
process. Given x(n) for 0 n N, where N is the number of
samples, rx(k) is estimated using the sample autocorrelation
() =

1
1
()( )

=0

(15)

Order selection is important in terms of to get less error

between signal and its AR model. And also there must be
enough number of coefficients, but not too much, to diagnose
faults conveniently. Akaikes Information Criteria (AIC) can be
used for order selection and given as follows
() = ( 2 ) + 2

(16)

2.4. SOFT COMPUTING METHODS

The relation matrix between fault types and fault features
extracted from vibration, current, temperature signals can be
too complicated to determine fault types by this matrix with
human inference. Intelligent methods like artificial neural
networks (ANN) and adaptive neuro-fuzzy inference system
(ANFIS) can be used for classification and pattern recognition.
2.4.1. ARTIFICIAL NEURAL NETWORKS
ANN is an interconnection of computational elements known
as neurons (27). Each neuron has multi inputs from other neurons
with assigned weights. The output of a neuron is computed by
summing of all the weighted inputs and then passing it through
a function. ANN consists of one or more layers of neurons in
interconnected topology (Fig.2).

158

The backpropagation learning algorithm is widely used to

determine input-output relationship of a system. It is an iterative
gradient search algorithm which adjust each weight in a
multilayer network so as to reduce the error in the outputs. It
works by propagating errors backward from the output layer.
For a three layer ANN with I inputs, one hidden layer with J
neurons and K outputs neurons, the error function is

1
=
( )2
2 =1 =1

(17)

where ykp and dkp are the actual and desired outputs of the
pattern respectively. Individual weight adjustment between the
hidden layer and the output layer for pattern p are computed by

=
(18)

where j=1,2,J and is a constant learning rate. The weights

between hidden and output layers are adjusted using the
recursive formula given as
( + 1) = () +

(19)

where n is the iteration number. Similar weight adjustment

formulas can be obtained between the input and output layers
by changing the indices from k to i.
2.4.2. ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM
Adaptive neuro-fuzzy inference system (ANFIS) is an
implementation of a fuzzy logic inference system with the
architecture of a five-layer feed-forward network (18,19,27,28).
With this way ANFIS uses the advantages of learning capability
of neural networks and inference mechanism similar to human
brain provided by fuzzy logic. The architecture of ANFIS with
two inputs, one output and two rules is given (Fig.3). Here x, y
are inputs, f is output, the circles represent fixed node functions
and squares represent adaptive node functions. This is a
Sugeno-type fuzzy system, where the fuzzy IF-THEN rules
have the following form:
Rule 1: If x is A1 and y is B1 then f1 = p1 x + q1 y + r1
Rule 2: If x is A2 and y is B2 then f2 = p2 x + q2 y + r2

Output

Layer 1

Output Layer

Layer 4
Layer 2

Layer 3
x y

A1
x
A2

Hidden Layer
Error
propagation
direction

Output
computation
direction

B1

w1

w2

w1

w1 f1

w2

w2 f 2

x y

B2

Input Layer

Layer 5

Fig.3. ANFIS architecture with two rules.

X1

X2

Input

X3

Fig.2. Topology of a feed forward neural network with three layers.

Copyright BAJECE

ISSN: 2147-284X

The operation of each layer is as follows:

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

Layer 1 called fuzzification layer and it fuzzify the values of
the input variables according to a membership function. This
layer forms the antecedents of the fuzzy rules (IF part).
Generalized bell membership function (Fig.4) is popular for
specifying fuzzy sets because of their smoothness and concise
notation. The parameters {ai, bi, ci} of the membership function
are called as premise parameters where i denotes the node
number.

caused by the dynamic eccentricity are given as

1
= [1 (
)] = | | ,
/2
= 1, 2, 3,

(20)

where fe is the electrical supply frequency, s is the per unit slip,

p is the number of poles, and fr is the rotor speed in Hz. Slip s
= (fs fr)/fs, where fs is the synchronous frequency.

slope = -b/2a

1.0
0.5
0

159

c-a

c
2a

c+a

The side-band effects of the motor current signals both of the

healthy and faulty cases by means of their power spectral
density variations at around supply frequency 60 Hz is shown
in (Fig.5a) and (Fig.5b).
150

Fig.4. Generalized bell function.

100

Amplitude [dB]

Layer 2 is rules layer and performs fuzzy AND operation at

the nodes. Thus the output is the product of all incoming signals
and represents the firing strength of each rule. Layer 3 is
normalization layer. Layer 4 is fuzzy inference or consequent
layer which estimates the rules output (THEN part). The
parameters {pi, qi, ri} of each rules output are referred as
consequent parameters. Layer 5 is called defuzzification layer
and calculates the sum of the outputs of all the rules. Learning
of ANFIS is done using hybrid learning procedure given in
Table 1. This algorithm combines backpropagation gradient
descent and least squares method for identification of premise
and consequent parameters in such a way that ANFIS output
matches the training data.

60

63.8
50

52.1
31

89

-50

-100
20

30

40

50

60
70
Frequency [Hz]

80

90

100

(a) APSD for healthy case.

150

3. CASE STUDIES
In this section example applications of mathematical
methods on experimental data for detection of mechanical and
electrical faults in induction motors are given. Detection of
eccentricity fault, bearing fault, and insulation fault is
considered. For this purpose 5 HP, three phases, four poles
induction motor is subjected to bearing damage and winding
insulation damage as well as thermal and chemical aging (6-7).
After each aging cycle the data is collected at 12 kHz sampling
frequency under 100 % load condition. Seven aging cycle is
performed until the motor is broken down.
3.1. MECHANICAL FAULTS AND THEIR DETECTION
Airgap eccentricity (Non-uniform air gap):
Air gap eccentricity may be the consequence of bearing, wear
or bearing failure, bad motor assembly with rotor unbalance or
a rotor which is not perfectly centered (4,8-9). This eccentricity
causes anomalies in the air gap flux density and creates
frequency side bands at around the supply frequency of the
stator current signal for each phase. The side-band frequencies

Copyright BAJECE

ISSN: 2147-284X

60

100

Amplitude [dB]

Table 1. Hybrid learning procedure.

Forward pass
Backward pass
Premise parameters
Fixed
Gradient descent
Consequent parameters Least squares
Fixed
Signals
Node outputs
Error signals

56

50

63.8
52.1

31

68.1
89

-50

-100
20

30

40

50

60
70
Frequency [Hz]

80

90

100

(b) APSD for aged case.

Fig.5. APSD variations of current signals

The big amplitudes defined as side-band frequencies

(Fig.5b), which are appeared at around the fundamental
frequency 60 Hz with difference of 4 Hz, indicate the rotor
eccentricity. Here the side-band frequencies (fsb) can be defined
by the following equation
= (1 2)

(21)

where k is an integer. In this application fr=1800/60=30 Hz and

fr=1742/60=29.03 Hz. Taking k=1,2 and s=0.032, from the Eq.
(21), all side-band frequency values can be found at 52, 56, 64
and 68 Hz (Fig.5b). Also, comparing the Fig.5a and Fig.5b the
September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

other important frequency components can be determined at 31
Hz and 89 Hz. These are the rotor eccentricity frequencies and
they can be defined between the rotational frequency (fr) and
fundamental frequency (fe) by Equation (20).
Bearing faults:
One cause of premature bearing failure is the passage of
electrical current through the bearing and it is known as
electrical discharge machining or fluting (1-3). A severe damage
is a source of audible bearing noise. The surface degradation of
bearing races and rolling elements results in extreme vibration
levels and eventual bearing failure.
Statistical properties of vibration signal can be investigated
for detecting the fault (10-11). The deviation between statistical
parameters for the healthy and the faulty cases are compared as
given in Table 2. The healthy and faulty vibration signals in
time domain and their amplitude probability density functions
(histograms) is shown in (Fig.6). These results indicate that the
signals have a normal distribution and the overall standard
deviation has increased by a factor of about 6, showing the
damage.
Table 2. Statistical parameters of vibration signal
Healthy
Faulty
Mean
0.0016
0.0030
Standard Deviation
0.1135
0.6040
Skewness
0.0591
-0.0060
Kurtosis
2.9266
3.0093

160

is Nb = 9. The characteristic frequencies are calculated as fc =

11.6 Hz, fo = 104.5 Hz, and fi = 156.7 Hz. Ball Spin Frequency
is calculated as fb = 136.9 Hz for the given rotational speed.
The characteristic bearing frequencies and its harmonics
occur when the bearing has a fault. The variations that are
produced by the air gap eccentricity due to bearing defect
generate stator currents that are related to these characteristic
bearing frequencies ( fv )
= | | , = 1, 2, 3,

(22)

where fv is one of the characteristic bearing frequencies namely

fc , fo , fi , fb .
The characteristic bearing frequencies are remarkable when
the bearings have single point defects. If the defects are
scattered on the bearing components, an increase is seen in the
high frequency region of the vibration spectrum. From the
power spectral density (Fig.7) of vibration signals for the
healthy and faulty case, it is clear that there is an increase in the
signal energy for the frequency ranges 1.5-4 kHz are due to
frosting on the surface of bearing elements, caused by material
removal by pitting. Data acquisition system has a low-pass filter
with the cutoff frequency at 4 kHz.
Healthy case
2.5

Amplitude [g]

1.5

1
1008 Hz
1395 Hz

0.5

1000

2000

Fig.6. Accelerometer signal waveforms: (a) Baseline. (b) Final aged cycle.
(c) Probability density functions of a) and b).

3000
Frequency [Hz]

4000

5000

6000

(a)
Faulty case

Copyright BAJECE

ISSN: 2147-284X

2.5

15xfi
2355 Hz

2
26xfo
2707 Hz

Amplitude [g]

Bearing faults can be diagnosed by frequency domain

analysis of vibration signals (12-14). When a fault occurs in any
bearing components, this creates vibrations at characteristic
frequencies defined by the bearing geometry. These
characteristic fault frequencies can be calculated by using either
the bearing geometry or approximate formulas for the bearings
which have the number of balls between six and twelve as
below:
Cage Frequency: fc = 0.4 fr
Ball Pass Frequency of the Outer Race (BPFO): fo = 0.4 Nb fr
Ball Pass Frequency of the Inner Race (BPFI): fi = 0.6 Nb fr
Ball Spin Frequency (fb) is calculated using the bearing data
provided by the bearing manufacturer. Here fr is the rotor
frequency and Nb is the number of balls. In this study, the motor
speed is 1742 min-1 and then the rotor frequency is calculated
as fr = 1742/60 = 29.03 Hz. The number of balls in the bearing

29xfb or 38xfo
3961 Hz

20xfb
2742 Hz

1.5

25xfb
3422 Hz

1008 Hz

7xfb
960.9 Hz
0.5

27xfb
3691 Hz

1395 Hz

13xfi
2039 Hz

5xfo
527.3 Hz

1000

2000

3000
Frequency [Hz]

4000

5000

6000

(b)

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

Fig.7. Power spectral densities of vibration signals (a) healthy, (b) aged cases.

The coherence function (Fig.8) indicates that the most

dominant frequency values, where motor current and vibration
signals are correlated, are located at 234 Hz and 469 Hz (8-9).
The side-band frequency due to the dynamic eccentricity is
fecc = 60 + 6(29.03) = 234 Hz.
The gap eccentricity generated current due to the bearing defect
is
fbng = 60 + 3(136.9) = 470.7 Hz.

mations
a1
a2
a3
a4
a5
a6
a7
a8

(Hz)
0 3000
0 1500
0 750
0 375
0 187.5
0 93.75
0 46.875
0 23.4375

161

d1
d2
d3
d4
d5
d6
d7
d8

(Hz)
3000
6000
1500 3000
750
1500
375
750
187.5 375
93.75 187.5
46.875 93.75
23.4375 46.875

MRA implementation is shown in (Fig.9) and (Fig.10).

According to these results, 3-6 kHz frequency band, which is
named as first detail (d1) in MRA, is the most dominant band in
terms of the similarity. The ratio, which can be calculated
between the RMS (root-mean-square) values of vibration
measurement and the RMS values of (d1), increases as the
motor bearing degrades toward failure. If looked at other subbands of the vibration measurement, a good trend cannot be
seen. For this reason, high frequency vibrations, which take
place between 3 kHz and 6 kHz, are very affective in bearing
fluting. Hence, a feature extraction from considered data could
be very effectively realized by using the multi-resolution
wavelet analysis technique.
(a)

(a)

(b)

(b)
Fig.8. The coherences for healthy (a) and faulty (b) cases.

The coherence function between the motor current and

accelerometer signals at the 234 Hz and 470.7 Hz has increased
by a factor of 2.5 in the 200-500 Hz frequency range. The
coherence between the motor current and vibration signals
indicates that rotor eccentricity and bearing faults are reflected
in the motor current spectrums. Also the location of bearing
fault is balls in the bearing as a result of coherence analysis.
In order to determine precisely which frequency band reflects
the bearing fluting damage, the sub-band or the MRA of the
faulty signals was performed by dividing them into eight subbands in the frequency range 0-6 kHz (7,12,14). These are given in
Table 3 in terms of details (di) and approximations (ai).

Fig.9. Details and approximations of vibration signal (s) after final aging
cycle. (a) Detail sub bands (d1-d8) vibration signal (s) for aged case.
(b) Approximation sub bands (a1-a8) vibration signal (s) for aged case.

Table 3. Frequency sub-bands of the vibration signal

Approxi- Sub-bands
Details
Sub-bands

Copyright BAJECE

ISSN: 2147-284X

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

162

Fig.10. RMS values of vibration signal and two sub-bands (d1 and d2) after
nal aging cycle.

And also, short-time Fourier transformation (STFT) (Fig.11)

of the first detail (d1) shows that the observed frequency band
is between 2 and 4 kHz because of the usage low-pass filter
which has a cut-off frequency at 4 kHz during the data
acquisition process.

Fig.12. Absolute values of continuos wavelet transform coefficients for scales

1, 2, 4, 8, 16, 32, 64, 128, 256 of vibration signals for healthy case.

If the first scale variation is taken outside to plot it

individually (Fig.13), it gives the high frequency components
which are represented by very small amplitudes, to indicate the
origin of the bearing damage as a potential defect.
-3

1.6

x 10

1395 Hz
1.4
1008 Hz

2426 Hz

1.2

20
10
0
0
6000

0.05
0.1

Amplitude [g]

Amplitude [g]

30

1
0.8
27xfb
3703 Hz

0.6
2367 Hz

4000
0.15
2000

0.2
Time [s]

0.25

Frequency [Hz]

Fig.11. STFT presentation of the sub-band (d1) related to bearing damage.

The fundamental feature, which is indicated (Fig.11) in the

frequency range 2-4 kHz, denotes the bearing damage as a
result of the bearing fluting test procedures. At this point, we
want to ask the question how to detect this before it becomes
severe. To answer this, the continuous wavelet transform
(Fig.12) is recommended to reveal the potential existence of the
bearing damage in early case using only healthy case data.

Copyright BAJECE

ISSN: 2147-284X

28xfb
3832 Hz

0.4 9xfb
1242 Hz
0.2
0

25xfi
3914 Hz

1535 Hz

29xfb or 38xfo
3973 Hz
0

1000

2000

3000
Frequency [Hz]

4000

5000

6000

Fig.13. First scale spectrum of vibration signals in healthy case as a potential

defect.

AR method is used to model vibration signals parametrically

and to determine how the model parameters changes with the
aging (15). The sixtieth order model is chosen to construct AR
models of vibration signals. The first AR coefficient and
variance of white noise input to drive the AR model is increased
with the aging (Fig.14 and Fig.15).

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

ANFIS is used to combine information from several sensors

to perform automatic fault detection of bearing failure (Fig.17).
The inputs of ANFIS are current, vibration and temperature as
given in Table 4, and the output is motors condition labeled as
healthy (H) or faulty (F).

a1 coefficient
4

AR model coefficient value

a1

-1.2

a2

-1.3

a3

2
1

a4

-1.4

a5

-1.5

a6

-1.6

163

a7

0
-1
-2

a8

-1.7

a9

-1.8

a10

-1.9
-2

-3

-2.1
-4
0

3
4
Aging cycle

-2.2

2 3 4 5
Aging cycle

Fig.17. ANFIS based fault detection structure.

Fig.14. Variation of first 10 AR coefficients (left) and the 1 coefficient

(right) with aging cycles.

Table 4. Inputs used in ANFIS.

Input #
Frequency band intervals
1
0 732 Hz
Vibration
2
2199 2930 Hz
3
2931 4028 Hz
4
0 53 Hz
Current
5
123 234 Hz
6
240 352 Hz
Temperature
7
-

0.03

Noise variance

0.025

0.02

0.015

0.01

The classification performance of ANFIS as healthy or faulty

condition is calculated by taking a threshold value of 0.5. Hence
97.8% of test data is classified correctly (Fig.18).

0.005

3
4
Aging cycle

7
M6 (test)
80 patterns

Fig.15. Variance of white noise input to drive the AR models.

M8
(train and test)
80 patterns

M7 (test)
80 patterns

Except from vibration and current signals, bearing

temperature gives valuable information about the health of the
bearings (17) since temperature is an important factor for
lubricant. High temperatures reduce the viscosity of lubricant
inside of the bearing, and cause early bearing failure. From the
relationship between bearing surface temperature and aging
(Fig.16), it is seen that the bearing temperature increase with
the degradation of the motor. The temperature rise is at most
5C for M8 by comparing the temperature values of the healthy
motors which is cycle #0 and faulty motors which is cycle #7.
The increase in temperature give information about anomaly,
but it is not enough for precise decision on the severity of the
fault.

64
62
Temperature [degC]

F: 0.9

T: 0.5

H: 0.1
0
ANFIS #3
Actual
Pattern

Fig.18. Training and test results for ANFIS.

H: Healthy, F: Faulty, T: Threshold

3.2. ELECTRICAL FAULTS AND THEIR DETECTION

66

60
58
56
54

M6
M7
M8
M11

52
50

M11 (test)
80 patterns

4
Aging Cycle #

Stator winding insulation faults are also a common source of

failure of electric motors. Especially the use of motor drives
creates some undesirable effects with the eventual failure of
motors (20). In this section, detection of stator winding insulation
fault of main-fed induction motor is addressed.
Stator current imbalances based upon the stator insulation
damage act on the stator magneto motor force (mmf) and cause
motor vibrations. In this sense the cross spectral approach
which is the calculation of coherence function between the
motor current and vibration signals for the initial and aged cases
(Fig.19) is used to detect the insulation damage.

Fig.16. Process-end bearing surface temperature.

Copyright BAJECE

ISSN: 2147-284X

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

164

4. CONCLUSIONS AND DISCUSSIONS

Fig.19. Coherence variation between motor current and vibration signal.

By comparing the initial and aged cases (Fig.19), the even

harmonics of line frequency between 4th and 16th in the range
0-1000 Hz, namely the frequencies 240, 360, 480, 600, 720,
840, and 960 Hz appear due to insulation aging.
A neuro-detector is designed for the automatic fault detection
using the coherence approach to determine the even harmonic
effects, which characterize the stator insulation damage. The
artificial neural network topology has an auto-associative
structure, which uses the same input-output pairs in the
frequency domain. Each input-output pairs obtained by
coherence approach cover the frequency band of 0-1000Hz,
which contains the most effective even harmonics, and they are
defined as training patterns of the training data set of the neural
structure. The feed forward neural structure has 50 nodes in
terms of the input and output nodes, while number of the hidden
nodes is 10. At the first step, the neural network is trained for
the normal condition that is healthy stator insulation case using
a learning algorithm. After that, the aged case is asked to it as
an unknown case. In this case, the neural network produces
different responses at the output nodes with a big error changes
(Fig.20). Even harmonic values related to the insulation
degradation can be detected looking through the error change.
0.7
Target
YSA output
Deviation

0.6

840 Hz

720
Hz

0.6
0.5

480
Hz

0.5

0.3
600
Hz

0.2

0.3
240
Hz

360
Hz

0.2

960
Hz

0.1
0

0.1

-0.1

100

200

300

400
500
600
Frekans [Hz]

700

800

900

-0.2
1000

Fig.20. Testing result of the neural network for the aged case.

Copyright BAJECE

ISSN: 2147-284X

Deviation

Coherence

0.4
0.4

In this paper fault detection and diagnosis in induction

motors using mathematical methods are discussed with theory
and example applications. Eccentricity detection, bearing fault
detection and stator winding insulation faults are considered.
Eccentricity is detected by spectrum of stator current and
coherence between current and vibration signals.
The statistical analysis of vibration signals give information
about the anomaly. Power spectral analysis of vibration signal
give the features of bearing fault as an increase of energy in the
high frequency region of spectrum The most dominant
frequency band is determined by multi-resolution wavelet
analysis of vibration signals. Potential bearing fault is detected
using continuous wavelet transform. Coherence calculation
revealed the location of fault in bearing. AR modeling give the
first coefficient of model can be used as a feature of bearing
fault. ANFIS application helps to make decision about the
condition of motor using spectral features of current and
vibration signals, and bearing temperature.
Features for stator winding insulation fault is determined as
even harmonics between 4th and 16th in coherence function. A
neuro-detector can be used to detect the changes of these
harmonics.
REFERENCES
[1]

S.Nandi, H.Toliyat, XLi, (2005). Condition monitoring and fault

diagnosis of electrical motors a review. IEEE Transactions on Energy
Conversion. 20 (4), pp.719-729.
[2] A.K.S.Jardine, D.Lin, D.Banjevic, (2006). A review on machinery
diagnostics and prognostics implementing condition-based maintenance.
Mechanical Systems and Signal Processing. (20), pp.14831510.
[3] A.H.Bonnett, (2000). Root Cause AC Motor Failure Analysis with a
Focus on Shaft Failures. IEEE Transactions on Industry Applications. 36
(5), pp.1435-1448.
[4] R.R.Schoen, T.G.Habetler, F.Kamran, R.G.Bartheld, (1995). Motor
Bearing Damage Detection Using Stator Current Monitoring. IEEE
Transactions on Industry Applications. 31 (6), pp.1274-1279.
[5] IEEE Standard Test Procedure for Evaluation of Systems of Insulation
Materials for Random-wound AC Electric Machinery, IEEE Std 1171974.
[6] A.S.Erbay, B.R.Upadhyaya, (1999) Multi sensor fusion for induction
motor aging analysis and fault diagnosis. Research Report,
UTK,UTNE/BRU/99-01. Knoxville, TN, USA.
[7] S.eker, E.Ayaz, B.R.Upadhyaya, A.S.Erbay, Analysis of Motor Current
and Vibration Signals for Detecting Bearing Damage in Electric Motors.
Maintenace And Reliability Conference. Knoxville, TN, USA, 8-10 May
2000, (1) pp.29.01-29.14.
[8] S.Seker, (2000). Determination of Air-Gap Eccentricity in Electric
Motors Using Coherence Analysis. IEEE Power Engineering Review.
July, pp.48-50.
[9] S.eker, E.Ayaz, E.Trkcan, (2003). Elmans Recurrent Neural Network
Applications to Condition Monitoring in Nuclear Power Plant and
Rotating Machinery. Engineering Applications of Artificial Intelligence.
16 (7-8), pp.647656.
[10] S.eker, E.Ayaz, (2003). A Reliability Model for Induction Motor Ball
Bearing Degradation. Electric Power Components & Systems. 31 (7),
pp.639-652.
[11] S.eker, E.Ayaz, (2002). A Study on Condition Monitoring for Induction
Motors Under the Accelerated Aging Processes. IEEE Power
Engineering Review. 22 (7), pp.35-37.
[12] S.eker, E.Ayaz, (2003). Feature extraction related to bearing damage in
electric motors by wavelet analysis. Journal of the Franklin Institute. 340
(2), pp.125-134.

September 2014

Vol:2 No:3

http://www.bajece.com

BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING, 2014, Vol.2, No.3

[13] E.Ayaz, S.eker, E.Trkcan, B.Barutu, Combination Of Spectral And
Multi-Resolution Wavelet Analysis For Fault Detection In Electric
Motors. ELECO2003, Third International Conference on Electrical and
Electronics Engineering. Bursa, Turkey, 3-7 December 2003, pp.94-98.
[14] E.Ayaz, A.ztrk, S.eker, B.R.Upadhyaya, (2009). Fault detection
based on continuous wavelet transform and sensor fusion in electric
motors. COMPEL: The International Journal for Computation and
Mathematics in Electrical and Electronic Engineering. 28 (2), pp.454470.
[15] E.Ayaz, (2014). Autoregressive modeling approach of vibration data for
bearing fault diagnosis in electric motors. Journal of Vibroengineering.
16 (5), pp.2130-2138.
[16] C.C.Wang, Y.Kang, P.C.Shen, Y.P.Chang, Y.L.Chung, (2010).
Applications of fault diagnosis in rotating machinery by using time series
analysis with neural network. Expert Systems with Applications. (37),
pp.16961702.
[17] M.S.Yilmaz, E.Ayaz, Adaptive Neuro-Fuzzy Inference System for
Bearing Fault Detection in Induction Motors Using Temperature,
Current, Vibation Data. EUROCON 2009-International IEEE
Conference. Saint-Petersburg, Russia, May 18-23, 2009.
[18] N.T.Nguyen, H.H.Lee, Bearing fault diagnosis using adaptive network
based fuzzy inference system. International Symposium on Electrical &
Electronics Engineering. Vietnam, 24- 25 October 2007.
[19] M.S.Ballal, Z.J.Khan, H.M.Suryawanshi, R.L.Sonolikar, (2007).
Adaptive neural fuzzy inference system for the detection of inter-turn
insulation and bearing wear faults in induction motor. IEEE Transactions
on Industrial Electronics. 54 (1).
[20] E.Ayaz, M.Uar, S.eker, B.R.Upadhyaya, (2009). Neuro-detector
based on coherence analysis for stator insulation in electric motors.
Electric Power Components and Systems. 37 (5), pp.533-546.
[21] V.S.Vaseghi, (1996) Advanced signal processing and digital noise
reduction, New York: John Wiley.
[22] T.K.Moon, W.C.Stirling, (1999) Mathematical Methods and Algorithms
for Signal Processing, Prentice Hall.
[23] S.Qian, D.Chen, (1996) The joint Time -Frequency Analysis-Methods
and Applications. Englewood Cliffs, NJ: Prentice-Hall.
[24] I. Daubechies, (1990). The wavelet transform, time-frequency
localization and signal analysis. IEEE Transactions on Information
Theory 36 (5) pp.961-1005.
[25] A.Papoulis, (1987) Probability, Random Variables and Stochastic
Processes. McGraw Hill International Edition, 5th printing, Singapore.
[26] M.H.Hayes, (1996) Statistical Digital Signal Processing and Modeling.
John Wiley&Sons, Inc.
[27] L.H.Tsoukalas, R.E.Uhrig, (1997) Fuzzy and neural approaches in
engineering. New York: Wiley.
[28] J.S.R.Jang, (1993). Adaptive-network-based fuzzy inference system.
IEEE Transactions on Systems, Man, and Cybernetics 23 (3).

Copyright BAJECE

ISSN: 2147-284X

165

ACKNOWLEDGMENT
The author gratefully thanks to Prof. Serhat Seker from the
Istanbul Technical University and Prof. Belle R. Upadhyaya
from the University of Tennessee, Knoxville for their support.

BIOGRAPHIES
EMINE AYAZ received the BS, MS and PhD.
degrees from the Istanbul Technical University
(ITU), Electrical Engineering Department, in
1993, 1997 and 2002 respectively. In 1999, she
joined to the University of Tennessee, Nuclear
Engineering Department and Maintenance &
Reliability Centre to do research on accelerated
aging studies of the electric motors. She is
currently associate professor in Electrical
Engineering Department of ITU. Her research
interests are signal processing, soft computing,
and condition monitoring in electric power systems.

September 2014

Vol:2 No:3

http://www.bajece.com