(Energy Science, Engineering and Technology) Mykhaylo v. Zagirnyak, Zhanna IV. Romashykhina, Andrii P. Kalinov - The Diagnostics of Induction Motor Broken Rotor Bars On The Basis of The Electromotive
(Energy Science, Engineering and Technology) Mykhaylo v. Zagirnyak, Zhanna IV. Romashykhina, Andrii P. Kalinov - The Diagnostics of Induction Motor Broken Rotor Bars On The Basis of The Electromotive
(Energy Science, Engineering and Technology) Mykhaylo v. Zagirnyak, Zhanna IV. Romashykhina, Andrii P. Kalinov - The Diagnostics of Induction Motor Broken Rotor Bars On The Basis of The Electromotive
THE DIAGNOSTICS OF
INDUCTION MOTOR BROKEN
ROTOR BARS ON THE BASIS OF
THE ELECTROMOTIVE
FORCE ANALYSIS
No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or
by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no
expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No
liability is assumed for incidental or consequential damages in connection with or arising out of information
contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in
rendering legal, medical or any other professional services.
ENERGY SCIENCE, ENGINEERING
AND TECHNOLOGY
THE DIAGNOSTICS OF
INDUCTION MOTOR BROKEN
ROTOR BARS ON THE BASIS OF
THE ELECTROMOTIVE
FORCE ANALYSIS
MYKHAYLO V. ZAGIRNYAK
ZHANNA IV. ROMASHYKHINA
AND
ANDRII P. KALINOV
Copyright © 2018 by Nova Science Publishers, Inc.
All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted
in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying,
recording or otherwise without the written permission of the Publisher.
We have partnered with Copyright Clearance Center to make it easy for you to obtain permissions to
reuse content from this publication. Simply navigate to this publication’s page on Nova’s website and
locate the “Get Permission” button below the title description. This button is linked directly to the
title’s permission page on copyright.com. Alternatively, you can visit copyright.com and search by
title, ISBN, or ISSN.
For further questions about using the service on copyright.com, please contact:
Copyright Clearance Center
Phone: +1-(978) 750-8400 Fax: +1-(978) 750-4470 E-mail: info@copyright.com.
Independent verification should be sought for any data, advice or recommendations contained in this
book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to
persons or property arising from any methods, products, instructions, ideas or otherwise contained in
this publication.
This publication is designed to provide accurate and authoritative information with regard to the subject
matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in
rendering legal or any other professional services. If legal or any other expert assistance is required, the
services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS
JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A
COMMITTEE OF PUBLISHERS.
Additional color graphics may be available in the e-book version of this book.
Preface vii
List of Abbreviations xi
List of Symbols xiii
Chapter 1 The Contemporary State of the Problem of the
Diagnostics of Induction Motor Broken Rotor
Bars 1
Chapter 2 Theoretical Foundation for the Research of
Induction Motor Broken Rotor Bars in the Self-
Running-Out Condition 25
Chapter 3 Mathematical Models for the Research of the
Method of Induction Motor Broken Rotor Bars
Diagnostics 69
Chapter 4 The Method of Induction Motor Broken Rotor
Bars Diagnostics with the Use of Wavelet-
Transform 105
vi Contents
the basic causes of breakages are stated, the advantages and drawbacks
of the known methods of diagnostics are determined.
The second chapter contains research concerning the choice of the
testing condition and diagnostic signal for the broken rotor bars
diagnostics, substantiation of the methods for electromagnetic field
calculation, a comparative analysis of methods for diagnostic signals
processing, and also the investigation of factors influencing the
generation of electromotive force in stator windings. Mathematical
models for the research of different operating modes of an induction
motor with broken rotor bars are presented in the third chapter of the
monograph.
The fourth chapter deals with a method for the induction motor
broken rotor bars diagnostics on the basis of the analysis of
electromotive force in the stator windings with the use of wavelet-
transform. Basing on the developed methods, a method of
decomposition of the phase electromotive force signal into the signals
of the electromotive forces of the active sides of the coil with the use of
the theory of inverse z-transform is proposed.
The fifth chapter of the monograph is devoted to experimental
research of the developed method of the diagnostics of induction motor
broken rotor bars.
The monograph has been written in Kremenchuk Mykhailo
Ostrohradskyi National University. The authors are grateful to
Professors V. I. Milykh, V. Yu. Kucheruk, Associate Professor
О. V. Kachura for reviewing and support of the monograph, Associate
Professors D. G. Mamchur and V. O. Melnykov for the assistance in the
performance of experimental research, and also to a worker of
Kremenchuk Mykhailo Ostrohradskyi National University
K. V. Kovalenko for the assistance in the preparation of the monograph
in the English language.
LIST OF ABBREVIATIONS
IM Induction Motor
EW Exciting Winding
ADC Analog-Digital Converter
DCG Direct Current Generator
BVS Block Of Voltage Sensors
WB Wavelet-Basis
VMP Vector Magnetic Potential
WT Wavelet-Transform
ShCR Short-Circuited Ring
SCR Squirrel-Cage Rotor
FDM Finite Difference Method
FEM Finite Element Method
CWT Continuous Wavelet Transform
CFMM Circuit-Field Mathematical Model
EMF Electromotive Force
EM Electric Machine
EMFl Electromagnetic Field
ED Electric Drive
LIST OF SYMBOLS
To confirm the credibility of the statistic data given in Table 1.1 the
statistic data of squirrel-cage IM failures at one of the enterprises of
Poltava region, Ukraine (PJSC «AutoKrAZ») during 2016 were
analyzed. The data are given for four production shops that use IMs in
their technological processes. The data of IM failures are given in Table
1.2, and the distribution according to damage types – in Table 1.3.
The analysis of the given data of IM failures and the distribution
according to the damage types revealed that the number of motors of
some production shops at PJSC “AvtoKrAZ” , that were repaired due to
broken rotor bars, makes on average 8.5%.
aluminum winding on the rotor the index of bar damaging is 31.9%, for
motors with cast aluminum winding – 8.9%.
The following types of damages are distinguished according to the
character of rotor bar breakage:
0 10 20 30 40 50 60
%
Figure 1.2. The distribution of rotor breakages for series A motors, size 11.
6 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
0 5 10 15 20 25 30 35 40
%
Figure 1.3. The distribution of rotor breakages for series A motors, size 12.
0 5 10 15 20 25
%
Figure 1.4. The distribution of rotor breakages for series A motors, size 13.
The analysis of the statistic data confirms the necessity for the
diagnostics of the mentioned IM damages.
In this case a usual value of current I br comes into the bar, but
along the length of the bar the current flows into adjacent bars ( I adj1 ,
I adj 2 ) through the rotor magnetic circuit (Figure 1.5).
I a d j1
I br
I adj 2
broken bar
Criterion 1
Information value. The criterion determines a possibility to separate
the types of the defects, to locate them.
Criterion 2
The degree of automation of the diagnostics process. This criterion
determines the level of soft- and hardware, the structure and
composition of measuring-diagnostic equipment during experimental
research.
Criterion 3
The expenditure of time for carrying out the diagnostics preparation
operations. According to this criterion, inefficient methods include
those that require the withdrawal of IM from the technological process,
its disassembling and the installation of the required measuring sensors
in the motor gap, etc.
Criterion 4
The expenditure of time for processing of information obtained as a
result of the research and making a decision concerning the existing
damage. The use of a mathematical apparatus with ready software
12 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Criterion 5
A necessity for specialized personnel for the analysis of the data. In
this case the cost of service personnel labor should be taken into
account.
Thus, correspondence of the methods for IM broken rotor bar
diagnostics to the proposed criteria enables determination of their
feasibility study for the use under different conditions.
It is expedient to mention a number of widespread methods for
induction motor broken bar diagnostics.
Method 1
A method of continuous monitoring [12] of the state of the stator
and rotor windings of IM with SCR according to the data of
measurement of the phase currents and voltages. To assess the IM
technical condition the symmetric components of the stator currents and
voltages, as well as the consumed active power and the angle of slope
of the electromotor mechanical characteristic in the domain of operating
slip are used. This method makes it possible to avoid diagnostics errors
in the presence of pulsations and harmonic components in supply
voltage.
Drawback: the results of the measurement are assessed according to
the complex criterion of the diagnostics, which prevents one from the
localization of the damages and, in its turn, the simplification of IM
repair.
The Contemporary State of the Problem of the Diagnostics … 13
Method 2
This method [13] provides for the measurement of the phase current
and voltage of the stator windings, and a conclusion as to the degree of
bar breakage is made according to the size of pulsation of the third
harmonic of the measured value.
Drawback: the supply mains voltage quality, imbalance and other
damages influence on the diagnostics results.
Method 3
The method is based on the analysis of starting current in the stator
in one of the motor phases [14]. During the diagnostics process every
previous amplitude value of the phase current is compared with the
following one. It is possible to judge about the presence of winding
defects by the obtained difference.
Drawbacks: it can be realized only in the starting mode, the analysis
is difficult because of the influence of IM electromagnetic parameters
during the start, low reliability, especially for low- and medium-power
IMs.
Method 4
A method with the measurement [15] of the instantaneous values of
two phase currents in the constant operation mode under load. The
presence of damage is determined by the appearance of phase portraits
of IM phase currents instantaneous values.
Drawback: the diagnostics is carried out under load, clear criteria
for the determination of broken rotor bars are not stated.
Method 5
The essence of methods based on the analysis of currents spectra,
Motor Current Signature Analysis (MCSA) [16–23] consists in the fact
that the presence of damage causes variation of magnetic field in the
motor air gap and, consequently, weak modulation of the current
consumed by the motor. The damages of IM electric or mechanical part,
14 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Method 6
Methods on the basis of the analysis of current and voltage
envelopes spectra [17, 24] provide for the record of the instantaneous
values of currents and voltages in three stator phases, the separation of
typical frequencies of the electric motor, the comparison of amplitudes
values at typical frequencies with the value of constant component.
Rotor winding defect is determined by the presence of two symmetric
peaks in the current spectrum in relation to supply mains frequency.
This method has the same drawbacks as the previous one.
Method 7
In paper [10] the spectra of IM vibration in the axial direction are
analyzed. The value of the amplitude of signal harmonics for a
damaged rotor is determined, and a conclusion as to the presence of a
damage is made on the basis of the increase of corresponding
components.
Drawback: the necessity for the installation of vibration sensors, the
high cost of vibration complexes.
Method 8
A method based on the analysis of the applied magnetic field
(AMF) [25] consists in the analysis of the variation of the magnetic
induction of IM applied magnetic field representing a combined
The Contemporary State of the Problem of the Diagnostics … 15
Method 9
A method based on the thermal action of electric current [9] and
input of such voltage to the rotor rings, at which the current value in
bars exceeds the rated value. A thermal imager is used as a measurer of
rotor bars thermal condition. The state of rotor bars can be judged by
the heating temperature when current flows in the bars: the healthy bars
are heated more than the broken ones.
Drawbacks: the necessity for the input of high values of currents to
the rotor winding, the necessity for motor disassembling.
Method 10
Methods based on the stator current analysis using wavelet
transform [26–33]. The presence of broken rotor winding is determined
according to the corresponding values of typical coefficients of IM
stator current wavelet-spectra.
Drawbacks: analogous to method 5, and also the impossibility of
damages localization.
Method 11
Methods [34, 35] provide for the use of the variation of rotor field
magnetic induction, caused by machine double-sided serration
influence on the fundamental harmonic in the electromotor gap, as a
diagnostics signal. The measurement of magnetic induction is realized
on the basis of hall-effect sensor. According to the analysis of the
16 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Method 12
Methods with the use of neural networks [36–42] make it possible
to use especially educated systems of artificial intellect for IM
diagnostics and forecasting of breakage occurrence. The signal of IM
consumed power in each phase is used as a diagnostic signal; it is
analyzed with the help of an artificial neural network. This method
enables the detection of damage in both IM electrical and mechanical
parts.
Drawback: the set of fuzzy logic rules is formed on the basis of the
performed experimental research, i.e., the assessment of the results is of
an individual character.
Method 13
The method is based on the analysis of electromagnetic torque
spectrum [43]. The method provides for the measurement of the stator
phase currents in IM idle mode, the determination of the
electromagnetic torque and the comparison of a considerable number of
the spectrum harmonics that change for a certain frequencies range.
Drawback: the necessity for taking into account instantaneous
losses in the motor steel for calculations, which cause the complexity of
electromagnetic torque calculation.
Method 14
According to this method [7], an alternating current electromagnet
with a magnetizing winding and a measuring winding is connected to
the tested bar. The diagnostic feature of the bar state consists in the
value of magnetomotive force at constant supply voltage.
The Contemporary State of the Problem of the Diagnostics … 17
Method 15
The method with the use of the wavelet-analysis of IM each phase
start currents [31] makes it possible to reveal broken rotor bars
independently of the level of load on the motor shaft. The analysis of
the signal of IM currents transient process is performed on the basis of
the algorithm that singles out the signal basic component according to
both amplitude and frequency.
Drawback: the impossibility of breakage location, realization only
in start modes, the complexity of the analysis due to the influence of IM
electromagnetic parameters measurement during start-up. The
considered methods of IM broken rotor bars diagnostics can be
classified according to the following signs. Depending on the
conditions according to which the diagnostics is carried out, the
methods of IM broken rotor bars diagnostics can be divided into several
groups:
electromagnetic torque
speed vibration
electric electromagnetic
magnetomotive force
electromotive force (of stator
phase, turn) magnetic induction of the rotor
field
active consumed power
signal of the external magnetic
field
steady dynamic
Figure 1.7. IM operation modes during the diagnostics of broken rotor bars.
20 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 1.8. Methods for information initial processing and methods for
making decisions.
2.1. Comparison with the threshold value of the signal. In this case
comparison with the threshold values causes low information
value.
2.2. Comparison of processes research results obtained on
standard models. In this case idealized standard models are
used; they do not take into account all the real physical
properties of the system.
2.3. Making a decision on the basis of logical rules. When logical
rules are made, it is necessary to take into consideration the
influence of many factors, which causes the complexity of the
analysis.
2.4. Making a decision with the use of artificial neural networks
and fuzzy logic. Neural networks have complex architecture; in
this case there occurs a necessity for instruction in the
networks or making expert rules.
2.5. Statistical analysis. The results of signals processing are
reliable only to a certain level of probability assigned by the
researchers before starting statistical data processing; as a rule,
to obtain results a considerable volume of processing is
required and breakage localization is impossible.
1.4. CONCLUSION
Idle Condition
When the diagnostics is performed in the idle condition, the motor,
as a rule, is to be withdrawn from the technological process.
Short-Circuit Condition
When research is performed in a short-circuit condition, there
appear limitations. They are caused by the fact that only equivalent
rotor resistance can be determined by the results of IM parameters
identification. That is why the diagnostics results are not very reliable
in such condition.
Start-Up Conditions
The diagnostics under start-up conditions is limited due to the
complexity of the analysis of the diagnostics results. Besides, the
analysis of diagnostics results is complicated because of the influence
of the efficiency of diagnostics depends on variation of IM
electromagnetic parameters during start-up.
Self-Running-Out Condition
Carrying out diagnostics in self-running-out condition, unlike the
above mentioned conditions of IM operation, has a number of
advantages:
Theoretical Foundation for the Research of Induction Motor … 27
Z2
k 1, k 0, 1, 2, 3 ... , (2.1)
p
Z2
k 1, k 1, 2, 3 ... . (2.2)
p
2 p2 , (2.3)
iZ1
1 1 i cos ;
i p
(2.4)
1 cos jZ 2 t ,
2 j 2
j p
iZ
(t ) 1 2 1 i cos 1 1
i p
jZ
j cos 2 ( 2t )
j p
(2.5)
1 Z
i j cos ( jZ 2 iZ1 ) j 2 2t
2 i j p p
Z
cos ( jZ 2 iZ1 ) j 2 2t .
p p
B,
r.u.
0.6
0.2
t, s
-0.2 0 0.008 0.016 0.024 0.032
-0.6
rotH j ;
B
rotE ;
t
B a H ; (2.7)
1
j E jout ;
divB 0,
2 Az 2 Az Az
J z _ out , (2.8)
x 2
y 2 t
Az
Bx ;
y
(2.9)
A
By z .
x
In the integral form the expression for the vector magnetic potential
determines its physical sense.
The circulation of the vector magnetic potential along a closed
contour is equal to magnetic flux Φ that pierces this contour:
Adl=Φ. (2.10)
The relation between the vector magnetic potential and the full flux
linkage of the stator winding phase is written by means of equation:
2l1w
ph
Ss S
Az dS , (2.11)
ph
where l1 – the active length of the stator; w – the number of turns in the
slot connected in series; S s – sectional area of the stator slot; Az – the
total arithmetic value of VMP in all the slots of the phase; S ph – the
total area of cross section of all phase coils connected in series.
By the law of electromagnetic induction (according to Maxwell’s
equations) [49–50], the stator winding phase EMF is equal to:
d ph t
e ph (t ) . (2.12)
dt
E,
V
2.5
0
0.01 0.02 0.03 0.04 0.05 t, s
-2.5
a
Figure 2.2. (Continued)
Theoretical Foundation for the Research of Induction Motor … 35
E,
V
0
0.01 0.02 0.03 0.04 0.05 0.06 t, s
-1
Figure 2.2. EMF in the measuring winding of a healthy IM (а) and an IM with three
broken rotor bars (b) in the motor self-running-out mode.
f p f1 (1 2ks), (2.13)
0
s . (2.14)
0
I,A
1
0.1
0.01
10 3
10 4
0 50 100 150 f, Hz
0 50 100 150 f, Hz
Figure 2.6. The drawbacks of Fourier transform at diagnostics of IM broken rotor bars.
biorthogonal wavelets
f
La log 2 s 1 log 2 2 f m t 1 , (2.15)
fm
a b
φ ψ
1
wavelet function psi
0.5
0
5 10 15 20 j
scaling
-0.5 function phi
Figure 2.9. Daubechies wavelets (а), Symlet wavelets (b) and Coiflets wavelets (c)
of the fourth order.
a
Analyzed Signal (length = 1001)
1
-1
100 200 300 400 500 600 700 800 900 1000
Ca,b Coefficients - Coloration mode : init + by scale + abs
127
120
113
106
99
92
85
78
71
64
57
50
43
36
29
22
15
8
1
b
Analyzed Signal (length = 1001)
1
-1
100 200 300 400 500 600 700 800 900 1000
Ca,b Coefficients - Coloration mode : init + by scale + abs
127
120
113
106
99
92
85
78
71
64
57
50
43
36
29
22
15
8
1
Figure 2.10. Sinusoid signal with a high-frequency component and its wavelet-
spectrum with the use of Daubechies bases (а), Symlet bases (b) and Coiflets bases (c).
Theoretical Foundation for the Research of Induction Motor … 51
In this case, the use of each of the wavelets as the basis allows
determination of a high-frequency component caused by the presence of
tooth harmonics. So, it can be stated that one of the mentioned bases
can be used for the analysis of sinusoidal signals containing a high-
frequency component.
The paper contains an analysis of the testing signal approximate to
the anticipated signal of EMF in IM stator with broken rotor bars. Let
us simulate testing signals of the stator windings coils EMF, taking into
consideration the fact that only the motor rotor has a stepped
appearance.
We assume that the stator winding coil group consists of three coils
whose EMF vectors are shifted by angle θ (Figure 2.11).
Let us assume that coils testing EMFs are calculated by the
following expressions:
etest1 t A1 t cos 2 t ;
etest 2 t A2 t cos 2 t ; (2.16)
etest 3 t A3 t cos 2 t 2 ,
E test 1
E test 2
E test 3
1.8
1.2
A1 ( t )
0.6 A2 ( t )
A3 ( t )
t, s
0 0.008 0.016 0.024 0.032
Figure 2.12. Functions assigning the rotor breakage and a stepped appearance.
0.6
t, s
0 0.008 0.016 0.024 0.032
-0.6
-1.8
Figure 2.13. The test signals of stator winding coils EMF, approximate to the
anticipated EMF signals.
54 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 2.14. The test signal of coil EMF etest1 with superimposed excitation that
corresponds to one broken bar and its wavelet-spectrum with the use: of Daubechies
wavelet (a), of Symlet wavelet (b), of Coiflets wavelet (c).
Theoretical Foundation for the Research of Induction Motor … 55
Figure 2.15. The test signal of coil EMF etest2 with superimposed excitation that
corresponds to one broken bar and its wavelet-spectrum with the use: of Daubechies
wavelet (a), of Symlet wavelet (b), of Coiflets wavelet (c).
56 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 2.16. The test signal of coil EMF etest3 with superimposed excitation that
corresponds to one broken bar and its wavelet-spectrum with the use: of Daubechies
wavelet (a), of Symlet wavelet (b), of Coiflets wavelet (c).
Theoretical Foundation for the Research of Induction Motor … 57
ˆ ei / 2 m0 / 2 ˆ / 2 ,
(2.18)
ˆ 2
1/ 2
m0 2 j ;
j 1
(2.19)
m0 – trigonometric polynomial:
N
1 ei
m0 L ,
2 (2.20)
sin 2 ξ
L ξ P ; (2.21)
2
P y – polynomial of the type:
N 1 N 1 k k
P y y . (2.22)
k 0 k
Figure 2.17. The test signal of stator winding coil group EMF etestΣ and its
wavelet-spectrum.
MMF MMF
a b
Figure 2.18. The simplified MMF curves of concentrated (а) and distributed
(b) windings.
Every phase of the stator winding consists of two coil groups, each
of which, in its turn, contains three coils. A winding phase coil is
formed by a group of turns connected in series and put into the same
slots.
Theoretical Foundation for the Research of Induction Motor … 61
Ec wc Et , (2.23)
where wc – the number of the coil turns; Et – the EMF of the winding
turn.
Ec 2
Ec1 Ec 3
/2 /2
q
q 3
0
a b
Ec 3
Eq
Ec 2
0
Ec1
Figure 2.19. Coil group in the magnetic field (а), coils EMF vectors (b) and vector
diagram for determination of EMF of the coil group (c) of IM stator winding.
62 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
. 6 .
E ph Eci , (2.24)
i 1
.
where E ci – the EMF of the i-th coil; i – coil number.
Coils EMFs are shifted in relation to each other. So, information
signs available in the signals of the EMF of every separate coil (caused
by the presence of broken rotor bars) are mutually superimposed when
EMFs are summed up.
Another factor influencing the amplitude of EMF signal is rotor
slots skewing. In a small machine in which increase of q is
complicated skewed slots are made to eliminate stepped harmonics.
Stator or rotor slots are located not parallel to the machine axis but
under a certain angle bev to it, which is called a skew angle.
The slots skew is assessed in linear bbev or relative bev dimensions
(Figure 2.20). These dimensions demonstrate by how many millimeters
or by what part of the step pitch along the air gap circular arc the
Theoretical Foundation for the Research of Induction Motor … 63
stator slots
bbev
rotor slots
The slots skew decreases the value of the EMF induced in the
winding turns.
Thus, the skew of rotor slots, as well as distribution of stator
winding by the slots, causes the decrease of EMF amplitude in the
winding.
2.7. CONCLUSION
Rotor geometric
parameters Currents in the rotor bars
For specification
Number of the rotor of distribution of currents
bars in the rotor bars at the
Circuit mathematical model on the basis moment of IM disconnection
Resistance of the of IM rotor equivalent circuit from the supply mains taking
rotor bars and into account the broken bars
ShCR relative position
where I 2(t 0) – the rotor current of the previous steady condition at the
moment of IM disconnection from the supply mains; k2 – rotor
coupling coefficient; I1(t 0) – the stator current of the previous steady
condition at the moment of IM disconnection from the supply mains.
At the following time moments the currents in the rotor bars change
according to exponential law with time constant:
L2 / R2 , (3.2)
It is known that, in case when one or several rotor bars are broken
in IM, the currents in the rotor bars redistribute. Besides, magnetic flux
distribution around the broken bar changes – the flux increases at one
end of the bar and decreases at the other end.
To determine the initial values of the current in the rotor bars at the
moment of motor disconnection from the supply mains an IM
mathematical model in a three-phase coordinate system is improved.
The model is based on the known IM mathematical model in a
three-phase coordinate system [66–67]. The improvement of the known
model consists in the fact that the rotor is simulated in the form of a
system of short-circuited bars. Besides, IM electromagnetic part is
presented as a system of magneto-connected windings located on the
stator and rotor.
The system of equations of electric balance of the stator circuit is of
the form:
d A (t )
u A (t ) i A (t ) RA dt ;
d B (t )
u B (t ) iB (t ) RB ; (3.3)
dt
d C (t )
uC (t ) iC (t ) RC dt ,
d a t
0 ia t Ra ;
dt
d b t
0 ib t Rb ; (3.4)
dt
d c t
0 ic t Rc ,
dt
where ia t , ib t , ic t – rotor currents; a t , b t , c t – rotor
phases flux linkages; Ra Rb Rc Rr – rotor phases resistances.
Taking into account the number of bars of the analyzed IM rotor the
system of equations of electric balance of the rotor circuit takes the
form:
d 1 (t )
0 i1 (t ) R1 dt ;
0 i (t ) R d 2 (t ) ;
2 2
dt
0 i (t ) R d 3 (t )
3 3 ;
dt (3.5)
...
0 i (t ) R d Z 1 (t ) ;
Z 1 Z 1
dt
0 i (t ) R d Z ( t )
,
Z Z
dt
M AB M AC M BC M S (3.7)
M A1 M cos ;
2
M A2 M cos ;
Z
2
M A3 M cos 2 ;
Z
(3.9)
...
2
M A Z 1 M cos Z 2 ;
Z
2
M AZ M cos Z 1 ,
Z
2
M B1 M cos ;
3
2 2
M B 2 M cos ;
3 Z
2 2
M B 3 M cos 2 ;
3 Z (3.10)
...
2 2
M B Z 1 M cos Z 2 ;
3 Z
2 2
M BZ M cos Z 1 .
3 Z
2
M C1 M cos ;
3
2 2
M C 2 M cos ;
3 Z
2 2
M C 3 M cos 2 ;
3 Z (3.11)
...
2 2
M C Z 1 M cos Z 2 ;
3 Z
2 2
M CZ M cos Z 1 .
3 Z
2
A LAi A M s iB M s iC Mi1 cos Mi2 cos
Z
2 2
Mi3 cos 2 ... MiZ 1 cos Z 2 (3.12)
Z Z
2
MiZ cos Z 1 .
Z
2
A LA M s i A Mi1 cos Mi2 cos
Z
2 2
Mi3 cos 2 ... MiZ 1 cos Z 2 (3.13)
Z Z
2
MiZ cos Z 1 .
Z
Mathematical Models for the Research of the Method … 77
2 2 2
B LB M s iB Mi1 cos Mi2 cos
3 3 Z
2 2 2 2
Mi3 cos 2 ... MiZ 1 cos Z 2
3 Z 3 Z
2 2
MiZ cos Z 1 .
3 Z
2 2 2
C LC M s iC Mi1 cos Mi2 cos
3 3 Z
2 2 2 2
Mi3 cos 2 ... MiZ 1 cos Z 2
3 Z 3 Z
2 2
MiZ cos Z 1 .
3 Z
d A
dtA M dt1 cos M i1 sin
di di
LA M s
dt
di2 2 2
M cos M i2 sin
dt Z Z
di 2 2
M 3 cos 2 M i3 sin 2
dt Z Z
diZ 1 2
... M cos Z 2
dt Z
2
M iZ 1 sin Z 2
Z
di 2 2
M Z cos Z 1 M iZ sin Z 1 ,
dt Z Z
dB di di 2 2
LB M s B M 1 cos M i1 sin
dt dt dt 3 3
di 2 2 2 2
M 2 cos M i2 sin
dt 3 Z 3 Z
di3 2 2 2 2
M cos 2 M i3 sin 2
dt 3 Z 3 Z
diZ 1 2 2
... M cos Z 2
dt 3 Z
2 2
M iZ 1 sin Z 2
3 Z
di 2 2 2 2
M Z cos Z 1 M iZ sin Z 1 .
dt 3 Z 3 Z
d C di di 2 2
LC M s C M 1 cos M i1 sin
dt dt dt 3 3
di 2 2 2 2
M 2 cos M i2 sin
dt 3 Z 3 Z
di 2 2 2 2
M 3 cos 2 M i3 sin 2
dt 3 Z 3 Z
di 2 2
... M Z 1 cos Z 2
dt 3 Z
2 2
M iZ 1 sin Z 2
3 Z
di 2 2
M Z cos Z 1
dt 3 Z
2 2
M iZ sin Z 1 .
3 Z
Mathematical Models for the Research of the Method … 79
2 2
1 L1 M r i1 Mi2 cos Mi3 cos 2
Z Z
2 2
... MiZ 1 cos Z 2 MiZ cos Z 1
Z Z
2 2
Mi A cos MiB cos MiC cos .
3 3
2 2
2 L2 M r i2 Mi1 cos Mi3 cos
Z Z
2 2
Mi4 cos 2 ... MiZ 1 cos Z 3
Z Z
2 2
MiZ cos Z 2 Mi A cos
Z Z
2 2 2 2
MiB cos MiC cos .
3 Z 3 Z
…
2 2
Z LZ M r iZ Mi1 cos Mi2 cos 2
Z Z
2 2
... MiZ 2 cos Z 2 MiZ 1 cos Z 1
Z Z
2 2 2 2 2
Mi A cos MiB cos MiC cos .
Z 3 Z 3 Z
80 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
The value of shift angles between the rotor bars can be written
down in the following general form:
N 1 K ph , (3.14)
d 1 di di 2 2
L1 M r 1 M 2 cos M i2 sin
dt dt dt Z Z
di 2 2
M 3 cos 2 M i3 sin 2
dt Z Z
diZ 1 2 2
... M cos Z 2 M iZ 1 sin Z 2
dt Z Z
diZ 2 2
M cos Z 1 M iZ sin Z 1
dt Z Z
di di 2
M S A cos M S i A sin M S B cos
dt dt 3
2 diC 2
M S iB sin MS cos
3 dt 3
2
M S iC sin .
3
The equations of electrical balance for all the rotor bars are obtained
analogously.
Stator phases voltage:
Mathematical Models for the Research of the Method … 81
UA=Umcosγ;
UB=Umcos(γ+2π/3); (3.15)
UC=Umcos(γ–2π/3);
2p
Me C B iA A C iB B A iC . (3.16)
3 3
d 1
p Me Mc . (3.17)
dt J
Taking into account the above said, the block diagram of the
mathematical model will be of the form (Fig. 3.2). The block diagrams
of separate blocks of IM mathematical model are given in appendix B.
Figure 3.2. The block diagram of IM mathematical model for the determination of
currents in rotor bars at the moment of motor disconnection from the supply network.
82 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
The above given expressions are true for IM operating mode. In the
motor self-running-out mode the flux linkages for every phase of the
stator will depend only on the value of currents in the bars of the rotor
that continues rotating:
2 2
A Mi1 cos Mi2 cos Mi3 cos 2
Z Z
2 2
... MiZ 1 cos Z 2 MiZ cos Z 1 .
Z Z
2 2 2
B Mi1 cos Mi2 cos
3 3 Z
2 2
Mi3 cos 2 ...
3 Z
(3.18)
2 2
MiZ 1 cos Z 2
3 Z
2 2
MiZ cos Z 1 .
3 Z
2 2 2
C Mi1 cos Mi2 cos
3 3 Z
2 2 2 2
Mi3 cos 2 ... MiZ 1 cos Z 2
3 Z 3 Z
2 2
MiZ cos Z 1 .
3 Z
2 2
1 Lr M r i1 Mi2 cos Mi3 cos 2
Z Z
2 2
... MiZ 1 cos Z 2 MiZ cos Z 1 .
Z Z
2 2
2 Lr M r i2 Mi1 cos Mi3 cos
Z Z
2 2
Mi4 cos 2 ... MiZ 1 cos Z 3
Z Z
2
MiZ cos Z 2 .
Z
…
2 2
Z Lr M r iZ Mi1 cos Mi2 cos 2
Z Z
2 2
... MiZ 2 cos Z 2 MiZ 1 cos Z 1 .
Z Z
The mathematical model is rather extensional and contains a lot of
cross links, but it has a number of advantages:
Ir, A
1.035
0.623
Ir, A 0.212
20
Figure 3.3. The transient processes of currents in IM rotor bars at no-load start and
the motor self-running-out mode ( t 0.5 s).
Ir, A
1.608
Ir, A
2.824
20
20
Figure 3.4. The transient processes of currents in IM rotor bars at no-load start,
subsequent load-on ( t 0.3 s) and the motor self-running-out mode ( t 0.5 s).
Mathematical Models for the Research of the Method … 85
Ir,A
0 10 20 30 Z
-1
Ir,A
0 10 20 30 Z
-1
Figure 3.6. The distribution of currents in the rotor bars of an IM with one broken
bar at the moment of motor disconnection from the supply mains.
86 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
E1 Zb1
Zscr1 Zscr1
E2 Zb2
Zscr2 Zscr2
E3 Zb3
Zscr3 Zscr3
E34
Zb34
Zscr34 Zscr34
4
bar resistance Rb.r. 1.985 10 Ohm;
4
bar inductive reactance X b 2.158 10 Ohm;
5
short-circuited ring resistance Rscr 1.472 10 Ohm.
Rotor bar EMF changes by the sinusoid law and is written down in
a complex form as:
E I b Zb , (3.19)
Zb Rb.r. jX b . (3.20)
model can be used for modeling rotors for IMs of different power with
different number of broken bars. The distribution of currents in the
rotor bars at the moment of motor disconnection from the supply mains
for a healthy IM and an IM with one, two and three broken bars is
shown in Figures 3.9.–3.12.
1 2
14
Ib, A
15
10
5
0
0 5 10 15 20 25 30 35 Z2
-5
-10
-15
Figure 3.9. The distribution of currents I b in the rotor bars for a healthy IM at
the moment of motor disconnection from the supply mains.
Ib, A
15
10
5
0
5 10 15 20 25 30 Z2
-5
one broken bar
-10
-15
Figure 3.10. The distribution of currents I b in the rotor bars for an IM with one
broken bar at the moment of motor disconnection from the supply mains.
Mathematical Models for the Research of the Method … 89
Ib, A
15
10
5
0
0 5 10 15 20 25 30 Z2
-5
-10 two broken bars
-15
Figure 3.11. The distribution of currents I b in the rotor bars for an IM with two
broken bars at the moment of motor disconnection from the supply mains.
Figure 3.12. The distribution of currents I b in the rotor bars for an IM with three
broken bars at the moment of motor disconnection from the supply mains.
EMF signals in the stator windings (for two complete rotor revolutions)
in the motor self-running-out mode. These signals are used for the
assessment of electromagnetic field distortion. The EMF signals of one
active side of the coil, the coil, the coil group and the winding phase of
the stator of a healthy IM, type АIR80V4U2, obtained as a result of the
calculation, are shown in Figures 3.16–3.19.
а b
Figure 3.15. The distribution of magnetic flux lines at the initial moment of the
motor disconnection from the supply mains for a healthy IM (а) and for an IM
with three broken rotor bars (b).
96 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Е, V
20
10
10
Figure 3.16. The EMF signal of one active side of the coil of a healthy IM
stator winding.
Е, V
40
20
-20
Figure 3.17. The EMF signal of the coil of a healthy IM stator winding.
Е, V
100
100
Figure 3.18. The EMF signal of the coil group of a healthy IM stator winding.
Mathematical Models for the Research of the Method … 97
Е, V
200
100
100
10
10
20
Figure 3.20. The EMF signal of one active side of the coil of the stator winding of IM
type АIR80V4U2 with one broken rotor bar.
20
20
Figure 3.21. The EMF signal of the coil of the stator winding of IM type АIR80V4U2
with one broken rotor bar.
98 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 3.22. The EMF signal of the coil group of the stator winding of IM type
АIR80V4U2 with one broken rotor bar.
Е, V
200
100
100
200
Figure 3.23. The EMF signal of the stator winding phase of IM type АIR80V4U2
with one broken rotor bar.
Then modeling for IM with one broken rotor bar was performed.
The obtained EMF signals of one active side of the coil, the coil, the
coil group and the stator winding phase of IM, type АIR80V4U2 with
one broken rotor bar are shown in Figure 3.20–3.23.
The results of the analysis revealed that the EMF signal of one
active side of the coil contains both tooth kink and signal shape
distortions caused by the presence of broken rotor bars. A visual
analysis showed that in the EMF signal of the coil (Figure 3.20) the
information signs that manifest in the signal shape distortion, become
Mathematical Models for the Research of the Method … 99
less noticeable and in the EMF signals of the coil group and the
winding phase (Figure 3.22–3.23) - they are practically absent.
Modeling of IM with two adjacent broken bars was performed in an
analogous way. The results of modeling are shown in Figures 3.24–
3.27.
The analysis of the results revealed that with the growth of the
number of broken bars (adjacent), information signs that are available
in the EMF signal of one active side of the coil intensify and manifest
in greater distortion of EMF signal shape.
10
10
Figure 3.24. The EMF signal of one active side of the coil of the stator winding of IM
type АIR80V4U2 with two broken rotor bars.
20
20
Figure 3.25. The EMF signal of the coil of the stator winding of IM type АIR80V4U2
with two broken rotor bars.
100 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Е, V
50
50
Figure 3.26. The EMF signal of the coil group of the stator winding of IM type
АIR80V4U2 with two broken rotor bars.
Е, V
200
100
100
Figure 3.27. The EMF signal of the stator winding phase of IM type АIR80V4U2
with two broken rotor bars.
10
20
Figure 3.28. The EMF signal of one active side of the coil of the stator winding
of IM type 4АN200L2U3 with one broken rotor bar.
20
40
Figure 3.29 – The EMF signal of the coil of the stator winding of IM type
4АN200L2U3 with one broken rotor bar.
Е, V
200
100
200
Figure 3.30. The EMF signal of the stator winding phase of IM type 4АN200L2U3
with one broken rotor bar.
e t Eme
t / a
Eres sin pte
t / s
0 , (3.21)
Е, V Е, V
1 2
200 150
2
150 1
100
100
50 50
Figure 3.31. The fragments of the initial (1) and approximated (2) signals of EMF
in the stator windings of a healthy IM (а) and an IM with three broken rotor bars (b).
Mathematical Models for the Research of the Method … 103
Е, V 2
40
1
20
0 0.02 0.04 t, s
-20
Figure 3.32. The differences of the approximated and calculated EMF signals
in the stator windings: 1 – for a healthy IM, 2 – for an IM with three broken rotor
bars.
3.5. CONCLUSION
The performed research (p. 2.4) made it possible to find out that
orthogonal wavelets with a compact support can be used for the
analysis of sinusoid-shape wave signals. To reveal the local features of
EMF signal in the stator winding taking into account wavelets
properties a wavelet analysis with the use of the Symlet wavelet was
carried out.
As stated above, the EMF signal of the stator winding phase does
not contain explicit signs typical of broken rotor bars, unlike EMF
signals of one active side of the coil. To confirm this fact an analysis of
the obtained signals for an IM with one broken rotor bar was performed
with the use of continuous WT (Figures 4.1–4.4).
106 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 4.1. The EMF signal of one active side of the coil of the stator winding of an IM
with one broken rotor bar and its wavelet-spectrum.
Figure 4.2. The EMF signal of the stator winding coil of an IM with one broken rotor
bar and its wavelet-spectrum.
Figure 4.3. The EMF signal of the stator winding coil group of an IM with one broken
rotor bar and its wavelet-spectrum.
The Method of Induction Motor Broken Rotor Bars … 107
Figure 4.4. The EMF signal of the stator winding phase of an IM with one broken rotor
bar and its wavelet-spectrum.
Figure 4.5. The EMF signal of one active side of the coil of the stator winding of an IM
with two broken rotor bars and its wavelet-spectrum.
Figure 4.6. The EMF signal of the stator winding coil of an IM with two broken rotor
bars and its wavelet-spectrum.
108 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 4.7. The EMF signal of the stator winding coil group of an IM with two broken
rotor bars and its wavelet-spectrum.
Figure 4.8. The EMF signal of the stator winding phase of an IM with two broken rotor
bars and its wavelet-spectrum.
Figure 4.9. The EMF signal of one active side of the coil of the stator winding of an IM
with three broken rotor bars and its wavelet-spectrum.
The Method of Induction Motor Broken Rotor Bars … 109
Figure 4.10. The EMF signal of the stator winding coil of an IM with three broken
rotor bars and its wavelet-spectrum.
Figure 4.11. The EMF signal of the stator winding coil group of an IM with three
broken rotor bars and its wavelet-spectrum.
Figure 4.12. The EMF signal of the stator winding phase of an IM with three broken
rotor bars and its wavelet-spectrum.
110 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
k a
K a a
, (4.1)
l
Ka
200
100
-100
-200
healthy IM
one broken rotor bar
three broken rotor bars
K a
Figure 4.14 contains function K * a 100 % created in relative
amax
units reduced to the maximum value of the scale amax 64 for an IM
with broken rotor bars. Research demonstrated that the value of
coefficient K a in the presence of several broken bars grows
approximately proportionally to their number.
112 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
K
19.2 % 1 – one broken rotor bar
a , 2 – two broken rotor bars
%
2 2
13.7 8.9 %
1
1
4
0 0.01 0.02 0.03 t, s
-5.8
a b
Figure 4.15. The surfaces of the coefficients of the wavelet-expansion of EMF signals
in the stator windings: (а) – for a healthy IM, (b) – for an IM with three broken
rotor bars.
EA
EC E AB
ECA EB
120
EA
120
EB
E BC
EC
Figure 4.16. The vector diagram of EMF with the stator windings connection
according to a “star” scheme.
K
e k e k t TZ e k t ek z k E z , (4.2)
k 0
K K K
E z e k zk e k n zk z n e k n z k n z n E z .
k 0 k 0 k 0
E, V phase EMF
1st stage
100
nd
2 stage -100 -100
3rd stage
EMF of the coil active sides 1 E, EMF of the coil active sides 2
E,
V V
broken bars
Division of the coil 10
10
EMF signal into
0 0 0.01 0.02 0.03 0.04 t, s
0.01 0.02 0.03 0.04 t, s
signals of EMF of the -10
-10
coils active sides
Figure 4.17. The block diagram of the decomposition of the signal of the EMF of IM
stator winding phase.
E ph z
k
z
Eq2 z
E q1 z Eq2 z
Figure 4.19. The comparison of the initial and detached signals of the EMF of
the stator winding coil group of an IM with one broken rotor bar.
Figure 4.20. The Initial signal of the EMF of the coil group
and its wavelet-spectrum.
The Method of Induction Motor Broken Rotor Bars … 119
Figure 4.21. The detached signal of the EMF of the coil group and its wavelet-
spectrum.
The Decomposition of the Signal of the EMF of the Coil Group into
the Signals of Coils EMF
One of the signals of the EMF of the coil group Eq z obtained as
a result of the decomposition is divided into the signals of coils EMF.
For the analyzed IM the decomposition of the signal of the EMF of the
coil group Eq1 z into the signals of the EMF of the coils Ecm z was
performed, where m – the number of coils forming a coil group (for
the analyzed IM m 3 ), k – the number of increments corresponding
to the angle of shift between the coils (for the analyzed IM at this stage
of decomposition k 2 ) (Fig. 4.22).
Z1
E q1 z
Ec2 z k k
z z
E c1 z Ec2 z E cm z
E cm z
Figure 4.22. The block diagram of the decomposition of the signal of the EMF of
the coil group.
120 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
20
E, V 10
initial signal detached signal
40
0.04 0.042 0.044
20
0
0.02 0.04 0.06 t, s
-20
-40
Figure 4.23. The comparison of the initial and detached signals of the EMF of the
stator winding coil of IM with one broken rotor bar.
Figure 4.24. The initial signal of coil EMF and its wavelet-transform.
Figure 4.25. The detached signal of coil EMF and its wavelet-transform.
The Method of Induction Motor Broken Rotor Bars … 121
The analysis of the obtained signals of the coil EMF revealed that
the distortion of the shape of the detached EMF signal certifies the
presence of broken rotor bars, which is confirmed by corresponding
research in Chapter 3.
To reveal the information signs of the breakage of the initial and
detached signals of EMF of the coil a continuous wavelet-transform is
performed. The results of the CWT are shown in Figures 4.24–4.25.
The obtained results demonstrated that the wavelet-spectrum of the
detached signal of the IM stator winding coil EMF contains typical
sections corresponding to the location of broken rotor bars. At the same
time the “duplication” of the said sections can be seen on the wavelet-
spectrum of the detached signal of the coil EMF (Figure 4.25), which is
also typical of the coil EMF signal obtained according to the results of
calculation of IM electromagnetic field (Figure 4.2). So, the obtained
results of the wavelet-transform (Figure 4.25) coincide with the results
of the wavelet-expansion for the initial signal of EMF of the stator
winding coil (Figure 4.2).
Thus, at the stage of singling out the signals of the coil EMF it is
possible to come to the conclusion that the use of the method for
decomposition of the winding phase EMF signal into the signals of the
EMF of its elements allows the determination of the information signs
of broken rotor bars.
The Decomposition of the Coil EMF Signal into the Signals of the
EMF of Two Active Sides of the Coil
The final stage of the decomposition of the signal of EMF of the
stator winding phase consists in singling out the signals of EMF of two
active sides of the coil Et1 z and Et 2 z from the signal of EMF of
coil Ec1 z (Figure 4.26).
122 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
E c1 z
k
z
Et 2 z
E t1 z Et 2 z
Figure 4.26. The block diagram of the decomposition of the coil EMF signal.
E, V detached signal
20 initial signal
10
-20
Figure 4.27. The comparison of the initial and detached signals of the EMF of one
active side of the coil of the stator winding of an IM with one broken rotor bar.
Figure 4.28. The initial signal of the EMF of one active side of the coil and its wavelet-
transform.
The Method of Induction Motor Broken Rotor Bars … 123
Figure 4.29. The detached signal of the EMF of one active side of the coil and its
wavelet-transform.
The results of the research for step-by-step singling out the signals
of the EMF of the stator winding elements at the relative position of the
broken rotor bars at a distance of the space angle of br.b. 84.7 are
given in Figures 4.30–4.34.
Е, V initial signal
detached signal
100
0
0.02 0.03 0.04 0.05 t, s
-100
Figure 4.30. The comparison of the initial and detached signals of the EMF of the
stator winding coil group of IM with two broken rotor bars ( br .b. 84.7 ).
Е, V detached signal
50 initial signal
25
0.04 0.08 t, s
-25
-50
Figure 4.31. The comparison of the initial and detached signals of the EMF of the
stator winding coil of IM with two broken rotor bars ( br .b. 84.7 ).
The Method of Induction Motor Broken Rotor Bars … 125
Figure 4.32. The detached signal of the EMF of the stator winding coil of an IM
with two broken rotor bars ( br .b. 84.7 ) and its wavelet-spectrum.
Е, V
initial signal detached signal
20
10
-20
Figure 4.33. The comparison of the initial and detached signals of the EMF
of one active side of the stator winding coil of an IM with two broken rotor
bars ( br .b. 84.7 ).
Figure 4.34. The detached signal of the EMF of one active side of the stator
winding coil of an IM with two broken rotor bars ( br .b. 84.7 ) and its wavelet-
spectrum.
126 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Figure 4.35. The detached signal of the EMF of one active side of the coil of the
stator winding with two broken rotor bars ( br .b. 31.8 ) and its wavelet-spectrum.
Figure 4.36. The detached signal of the EMF of one active side of the coil of
the stator winding with two broken rotor bars ( br .b. 169.4 ) and its wavelet-
spectrum.
The Method of Induction Motor Broken Rotor Bars … 127
of the signal of the winding phase EMF (paragraph 4.2.1). In this case
the function of the average value of the sum of wavelet-expansion
coefficients for the medium frequency area is used as an initial signal
(Figure 4.13). The signals used as a result of decomposition are shown
in Fiure 4.37.
Ka
200
100
-200
-100
Ka
10 detached signal initial signal
of the coil group of the phase
5
-10
0
0.028 0.036 0.044 0.052 t, s
-5
Figure 4.39. The functions of the average values of the sums of the wavelet-expansion
coefficients for the area of the medium frequencies of the signals of the EMF of the
phase and the coil group.
Ka
6
2
4
1
2
0 t, s
0.028 0.036 0.044 0.052
-2
-4
-6
Figure 4.40. The functions of the average values of the sums of the wavelet-expansion
coefficients for the area of the medium frequencies of the coil EMF signal.
of the value of the amplitude of the surge reflecting the degree of rotor
breakage (analogously to Figure 4.14).
Ka
6
4
2
2
1
0
0.028 0.036 0.044 0.052 t, s
-2
-4
-6
Figure 4.41. The functions of the average values of the sums of the wavelet-expansion
coefficients for the area of the medium frequencies of the signal
of the EMF of one active side of the coil.
4.3. CONCLUSION
Beginning
End
Characteristic Value
Pass band 85 kHz
Input voltage ±200 mV (±300 mV max)
Input voltage shift, max. 0.9 mV
Nonlinearity, max 0.3%
Measurement accuracy 1%
Initial voltage:
- minimum 1.18 V
- medium 2.39 V
- maximum 3.61 V
Thermal drift 4.6 mcV/°С
Supply voltage +5V
Input resistance 530 kOhm
136 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
~ 380 V
QF1
Ua Ub Uc
PC ADC BVS
A В C
DCG
IM
RDCG
EW
UEW
V1
DA 1
DC/DC 5/5
XS 4
1 2 4 6
C1 № Circuit
1 U1out
R5
2 U2out
+ 3 U3out
C4
C2 C3 4 U4out
C5
5 +12V
DA 3
6 -12V
DA 2 1 8
R1 7 GND
1 8 2 7
R3 UD140 8 +5V
2 7 3 U1408А 6
XS 2 HCPL
3 7800 6 4 5
№ Circuit R2 4 5
R4
1 U1.1in
C6
2 U1.2in
R6
3 U2.1in
4 U2.2in
5 U3.1in
6 U3.2in V2
7 U4.1in
DA 4
8 U4.2in
1
DC/DC
V3 5/± 12 V 4
V4
Figure 5.7. The oscillogram of the stator phase voltages of a healthy IM.
U, V
I, A
500 IM disconnection
Ua Ub Uc
Ib
Ia
10.262 10.265 10.267 t, s
Ic
-500
Figure 5.8. Voltage and current during IM disconnection from the supply mains.
140 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
E, V
100
0
0 0.05 0.01 0.015 t, s
-100
Figure 5.9. The experimental signal of the EMF of the stator winding phase
of IM with broken rotor bars.
Figure 5.10. The experimental signal of the EMF of the stator winding phase
of IM with rotor broken bars and its wavelet-spectrum.
The Experimental Verification of the Method … 141
E,
V
broken bars
7.5
-7.5
-15
Figure 5.11. The EMF signal of one active side of the coil, singled out
from the experimental signal of the EMF of the stator winding phase.
Figure 5.12. The EMF signal of one active side of the coil, singled out from the
experimental signal of the EMF of the stator winding phase, and its wavelet-spectrum.
142 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
So, the analysis of the obtained results revealed, that the number of
broken bars about 10% is critical for IM. The analysis of IM starting
torque multiplication factor Ì st showed that the value of this index
decreases with the increase of the number of broken bars. Besides,
144 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Thus, in the presence of three broken rotor bars (9%) the value of
losses in the stator and rotor windings grows almost by 1.5 times; IM
load-carrying capacity reduces to the value of 0.95; the motor
acceleration time is twice as much; the temperature of heating of the
stator windings and the rotor bars reaches the values approaching the
146 M. V. Zagirnyak, Zh. Iv. Romashykhina and A. P. Kalinov
Table 5.4. The relation of losses in the stator and rotor windings
to the rated value of losses
Thus, with 3% of broken rotor bars the value of heat losses grows
by 4%, which causes the increase of windings temperature also by 4%.
So, taking into account the growth of windings temperature, the
insulation durability is:
5.4. CONCLUSION
The developed hard- and know are enable the efficient realization
of the proposed methods for the diagnostics of induction motor broken
rotor bars.
The experimental research of the method for the diagnostics of IM
broken rotor bars at the devised laboratory computer-aided stand
confirmed the efficiency of the proposed diagnostics method.
It is demonstrated that, when there are about 10% of broken rotor
bars, the stator and rotor windings are greatly overheated, which is a
reason for the withdrawal of the motor from the technological process
and the necessity for its repair.
CONCLUSION
APPENDIX A
Parameter Value
Rated power, kW 1.5
Rated rotational speed, rev/min 1395
Efficiency,% 77
Power coefficient, r.u. 0.81
Rated voltage, V 220/380
Rated current, А 6.3/3.6
Winding connection Δ/Υ
Stator windings resistive impedance, Ohm 5
Rotor windings reduced resistive impedance, Ohm 3.69
Stator winding inductive reactance, Ohm 4.35
Rotor windings reduced inductive reactance, Ohm 4.01
164 Appendices
Parameter Value
Rated power, kW 75
Rated rotational speed, rev/min 2940
Efficiency,% 92
Power coefficient, r.u. 0.9
Rated voltage, V 220/380
Rated current, А 79.5/54.5
Winding connection Δ/Υ
Parameter Value
Stator windings resistive impedance, Ohm 0.14
Rotor windings reduced resistive impedance, Ohm 0.075
Stator winding inductive reactance, Ohm 0.585
Rotor windings reduced inductive reactance, Ohm 0.643
Appendices 165
APPENDIX B
Figure B1. The test signal of the EMF of coil etest1 with alias disturbance, which
correspond to two broken bars and their corresponding wavelet-spectra with the use: of
Daubechies wavelet (a), of Symlet wavelet (b), of Coiflets wavelet (c).
Appendices 167
Figure B2. The test signal of the EMF of coil etest2 with alias disturbance, which
correspond to two broken bars and their corresponding wavelet-spectra with the use: of
Daubechies wavelet (a), of Symlet wavelet (b), of Coiflets wavelet (c).
168 Appendices
Figure B3. The test signal of the EMF of coil etest3 with alias disturbance, which
correspond to two broken bars and their corresponding wavelet-spectra with the use: of
Daubechies wavelet (a), of Symlet wavelet (b), of Coiflets wavelet (c).
Appendices 169
Figure B4. The test signals of the EMF of coils with alias disturbance, which
correspond to two broken bars relatively located at the angle of 95o and their
corresponding wavelet-spectra with the use: of Daubechies wavelet (a), of Symlet
wavelet (b), of Coiflets wavelet (c).
170 Appendices
APPENDIX C
y
1
[y]
cos K 34
From 1 1 8
1
Gain 8
K 34 cos 2
[y]
2
From 18
cos
cos
K 34 2
3
3
Gain 3 K 34 9
cos
4 Gain 10
cos
[y] 4
From 2 [y]
From 14
K 34
3
Gain 1
cos
10
K 34
[y]
Gain 9
From 4
cos
[y]
Figure C1. A block diagram for simulating the shift angles between the stator From 15
K 34
phase A winding and the rotor
4
bars.
Gain 4
K 34 11
[y]
Gain 7
From 10
cos
5
[y]
5
K 34 5 From 17
cos 6
Gain 6 6
[y]
From 6 cos
12
K 34
Gain 12
cos cos
6 [y]
K 34 7
7 From 22
Gain 5
[y]
K 34 7
From 24
Gain 2
[y]
From 9
Appendices 171
2
34 PsiA
UA
dPsiA /dt
PsiA
1 IA
1/Ls
s
1
Integrator 1 Gain 3 IA
Gain 2
Rs
Gain 1
I1
k1 1
I2
Product 4
2
Gain 4
Product 1 I3
Lm 3
Product 2
I4
4
Gain 5 Product 3 I5
5
1/11
Product 5 I6 cosy
6
1
Product 6 I7
7 2
3
I8
Product 8
8 4
5
I9
Product 7
9 6
I10 7
Product 9 10
8
9
Product 10 I11
11 10
I12 11
Product 11 12
12
I13 13
Product 12 13
14
Product 13 I14 15
14
16
y
Product 14 17 y 35
I15
18
15
Product 15 19
I16 20
Product 16 16
21
I17 22
17
Product 17 23
I18
18 24
Product 18
I19 25
19
26
Product 19
27
I20
Product 20 20 28
I21 29
21
Product 21 30
I22
22 31
Product 22 32
I23 33
Product 23 23
I24
24
Product 24
I25
25
Product 25
I26
26
Product 26 I27
27
I28
Product 27
28
I29
Product 28
29
I30
Product 29
30
I31
Product 30
31
I32
Product 31 32
I33
Product 32 33
Product 33
APPENDIX D
I1n=-6.176643337
I2n=-1.048275981
I3n=4.334702638
I4n=9.252529795
I5n=0
I6n=15.01380419
I7n=15.08149291
I8n=13.23742675
I9n=9.726220843
I10n=5.010318713
I11n=-0.2884278
I12n=-5.470197361
I13n=-9.850188358
I14n=-12.85026538
I15n=-14.07665185
I16n=-13.37301515
I17n=-10.8416652
I18n=-6.829928836
I19n=-1.883488046
I20n=3.327053568
I21n=8.096487157
I22n=11.780023
I23n=13.88020768
I24n=14.11400968
I25n=12.45102255
I26n=9.117630245
Appendices 173
I27n=4.566569145
I28n=-0.584012643
I29n=-5.633788235
I30n=-9.894566936
I31n=-12.78299851
I32n=-13.89921019
I33n=-13.08084526
I34n=-10.42532979
T=0.159
tn=0
tk=(0.041*2)
k=(360*2) the number of models for calculation is assigned
alfa=0
n=1
h=tk/k
for t=tn,tk,h do
open("34.FEM")
mi_saveas("temp.fem")
mi_selectgroup(37)
mi_move_rotate(0,0,alfa)
I1=I1n*exp(-t/T)
I2=I2n*exp(-t/T)
I3=I3n*exp(-t/T)
…
I34=I34n*exp(-t/T)
mi_addcircprop("1",I1,1) the values of current in the rotor bars
are assigned
mi_addcircprop("2",I2,1)
mi_addcircprop("3",I3,1)
…
mi_addcircprop("34",I34,1)
alfa=alfa+1
174 Appendices
mi_analyze(1)
mi_loadsolution()
mo_groupselectblock(1) stator slots are chosen
A1=mo_blockintegral(1) the values of vector magnetic
potential in the chosen elements of
the stator winding are calculated
handle=openfile("AD1.txt","a");
write(handle,A1,"\n");
closefile(handle);
mo_clearblock()
mo_groupselectblock(2)
A2=mo_blockintegral(1)
handle=openfile("AD2.txt","a");
write(handle,A2,"\n");
closefile(handle);
mo_clearblock()
mo_groupselectblock(3)
A3=mo_blockintegral(1)
handle=openfile("AD3.txt","a");
write(handle,A3,"\n");
closefile(handle);
mo_clearblock()
mo_groupselectblock(10)
A10=mo_blockintegral(1)
…
mo_groupselectblock(30)
A30=mo_blockintegral(1)
handle=openfile("AD30...txt","a");
write(handle,A30,"\n");
closefile(handle);
mo_clearblock()
n=n+1
end
Appendices 175
APPENDIX E
broken bars
Е, V
10
10
Figure E1. The signal of the EMF of one active side of the coil of the stator winding of
АIR80V4U2 IM with three broken rotor bars.
broken bars
Е, V
20
20
Figure E2. The signal of the EMF of the coil of the stator winding of АIR80V4U2 IM
with three broken rotor bars.
176 Appendices
Е, V
50
50
Figure E3. The signal of the EMF of the coil group of the stator winding of
АIR80V4U2 IM with three broken rotor bars.
Е, V
200
100
100
Figure E4. The signal of the EMF of the phase of the stator winding of АIR80V4U2
IM with three broken rotor bars.
Е, V broken bars
20
10
20
Figure E5. The signal of the EMF of one active side of the coil of the stator winding of
4АN200L2U3 IM with two broken rotor bars.
Appendices 177
Е, V broken bars
40
20
40
Figure E6. The signal of the EMF of the coil of the stator winding of 4АN200L2U3 IM
with two broken rotor bars.
Е, V
200
100
200
Figure E7. The signal of the EMF of the phase of the stator winding of 4АN200L2U3
IM with two broken rotor bars.
Е, V broken bars
20
10
20
Figure E8. The signal of the EMF of one active side of the coil of the stator winding
of 4АN200L2U3 IM with three broken rotor bars.
178 Appendices
Е, V broken bars
40
20
20
Figure E9. The signal of the EMF of the coil of the stator winding of 4АN200L2U3 IM
with three broken rotor bars.
Е,V
200
100
200
Figure E10. The signal of the EMF of the phase of the stator winding of 4АN200L2U3
IM with three broken rotor bars.
AUTHORS’ CONTACT INFORMATION
Mykhaylo Zagirnyak
Rector, D. Sc. (Eng.), professor
Kremenchuk Mykhailo Ostrohradskyi National University
20, Pershotravneva ul, Kremenchuk, Ukraine
Email: mzagirn@kdu.edu.ua;mzagirn@gmail.com
Zhanna Romashykhina
Senior Lecturer, PhD (Eng.)
Kremenchuk Mykhailo Ostrohradskyi National University
20, Pershotravneva ul, Kremenchuk, Ukraine
Email: romashykhina.zhanna@gmail.com
Andrii Kalinov
Associate Professor, PhD. (Eng.).
Kremenchuk Mykhailo Ostrohradskyi National University
20, Pershotravneva ul, Kremenchuk, Ukraine
Email: andrii.kalinov@gmail.com
INDEX
A D
air gap, vii, xiii, 8, 13, 16, 28, 29, 30, 31, Daubechie wavelets, 48, 49
38, 62, 94 decomposition of the coil EMF signal, 121,
angle of shift of stator winding coils, xviii, 122
51 decomposition of the signal of winding
phase EMF, 117
diagnostic signal, ix, 16, 20, 21, 25, 27, 35,
B
39, 42, 45, 65, 66
disconnection from the supply main, xiv,
bar inductive reactance, 87
71, 72, 84, 85, 86, 88, 89, 94, 95, 104,
block of voltage sensors, 133, 134, 135,
139
136
discretization frequency, xiv, 48
discretization period, 48
C distribution of currents, 85, 88, 89
electromotive force, viii, ix, xi, 31, 58, 94, magnetic induction, xiii, 14, 15, 28, 30, 31,
105, 113, 114, 127, 149, 160, 162, 166, 32, 62, 92, 94
175 mathematical model, ix, 69, 70, 71, 72, 79,
electromotive force of the winding phase, 81, 83, 85, 86, 89, 90, 100, 103, 104,
127 143, 145, 170
experimental signal, 140, 141 mutual inductance, xv, 74, 75
F N
finite difference method, xi, 36 nominal condition, xvi, xviii, 142, 144
finite element method, xi, 36, 66, 69, 92, number of IM poles pairs, xvi, 29
93, 104, 158, 160 number of poles, 60
finite elements grid, 38, 94 number of slots per a pole and a phase, 59,
flux linkage, xix, 33, 72, 73, 74, 76, 77, 78, 60
79, 80, 82 number of stator slots, 60, 150
Fourier transform, viii, 14, 40, 41, 42, 45, number of the coil turns, xvii, 61
67, 156 number of turns in the slot, xvii, 33
number of wavelet-expansion coefficients,
xv, 111
H
L
R
losses in IM windings, xvi, 144
losses in the rotor of IM, xvi, 142 relative permeances of IM stator, xviii
losses in the stator windings of IM, xvi, 142 rotor bar current, 84, 87
rotor current, 71, 73
rotor rotation angle, xvii, 51, 75
M
rotor time constant, 102
magnetic field, vii, viii, xiv, 13, 14, 18, 28,
29, 30, 31, 34, 61, 90, 156, 157, 161 S
magnetic flux, 8, 27, 33, 72, 94, 95
scalar magnetic potential, xviii, 36
Index 183
self-running-out mode, xix, 34, 35, 39, 42, temperature of heating of the stator
63, 65, 66, 71, 82, 83, 84, 92, 93, 94, windings, xviii, 143, 145
103, 104, 140 temperature of insulation heating, xviii, 146
short-circuited ring, xi, 5, 7, 85, 86, 87, 91, three-phase coordinate system, 71, 72, 85,
138 103
skew angle, xviii, 62, 63 time psi-function, 43
spectral analysis, viii, 20, 22, 39, 40, 41, 42,
67
V
squirrel-cage rotor, vii, xi, 7, 9, 23
stator current, vii, 12, 15, 71, 72, 142, 153,
vector magnetic potential, xi, xiii, 32, 33,
155, 158, 160
92, 94, 174
stator phase voltage, 138, 139
stator slot, xvi, 33, 34, 63, 116, 117, 164,
165, 174 W
Symlet wavelets, 48, 49, 53
wavelet order, 57
wavelet-basis, xi, 44, 45, 47, 53, 166
T winding pitch in slots, 60
winding type, 60, 150
temperature of heating of the rotor bars,
xviii, 143