TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
On Line Fault Identification of Induction Motor
using Fuzzy System
1
D. K. Chaturvedi, 2Akash Gautam, 3Mayank Pratap Singh, 4Md. Sharif Iqbal
Dept. of Electrical Engineering, Faculty of Engineering, Dayalbagh Educational Institute, Agra-282110, India.
Email: dkc.foe@gmail.com
motor may suffer severe damage. Thus, undetected motor
faults may cascade into motor failure, which in turn may
cause production shutdowns. Such shutdowns are costly
in terms of lost production time, maintenance costs, and
wasted raw materials. The motor faults are due to
mechanical and electrical stresses. Mechanical stresses are
caused by overloads and abrupt load changes, which can
produce bearing faults and rotor bar breakage. On the
other hand, electrical stresses are usually associated with
the power supply. Induction motors can be energized from
constant frequency sinusoidal power supplies or from
adjustable speed ac drives.
Abstract- It is well known that Induction motors are used
worldwide as the “workhorse” in industrial applications.
Although, these electromechanical devices are highly
reliable, they are susceptible to many types of faults. Such
fault can become catastrophic and cause production
shutdowns, personal injuries and waste of raw material.
However, induction motor faults can be detected in an initial
stage in order to prevent the complete failure of an induction
motor and unexpected production costs. The motive of this
project is to analyse the fault in induction motor through
sound and electrical signature produced during the specific
fault existing in the induction motor and then to analyse it
through various technique. In this paper, a method for
mechanical and electrical fault diagnosis in induction motor
through sound and electrical signature analysis has been
proposed. The work reported in this project uses
noninvasively method for sound signature for diagnosing
different mechanical faults. For the electrical fault diagnosis
the current signature of 3-phase induction motor has been
recorded. The recorded sound signature and current
signature of the faulty induction motor and a healthy
induction motor during different faults have been analysed
using Fourier transform. The magnitude and frequency of
FFT of these signatures have been used for identification of
different faults using fuzzy system.
This paper proposes a mechanical fault and electrical
diagnosis of induction motor system using sound
signature and current signature analysis.
II. THE PROPOSED SYSTEM
The Proposed method is for diagnosis of mechanical and
electrical faults in 3-phase induction motor through
electrical and sound signal analysis. It is a complete fault
detection system. Through this method the mechanical
fault prevailing in the induction motor can be diagnose,
and for electrical faults the electrical current waves during
different faults are also recorded. The main advantage of
this report is a noninvasive fault diagnosis in induction
motor. An experiment on 3-phase induction motor for
fault diagnosis was conducted to record sound signal of
the faulty induction motor and a healthy induction motor
then we analyze it through the use of software i.e.
MATLAB 7.5 by using Fourier transform technique.
After the analysis of the captured data it should be
compared using fuzzy system and then final results will
be made.
Keywords—Induction motor, fuzzy logic
I. INTRODUCTION
INDUCTION motors are electro-mechanical devices
utilized in most industrial applications for the conversion
of power from electrical to mechanical form. Induction
motors are used worldwide as the workhorse in industrial
applications. Such motors are robust machines used not
only for general purposes, but also in hazardous locations
and severe environments. General purpose applications of
induction motors include pumps, conveyors, machine
tools, centrifugal machines, presses, elevators, and
packaging equipment. On the other hand, applications in
hazardous locations include petrochemical and natural gas
plants, while severe environment applications for
induction motors include grain elevators, shredders, and
equipment for coal plants. Additionally, induction motors
are highly reliable, require low maintenance, and have
relatively high efficiency. Moreover, the wide range of
power of induction motors, which is from hundreds of
watts to megawatts, satisfies the production needs of most
industrial processes. However, induction motors are
susceptible to many types of fault in industrial
applications. A motor failure that is not identified in an
initial stage may become catastrophic and the induction
III. EXPERIMENTAL SETUP
The experimentation done in Electrical Engineering lab at
Dayalbagh Educational Institute (Deemed University)
Agra, India. The experiment consisting of two induction
motor (one is healthy IM and other is faulty IM), A laptop
preloaded with MATLAB software and a sound recorder.
Induction motor specifications are 240V, 50 Hz, 500W.
The real time data is acquired with the help of recorder
from both the IM one by one and sends them to computer.
MATLAB imports these signals and signal analysis using
FFT is done shown in fig.5.
964
TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
The estimated parameters compared with the healthy IM
under normal operating conditions
Fig. 2: Healthy bearing
IV. FLOW CHART FOR PROPOSED METHOD
Fig. 3: Faulty bearing
A flow chart for proposed method is shown below in fig.1
B. SIGNAL ACQUISITION
A. SYSTEM
In signal acquisition a signal is acquired from the system.
Firstly the signal is acquired from the healthy induction
motor by using sound recorder then similarly the signal is
acquired for the faulty induction motor through sound
recorder.
For the proposed method we have a system i.e. induction
motor. There are two identical induction motor out of
which in one induction motor there no fault exist i.e. it is a
healthy induction motor and another induction motor in
which fault is there. Healthy and faulty induction motor
has been shown in fig 2 and fig 3.
Fig. 4a: Original healthy signal
Fig.1: Flow chart
965
TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
Fig. 5b: nFFT of faulty induction motor
PERIODOGRAM
Fig. 4b: Original Faulty signal
A plot of power versus frequency is called a periodogram.
C. DATA PRE-PROCESSING
E. EXPERIMENTAL RESULTS AND COMPARISON
The results obtained after signal analysis of healthy and
faulty system have been compared to identify faults. It is
clearly seen in FFT of the healthy induction motor there
are many peaks out of which two major peaks are of
approx. 756.2 and 631.6 amplitude at approximate 555
and 832 Hz frequency respectively. Similarly in faulty
induction motor there are two major peaks are of approx.
2503 and 1128 amplitude at approximate 927 and 790 Hz
frequency respectively. These are the characteristics of
healthy and faulty induction motor signal.
In data pre-processing stage the signal is prepared for
further analysis process i.e. if the signal is need to be
amplified or the conversion of signal to an appropriate
format for further processing.
Similarly in the periodogram there is one major peak
between 0 to 5 Hz whereas in faulty induction motor
along with that sharp peak there are some distinct spikes
between 5 to 10 Hz.
D. SIGNAL ANALYSIS
V. EXPERIMENTAL SETUP FOR RECORDING SOUND
SIGNALS DURING ELECTRICAL FAULTS
After the data pre processing part the signal analysis is
done. In signal analysis the signal which is obtained after
pre processing, it is analysed through the use of
MATLAB software by taking Fourier transform.
The experiment consisting of 3-phase induction motor, a
laptop preloaded with MATLAB software and a sound
recorder. The real time data is acquired with the help of
recorder from IM under different fault conditions and then
they are sends to computer for further analysis.
MATLAB imports these signals and signal analysis using
FFT is done then the estimated parameters compared with
the healthy IM under normal operating conditions.
Various diagrams whose waveforms have been analysed
are as follows.
1.
Fast Fourier Transform.
2.
Periodogram.
Fig. 7: Setup in Electrical Machines Lab, DEI Dayalbagh,
Fig.5a: aFFT of healthy induction motor
966
TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
A. Healthy Sound Signal in Time Domain
1.
C. FFT of the signal when 1/3 Coil Is Short Circuited
Star connected
1.
Star Connected
2.
Delta Connected
1)
2.
Delta connected
D. FFT of the signal in single phasing fault
B. Healthy Sound Signal in Frequency Domain
1.
Star connected
2.
Delta connected
967
1.
Star Connected
2.
Delta Connected
TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
VI. COMPARISON OF EXPERIMENTAL RESULTS
TABLE I: Frequency and amplitude during different faults in star
connection
C. Rule Base:
If the highest amplitude lies in low amplitude and at
medium frequency range then it is healthy motor in star
connection.
If highest amplitude lies in medium amplitude and at low
frequency range then it is healthy motor in delta
connection.
TABLE.II: Frequency and amplitude during different faults in delta
connection
If the highest amplitude lies in high amplitude and at low
frequency range then it is single phasing fault.
If the highest amplitude lies in medium amplitude and at
medium frequency range then it is short circuit fault in
star connected induction motor
If highest amplitude lies in high amplitude and at medium
frequency range then it is short circuit fault in delta
connected induction motor.
VII. FUZZY SYSTEM FOR IDENTIFYING FAULTS
D. Defuzzification:
Incoming FFT signals are fuzzified. The sound signals
were given to fuzzy system for identifying the different
faults based on different amplitude and frequency in 3phase induction motor.
Then defuzzification will be done using centre of area
method.
VIII. EXPERIMENTAL SETUP FOR RECORDING ELECTRICAL
SIGNATURE
The experimentation is done in Electrical Engineering Lab
Faculty of Engg. at Dayalbagh Educational Institute
(Deemed University) Agra, India.
A. Fuzzification:
Input to the fuzzification block is frequency and
amplitude of different signals which should be recorded
during electrical faults which are fuzzified.
B. Membership Function:
Let FREQ and AMP be linguistic variables with the label
“frequency” and “amplitude”. The unit of frequency is
hertz.
Fig.8: setup in Electrical Machines Lab, DEI, Dayalbagh
FREQ= [Low, Medium, High]
A. 3-phase Induction Motor specifications:
Rating- 3 HP, 50Hz
3-phase, 4 pole, 1500 R.P.M, 400/440volts.
Total No. of coils or slots 36.
Coils/pole-9.
Coils/pole/phase-3
AMP = [Low, Medium, High]
968
TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
IX. STATOR WINDING
It consists of conventional squirrel cage rotor and a stator
with 36 slots and wound in double layer with 36 multiturn
coils, all the 72 ends of which are brought out to a circular
Bakelite board on the face of which is inlaid a
representation of the 36 coils of the stator as shown in
diagram which gives the full details of the terminal plate
except that the arcs showing the coils are only 12 instead
of 36 to avoid congestion.
The coil pitch is 9 slots or 19 coils sides i.e. if one of a
coil is in the 1st slot then the other coil side is in 10th slot
and so on
Fig.10b : wave forms of line current ir during starting of 3-phase
induction motor
Fig.9: stator with 3-phase winding
Fig.11a: wave forms of line current IR when 1/3 coil of r-phase is short
circuited
X. EXPERIMENTAL RESULTS
TABLE III: Effect of Faults
XI. TRANSIENTS DURING ELECTRICAL FAULTS
Fig.11b: wave forms of line current IR when 1/3 coil of r-phase is short
circuited
Fig.10a: wave forms of line current IR during starting of 3-phase
induction motor
Fig.12a: wave forms of line current IR when r-phase is open circuited in
star connection
969
TECHNIA – International Journal of Computing Science and Communication Technologies, VOL.6 NO. 2, January. 2014 (ISSN 0974-3375)
[7] N. Paterson, “The Analysis and Detection of Faults in
Three Phase Induction Machines Using Finite Element
Techniques,” Doctoral Thesis, Robert Gordon Univ.,
Wetherby British Library, Aberdeen, 1998, 180 p.
[8] W. T. Thomson, R. J. Gilmore, “Motor Current Signature
Analysis to Detect Faults in Induction Motor Drives–
Fundamentals, Data Interpretation, and Industrial Case
Histories,” Proceedings of the 32nd Turbo machinery
Symposium, Houston, TX, USA, Sept. 8-11, pp. 145 156,
2003.
[9] S. Nandi, H. A. Toliyat, “Detection of Rotor Slot and Other
Eccentricity Related Harmonics in a Three Phase Induction
Motor With Different Rotor Cages,” Proceedings of
International Conference on Power Electronic Drives and
Energy Systems for Industrial Growth–PEDES ’98,
Perth,Australia, Nov. 30-Dec. 3, Vol. 1, pp. 135-140, 1998.
Fig.12b: wave forms of line current IR when y-phase is open circuited in
delta connection
XII. CONCLUSION AND FUTURE WORK
The purpose of this paper is to diagnose electrical and
mechanical faults in induction motor systems. The
analysis results showed that the proposed method can
identify the faults. The method described provides a
promising way to establish potential metrics for the
description of motor health degradation. Therefore, it is
desirable to develop a condition monitoring system based
on the above method and realize on-line health evaluation
of induction motor. With such a function, the critical
failure of induction motor systems can be avoided, and the
reliability and availability of motor can be guaranteed
[10] Peter Vas, “Parameter estimation, condition monitoring,
and diagnosis of electrical machines”, Clarendon Press
Oxford., 1993.
REFERENCES
[13] O. I. Okoro, “Steady and transient states thermal
analysis of a 7.5-kW squirrel- cage induction machine at
rated-load operation,” IEEE Transactions on Energy
Conversion, Vol. 20, No. 4, pp. 730-736, 2005.
[11] P. J. Tavner and J. Penman, “Condition monitoring of
electrical machines". Hertfordshire, England: Research
Studies Press Ltd, ISBN: 0863800610, 1987
[12] P. H. Mellor, D. Roberts and D. R. Turner, “Lumped
parameter thermal model
for electrical machines of
TEFC design,” IEEE Proc. Electric Power Application,
Vol. 138, pp. 205-218, 1991.
[1] D. K. Chaturvedi and H. Vijay, “Parameters Estimation of
an Electric Fan Using ANN”, Journal of Intelligent
Learning Systems and Applications, 2010, 2: 33-38.
[14] John S. Hsu, “Monitoring of defects in induction
motors through air-gap
torque observation” IEEE
Transactions on Industry Applications, Vol. 31, No. 5,
pp.1016- 1021, 1995.
[2] D.K. Chaturvedi, “Modeling & simulation of systems using
Matlab/simulink”, CRC Press, 2009
[3] P. F. Albrecht, J. C. Appiarius, and D. K. Sharma,
"Assessment of the reliability of Motorsutility applicationsUpdated," IEEE Transactions on Energy Conversion,
vol.1,pp. 39-46, 1986.
[15] D. G. Dorrell and A. C. Smith, “Calculation and
measurements of unbalance
magnetic pull in cage
induction motors with eccentric rotors, part 2: experimental
investigation”, IEEE Proceedings Electric Power
Applications, Vol. 143, No. 3, May, pp. 202-210, 1996.
[4] A. Siddique and G. S. Yadava, “A Review of Stator Fault
Monitoring Techniques of
Induction Motors,” IEEE
Trans. Energy Convers., vol. 20, no. 1, pp. 106–114, 2005.
[16] A. Belahcen, A. Arkkio, P. Klinge, J. Linjama, V.
Voutilainen and J. Westerlund, 1999, “Radial forces
calculation in a synchronous generator for noise
analysis”, Proceeding of the Third Chinese International
Conference on Electrical Machines, Xi’an, China, pp.
199-122, August 1999
[5] P. J. Tavner and A. F. Anderson, “Core Faults in Large
Generators’, IEEE Proceedings on Electric Power
Applications, Vol. 152, Issue 6, pp. 1427-1439, 2005.
[6] J. Ramirez-Nino, A. Pascacio, “Detecting Interturn Short
Circuits in Rotor Windings,” IEEE Comp. Appl. in Power,
vol. 14, issue 4, pp. 39 - 42, 2001.
[17] D.K. Chaturvedi, “Soft computing techniques and its
applications to electrical engineering”, Springer Verlag,
Germany,2008
970