Sunteți pe pagina 1din 5

INTERNATIONAL JOURNAL OF SYSTEMS APPLICATIONS, ENGINEERING & DEVELOPMENT Volume 1, Issue 1, 2007

An Approach of Condition Monitoring of Induction


Motor Using MCSA
Neelam Mehla, Ratna Dahiya

Abstract—With the industrial growth, it has become in low voltage stator windings, etc. MCSA can
necessary to monitor the condition of the detect these problems at an early stage and thus
machine/system. Electrical machine being the most avoid secondary damage and complete failure of
sensitive part has great importance for the researcher the motor [2, 4, 7, and 6]. It is true that broken
to monitor the faults diagnosis. Three phase squirrel rotor bars will result in a change to the vibration
cage motor is normally use for industrial purposes.
Various techniques are used to control the speed such
spectrum, but vibration is traditionally sensed at
as DTC (Direct Torque Control), Vector Control, the bearings. And for each motor there is a
Close Loop Feedback Control etc. Small single phase different mechanical stiffness between the
Induction machine are used for home appliances hence electromagnetic forces caused by broken bars
the machine monitoring plays an important role for and the position where the vibration is sensed.
industrial as well as domestic appliances growth. This adds an additional complexity when
Various fault detection method has been used in past attempts are made to quantify the severity of the
two decades. Special attention is given to non-invasive problem via vibration analysis. Electromagnetic
methods which are capable to detect fault using major
forces are proportional to the flux density
data without disassembly the machine. The Motor
Current Signature Analysis (MCSA) is considered the squared waveform in an induction motor [6, 7].
most popular fault detection method now a day Hence, the vibration from unique
because it can easily detect the common machine fault electromagnetic forces from broken bars, etc., is
such as turn to turn short ckt, cracked /broken rotor a second order effect compared to current
bars, bearing deterioration etc. The present paper components directly induced from the specific
discusses the fundamentals of Motor Current rotating flux waves. In many cases, the fault
Signature Analysis (MCSA) plus condition monitoring severity (e.g., number of broken rotor bars) has
of the induction motor using MCSA. to be serious before it can be detected by
vibration analysis, and even then the prediction
I. INTRODUCTION of fault severity is another order of magnitude

T
more difficult. This is not the case with MCSA
he operators of induction motor drives are as has been proven via numerous industrial case
under continual pressure to reduce histories. With respect to detecting airgap
maintenance costs and prevent unscheduled eccentricity problems, a similar reasoning
downtimes that result in lost production and applies as reported by Cameron, et al. [8],
financial income. Many operators now use Tavner and Penman [9], and as demonstrated via
online condition-based maintenance strategies in industrial case histories by Thomson and
parallel with conventional planned maintenance Barbour, [10] and Thomson, et al. [11]. With
schemes. However, it is still the operator who respect to detecting shorted turns in low voltage
has to make the final decision on whether to stator windings then Thomson [12] has shown
remove a motor from service or let it run based that MCSA can detect the fault before a phase-
on information from condition monitoring to-phase or phase-to-earth failure. It is therefore
systems. A crucial point about motor current possible with a low voltage (LV) stator winding
signature analysis (MCSA) is that it is sensing an to have some lead time between shorted turns.
electrical signal that contains current components
that are a direct by-product of unique rotating
flux components caused by faults such as broken
rotor bars, air gap eccentricity, and shorted turns

Manuscript Received April 6, 2007; Revised Received July 27, 2007.


The authors are with the National Institute of Technology, Kurukshetra, Haryana, India

13
INTERNATIONAL JOURNAL OF SYSTEMS APPLICATIONS, ENGINEERING & DEVELOPMENT Volume 1, Issue 1, 2007

Figure 1: Basic MCSA Instrumentation System.

developing and actual failure. In comparison to a 4.3. Review high frequency demodulated current
high voltage (HV, e.g., 4160 V and above) and voltage in order to determine other faults
induction motor, the time to failure with an inter including electrical and mechanical health.
fault will be very short indeed Most faults can be determined at a glance, with
many rules being similar for both MCSA and
II. MOTOR CURRENT SIGNATURE vibration analysis. In addition, there are several
ANALYSIS rules that should be considered:
1. Pole pass frequency (ppf) sidebands around
Motor Current Signature Analysis (MCSA) is a the line frequency indicate rotor bar faults. The
system used for analyzing or trending dynamic, higher the peaks, the greater the faults.
energized systems. Proper analysis of MCSA 2. Harmonic pole pass frequencies often relate to
results will assist the technician in identifying: casting voids or loose rotor bars.
1. Incoming winding health 3. Non-ppf sidebands that cause a ‘raised noise
2. Stator winding health floor’ around the line frequency peak normally
3. Rotor Health relate to driven load looseness or other driven
4. Air gap static and dynamic eccentricity problems.
5. Coupling health, including direct, belted and 4. ‘Raised noise floor’ signatures relate to such
geared systems things as looseness or cavitation.
6. Load issues 5. Peaks that show in current and voltage relate
7. System load and efficiency to electrical issues, such as incoming power.
8. Bearing health Peaks that show in current only relate to winding
and mechanical faults.
III. BASIC STEPS FOR ANALYSIS 6. Peak pairs that do not relate to running speed
or line frequency are most often bearing related
There are a number of simple steps that can be problems.
used for analysis using MCSA. The steps are as Induction motors are the most widely used
follow: electrical machines. Therefore, many researchers
1. Map out an overview of the system being have studied motor diagnosis methods to prevent
analyzed. sudden stop in motor system.
2. Determine the complaints related to the If there are slightly damaged motors, which have
system in question. For instance, is the reason for been operating for a long time without
analysis due to improper operation of the symptoms, entire system including these motors
equipment, etc. and is there other data that can be can be very dangerous and these motor may be
used in an analysis. seriously damaged .Preventive measures should
3. Take data. be periodically taken in order to protect motors
4. Review data and analyze: and systems including motors. This is the most
4.1. Review the 10 second snapshot of current to efficient way to keep motor operating
view the operation over that time period. continuously in healthy conditions.
4.2. Review low frequency demodulated current
to view the condition of the rotor and identify
any load-related issues.

14
INTERNATIONAL JOURNAL OF SYSTEMS APPLICATIONS, ENGINEERING & DEVELOPMENT Volume 1, Issue 1, 2007

Figure 2: Overall MCSA Strategy.

IV. THEORY OF MCSA respect to the rotating rotor. When the cage
winding is symmetrical, there is only a forward
A full mathematical analysis (with experimental rotating field at slip frequency with respect to the
verification) of a three-phase induction motor rotor. If rotor asymmetry occurs, then there will
operating with broken rotor bars was published be a resultant backward rotating field at slip
by Williamson and Smith (1982)—this gives an frequency with respect to the forward rotating
excellent in-depth analysis. A conceptual rotor. The result of this is that, with respect to the
explanation is now presented to assist the reader stationary stator winding, this backward rotating
in gaining a physical understanding of what field at slip frequency with respect to the rotor
happens in an induction motor with broken rotor induces a voltage and current in the stator
bars. It is well known that a three-phase winding at
symmetrical stator winding fed from a
symmetrical supply will produce a resultant fsb =f1 (1-2s) Hz (1)
forward rotating magnetic field at synchronous
speed, and, if exact symmetry exists, there will This is referred to as a twice slip frequency
be no resultant backward rotating field. Any sideband due to broken rotor bars. There is
asymmetry of the supply or stator winding therefore a cyclic variation of current that causes
impedances will cause a resultant backward a torque pulsation at twice slip frequency (2sf1)
rotating field from the stator winding. Now apply and a corresponding speed oscillation that is also
the same rotating magnetic field fundamentals to a function of the drive inertia. This speed
the rotor winding, the first difference compared oscillation can reduce the magnitude (amps) of
to the stator winding is that the frequency of the the f1 (1- 2s) sideband, but an upper sideband
induced voltage and current in the rotor winding current component at f1(1+2s) is induced in the
is at slip frequency and not at the supply stator winding due to the rotor oscillation. This
frequency: s = per unit slip, f1= supply frequency upper sideband is also enhanced by the third time
Hz, f2= sf1 Hz, f2= slip frequency of rotor harmonic flux. Broken rotor bars therefore result
currents Hz. The rotor currents in a cage winding in current components being induced in the
produce an effective three- phase magnetic field, stator winding at frequencies given by:
which has the same number of poles as the stator
field but it is rotating at slip frequency (f2) with fsb =f1 (1±2s) Hz (2)

15
INTERNATIONAL JOURNAL OF SYSTEMS APPLICATIONS, ENGINEERING & DEVELOPMENT Volume 1, Issue 1, 2007

from the supply frequency line. The left-side


This gives ±2sf1sidebands around the supply component is caused directly by the fault, while
frequency component f1. These are the classical the right-side component is caused by the
twice slip frequency sidebands due to broken consequent speed ripple. The sum of the
rotor bars. amplitudes of these two components was proven
These are sometimes referred to as the pole pass to be a very good diagnostic index, suitably
frequencies by condition monitoring correlated to the fault severity for fabricated
practitioners, but this is not really an appropriate rotors. A drawback of this diagnostic procedure
terminology and can cause confusion. The is the possible confusion with the motor current
publications by electrical machine designers, modulation produced by other events. As an
researchers, and manufacturers always refer to example, pulsating load and particular rotor
the twice slip frequency sidebands due to broken design also cause sideband current components.
bars, as can be verified by reading the references If the load variation frequency is near, the
in this paper. Due to the variables that affect the resulting current spectrum is similar to that of a
frequency of these sidebands and their faulted rotor, but the two causes can still be
magnitude in amps (normally in dB in a MCSA distinguished. A more difficult issue is that of
system), the diagnostic strategy has to consider the particular design of the rotor structure. In
the following: large motors, a spidered structure with the same
• Different rotor designs (effect of pole number number of legs and poles produces a magnetic
and number of rotor slots, etc.). asymmetry whose effect is the same of rotor
• A wide range of power ratings. electrical asymmetry.
• Different load conditions.
• Mechanical load characteristics. VII .FAULT DIAGNOSIS AND FUZZY
• Mechanical components in the drive train. LOGIC
These factors can significantly affect the
diagnosis and need to be considered in the Fuzzy logic can systematically translate
development of reliable MCSA instrumentation linguistic concepts to numbers and associate
systems for three-phase induction motors. elements from a number set to concepts. This
capability provides a simple method to analyze
V. LOW VOLTAGE STATOR WINDING and interpret the frequency spectrum obtained
FAULTS for the stator current of induction machines.
Fuzzy based algorithms and Fuzzy logic are well
The most common kind of faults related to stator adapted to situations where no clear distinction
winding of induction motors are: phase-to- between the concepts of true and false exists.
ground, phase-to-phase and short-circuit of coils Fuzzy logic can handle situations where the
of the same or different phase. The last kind of answer lies somewhere in-between. This is the
fault is also called turn to turn fault. All these typical case of machine fault diagnosis. In fact,
faults are classified as isolation faults and have in general it is a difficult task to establish the
several causes: hot spots in the stator winding (or actual condition of a machine in terms of the
stator core) resulting in high temperatures, existence or not of a defect. However, it makes
loosening of structural parts, oil contamination, more sense to classify a fault in terms of its
moisture and dirt, electrical discharges (in case degree of severity. Fuzzy logic permits to infer
of high voltage windings), slack core lamination, about the machine state and to establish its
abnormal operation of the cooling system. Short- condition in different degrees of faulty condition.
circuit related faults have specific components in An important feature of Fuzzy based systems is
the stator current frequency spectrum. Incipient that the human knowledge and experience can be
faults can be detected sampling the stator current integrated into the systems in a systematic way.
and analyzing its spectrum. This procedure is on This can be done when the Fuzzy sets and Fuzzy
the base of MCSA Method. rules are being defined. This feature is even
more important because the fault detection is in
VI. ROTOR FAULTS practice based on the human knowledge and
experience, at least up to some degree.
Motor current signature analysis (MCSA) is one
of the most spread procedures to detect rotor
faults. In fact, a rotor bar breakage introduces
two anomalous lines in the current spectrum far

16
INTERNATIONAL JOURNAL OF SYSTEMS APPLICATIONS, ENGINEERING & DEVELOPMENT Volume 1, Issue 1, 2007

International Conference on Condition


VIII. CONCLUSION Monitoring, University College of Swansea,
Motor Current Signature Analysis is an electric Wales, United Kingdom, 1987.
machinery monitoring technology. It provides a 5. Kliman, G. B. and Stein, J. “Induction
highly sensitive, selective, and cost-effective Motor Fault Detection via Passive Current
means for online monitoring of a wide variety of Monitoring,” Proceedings of International
heavy industrial machinery. It has been used as a
test method to improve the motor bearing wear Conference (ICEM’90), Massachusetts
assessment for inaccessible motors during plant Institute of Technology, Boston,
operation. This technique can be fairly simple, or Massachusetts, pp. 13-17, 1990.
complicated, depending on the system available 6. Alger, P. L., Induction Machines,
for data collection and evaluation. MCSA New York, New York: Gordon and Breach,
1965.
technology can be used in conjunction with other
technologies, such as motor circuit analysis, in
7. Yang, S. J.,“Low Noise Electric Motors,”
order to provide a complete overview of the Monographs in Electrical and Electronic
motor circuit. The result of using MCSA as part Engineering, IEE, Savoy Place, London,
of motor diagnostics program is a complete view England, 1981.
of motor system health. 8. Cameron, J. R., Thomson, W. T., and Dow,
A way forward is for the technology
and intelligent diagnosis to be integrated into a A. B., “Vibration and Current Monitoring
technologically advanced hand-held instrument for Detecting Airgap Eccentricity in Large
that is applicable to a diverse range of induction Induction Motors,” IEE Proceedings, Vol.
motor derives. It has to be appreciated that the
133, no. 3, 1986.
operators requirements can be differ widely and
an MCSA instrument must be able to cope with 9. Tavner, P. and Penman, J., “Condition
induction motor drives in power stations, Monitoring of Electrical Machines”,
Petrochemical refineries, offshore oil and gas Research Studies Ltd., London, England:
production platforms, mining industry, paper John Wiley & Sons, 1987.
mills and car industry.
10. Thomson, W. T. and Barbour, A., “On-line
REFERENCES Current Monitoring and Application of a
1. Thomson, W. T. and Orpin, P., “Current Finite Element Method to Predict the Level
and Vibration Monitoring for Fault of Airgap Eccentricity in 3-Phase Induction
Diagnosis and Root Cause Analysis of Motors,” IEEE Transactions on Energy
Induction Motor Drives,” Proceedings of Conversion, Vol.13, no.4, pp. 347-357,
Thirty-First Turbomachinery Symposium, 1998.
Turbo machinery Laboratory, Texas A&M 11. Thomson, W. T., Rankin, D., and Dorrell, D.
University, College Station, Texas, pp. G.,“On-line Current Monitoring to Diagnose
61-67, 2002. Airgap Eccentricity—An Industrial Case
History of Large HV, 3-Phase Induction
2. Hargis, C., Gaydon, B. G., and Kamish, K.,
Motors,” IEEE Transactions on Energy
“The Detection of Rotor Defects in
Conversion, 1999.
Induction Motors,” Proceedings of IEE
12. Thomson, W. T., “On-Line MCSA to
EMDA Conference, London, England, pp. Diagnose Shorted Turns in Low Voltage
216-220, 1982. Stator Windings of 3-Phase Induction
3. Thomson, W. T., “Diagnosing Faults in Motors Prior to Failure,” Proceedings of
Induction Motors Engineering Ideas,” IEEE Conference on Electrical Machines
Electrical Review, Vol. 215, no. 17, 1984. and Drives (IEMDC), Massachusetts
4. Thomson, W. T. and Rankin, D.,“Case Institute of Technology, Boston,
Histories of Rotor Winding Fault Diagnosis Massachusetts, 2001.
in Induction Motors,” Proceedings Second

17

S-ar putea să vă placă și