March 5, 2015
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 22
Introduction
Meta-Trends
OMICRON
Page 3
March 10,
2015
Page: 4
Blackouts
millions of
location
people affected
date
620
India
230
India
2 January 2001
150
Bangladesh
1 November 2014
140
Pakistan
26 January 2015
100
Indonesia
18 Aug 2005
97
Brazil
11 March 1999
87
Brazil, Paraguay
55
55
28 Sep 2003
40
Thailand
18 Mar 1978
30
9 Nov 1965
March 10,
2015
Page: 5
What to do ?
maintenance
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 77
Insulation strength
New condition
Run to Failure
Preventive
Maintenance
Predictive
Maintenance
Real Time
Monitoring
Ease of repair/replacement
Reliability
cantered
Maintenance
RCM
Preventive
Maintenance
PM
Condition
Based
Maintenance
Run to Failure
CBM
Real Time
Monitoring
RTM
RTF
Predictive
Maintenance
PdM
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 13
13
OMICRON
Page 14
OMICRON
Page 15
OMICRON
Page 16
OMICRON
Page 17
OMICRON
Page 18
Diagnostic monitoring
> Continuous or periodic monitoring of HV equipment is an essential tool for
effective maintenance management
> Monitoring answers questions about the present condition of the
equipment and its future performance
> The large amount of real-time data gathered by monitoring systems can
be used for precise insulation condition assessment
> Effective comparison of historical data is enabled via an easy-to-use and
extendable database solution
> Exact knowledge of insulation state at any time saves money, as
maintenance schedules can be specifically optimized and the service life
of assets can be extended
OMICRON
Page 19
Diagnostic monitoring
> Involves no operational variables
> Provides information to evaluate the aging condition of equipment
> Allows assessment of future developments in the aging process
> Supports decision making for ongoing maintenance or the replacement
of components
> is a "planning tool (asset management)
OMICRON
Page 20
OMICRON
Page 21
OMICRON
Page 22
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 23
23
Seite 24
OMICRON
Seite 25
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 26
26
all above may be correct what we learn: prediction is difficult ... impossible,
therefore diagnostically methods or monitoring is needed
OMICRON
Seite 27
Glew C.N.:The Next Generation A Review of the Factors influencing the Output
of an Electrical Machine in the New Millenium., INSUCON/ISOTEC 98, p. 231-242
OMICRON
Seite 28
Generator
OMICRON
Seite 29
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 30
30
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 31
OMICRON
Seite 32
OMICRON
Seite 33
OMICRON
Seite 34
Covering Tape
Spacer, Coil-End Bracin
Groundwall Insulation, Mica Tapes
Grading/ Silicon Carbide Coating
E.
F.
G.
H.
I.
OMICRON
Seite 35
OMICRON
Seite 36
IEC 60034-18
IEC 60034-18-31 Test procedures for form-wound windings Thermal evaluation and classification of
insulating systems used on machines up to and including 50 MVA and 15 kV
IEC 60034-18-33 Test procedures for form-wound windings Multifactor functional evaluation; Endurance
under combined thermal and electrical stresses of insulating systems used in machines
up to and including 50 MVA and 15 kV
IEC 60034-18-34 Test procedures for form-wound windings - Thermomechanical cycle endurance evaluation
IEC 60085
IEC 60216
IEC 60371
IEC 60270
High-voltage test techniques Partial discharge measurement, Version 2000, 3rd Edition
IEC 60034-27
TS, Ed.1, Rotating electrical machines Part 27: Off-line partial discharge measurements
on the stator winding insulation of rotating electrical machines, Version 2006
IEEE 1434-2000 IEEE Trial-Use Guide to the Measurement of Partial Discharges in Rotating Machinery,
Version 2000
OMICRON
Seite 37
OMICRON
Seite 38
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 39
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 40
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 41
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 42
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 43
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 44
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 45
Source: Brtsch et al. "Insulation Failure Mechanisms of Power Generators", DEIS July/August 2008
OMICRON
Seite 46
OMICRON
Seite 47
ambient (chemical)
> Moisture + (water with nitrogen oxides under PD influence generates nitric acid)
> Oxidation by aggressive environment
electrical
> Overvoltages, transients or permanent stress
mechanically
>
>
>
>
>
OMICRON
pollution
misalignment
Vibrations / relaxations
Exposure by foreign object damage
Forces caused by short-circuits & faulty synchronization actions
Seite 48
OMICRON
Rated Voltage
Test Voltage
Over Voltage
water, ice
aggressive and reactive chemicals
abrasive particles: metal parts, ash,
carbon, lubricants
Seite 49
Topics
> Introduction
> Maintenance
> Measurement, Diagnostics and Diagnostic Monitoring
> Rotating Machines (Generators / Motors)
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Insulation Resistance
Polarization Index
C / tan (power factor)
Partial discharge
Page
Page 50
50
Diagnosis and Monitoring can not avoid faults, but it helps to plan and
minimize the cost
A suddenly popped up insulation damage is not repaired on average less than
6 to 8 weeks (open end) - if all goes perfect.
A well-planned repair of a machine takes 2-3 weeks till re-commissioning
(depending on the size of the machine)
OMICRON
Seite 51
OMICRON
Seite 52
OMICRON
Seite 53
OMICRON
Source: electrical-engineering-portal.com
Seite 54
OMICRON
Seite 55
OMICRON
Seite 56
Source: electrical-engineering-portal.com
OMICRON
Seite 57
<1.0
Hazardous
1 - 1.5
Bad
1.5 - 2.0
Doubtful
2.0 - 3.0
Adequate
3.0- 4.0
Good
> 4.0
Excellent
OMICRON
Seite 58
60 s
Time
Current
600s
3.7637 nA
OMICRON
Seite 59
25.9809 nA
electrical
optics
engineering
boundary layer polarization
refraction index n
r = n
orientation polarization
grid polarization
atom polarization
1
0
Hz
polarization
losses tan()
kHz
MHz
wavelength
GHz
m
frequency
IR
losses caused
by reloading of
partial
capacitances
thermal losses
dipoles follow
the field with
delay
mm
light
nm
UV
x-rays
atoms get
excited
Conductivity
insulation
surface
current
current through
insolation
insulation
cavity
with PD
Vacuum capacity
> Capacity of a test object which has as dielectric vacuum
vacuum
vacuum
E-field
polarization
conductivity
equivalent circuit
Equivalent circuit
> The equivalent circuit can be
summarized to two elements:
> A capacitor
> A resistor
> This gives the equivalent circuit
for the tangent() measurement
> As parallel circuit
> A series circuit is also possible
IC
test
object
IR
Itest
i(t)
IR
IC
= phase difference
test object
current
capacitive
current
Itest
Insolation material
Typical tangent()
(50Hz)*
8*10-5 - 4*10-4
PTFE / Teflon
~1*10-4
~2*10-2
5*10-3 - 2*10-1
2*10-3 - 1
1*10-3 - 8*10-2
Silicone oil
~2*10-4
Mica
Glass
~1*10-4
Pressboard
~3*10-3
> The losses due to polarization may sink with higher temperatures due to
better dipole movement
> This gives a lower dissipation factor
> In combination this mostly gives a rising dissipation factor with rising
temperature, but other results may also occur
tan G
Dissipation Factor
cos I
I
Insulation
tan G
OMICRON
cos M
Seite 71
Power Factor
P
(Dissipation Factor DF)
Q
P
(Power Factor PF)
S
.....
Conductive Losses
OMICRON
Polarization Losses
Seite 72
OMICRON
Seite 73
tan
Sum
Polarization Losses
Conductive Losses
f [Hz]
OMICRON
Seite 74
50Hz
5.000
DF
2.000
1.000
0.500
0.200
0.12
0.100
Moderate
New
Aged
0.050
0.020
0.010
0.0036
0.005
0.0024
0.0001
OMICRON
0.001
0.010
0.10
1.0
10.0
Seite 75
Freq/Hz 1000
Measuring Principle
25 mm
2.5 km
tan G =
conductor (tan GC)
polarization (tan GP)
losses caused by PD (tan GPD)
tan G = f (U)
Seite 77
The tip-up test and the delta tan delta diagram show the same thing.
The delta tan delta diagram is frequency independent.
OMICRON
Seite 78
OMICRON
Seite 79
Important units
electron
2nd orbit
nucleus
The more electrons per TIME interval, the higher is the current
Current
Charge
per
Time
[A]
[Coulomb]
per
[s]
[A]
[Coulomb]
[s]
Charge
Current
times
Time
Charge
Current
Time
q =
1st orbit
q =
SI prefixes
kilo Volts
[kV]
Types of PD
> Internal PD
> Void discharges, electrical treeing
> External PD
> Corona
> Surface discharges
Outer semicon
Solid
insulation
Inner
semicon
Image Source: Andreas Kchler Hochspannungstechnik
Solid insulation
External PD
Surface discharge
Corona discharge
Internal PD
Internal discharge
in laminated material
Treeing
PD classification
En = 2
Lines of
electrical field
Void
(field strength doubling)
2. Thermal breakdown
> thermal instability of the insulation caused by high temperature
Thermal breakdown
> dielectric losses are caused by conductivity and polarization phenomena
> material heats up (conductivity raises exponential with temperature)
tan()
polarization losses
conductivity losses
Electrical insulations
are at the same time
thermal insulations
> breakdown
cracks cleaving
voids
cavity
delamination of fibers
PD treeing
Acrylic glass
Epoxy resin
PD treeing stopped
How to analyze PD
> Well known approaches:
> PRPD
> Trend
> Q(U)
> PSA (Pulse Sequence Analysis)
> TDR and STDR (PD localization)
Trigger
Trigger
Trigger
Trigger
Trigger
Trigger
50
...
Time
Amplitude
Amplitude
Time (ms)
10
20
T/W map
External disturbance
PD
Noise
850pC
900pC
500pC
Center Phase
850pC
900pC
Right Phase
Left Phase
Noise
PD source 1
PD source 2
origin
reflection
Ut(t)
U1(t)
tt
Bigger cavities
Bubble in oil
Surface discharges
Contact PD
When the bad contact becomes better, the contact PD can disappear.
Floating potential
Hopping particle
Hopping particles
Hopping particles
Source: J. Fuhr, Procedure for Identification and Localization of PD, IEEE Transactions 2005
Source: J. Fuhr, Procedure for Identification and Localization of PD, IEEE Transactions 2005
PDIV
Specified threshold
for PDIV and PDEV
noise level
U/Umax
PD magnitude as a function of the normalized test voltage Qm = f (U/Umax)
Full bandwidth
Low-pass
High-pass
Band-pass
> key element of the PD
measuring system
50 Hz: 1 V
500 Hz: 0.1 V
50 Hz: 1 V
150 Hz: 0.3 V
50 Hz:
150 Hz:
250 Hz:
350 Hz:
450 Hz:
1V
0.3 V
0.1 V
0.03 V
0.01 V
N x 50 Hz:
1V
time (t)
The area below the curve is 1.
frequency (f)
area = q = 10pC
area = q = 10pC
area = q = 10pC
area = q = 10pC
t
cut-off frequency
charge = q =
10pC
charge = q =
10pC
b)
i (t)
1,5
3
imax
1
imax/2
F()
0,5
2
F(0)
T1
T2
ideal PD-impulse:
0
103 104 105 106 107 Hz
109
This one...
Introduction
> ...For many years, the measurement of PD has been employed as a
sensitive means of assessing the quality ... and localize sources of PD in
used electrical winding insulation ...
> ... Compared with other dielectric tests (i.e. the measurement of dissipation
factor or insulation resistance) the differentiating character of partial
discharge measurements allows localized weak points of the insulation
system to be identified...
Introduction
> ...The PD testing ... is also used when inspecting the quality of new
assembled ... stator windings, new ... components (e.g. form-wound coils
and bars, HV bushings, etc..) and fully impregnated stators...
Introduction
> Partial discharge testing of stator windings can be divided into two broad
groups:
a. Off-line measurements, in which the stator winding is isolated from the power
system and a separate power supply is employed to energize the winding;
b. On-line measurements, in which the rotating machine is operating normally and
connected to the power system.
Annex A
and
IEC 60034-27-2
Introduction LIMITATIONS
> ... different types of PD measuring instruments will inevitably produce
different results ...
> ... PD measurements will only be comparable under certain conditions ...
> ... absolute limits for the windings of rotating machines, for example as
acceptance criteria for production or operation, are difficult to define ...
>
>
>
>
>
Introduction LIMITATIONS
> ... Empirical limits verified in practice can be used as a basis for evaluating
test results ...
> ... PD trend evaluation and comparisons with machines of similar design
and similar insulation system measured under similar conditions, using the
same measurement equipment
> ... not all insulation-related problems in stator windings can be detected
by measuring PD (e.g. insulation failures involving continuous leakage
currents due to conductive paths between different elements of the
insulation or pulseless discharge phenomena) ...
1 Scope
> Measuring techniques and instruments,
> The arrangement of test circuits,
> Normalization and testing procedures,
> Noise reduction,
> The documentation of test results,
> The interpretation of test results
2 Normative references
> IEC 60060-1, High-voltage test techniques Part 1: General definitions and
test requirements
3 Definitions
> ... general terms and definitions for partial discharge measurements given
in IEC 60270 apply...
3 Definitions (Coating)
> Stress control coating
> Paint or tape on the surface of the groundwall insulation that extends beyond
the conductive slot portion coating in high-voltage stator bars and coils
> The stress control coating reduces the electric field stress along the winding
overhang to below a critical value that would initiate PD on the surface. The
stress control coating overlaps the conductive slot portion coating to provide
electrical contact between them.
3 Definitions (Evaluation)
> Pulse height distribution
> The number of pulses within a series of equally-spaced windows of pulse
magnitude during a predefined measuring time
3 Definitions (Evaluation)
> Pulse height distribution
3 Definitions (Hardware)
> Coupling device
> Usually an active or passive four-terminal network that
converts the input currents to output voltage signals
> These signals are transmitted to the measuring instrument by a transmission
system. The frequency response of the coupling device is normally chosen at
least so as to efficiently prevent the test voltage frequency and its harmonics
from reaching the measuring instrument.
Internal voids
Internal delamination
Delamination between conductors and insulation
Slot discharges
End-winding surface discharges
Conductive particles
PD pulse propagation
Damping, reflection, attenuation
Calibration
>
Ratio CK / CA
>
Center frequency
>
Propagation path
Two frequency components of the PD signals in a stator winding
Superposition
errors!
6 Visualization of measurements
6.2 Data presentation
> PD magnitude [pC or mV]
> r.m.s. value of test voltage
> Inception voltage Ui (PDIV)
PDEV PDIV
Specified threshold
for PDIV and PDEV
Noise level
U/Umax
6 Visualization of measurements
6.3 Additional means of PD data
representation
> ... Additional quantities ... like integrated
charge, discharge current, quadratic rate,
PD power, and PD energy ...
> ... pulse height distribution, pulse phase
distribution, phase resolved pulse height
distribution, oscillograms of pulse trains, PD
distribution maps, etc.... CIGRE technical
brochure 226 ...
Part V: Generator
Stator Insulation
6 Visualization of measurements
> 6.3.2. PRPD pattern
> ... a 2-dimensional PD distribution map (-q-n pattern) is employed for
visualization ...
7 Test circuits
IEC 60060-2:
Measuring systems
Example:
IEC 60060-1
7 Test circuits
> ... sufficiently PD free...
> ... To ensure that the test circuit does not influence the measurement of
partial discharges from the test object, the arrangement should first be
tested up to the maximum test voltage in accordance with the test
procedure given in 9.1.6. The noise level produced by the complete test
circuit at maximum required test voltage shall not exceed 100 pC when
using the normalization procedure in accordance with Clause 8...
> ... The whole test circuit should be of a low-inductance arrangement.
It is essential that ground loops are avoided. Low inductance leads are
recommended as ground connections...
7 Test circuits
Z
Ck
U~
Ca
CC
Zmi
CD
MI
7 Test circuits
Z
Ca
U~
Ck
CC
Zmi
CD
MI
Phases
Phasesnot
nottoto
be
beseparated
separated
Double length as
filter
8 Normalization of measurements
> ... Due to pulse propagation, resonance and mutual cross coupling in
machine windings, mentioned in 4.3, calibration is not possible...
> ... The aim of normalization is to ratio out various influences of the test
circuit, for example power supply connections, stray capacitance, coupling
capacitance and test object capacitance...
> ... Normalization is to ensure that the PD measuring system provides
sufficient sensitivity to measure a specified value of PD magnitude correctly,
as it appears at the machine terminals during the measurement, and to
show that the PD detection system used, is responding in a repeatable
fashion...
> ... comparisons between measurements on objects having the same
design, taken with the same PD device...
> ... Normalization of the test circuit should be performed by injecting shortduration current pulses of known magnitude by means of a reference pulse
generator (calibrator) ...
9 Test procedures
> Tests on complete windings
> Individual phases
> Individual winding components
> DUT disconnected from all external power supplies
>
>
>
>
Bus work
Surge arrestors
Surge capacitors
Excitation systems
> ... sufficient air clearance between ...connections and ... cables ...
> ... check the insulation resistance before starting the PD test ... >100M ...
[Mainly dirt, as dirt may influence PD behaviour]
> ... conductive slot coating should be in contact with ground potential
throughout its whole length ... [for winding components]
> CONDITIONING!
> ... PD will typically decrease during the first minutes ...
> ... 5 min at the maximum test voltage is recommended ...
> ... voltage may then be re-applied to start the partial discharge measurements
...
Paschens Law
9.1.6.3 PD testing
> ... each voltage step, or during continuous ramping, the PD data ... should
be recorded and processed ...
> ... Qm = f(U) curve ...
> PDIV
> PDEV
> ... distribution of pulse magnitudes, phase resolved distributions or specific
PD patterns ... at various levels during test voltage increase and decrease
...
> ... Any comments or observations during the test should be
recorded ...
> ... an appropriate sequence of standard measurements and ... extended
measurements ... should be taken ...
10 Interpretation
> ... it has to be decided whether there are ... defects and ... what they imply
regarding the performance of the insulation system, whether any
supplementary tests are needed and the planning and/or implementation of
any essential corrective maintenance. ...
> ... Neither is it possible to establish any absolute limits for complete
windings ...
10 Interpretation
> ... the PD site, which produces the highest PD magnitude is not necessarily
the location in the winding being at most risk ...
> ... the higher the PDIV and PDEV ... the less insulation deficiencies ...
10 Interpretation
Trend in PD in a machine over time
> ... Initial fingerprint ... when the winding is new ...
> ... If the winding deteriorates ... Qm will usually increase over time ...
> ... Compare the PD quantities ... between windings ...
> Limitations
> ... a new stator may have relatively high PD that decreases after the first 5000
to 10000 equivalent operating hours ...
> ... same voltage, temperature and similar humidity conditions ...
> ... variations of Qm of a certain percentage, for example 25 %, are normal
...
> Why?: ... That is, for example, Qm for the winding at the specified test
voltage is lower than 95 % of the mean Qm magnitudes achieved by the
manufacturer on the same windings they have made in the past....
3CFRD
Disturbance
PD
Noise
L2
L1
L3
L2
L1
Inner PD Source in
L1
L2
L3
L3
L1
timeframe
1 s
3PARD
L2
L1
Outer Noise
L2
L3
L3
L1
timeframe
1 s
D L Signal Velocity
TDD
t1
2
Signal Velocity
213
L Signal Velocity t
t unit1.1
unit 1.2
c0
Hr Pr
March 5, 2015
Topics
> Transformer
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
> HV Cables
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Page
Page 22
OMICRON
Seite 3
OMICRON
Seite 4
Transformer Breakdowns
OMICRON
Seite 5
Accessories
Core
Bushing
Winding
Tap
Changer
Faults in Transformers
OMICRON
Seite 6
Insulation strength
New condition
OMICRON
Seite 8
OMICRON
Seite 9
Maintenance Intervals
Action
Task Interval
Light
Regular
Remark
Intensive
Visit
6m
1m
1d
In service
Detailed inspection
(visual)
1y
3m
2w
In service
DGA
2y
1y
3m
Oil Test
6y
2y
1y
Cooling system
cleaning
Conditional
Conditional
Any interval
Accessories
verification
12 y or
conditional
6-8 y
1-2 y
Outage required
Conditional
Conditional
Any interval
Outage required
Conditional
6-8 y
2-4 y
Outage required
OLTC inspection
12 y
6-8 y
4y
Diagnostic Technique
Basic Electrical
Winding Ratio
Winding Resistance
Magnetisation Current
x
x
x
x
Leakage Reactance
x
Advanced Electrical
x
x
x
x
Polarisation/Depolarisation
Electrical Detection of PD
Acoustical detection of PD
UHF Detection of PD
OMICRON
Insulation Resistance
Seite 11
RBP
OIP
RIP
increase of capacitance
oil in cracks
or partial breakdowns
partial breakdowns
partial breakdowns
partial breakdowns;
insulator surface wet or
dirty (clean the insulator);
ageing of the inner
insulation;
water in the inner
insulation;
partial breakdowns;
insulator surface wet or
dirty (clean the insulator);
ageing of the inner
insulation;
water in the inner
insulation;
partial breakdowns;
insulator surface wet or dirty
(clean the insulator);
ageing of the inner
insulation;
water in the inner insulation;
dissipation factor is
decreasing with
increasing voltage
dissipation factor is
strongly increasing with
increasing temperature
normal, if constant
Discharges produce
gasses;
Errosion of the cellulose;
production of x-wax
partial breakdowns;
cracks or voids after
electrical or mechanical
stress;
partial discharges
OMICRON
Seite 13
Capacitance
(voltage sweep)
30nF
0.45%
25nF
V
H(V)
OMICRON
V
HL(V)
L(V)
H(V)
HL(V)
L(V)
14000V
12000V
10000V
14000V
12000V
10000V
8000V
0nF
6000V
0.2%
4000V
5nF
2000V
0.25%
8000V
10nF
6000V
0.3%
15nF
4000V
0.35%
20nF
2000V
0.4%
0V
Capacitance
0.5%
0V
Tan delta
(voltage sweep)
Advanced diagnostics
Transformer power and dissipation factor (tan delta)
with variable frequency from 15 Hz to 400 Hz
New transformer
f
H(f)
HL(f)
L(f)
H(f)
OMICRON
HL(f)
L(f)
450Hz
400Hz
350Hz
0%
300Hz
450Hz
400Hz
350Hz
300Hz
250Hz
200Hz
150Hz
100Hz
50Hz
0Hz
0%
0.2%
0.1%
250Hz
0.1%
200Hz
0.2%
0.4%
0.3%
150Hz
0.3%
100Hz
0.4%
0.6%
0.5%
50Hz
0.5%
0.7%
0Hz
0.6%
Advanced diagnostics
Transformer power and dissipation factor (tan delta)
with variable frequency from 15 Hz to 400 Hz
Advantages
OMICRON
Bad bushing
0.32%
0.31%
0.3%
0.29%
0.28%
0.27%
0.26%
0.25%
1.1%
1%
0.9%
0.8%
0.7%
0.6%
OMICRON
14000V
12000V
10000V
8000V
6000V
4000V
14000 V
12000 V
10000 V
8000 V
6000 V
4000 V
2000 V
0V
2000V
0.5%
0.24%
0V
New bushing
Advanced diagnostics
Bushing power and dissipation factor (tan delta)
with variable frequency from 15 Hz to 400 Hz
0.85%
0.75%
only fNOM values
0.65%
Aged bushing
0.55%
Medium-aged bushing
0.45%
New bushing
0.35%
0.25%
0Hz
100Hz
200Hz
300Hz
400Hz
500Hz
OMICRON
Advanced diagnostics
Bushing power and dissipation factor (tan delta)
with variable frequency from 15 Hz to 400 Hz
0.6 %
0.55 %
0.5 %
0.45 %
450 Hz
400 Hz
350 Hz
300 Hz
250 Hz
200 Hz
150 Hz
100 Hz
50 Hz
0 Hz
0.4 %
f
1
OMICRON
OMICRON
0.0045 A
Exciting current
0.004 A
0.0035 A
0.003 A
0.0025 A
0.002 A
0.0015 A
0.001 A
0
10
15
20
25
OMICRON
30
Exciting current
100 mA
80 mA
60 mA
40 mA
B
C
20 mA
0 mA
1
10
11
1000 W
Watt losses
800 W
600 W
400 W
B
C
200 W
0W
1
OMICRON
10
11
Phase A
I out
[mA]
Watt
Loss
[W]
Reactance Phase B
(+/-jX)
I out
[k]
[mA]
Watt
Loss
[W]
Reactance Phase C
(+/-jX)
I out
[k]
[mA]
Watt
Loss
[W]
Reactance
(+/-jX)
Pattern
[k]
Rating
61,56
573,22
67,62
42,81
396,23
88,42
59,00
540,12
67,43
NONE
62,62
569,87
66,09
44,31
410,38
86,00
61,12
559,15
66,15
NONE
64,79
588,69
64,45
45,79
422,97
83,65
63,23
577,36
64,53
NONE
67,05
608,41
62,72
47,44
438,60
81,21
65,46
596,85
62,81
NONE
70,51
625,79
60,10
49,08
452,48
78,84
67,80
617,27
61,04
NONE
71,89
651,07
59,04
50,92
469,99
76,44
70,24
638,84
59,23
NONE
74,44
673,86
57,11
52,74
486,24
73,79
72,79
661,55
57,33
NONE
77,18
698,20
55,26
54,69
503,53
71,31
75,56
685,89
55,56
NONE
80,07
724,07
53,37
56,78
522,56
68,86
78,47
711,65
53,72
NONE
10
82,97
750,83
51,35
58,97
542,90
66,22
81,55
739,10
51,88
NONE
11
86,33
780,10
49,64
61,28
564,43
63,40
84,81
768,08
50,05
NONE
Ratio
16
15
A
B
14
13
12
000
005
010
Tap
OMICRON
015
020
Ratio deviation
0.04 %
TRRatio:
0.02 %
0.00 %
-0.02 %
-0.04 %
C
-0.06 %
-0.08 %
0
10
15
20
Taps
Exciting current
0.01 A
NO
Assessment:
Range:
0.004 A
0.002 A
0
10
Taps
15
20
n/a
AC 2kV
Nominal
values:
Frequency:
50,0Hz
V test:
1000,0V
Tap
V prim. Nom.
Vsec.
nom.
Ratio
nom.
V prim.
001
449610,0/3V
15750,0V
16,4814:1
999,59V
0,0
60,636585V
0,0
16,4849:1
-0,02%
0,004566A
-40,71
002
444890,0/3V
15750,0V
16,3084:1
999,74V
0,0
61,285938V
0,01
16,3127:1
-0,03%
0,004625A
-41,18
003
440170,0/3V
15750,0V
16,1354:1
999,79V
0,0
61,947819V
0,0
16,1392:1
-0,02%
0,004721A
-41,48
004
435440,0/3V
15750,0V
15,962:1
999,85V
0,0
62,622185V
0,0
15,9664:1
-0,03%
0,004791A
-41,86
005
430720,0/3V
15750,0V
15,789:1
999,83V
0,0
63,309776V
-0,01
15,7927:1
-0,02%
0,004902A
006
426000,0/3V
15750,0V
15,616:1
999,81V
0,0
64,010597V
-0,01
15,6194:1
-0,02%
0,005008A
-42,35
007
421280,0/3V
15750,0V
15,4429:1
999,78V
0,0
64,727524V
-0,01
15,446:1
-0,02%
0,005097A
-42,77
008
416560,0/3V
15750,0V
15,2699:1
999,89V
0,0
65,464005V
-0,01
15,2739:1
-0,03%
0,005211A
-42,96
009
411830,0/3V
15750,0V
15,0965:1
999,6V
0,0
66,197601V
-0,01
15,1002:1
-0,02%
0,005334A
-42,71
010
407110,0/3V
15750,0V
14,9235:1
999,78V
0,0
66,981621V
0,01
14,9262:1
-0,02%
0,005422A
-43,33
0,00568A
-45,03
15750,0V
14,5775:1
999,83V
0,0
68,570297V
0,0
14,5811:1
-0,02%
0,005816A
-45,37
15750,0V
14,4044:1
999,78V
0,0
69,396141V
-0,02
14,4069:1
-0,02%
0,005939A
-45,7
014
388230,0/3V
15750,0V
14,2314:1
999,74V
0,0
70,235367V
0,0
14,2341:1
-0,02%
0,006073A
-45,83
0,0
14,0614:1
-0,02%
0,006229A
-46,1
016
378780,0/3V
15750,0V
13,885:1
999,72V
0,0
71,986328V
0,0
13,8876:1
-0,02%
0,006361A
-46,33
017
374060,0/3V
15750,0V
13,712:1
999,78V
0,0
72,902046V
-0,01
13,714:1
-0,01%
0,00651A
-46,59
018
369340,0/3V
15750,0V
13,539:1
999,79V
0,0
73,832764V
-0,01
13,5413:1
-0,02%
0,006659A
-46,73
019
364620,0/3V
15750,0V
13,3659:1
999,75V
0,0
74,779701V
-0,01
13,3693:1
-0,02%
0,006823A
-46,88
14,058:1
999,73V
999,79V
0,0
0,0
67,760643V
71,101517V
0,0
14,7538:1
-0,02%
-42,1
397670,0/3V
15750,0V
14,7505:1
I prim.
392950,0/3V
383500,0/3V
15750,0V
Ratio
012
015
402390,0/3V
Vsec.
013
0.006 A
OMICRON
02.05.2008 13:22
Overload:
011
0.008 A
Date/Time:
OMICRON
OMICRON
Leakage flux
Forces
10 kV winding
220 kV winding
OMICRON
3-Phase Measurement
SN [kVA]
Pk [kW]
f [Hz]
Tap Position:
UHV [kV]
ULV [kV]
uk [%]
Uk [kV]
6500
34.5
50
51.7
27.2
6.250
1.9
f [Hz]
V1 AC [V]
I AC [A]
Z []
Phi []
R []
X []
L [mH]
A-B
50
55.224
1.049
25.7019
85.01
2.2370
25.6044
81.50
B-C
50
54.675
1.048
25.8922
85.24
2.1470
25.8030
82.13
A-C
50
54.464
1.048
25.2547
84.96
2.2196
25.1570
80.08
Average Impedance
f [Hz]
uk [%]
Z []
Phi []
R []
X []
xk [%]
L [mH]
50
6.250
25.7009
85.13
2.1826
25.6080
6.227
81.51
Measured Values:
50
6.229
25.6162
85.07
2.2012
25.5215
6.206
81.24
-0.33
-0.33
-0.07
0.85
-0.34
-0.34
-0.34
V1 AC [V]
I AC [A]
Z []
Phi []
R []
X []
L [mH]
A-B
50
25.061
1.010
24.5892
84.37
2.4102
24.4708
77.89
B-C
50
25.337
1.011
24.8397
84.54
2.3641
24.7270
78.71
A-C
50
25.485
1.010
24.9888
84.44
2.4224
24.8711
79.17
24.8059
2.3989
24.6896
78.5896
0.87
1.45
0.89
0.89
Average Impedance
OMICRON
Measurement in accordance
to IEEE Std C57.12.90
Per-phase test
Advanced diagnostics
Frequency response of stray losses (FRSL)
Measured resistance
Frequency
OMICRON
Advanced diagnostics
Frequency response of stray losses (FRSL)
HV winding
LV winding
Losses through
induced currents
OMICRON
Advanced diagnostics
Frequency response of stray losses (FRSL)
B
HV winding
LV winding
Parallel strands
with twisting
Induced currents
are compensated
OMICRON
Advanced diagnostics
Frequency response of stray losses (FRSL)
B
HV winding
LV winding
Additional losses through
induced currents
OMICRON
Advanced diagnostics
Frequency response of stray losses (FRSL)
5
20.4mH
4
20mH
3
19.6mH
2
19.2mH
1
18.8mH
0
0Hz
100Hz
200Hz
300Hz
400Hz
0Hz
100Hz
200Hz
ZkBC
ZkAC
ZkAB
ZkBC
OMICRON
400Hz
f
ZkAB
300Hz
ZkAC
Tap changer
OMICRON
Connection clamps
Measurement example
Tap changer in
good condition
Winding resistance
0.19
0.18
0.17
0.16
0.15
0.18
0.17
0.16
0.15
0.14
0.14
0
10
15
20
25
30
10
Taps
20
Taps
Rmeas
OMICRON
15
Rref
Rmeas
25
30
Influence of remanence
Exciting current
0.02A
0.018A
0.016A
0.014A
U with remanence
V with remanence
W with remanence
U without remanence
V without remanence
W without remanence
0.012A
0.01A
0.008A
0.006A
0.004A
0.002A
0.0A
OMICRON
Advanced diagnostics
Dynamic LTC diagnostics with CP SB1 (OLTC test)
OMICRON
Advanced diagnostics
Dynamic LTC diagnostics with CP SB1 (OLTC test)
Current
5.0 A
4.9 A
Ripple
4.8 A
4.7 A
-0.05 s
0s
OMICRON
Advanced diagnostics
Dynamic LTC diagnostics with CP SB1 (OLTC test)
2.5%
5.5%
2%
5%
Ripple
Ripple
1.5%
1%
0.5%
4.5%
4%
3.5%
0%
3%
0
10
15
Taps
15
20
Taps
AUP
ADOWN
BUP
AUP
ADOWN
BUP
BDOWN
CUP
CDOWN
BDOWN
CUP
CDOWN
OMICRON
10
OMICRON
Seite 40
Topics
> Transformer
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
> HV Cables
Lifetime expectations
Faults and Aging
Measurement & monitoring parameters /quantities
Page
Page 41
41
Cigre WG 37 -27, Aging of the System - Impact on Planning, Cigre Brochure 2001
OMICRON
Seite 42
Popma J. and Pellis J., Diagnostics for high voltage cable systems, proceedings ERA
conference on HV plant life extension, Belgium, 23-24 November, 2000
OMICRON
Seite 43
Diagnostic Parameters
Contribution of different on-site diagnostics to short and/or long term
condition assessment
a) measurement of dielectric
losses/response
b) PD measurements
c) Dissolved gases analysis
d) DC sheath test
e) visual inspection of accessories
f) Inspection of earthing system
g) Inspection of hydraulic system
h) determination of impregnation
coefficient
Popma J. and Pellis J., Diagnostics for high voltage cable systems, proceedings ERA conference on
HV plant life extension, Belgium, 23-24 November, 2000
OMICRON
Seite 44
Diagnostic Parameters
OMICRON
Seite 45
March 5, 2015
Topics
> Diagnostic Monitoring Systems
> Permanent and Temporary PD Monitoring System for Rotating
Machines
> PD, C, TanDelta, transient Monitoring System for Transformers
> Permanent and Temporary PD Monitoring System for HV Cables
Page
Page 22
OMICRON
Page 3
Diagnostic monitoring
What is diagnostic monitoring?
> The large amount of real-time (trend) data gathered by monitoring
systems can be combined with precise insulation condition assessment
> Effective comparison of historical data is enabled via an easy-to-use and
extendable database solution
Benefits for asset management
> Monitoring answers questions about the present condition of the
equipment and its future performance
> Exact knowledge of insulation state at any time saves money, as
maintenance schedules can be specifically optimized and the service life
of assets can be extended
> Periodic or continuous monitoring of HV equipment is an additional,
essential tool for effective maintenance management
OMICRON
Page 4
Diagnostic monitoring
> Involves no operational variables
> Provides information to evaluate the aging condition of equipment
> Allows assessment of future developments in the aging process
> Supports decision making for ongoing maintenance or the replacement
of components
> is a "planning tool (asset management)
OMICRON
Page 5
Diagnostic monitoring
> Continuous or periodic monitoring of HV equipment is an essential tool for
effective maintenance management
> Monitoring answers questions about the present condition of the
equipment and its future performance
> The large amount of real-time data gathered by monitoring systems can
be used for precise insulation condition assessment
> Effective comparison of historical data is enabled via an easy-to-use and
extendable database solution
> Exact knowledge of insulation state at any time saves money, as
maintenance schedules can be specifically optimized and the service life
of assets can be extended
OMICRON
Page 6
Diagnostic monitoring
> Involves no operational variables
> Provides information to evaluate the aging condition of equipment
> Allows assessment of future developments in the aging process
> Supports decision making for ongoing maintenance or the replacement
of components
> is a "planning tool (asset management)
OMICRON
Page 7
OMICRON
Page 8
OMICRON
Page 9
Topics
> Diagnostic Monitoring Systems
> Permanent and Temporary PD Monitoring System for Rotating
Machines
> PD, C, TanDelta, transient Monitoring System for Transformers
> Permanent and Temporary PD Monitoring System for HV Cables
Page
Page 10
10
Seite 11
OMICRON
Seite 12
OMS 600
Permanent PD monitoring for rotating machines
OMICRON
10 March 2015
OMICRON
Page 14
1st machine
Capacitive
sensors
Acquisition
unit
Fiber optic
cable
Central computer
and software
OMICRON
Page 15
2nd machine
Additional
machines
Capacitive sensors
5
1
1
2
3
OMICRON
Seite 16
Kapazitiver Sensor
OMICRON
Seite 17
OMICRON
Seite 18
OMICRON
Seite 19
OMICRON
Seite 20
UHF
OMICRON
Seite 21
350
100
MHz
40
20
0.5
0.025
10
15
20
25
30
35
40
45
50
% Relative Coverage
OMICRON
Seite 22
55
60
65
70
75
80
85
90
95
100
OMICRON
Seite 23
OMICRON
Seite 24
G) Turn Insulation
OMICRON
C
C
Seite 25
OMICRON
Seite 26
OMICRON
Seite 27
high
Delamination: conductor
main insulation
normal
J&L
corona
normal
medium
low
high
low
low
normal
Micro void
noise
high
noise
low
low
Contact PDs
200 pC
medium
900 pC
Alarm
high
500 pC
Delamination of tape layersbigger inner micro void
OMICRON
Time
Seite 29
> 2015-03-10
E
E
36
35
34
33
32
31
30
TE
CORE
L1
L2
L3
neutral
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
36
35
34
33
32
Inner PD-source
L1>L2>L3
L1
L2
Inner PD source in L1
L2
L3
L3
L1
timeframe
1 s
Because of the complex propagation and coupling characteristics of PD impulses, every source
of PD impulses has unique impulse to couplings ratio and consequently a characteristic 3PARD
signature
Seite 31
OMICRON
3/10/2015
Noise
L1 L2 L3
L1
L2
noise
L2
L3
L3
L1
timeframe
1 s
Electrical interferences and outer noise usually have the pulse-like nature of PD but
occur with almost same amplitude on all three phases of the machine.
OMICRON
Seite 32
OMICRON
Seite 33
OMS 605
Temporary PD monitoring for rotating machines
OMICRON
10 March 2015
OMICRON
Page 36
Seite 37
Coupling
capacitors
Terminal box
OMS 605
Notebook
with software
OMICRON
Seite 38
Signal cable
Grounding band
USB cable
Grounding clamp
2
1
4
3
5
6
OMICRON
Seite 39
3
4
OMICRON
Seite 40
OMICRON
Page 41
OMICRON
Seite 42
3-channel
OMICRON
Seite 43
4-channel
OMICRON
Seite 44
OMICRON
Seite 45
OMICRON
Seite 46
PRPD
3PARD
OMICRON
Seite 47
OMICRON
Seite 48
Topics
> Diagnostic Monitoring Systems
> Permanent and Temporary PD Monitoring System for Rotating
Machines
> PD, C, TanDelta, transient Monitoring System for Transformers
> Permanent and Temporary PD Monitoring System for HV Cables
Page
Page 49
49
OMICRON
Seite 50
OMICRON
Seite 51
Theory
Supplementary bushing monitoring is vital
> Bushing failures contribute significantly
to failure statistics of power transformers
> With dissolved gas analysis (DGA), only
by-products of insulation degradation
inside the tank can be detected not at
the bushings
Tap changer
26%
Other
1%
Bushings
17%
Isolation
1%
OMICRON
Page 52
Lead exit
7%
Winding
45%
Theory
Bushing defects and their indicators
> Partial breakdowns
> Capacitance
> Partial discharges
OMICRON
Page 53
Theory
Capacitance and dissipation/power factor
> Critical indicators of partial breakdowns,
cracks, aging and moisture in bushings
> With online monitoring, these indicators
are measured under real load conditions
> Continuous information about insulation
status allows early corrective action
> Reliable and accurate measurement
results enable comparison with
acknowledged standards
OMICRON
Page 54
Capacitance
Dissipation/
power factor
OMICRON
Page 55
Partial
discharge
Transient
over-voltages
OMICRON
Page 56
OMICRON
Page 57
OMICRON
Page 58
UHF sensors
1
6
3
4
2
OMICRON
Page 59
OMICRON
Page 60
OMICRON
Page 61
OMICRON
Page 62
OMICRON
Page 63
OMICRON
Page 64
OMICRON
Page 65
MONTRANO software
System configuration and control
> Topology views of the monitored assets
> Configuration of all monitoring-related parameters (admin only)
> Configuration of alarm rules (admin only)
OMICRON
Page 66
MONTRANO software
Warnings and alarms
> Defined threshold settings for
alarm notifications
> Real-time display of current
alarm level
> View of events that triggered
an alarm
> Alarm confirmation by operators
> Confirmed alarms change color or
> Are taken out of viewing window
(optional settings)
> All alarm data saved
OMICRON
Page 67
MONTRANO software
Data visualization
> Configurable trend charts show
data for each monitored parameter
> Tooltip pop-up for easy accessable
detailed information of each data
point
> All data records and corresponding
charts can be exported for
reporting
OMICRON
Page 68
Technical specifications
Capacitance, dissipation/power factor
>
Up to 3x Ur
5 MS/s
300 ns (20%...80%)
1.5 %
20 ms...2 s
Frequency range
Sampling rate
Frequency range UHF
System noise
OMICRON
16 kHz...30 MHz
125 MS/s
100 MHz...2000 MHz
< 1 pC
Page 69
References
"Many other systems and techniques had
been tried previously and MONTRANO was
the only system to successfully prove that
the PD was real.
Mr. Colin Feely Plant Maintenance
Engineer, Powercor Australia Ltd.
OMICRON
Page 70
Standards
Offline bushing tests
> Capacitance:
> Cigre working group A2.34: Guide for transformer
maintenance presents capacity changes when
one control field layer is short-circuited
1.0
IEC 60137
0.8
max: 0.7%
0.6
0.4
0.2
0.0
typ. new
OIP
typ. new
RIP
Dissipation Factor(%)
1.0
0.4
IEEE C57.19.01
max: 0.85%
0.2
max: 0.5%
typ. new
OIP
typ. new
RIP
0.0
Acceptance level of dielectric losses for bushings of different design
OMICRON
Page 71
No. of
layers
Capacitance
change in %
123
14
7.1
245
30
3.3
420
40
2.5
550
55
1.8
0.8
0.6
Voltage
[kV]
Methods
Relative C, DF/PF measurement
> Changes in grid unbalance (phase shifts) can have a misleading impact on
online C, DF/PF measurements when the sum-of-current approach or a
bushing-to-bushing comparison is used to determine the values
> MONTRANO uniquely offers alternative options to reduce the uncertainty
of measurement
Sum of voltages
Page 72
Methods
Benefits of highly accurate C, DF/PF measurement
tg in %
Capacitance (pF)
500
0.8
467 2 pF
measurement
DF (%)
478 pF warning
0.6
Temperature in C
0.4
0.2
400
0.27 0.05 %
measurement
RIP
OIP
0.0
Capacitance and DF measurement over 1.75 years, and alarm levels
OMICRON
Page 73
Temperature dependance
of the dissipation factor
[Source: Karl Frey, Micafil]
Methods
Reducing measurement uncertainty
> Grid unbalance impacts are minimized
with the use of a 2nd transformer for
comparison or a VT as reference
> Using a VT as reference, the smallest
uncertainty is achieved
1
Bushing
aging
Reference
aging
Grid
unbalance
Relative
measurement
Highest
uncertainty
OMICRON
Page 74
2
Bushing
aging
Reference
aging
Dual transformer
comparison
3
Bushing
aging
Voltage transformer
reference
Lowest
uncertainty
Methods
Measurement configuration options
3
Relative measurement
(bushing-to-bushing)
Dual transformer
comparison
Lowest
uncertainty
Highest
uncertainty
OMICRON
Absolute measurement
(VT reference)
Page 75
IB
2 S f CB U High CB
IB
2 S f U High
The diagnosis of bushings based on their capacitance is done by comparing the values from
on-line monitoring with the values from off-line measurements. The difference between them
gives the capacitance variation C. This quantity (C) has to remain within certain limits. The
example of capacitance change when one control field layer is short-circuited for bushings of
different rated voltage is presented in table 2 [1]. The failure of such a layer does not
necessarily lead the total failure of the bushing but it creates a higher field distribution on the
healthy layers.
Table 2. Voltage class and change of capacitance for condenser type bushings [1]
Voltage in kV
123
245
420
550
No. of layers
14
30
40
55
Change in %
7.1
3.3
2.5
1.8
where: is the bushing temperature and
cMat is a correction factor which depends on
the material of the bushings insulation
system (OIP, RIP, RBP-Drysomic and RBPSuperresocel) and has to be calculated for
each bushing separately. The temperature
dependence of the dissipation factor for
different type of bushings is presented below:
tgG i'
rCAL i tgG i
'
'G '
GU' GU'
2
As the bushings are not identical the angle 'G will not be zero. Furthermore, if the two
'
bushings get aged in a similar way, 'G will remain constant. This means, if the rate of ageing
is the same for both bushings, ageing effects cannot be detected.
'
'G ' > 'G '' - shows the degradation of the bushing 1
'G ' < 'G '' - shows the degradation of the bushing 2
IU 1
2 S f CU1off line U1 U1
IU 1
2 S f CU1off line
CU1on line
IU1
2 S f U1
where: U1 is the level of the voltage on the HV side and CU 1 off line is the capacitance of the
bushing from off line measurement (transformer 1 phase U ).
Formula above (left) is applied for each bushing to determine the voltages U1 ... U6. With the
values of the voltage determined using equation above (reight), the on line values of the
capacitance for each bushing are calculated.
A change in the capacitive leakage currents measured at the taps indicates a change in the
capacitance of the bushings. The system voltage variations do not affect the measurement
since both bushings are affected in the same way by these variations. The diagnosis of the
bushings is done based on the difference between the values of the capacitance determined
in off-line and on line mode C. This quantity (C) has to remain within certain limits.
Source: Proposal from OMICRON, GBEL and HSP for the Cigre Working Group A2.43 Reliability of HV Bushings
PAGE 90
Commissioning
Standards and recommendations regarding DF
Type
Main insulation
RIP
Resin
impregnated
paper
OIP
Oil
impregnated
paper
RBP
Resin bonded
paper
< 0.7 %
< 0.7 %
< 1.5 %
< 0.85 %
< 0.5 %
<2%
0.3-0.4 %
0.2-0.4 %
0.5-0.6 %
PD (IEC60137) at Um
1.5 Um /3
1.05 Um /3
< 10 pC
< 5 pC
< 5 pC
< 10 pC
< 5 pC
< 5 pC
< 300 pC
PAGE 91
RBP
OIP
RIP
increase of capacitance
oil in cracks
or partial breakdowns
partial breakdowns
partial breakdowns
partial breakdowns;
insulator surface wet or
dirty (clean the insulator);
ageing of the inner
insulation;
water in the inner
insulation;
partial breakdowns;
insulator surface wet or
dirty (clean the insulator);
ageing of the inner
insulation;
water in the inner
insulation;
partial breakdowns;
insulator surface wet or dirty
(clean the insulator);
ageing of the inner
insulation;
water in the inner insulation;
dissipation factor is
decreasing with
increasing voltage
dissipation factor is
strongly increasing with
increasing temperature
normal, if constant
Discharges produce
gasses;
Errosion of the cellulose;
production of x-wax
partial breakdowns;
cracks or voids after
electrical or mechanical
stress;
partial discharges
Methods
Partial discharge (PD) detection
> PD detection is the early indicator of defects in dielectric insulation
> In typical field environments of transformers, electrical PD detection
suffers from a high noise level
> Originated from outside the transformer, like corona
OMICRON
Page 93
Methods
PD denoising and source separation
Phase C
Phase C
Phase B
Phase B
Channel 2
3-band filter
Phase A
Phase A
Channel 1
3-band filter
PD diagrams of 3 phases
without separation
OMICRON
Channel 3
3-band filter
Page 94
Separated PD activity:
located in phase C
Methods
Recording of transient over-voltages
> Detection directly at bushings
> Where the dielectric stress occurs
1.5e+005
1.5e+005
1.0e+005
1.0e+005
5.0e+005
Voltage in kV
Voltage in kV
5.0e+005
0.0e+005
-5.0e+005
-1.0e+005
0.0e+005
-5.0e+005
-1.0e+005
-1.5e+005
-1.5e+005
0.095
0.100
0.105
t in ms
0.110
0.115
0.085
OMICRON
Page 95
0.090
0.095
0.100
t in ms
0.105
0.110
0.115
0.120
0.125
Topics
> Diagnostic Monitoring Systems
> Permanent and Temporary PD Monitoring System for Rotating
Machines
> PD, C, TanDelta, transient Monitoring System for Transformers
> Permanent and Temporary PD Monitoring System for HV Cables
Page 96
OMICRON
March 5, 2015
OMICRON
Seite 98
OMICRON
Page 99
Temporary monitoring
system
OMICRON
Page 100
OMICRON
10 March 2015
Page 102
System configuration
Extendable system for long and short link
HV cable systems
Acquisition unit
Fiber optic cable
Inductive
power supply
HF current trasformer
OMICRON
Page 103
System components
1
Current transformer
Acquisition unit
6
2
1
2
4
1
OMICRON
Page 104
System components
MCT 120 high-frequency current transformer (HFCT)
0 mm gap
180kHz-40MHz,
0.22 mm gap
290kHz-40MHz,
0.44 mm gap
380kHz-40MHz,
0.66 mm gap
MCT 120 HFCT
OMICRON
Seite 105
System components
MCT 120 high-frequency current transformer (HFCT)
> Fixation to wall or frame
Cable Clambing
Universal
Page 106
System components
UCS1 high-frequency UHF sensor (option)
> Measures PD in high frequency
UHF range
> Designed specifically for PD detection
at HV cable terminations
> Ensures sensitive measurements
in noisy environments
> Galvanically isolated signal output
prevents induced disturbance currents
Seite 107
System components
Data acquisition unit
> Synchronous multi-channel PD data acquisition
> Advanced technology like 3PARD and 3CFRD for PD source separation
> Optical isolation between partial discharge acquisition units and server
using fiber optic cable
OMICRON
Page 108
System components
IPS 820 inductive power supply
> Allows operating the acquisition units offside
> Ideal for cable tunnel applications where copper
wiring has to be avoided and voltage sources are
not available
> Electronic elements are filtered and optimized to
prevent measurement disturbance
OMICRON
Page 109
Software
Central computer and monitoring software
> State-of-the art database system for longterm data storage and retrieval
> Web-based data access & visualization
> Different user roles/logins:
> Operators:
> Overview of full system setup
> View all data with free configurable charts
and diagrams
> Confirm warnings and alarms
OMICRON
Page 110
OMS 605
Temporary PD monitoring for HV cable terminations and joints
OMICRON
10 March 2015
OMICRON
Page 112
Seite 113
OMS 605
Notebook
with software
HFCT
sensors
OMICRON
Seite 114
Notebook
with software
OMS 605
HFCT
sensors
OMICRON
Seite 115
0 mm gap
180kHz-40MHz,
0.22 mm gap
290kHz-40MHz,
0.44 mm gap
380kHz-40MHz,
0.66 mm gap
MCT 120 HFCT
OMICRON
Seite 116
Cable Clambing
Universal
Page 117
Seite 118
Seite 119
OMICRON
Seite 120
OMICRON
Seite 121
OMICRON
Seite 122
PRPD
3PARD
OMICRON
Seite 123
OMICRON
Seite 124
Presentation 25.1
Abstract
This
paper
describes
general
design
considerations, the system layout and practical
experiences with on-line monitoring of electrical
apparatus in large thermal power plants. The
availability of electrical equipment in power plants
is of utmost importance, since outages can easily
lead to financial loss of millions of Euros, even per
day. Therefore operators of power plants pay large
attention to the system health of their apparatus,
particularly these in the functional chain of power
generation, as e.g. generators and generator stepup transformers. On-line monitoring appears to be
an important tool to provide reliable status
information, detect failures early and utilize the full
life time.
For identifying possibly unreliable components,
general failure statistics provide first input. For
example in the work of Cigr WG A2.37
"Transformer Reliability Survey", most failures
have been attributed to the dielectric of power
transformers. The same failure mode applies for
generators. Experiences with specific grid
conditions and particular apparatus designs are
another input for selecting appropriate monitoring
systems. Finally, economic calculations prescribe
the frame of the extent of the applied technology.
For large turbo generators, it was decided to
monitor the temperature in the slots of the stator
and the partial discharges of the stator. The
measurement of partial discharges is carried out in
a frequency range of several 100 kHz to a few
MHz for being more sensitive towards signals
coming from the inside of the stator windings.
Rotating electrical machines generally have
several sources of PD, therefore source separation
is essential to discriminate between noise, PD with
low failure probability and dangerous PD. In the
context of this work, amplitude-based impulse
correlations were most effective (3 phase
amplitude relationship diagram - 3PARD). Alarms
are based on levels as well as on rate of change.
PD pattern classification is applied for a basic risk
assessment.
For the large generator step-up transformers it was
decided to monitor temperatures, dissolved gases
in oil (DGA), bushing capacitance and dissipation
factor, partial discharges and transient overvoltages. Transient over-voltages where suspect to
be the reason for a number of breakdowns in the
last decades, so their monitoring with true
waveform recording was of high importance. Actual
measurements show the interaction between the
transformer and the grid. For monitoring partial
OMICRON electronics Asia Limited 2013 Conference on Electrical Power Equipment Diagnostics
Presentation 25.2
Thermal;
17%
Mechanical
; 27%
Unknown;
14%
Chemical; 3%
Thermal;
11%
Electrical;
15%
Dielectric;
56%
Dielectric;
37%
Mechanical
; 20%
Component
Stator
winding
Rotor
bearings
Cooling
system
Monitored Parameter
Partial discharge monitoring
Slot temperature monitoring
Vibration monitoring
Hydrogen dew-point and
hydrogen purity monitoring
OMICRON electronics Asia Limited 2013 Conference on Electrical Power Equipment Diagnostics
Presentation 25.3
Disturbances
200 pC
medium
E1
900 pC
Alarm
Q
high
S2
500 pC
Component
Insulation
system
time
Windings
Bushings
OLTC
Frequency
6 or 12 months
Dissolved gas
analysis
Furanes
6 or 12 months
Dielectric response
analysis
Winding resistance
Transmission ratio
Short circuit
impedance
No load current
Frequency
Response Analysis
FRA
Capacitance and
dissipation factor
Dynamic resistance
2 years / After
indication
After indication
8 years
8 years
8 years
8 years
8 years / after
indication
4 years
4 years
Component
Insulation
system
Bushings
OLTC
Figure 2: An example of separation of PD sources by
3PARD
Diagnostic Test
Oil analysis
Monitored Parameter
Oil temperature
Dissolved gas analysis (gas
selective multi-gas systems)
Moisture in oil
Partial discharges
Transient over-voltages
Capacitance and dissipation factor
Motor current analysis
OMICRON electronics Asia Limited 2013 Conference on Electrical Power Equipment Diagnostics
Presentation 25.4
C2H2
CH4
H2
C2H4
CO
AmbientTemp
120
C2H6
CO2
OilTemp
100
1000
80
100
60
10
40
1
Temperature / C
10000
20
0,1
0
25-6
30-6
5-7
10-7
15-7
20-7
Date
25-7
Figure 4 shows the individual gases measured online over a time period of one month. At 03.July,
the hydrocarbons strongly increased. DGA
laboratory measurements from oil samples
confirmed the on-line readings. Various analysis
schemes were applied, pointing towards a local hot
spot, although not congruently. The very specific
conclusion of the Rogers assessment rule finally
agreed with the actual failure. In discussion with
the transformer manufacturer the possible location
of the hot spot was identified and later, after
opening the transformer, confirmed. A short circuit
between the tank and one field grading electrode
of the winding caused circular currents which lead
to over-heating and burned the oil-paper insulation.
For operation of a power plant, such early
detection of an ongoing failure is of outmost
importance. This allowed early planning of the
outage, searching for a replacement of the
transformer and performing the exchange in less
than 14 days. An unplanned outage e.g. based on
Buchholz tripping might easily result in an outage
of months duration with large consequent financial
losses.
1
40
0.9
35
0.8 IEC60137 max: 0,7 %
30
0.7
25
0.6
0.5
20
0.4
15
U-phase W-phase
0.3
10
0.2
V-phase
V
h
5
0.1
0
00
5
10
time / days 15
OMICRON electronics Asia Limited 2013 Conference on Electrical Power Equipment Diagnostics
temperature in C
Presentation 25.5
10
Ph. V
1
Ph. U
0.1
400
W Phase
0.01
0
10
t in ms
20
0
V Phase
-200
100
-60
12
t in ms
18
U in V
U Phase
P in dBm
Voltage in kV
200
-400
Ph. W
-100
-120
0.001
10
Ph. V
1
Ph. U
0.1
0.01
0
10
t in ms
Ph. W
20
10
t in ms
20
500
1000
f in MHz
OMICRON electronics Asia Limited 2013 Conference on Electrical Power Equipment Diagnostics
Presentation 25.6
Voltage Transformers
OMS843
- C/DF
(Reference)
Generator
PDM600
ca. 10 m
OMS843
- PD
- Transients
- C/DF
Transformer 1
Transformer 2
UHF620 + UVS
Bibliography
UHF620 + UVS
OMICRON electronics Asia Limited 2013 Conference on Electrical Power Equipment Diagnostics
INTRODUCTION
SYSTEM CONCEPT
CASE STUDY:
1160MVA, 27KV
TURBO
GENERATOR
Defects type
rmS1
rmS2
rmS3
rmS4
rmE1
rmE2
rmE3
a.
b.
c.
Figure 7: PD data evaluation using 3PARD feature
CASE STUDY:
180MVA, 16.5KV
TURBO
GENERATOR
Figure 8: Three phase PD trend and PRPD patterns before and after maintenance
CONCLUSIONS
REFERENCES
SUMMARY
The paper deals with on-site PD testing, diagnostics and continuous monitoring of HV XLPE cable
systems and describes the method that can detect and locate partial discharges at all accessories
simultaneously. The examples of after-installation PD measurement procedure applied on 220 kV
cable line as well as in-service PD measurements are shown in the paper. For continuous monitoring
an advance concept to assess and monitor the condition of the insulation of a 420 kV cable system in a
10 km underground tunnel is described. Partial discharges are continuously monitored at all 400 kV
joints and terminations. At the same time, the system performs measurement of oil pressure in 400 kV
terminations as well as the condition of all sheath voltage limiters placed in the 400 kV joint bays.
Major attention is paid to the description of the technology of the partial discharge acquisition unit and
the inductive power supply solution. The advanced features for the elimination of disturbances and for
separation of different types of insulation defects based on synchronous, multi-channel and multifrequency techniques are shown. The processing of monitoring data is implemented in a modular
software system that allows reliable long-term storage of monitoring data and provides remote access
via a web interface. The separate chapters are dedicated to the procedure of site acceptance for the
monitoring system and also to system maintenance strategy.
KEYWORDS
HV XLPE Cables, after-installation test, partial discharges, continuous monitoring
wojciech.koltunowicz@omicron.at
INTRODUCTION
Cross-linked polyethylene (XLPE) with its excellent insulating properties is used both for new cable
installations as well as for replacing old cable systems up to 500 kV. The correct design of the XLPE
cable and its accessories (terminations and joins) is checked by the type tests and the quality of
production is checked by routine tests at the manufacturers plant according to relevant standards.
Installation work on site has an additional risk of introducing faults, as conditions are not ideal for an
installation of EHV accessories. Small particles, dust, moisture, etc. might lead to defects in
electrically critical locations of the accessories. Dielectric tests performed on site do not replace type
tests and routine tests. They are supplementary to dielectric routine tests and aim at checking the
dielectric integrity of the fully assembled cable line in order to eliminate defects such as damages
during transportation and lay-out, or incorrect assembly of the accessories. On-site tests are not
intended to check the design but they shall mainly detect any critical defects after installation on site,
which may decrease the withstand voltages of the new installation below the coordination withstand
voltages during on-site tests and later in service.
The most important stress situations for high-voltage cables during normal operation arise from the
operational alternating voltage and from switching over-voltages. Consequently, the preferred voltage
for on-site tests is ac voltage of industrial frequency. For on-site testing a larger frequency tolerance is
acceptable but frequencytuned circuits can easily be handled. Voltage testing delivers only binary
results (withstand or breakdown). Therefore, it is recommended to combine ac voltage testing with
sensitive on-site partial discharge (PD) measurements [1, 2]
A major part of all in-service failures in HV XLPE cables can be attributed to the insulation system of
accessories (joints and terminations) [2]. These failures will normally develop over time, due to the
ageing process during the service life or due to local degradation in the material. In order to detect
these changes at an early stage, detailed information on the actual insulation condition is necessary.
With suitable sensors this information can be derived by monitoring partial discharge activity during
the operation of the equipment.
This paper describes the best practice for performing after installation testing, checks in service, as
well as continuous monitoring of PD to assess the quality of the HV XLPE cable system and presents
a key tool to significantly reduce the risk of forced outages.
II
HV tests are executed on site for all extruded HV cables and on-site test procedures have to be agreed
between the manufacturer and user on the basis of international and national standards.
Two IEC standards describe after installation tests of extruded cable systems: IEC 60840:2004 is
related to cables of rated voltages from 30 kV (Um = 36 kV) up to 150 kV (Um =170 kV) and
IEC 62067:2001 to cables of rated voltages from 150 kV up to 500 kV (Um = 550 kV) [3, 4]. The AC
test voltage level for the on-site test of new cable systems depends on the cable rated voltage and is
between 1.7 U0 and 2.0 U0 for rated voltages between 30 kV and 150 kV [2]. At higher rated voltages
the test voltage decreases from: 1.4 U0 (220-230 kV) to 1.3 U0 (275-345 kV), 1.2 U0 (380-500 kV) and
1.1 U0 for cables of 500 kV rated voltage.
High test power, especially required for long cable lines testing, can only be effeciently generated by
mobile resonant test systems, where the weight-to-power ratio and feeding power demand is relatively
low and the transport volume is acceptable (figure 1).
Fig.1.
The voltage during the dielectric test should be applied in steps (figure 2). PD measurements should be
taken at every voltage step. By increasing the test voltage in steps of e.g. 20% of the maximum test
voltage, critical defects are usually identified before breakdown.
Fig.2.
An example of voltage application in steps during dielectric on-site testing of 220 kV XLPE cable system
Because HV cables must be tested at the manufacturing facility prior to shipping to the installation
site, the on-site PD measurements focus on the field-installed accessories. For this purpose, each
accessory has to be equipped with special sensors to pick up the PD signal with high efficiency (figure
3). In one particular case of after installation testing of 220 kV XLPE cable line, 15 high frequency
current transformers (HFCTs) were used to pick up the PD signal [5]. They were equipped with an air
gap in the core to prevent magnetic saturation of the core.
Fig.3.
PD test arrangement
The PD system, HFCTs (MCTs100) connected to AUs (MPD600), was checked before installation
with calibration pulses injected to all the units connected in the daisy chain. On site, one MPD600 unit
was connected to the coupling capacitor and calibration was performed according to the IEC 60270
and IEC 60885 standards [1]. For the other 15 PD AUs mounted into the link boxes and close to the
terminations and connected in daisy chain with fibre optics (FO), the check of performance was
performed. A real calibration was not possible because impulses cannot be injected directly into the
closed and buried joint. The calibrator was connected to one turn of wire through the core of the
HFCT.
The resonance test set used IGBTs as switching semiconductors. They produce strong impulses,
which cause high interference impulses on the MPD600 instruments, especially to those units which
are rather close to the resonance test set. To eliminate this effect, PD signal gating was performed by
one extra MPD600 unit (with antenna) that was installed close to the IGBT circuit to receive the
switching impulses. The result of gating is shown in figure 4.
Fig.4. PD measurement at the coupling capacitor without gating (left) and with gating (right)
The final test voltage was 180kV, the test frequency was 25.95Hz and the test current was 68.1A. The
measurement frequency for all MPD600 units was set in the range from 2MHz to 3MHz. The
measurements at lower frequencies were not possible due to the strong interferences.
The PD results at 180kV of all MPD600 units are shown in figure 5. The tests could be finished within
3 days including the safety instructions, mounting of MPD600 to all link boxes and changing the
jumpers to all phases. All tests were successfully concluded and no partial discharges could be
detected above the noise level.
Fig.5. PD measurement at 180kV with 15 MPD units - 1.1=coupling capacitor, 1.3=joint 1-2, 1.4=joint 2-3,
1.5=joint 3-4, 1.6=joint 4-5, 1.7= joint 5-6, 1.8= joint 6-7, 1.9=joint 7-8, 1.10=joint 8-9, 1.11=joint 9-10,
1.12=joint 10-11, 1.13=joint 11-12, 1.14=joint 12-13, 1.15=end terminal GIS
Fig.6.
Test set up
Fig.7.
Multi-channel (3PARD) and multi-spectral (3CFRD) techniques were applied to analyse the results
and eliminate the background noise [6]. Figure 8a shows the synchronous PRPD patterns in three
phases of the GIS cable terminations and their equivalent 3PARD diagrams. Several clusters are
visible in the 3PARD diagram. The most relevant clusters were selected and the back transformation
to the PRPD patterns was performed. External disturbances have been successfully separated from
PD-like signal in phase S (figure 8b). As calibration of the set up was not possible in on-line
conditions, the PD values displayed in the PRPD patterns provide no significant information.
a)
b)
Fig.8. PRPD pattern before (a) and after separation and back transformation (b)
A monitoring session at the GIS side of the cable was initialized after the on-site PD measurements.
PD data was acquired for a centre frequency of 5 MHz and bandwidth of 1 MHz for 18 hours and had
been stored every 10 minutes. These values build up the PD trend diagrams (figure 9). In the figure,
the random appearing peaks of disturbances can be observed. The Besides the PD values, PRPD
patterns and 3PARD diagrams were stored in the database as pictures for each measuring point.
Fig.9.
In case of PD signals, the acquisition unit performs advanced pre-processing of the raw data. The
disturbances are removed and main characteristics of the PD signal are determined. The output of the
data pre-processing is transferred to a server that enables long-term data storage. Advanced intelligent
pre-processing reduces the amount of data to adequate levels for transmission over a communication
network.
HFCT sensors mounted on cross-bonding (CB) links are used to detect PD directly at the accessories
(figure 11). The PD acquisition system consists of one four-channel acquisition unit for each
accessory, connected to a data concentrator. One data concentrator collects monitoring data from two
or three acquisition units via optical fibres and routes them to a server. Pre-processing functions are
already performed in data concentrator like e.g. band passes integration, gating and denoising.
The active components of the monitoring system need electrical power for several processes. Beside the
computer and communication devices, also pre-amplifiers or signal converters (electrical to optical,
e.g.) located close to the sensors have to be supplied with power. The Inductive Power Supply provides
the necessary electronics to supervise and manage the dc current delivered at its output depending on
the various HV cable current load situations (figure 12). The electronics elements are filtered and
optimized to avoid any disturbances of the PD measurements close by.
Fig.11.
TREND ANALYSIS
The monitoring system provides data from each of the PD and oil pressure sensors in two time
intervals: permanent and periodic one (Table I). During the permanent mode, the data are acquired
every 2-3 seconds, compared with threshold values and in real time displayed in graphical user
interface (figure 14). In case those data are within normal margins, they will be coloured in green. If
the values exceed thresholds for "warning" or "alarm" they are coloured in yellow or red accordingly.
Periodic measurements are initiated in equidistant time spans, e.g. every hour. Duration of the periodic
measurement is normally 1 min. During this time span all mentioned scalar values are calculated and
PRPD (phase resolved PD) and 3PARD diagrams are acquired. These data are saved for later postprocessing and trend visualization. Unscheduled periodic measurements are triggered in case of one or
more measured quantities exceeding the threshold level.
Permanent
Periodic
Partial Discharge
for 1 every 1 h
every 2- 3 sec
once every 1 h
SVL status
--
once every 8 h
Maintenance scheme
Activity to be
performed
Visual check
Yearly
Functionality check
Every 3 years
Updates
Every 3 years
Owner
Owner &
system provider
System provider
Data evaluation
Periodic reports
System provider
Expert consultancy
In case of PD event
System provider
Periodicity
Responsibility
Regular checks and evaluation of data values and trending for the acquired partial discharges stored in
the database are also possible, if requested by the system owner. In case of repetitively reported
trending alerts or alarms or specific PD events detected by the monitoring system, PD Expert
consultancy and support can be requested by the system owner.
VI CONCLUSIONS
It can be recommended during after-installation testing to combine ac voltage testing with sensitive
on-site PD measurements;
PD measurements should focus on the field-installed accessories and the method should detect and
locate partial discharges at all accessories synchronously.
The monitoring system provider shall support the asset owner in all stages of the monitoring
project, from system design to maintenance of the monitoring system in service;
Separation of PD sources and suppress of external noise is performed by multi-channel evaluation
techniques;
The Inductive Power Supply provides the necessary power to the monitoring equipment when no
power supply is available;
To verify the status of SVL, the system utilizes the CB link loop impedance;
A modular, distributed monitoring software system allows reliable long term storage of monitoring
data and provides access via web interface.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
IEC 60885-3, Electrical test methods for electric cables Part 3: Test methods for partial
discharge measurements on lengths of extruded power cable
CIGRE Working Group 21-09 After Laying Tests on High Voltage Extruded Insulation Cable
Systems, ELECTRA No. 173 AUGUST 1997
IEC 60840, Edition 3.0 2004-04 Power cables with extruded insulation and their accessories
for rated voltages above 30 kV (Um= 36 kV) up to 150 kV (Um= 170 kV) Test methods and
requirements
IEC 62067, Edition 1.1 2006-03 Power cables with extruded insulation and their accessories
for rated voltages above 150 kV (Um= 170 kV) up to 500 kV (Um= 550 kV) Test methods
and requirements
M. Krger, R. Hummel, S. Bhler, M. Ufferhard, A. McGuigan, "Partial discharge measurement
and Monitoring on High Voltage XLPE Cables", CIGRE Symposium in Auckland, 2013
W. Koltunowicz and R. Plath, 2008, "Synchronous multi-channel PD measurements", IEEE
Transactions on Dielectrics and Electrical Insulation, vol.15, no.6, pp. 1715-1723, 2008
D. Gieselbrecht, W. Koltunowicz, A. Obralic, T, Ritz, P. Christensen, B. Schneider, K.H.
Cohnen, Monitoring of 420 kV XLPE Cable System in Underground Tunnel, published in
proceedings of the 2012 IEEE International Conference on Condition Monitoring and Diagnosis
(CMD), September 23-27, 2012, Bali, Indonesia
W. Koltunowicz, A. Obralic, A. Belkov, D. Gieselbrecht,"Continuous PD Monitoring of HV
XLPE Cable Lines", paper B1.9, in proceedings of: 8th International Conference on Insulated
Power Cables (Jicable), Paris-Versailles-France, 19-23 June, 2011.
voltage
I.
equipment;
monitoring;
partial
INTRODUCTION
III.
Fig. 6. The 3-phase PD trend for conventional measurements with PRPD and
3PARD diagrams (right).
Fig. 7
Fig. 9
Fig. 8
IV.
TABLE I.
Cluster
No.
6
7, 8
3, 4, 5
Defect
class
rmsS1
rmS2
rmS3
rmS4
rmE1
rmE2
rmE3
CLASSIFICATION OF PD DEFECTS
Defect type
Discharge in micron cavities
Delamination of the insulation layers
Delamination of the insulation on the copper side
Delamination of the insulation on the core side
End-winding surface discharge (tracking)
End-winding surface discharge (gap discharge)
Discharge between corona protection and stress
grading layers
V. CONCLUSIONS
The following monitoring system solutions, hardware
components and software features were successfully applied to
monitor HV equipment:
The Inductive Power Supply delivers the electrical
power to the active components of HV cable
monitoring system in tunnels.
The Monitoring Software evaluates the status of SVL;
no specific hardware elements were necessary.
The Multi-channel and multi-spectral classification is
used to separate PD sources and discriminates them
from noise.
The reference voltage signal taken from voltage
transformers of the substation allows the dissipation
factor and capacitance of the bushings to be calculated
with high accuracy.
The combination of conventional and unconventional
UHF PD measurements increases the sensitivity and
reliability of the measurements.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Monitoring of a Distribution
Transformer at Winchelsea Substation
by
L.V. Badicu, W. Koltunowicz, A. Piccolo,
A. McGuigan, C. Feely
Abstract
Background
The transformer is a 5/7 MVA ONAN/ONAF 66000/22000 V Dyn1 unit constructed in 1998. It
was initially put into service in a city environment in Melbourne where it operated
predominantly on a fixed tap and without any problems until removed from service in 2007
as part of a major upgrade project. Later in 2007, the transformer was relocated to
Winchelsea Zone substation (figure 1), a rural environment where it was placed in service to
replace a failed transformer. Once again operation was on a fixed tap until 2009 when the
tap changer was put into automatic operation and used to provide voltage regulation for
the station. The unit was not heavily loaded and gave no indication of any problems until
the annual oil sample dissolved gas analysis (DGA) result from February 2010, which
indicated 5400 ppm of hydrogen. The dissolved gas ratios indicated that partial discharge
(PD) was taking place.
In order to better monitor what was occurring, an on-line DGA monitor was installed giving
4 hourly readings. Initially the monitor recorded less H 2 ppm, but it was quickly realized that
this was actually a false reading due to the fact that the detection limit for hydrogen of 3000
ppm had been exceeded and, counter-intuitively, this was causing the monitor to read low.
The transformer was then partially de-gassed. The results for the next three years are
shown in figure 2. As can be seen, the hydrogen and methane levels showed a steady
increase, which seemed to indicate that PD was ongoing.
Figure 1
Monitored transformer at Winchelsea Zone substation
Figure 2
On-line DGA results
There have been several attempts to measure PD with limited success up until the
installation of the continuous OMICRON monitoring system, which has confirmed the PD
activity and facilitated some further analysis.
Figure 3
Parameters continuously monitored
Project-specific system design
Accurate measurement of capacitance and dissipation/power factor
Traditionally, the sum-of-currents method is the most common method used in monitoring
of bushing capacitance. The leakage currents measured at the three bushing taps are added
to obtain the imbalance current. The changes of the system voltage phase angle due to the
daily load variation lead to significant changes of the bushing capacitance and dissipation
factor, introducing a measuring error 1. To increase the sensitivity of the measurements,
reference and comparison methods have become more and more popular 2. Both these
methods require a reference signal for calculation of dissipation factor and capacitance.
Depending on the source of the reference signal, relative or absolute measurements can be
performed (figure 4) 3. With relative measurements, the bushing-to-bushing comparison or
dual transformer comparison is performed, while with absolute measurement reference
signal is taken from voltage transformers (VTs) of the substation. Grid unbalance, like phase
shifts or changes in the system voltages can have a misleading impact on the relative
measurements of capacitance and dissipation factor. The simultaneous ageing of the
reference and test bushing cannot be detected. To reduce the impact of these uncertainties
(figure 5), absolute measurements are recommended at the Winchelsea substation. The
reference signal is taken from the LV side of a group of VTs located near the monitored
transformer. This pure resistive signal (UVT) is compared with the mainly capacitive leakage
current (IB) measured at the bushing tap of the transformer. The corresponding vectors of
these two signals are rotating in the same four-quadrant coordinate system (figure 6) and
&
the capacitive current I B always leads in revolution. The angles between the reference axis
and the two vectors are permanently measured and, by computing the difference between
them, the angle and the power factor cos are obtained. The dissipation angle () and
dissipation factor (tg) can be calculated accordingly. The temperature correction of the
dissipation factor values is performed 4. It takes into account the OIP (oil impregnated
paper) type of the bushing. The formula (1) is used to calculate the bushing capacitance CB.
The IB current is measured by the monitoring system, while the U High value is directly taken
from VTs.
Figure 4
Architecture of the monitoring system applied for
comparison and reference method
Figure 5
Sensitivity/uncertainty
introduced by each method
Figure 6
Vector representation of the UVT and IB in a four-quadrant coordinate system
IB
2 S f C B U High C B
IB
2 S f U High
(1)
The accuracy of the monitoring system is < 0.8 pF for capacitance measurement and 0.01
% for dissipation factor measurement. The 0.5 class of VTs also has to be taken into
consideration (Table 1).
Table 1
Voltage error and phase displacement for measuring voltage transformers 5
Figure 7
Example of PRPD patterns measured by conventional method (left) and UHF method (right)
for a known internal PD defect.
Figure 8
Waveform of the transients when a cable or
GIL or OHTL is connected at the transformer 7
Figure 9
Variation of peak magnitude of transients
as a function of the length of terminals 7
Figure 9.
Real-time data displayed in the graphical user interface
System installation
The architecture of the monitoring system is shown in figure 10. For each bushing, one
capacitive tap sensor to measure PD, capacitance and dissipation/power factor is installed
(figure 11). It is equipped with multiple redundant protections to guarantee safe operation.
The ambient temperature and relative humidity are also measured within the same tap
sensor to compensate the measured values. The UHF antenna is put inside the oil drain
valve located in the upper part of the tank (figure 12) and is connected to UHF down
converter and finally to the fourth channel of the acquisition unit. The outdoor monitoring
server and communication modem are mounted on the tank of the transformer (figure 13).
Only two VTs (phase B and phase C) were available for reference. The calculation of the
dissipation factor for phases B and C is based on the reference signal taken from the VTs
installed on these phases. To calculate the dissipation factor of the bushing installed on
phase A, the reference signal is taken from both VTs. The resulting reference voltage vector
follows the changes of the other two phases in a dynamic way. During the commissioning,
calibration of the monitoring system was performed and the level of alarms and measuring
intervals was selected in the monitoring software. The threshold values for dissipation
factor and capacitance were chosen based on the recommendations in the standards and
experience:
Dissipation factor
The diagnosis of the bushing is performed by analyzing the trend of tg, the magnitude and
its rate of change. When tg reaches a predefined level of alarm 0.7% (table 2) 9
continuous operation is no longer recommended. On the other hand, bushings with values
of tg above a predefined level but with stable trend may stay in operation. When tg
doubles the value over six months, off-line investigations and additional measurements are
recommended.
Figure 10.
Architecture of the installed monitoring system
Figure 11.
Tap sensor for C, tg, transients
and PD measurements
Figure 12.
UHF antenna installed in oil
drain valve
Figure 13.
Server, acquisition units
and UPS
Capacitance
The diagnosis is performed by comparing the capacitance values from on-line monitoring
with the values from off-line measurements. The difference between them gives the
capacitance variation C that has to remain within certain limits. The change of bushing
capacitance when one insulation layer is short-circuited for bushings of different rated
voltage is presented in table 3 10. As the detailed information about the bushing design
(number of insulation layers) is not available, the limits of C variation was estimated to be
a maximum 10%, taking into account the type of the bushing and its rated voltage. Such a
defect does not necessarily lead to the total failure of the bushing, but it creates a higher
electric field on the healthy layers.
Table 2
Acceptance level of dielectric losses for bushings of
9
different design (at 20C)
Standards
DF (tan) IEC60137
PF IEEE C57.19.01
RIP
<0.7%
OIP
<0.7%
RBP
<1.5%
<0.85%
< 0.5%
<2%
Table 3
Voltage class and change of capacitance for
10
condenser type bushings
Voltage in kV
123
245
420
No. of layers
14
30
40
Change in %
7.1
3.3
2.5
PD detection with the conventional method requires a calibration of the measuring system
according to IEC60270. To fulfill this requirement, PD pulses of a known charge are injected
in the current path of the bushing and measured at their taps. The threshold for alarm level
has to be related to the routine acceptance test level, 500 pC at 1.5Um according to
IEC60067-3 and IEEE C.57.12.90. Different types of the defect generate signals of different
apparent charge magnitude and PD pulse repetition rate. The decision about the criticality
of measured PD signals requires analyses of PD patterns. Wide bandwidth or narrow
bandwidth UHF measurements can be applied. In wide bandwidth measurements, the
highest magnitude of the UHF signal is measured in the frequency range from 100 to 2000
MHz. In this case the noise can influence the result of measurements. To overcome this,
narrow band measurements are performed in the selected frequency range. With the
transformer out of service, the spectra of the UHF signal are acquired and the external noise
is identified. The central frequency and the bandwidth of the measurements are selected in
the "clean" area of the spectra. The spectra are verified when the transformer is back in
service and the frequency range of the measurements can be adjusted if necessary.
The amplitude of the transient over-voltages is calculated and expressed as a function of the
phase-to-ground voltage and is displayed in p.u.
Periodic measurements are initiated in equidistant, one-hour time spans. The duration of
the periodic measurement is normally 1 min. During this time, all scalar values (i.e.
dissipation factor, capacitance, charge, repetition rate of the PD pulses, etc.) as well as
images (i.e PRPD - phase resolved partial discharge patterns, 3PARD 3 phase amplitude
relation diagram 11, magnitude and shape of transient over voltages, etc.) are stored in the
database.
Figure 14
Three phase capacitance and ambient
humidity variation (red trend)
Figure 16
Three phase dissipation factor variation
Figure 15
Comparison between the highest capacitance
variation (green) and the threshold limit (red)
Figure 17
Comparison of the highest measured value
of dissipation factor with the threshold limit
PD monitoring
PD monitoring with the conventional method
The three-phase PD trend is presented in figure 18. At each point visible in the trend
diagram, PRPD patterns and 3PARD are available.
Figure 19 depicts the PRPD patterns of the PD signal acquired by the three-channel
synchronous system. They are complex patterns with several PD sources overlapped. In
order to separate clusters of different PD sources, a synchronous multi-channel PD
evaluation technique is applied 11. The 3PARD diagram (figure 20) visualizes the relationship
between amplitudes of a single PD pulse in one phase and its crosstalk generated signals in
the other two phases. By repetition of this procedure for a large number of PD pulses, PD
sources within the test object as well as outer noise appear as a clearly distinguishable
concentration of dots in a 3PARD diagram 11. By examining individual clusters in the 3PARD
diagram, a separation between noise and PD phenomena is possible.
Figure 21 shows the back transformation to PRPD pattern of the clusters 1 and 2. The
patterns of the clusters 1 and 3 appear to be generated by bubbles and surface discharge
with the highest amplitude in phase B (cluster 1) and phase A (cluster 3). Similar PRPD
patterns were reported in 12. The shape and phase position of the patterns of the clusters 2
(phase A) and 6 (phase C) may indicate partial discharge activity inside the voids of the
insulation system 12. The clusters 4 and 5 are generated by external interferences
Figure 18
Three phase PD trend
Figure 19
Three phase synchronous PRPD patterns
PD monitoring with unconventional UHF method
Performing a frequency sweep, two spectra of the signal are obtained (figure 22). The upper
spectrum is built based on the maximum amplitudes of the time domain signal acquired at
each value of the frequency during the sweep. The lower spectrum corresponds to their
minimum amplitudes. Internal PD activity is always visible on the upper spectrum while
external interferences (corona discharge, radio waves, GSM) are visible on both spectra.
Phase C
Figure 20
Equivalent 3PARD diagram
Phase B
Phase A
Figure 21
Individual PRPD patterns of the selected clusters
Internal PD activity was identified in the frequency range from 450 to 650 MHz. The PRPD
pattern corresponding to a central frequency of 600 MHz is presented in figure 23. The
signal detected during the on-line monitoring was synchronized with a 50 Hz signal taken
from the measuring tap of the phase A. It can be seen in figure 23 that the phase of the
voltage where the PDs occur is the one characteristic to internal discharges. Furthermore, it
indicates a possible location of the PD activity, namely, in the vicinity of phase A. The PRPD
patterns of frequencies between 1 GHz and 1.4 GHz were checked and no internal PD
activity was found.
Figure 22
Frequency sweep diagram
Figure 23
PRPD pattern
Transients
Thirty transient over-voltages with the amplitudes between 1.25 and 1.77 p.u. were
recorded in four months after installation of the system (figure 24).
The transient with the highest amplitude (1.77 p.u.) is shown in figure 24a. Figure 24b and c
show other transient over-voltages recorded during the monitoring period. Even with such
amplitudes, these transients represent a threat to the insulation of the bushings and
windings because of the resonant phenomena which can lead to higher voltage distribution
between the turns. The number of transients cannot be controlled but their monitoring can
help identify defective equipment which generates over-voltages, near the transformer.
Figure 24
Time stamp and shape of the transients detected at the bushings
Discussion
As the dissolved gas ratios in the DGA system indicated that PD is taking place, the
OMICRON on-line monitoring system for PD, bushing capacitance and tg was installed on
the distribution transformer of POWERCOR. The PD activity was confirmed with both
conventional and unconventional UHF measurements. The synchronous multi-channel
evaluation technique separated clusters of different PD sources. Furthermore, it indicated a
possible location of the PD source in the vicinity of phase A.
The bushing capacitance and tg values remain within acceptable levels indicated in the
literature.
Concerning DGA analysis, high concentrations of H2 and CH4 indicate oil decomposition
under PD activity 13. The presence of high quantities of CH4 and C2H6 indicate an extreme
overheating of the mineral oil and adjacent metals. A low concentration of CO suggests that
PD activity may not take place in the vicinity of the paper insulated parts.
It is intended to carry out internal inspections following the installation of another
transformer at the site planned for 2014.
Acknowledgement
OMICRON sincerely appreciates the high level of support and co-operation of POWERCOR in
enabling this trial to proceed and with the installation of the monitoring system.
References
1. Z. Berler, V. Prykhodko, J. Watson, J.S. Skinner, D. Bates, Analyzing Data from On-line
Continuous Bushing Monitoring Systems Paper 9, proceedings of the International
Conference on Condition Monitoring, Diagnosis and Maintenance CMDM2013,
Bucharest, Romania, pp.73-78, 2013.
2. M. Koch, M. Krger and W. Koltunowicz,A New Method for On-line Monitoring of
Bushings, Transients and PD of Power Transformers, Proceedings of The 4th International
Advanced Research Workshop on transformers, Baiona, Spain, pp. 287-290, 2013
3. ***MONTRANO Continuous Monitoring System for Power Transformers Technical
Brochure, OMICRON Energy Solutions, Berlin GmbH, 2013.
4. Transformer Bushings, type GOB Installation and Maintenance Guide, ABB Technical
Brochure,http://www.abb.de/productguide/product.aspx?&c=c12573e700330462c1256f
5c004a7dc9&db=db0003db004283
5. ***IEC60044-2:2003 Instrument transformers Part2: Inductive voltage transformers
6. X. Dong, S. Rosado, Y. Liu, N.C. Wang, E.L. Line and T.Y. Guo, Study of abnormal electrical
phenomena effects on GSU Transformers, IEEE Transactions on Power Delivery, vol. 18,
No. 3, July 2003.
7. A. Said, E.A. Badran, M.A. Abd-Allah, Mitigation of very fast transient overvoltages at the
more sensitive points of gas insulated substation, International Journal on Electrical
Engineering and Informatics, Vol. 4, No. 3, October 2012.
8. U. Riechert, H. Ito, E. Zaima, K. Uehara and W. Chen, Insulation Co-ordination for Very
Fast Transients in Gas-Insulated UHV Substations, presented at UHV Colloquium New
Delhi 2013, Session 2.3 Substations.
9. CIGRE TB 445 Guide for Transformer Maintenance CIGRE Working Group A2.34,
February 2011.
10. T. Stirl, R. Skrzypek, S. Tenbohlen, R. Vilaithong, Online Condition Monitoring and
Diagnosis for power Transformers their Bushings, Tap Changer and Insulation System
Paper? Proceedings of the International Conference on Condition Monitoring and
Diagnosis CMD 2006, Changwon, South Korea , April 2006.
11. W. Koltunowicz and R. Plath, Synchronous Multi-Channel PD Measurements IEEE
Transactions on Dielectrics and Electrical Insulation, Vol.15, No. 6, p. 1715-1723, 2008.
12. A. Carlson, J. Fuhr, G. Schemel, F. Wegscheider, Testing of Power Transformers
Routine tests, Type tests and Special tests, 1 st Edition, published by Pro Print, Zrich,
Switzerland, 2003.
13. ***IEC60599:1999, Mineral oil-impregnated electrical equipment in service Guide for
interpretation of dissolved and free gases analysis.
Biography
Laurentiu Viorel Badicu is HV Application Engineer at OMICRON Energy Solutions GmbH,
Berlin, Germany since 2012. He is responsible for maintenance of the installed on-line
monitoring systems, customer trainings, performing site measurements (partial discharge,
dissipation/power factor, capacitance etc.) and data evaluation. His interest covers also
aspects related to the ageing mechanism and condition assessment of the insulating
materials.
He received the Dipl.-Ing. and the Ph.D degrees in electrical engineering from University
Politehnica of Bucharest, Romania in 2008 and 2012, respectively.
Wojciech Koltunowicz is with OMICRON Energy Solutions GmbH, Berlin, Germany, where he
is involved in monitoring of HV equipment. From 1987 to 2007 he was with CESI, Italy,
where he was mainly involved in HV testing and diagnostics of HV equipment. From 1984 to
1987, he was a research scientist in the High Voltage Department at the Institute of Power
in Poland.
He received M.Sc., PhD and D.Sc. degrees in electrical engineering from the Warsaw
University of Technology in 1980, 1985 and 2004, respectively.
He is secretary of CIGRE Advisory Group D1.03 Insulating Gases, WG D1.25 and D1.37. He
is member of CIGRE AG D1.02 High Voltage and High Current Test and Measuring
Technique and Diagnostic and WG D1.51. He is also member of IEC TC42 WG14. He is
author of dozens of international reports.
Andrea Piccolo is HV Application Engineer at OMICRON Energy Solutions GmbH, Berlin,
Germany since 2013. From 2009 to 2011 he was with TechImp SpA as R&D engineer. From
2011 to 2013 he was responsible for electrical diagnostics within Nidec ASI (former Ansaldo
Sistemi Industriali). His main interests are related to testing and monitoring of MV rotating
machines technology, manufacturing, testing and monitoring.
He received M.Sc. in Electrical Engineering from University of Trieste, Italy, in 2010.
He is member of CIGRE WG A1.42 "Survey on Hydro Generator Instrumentation &
Monitoring".
Alan McGuigan has had a long and diverse career in all aspects of the electrical supply
industry. He is past member of the Institute of Engineers, Australia.
His previous experience includes extensive experience as a Test and Commissioning
Engineer covering construction and maintenance of medium to large substations, thermal
and hydro power stations. This was followed by 10 years experience in roles of design,
system planning and construction in the distribution area.
Alan commenced with OMICRON Australia in January 2008. He is Manager of Australian and
New Zealand offices and also fulfills the role Application Engineer. In this role he provides
technical support, training and demonstrations for all Omicron primary test equipment. He
also organises special focus symposiums and workshops.
Colin Feely is an Asset Strategy Engineer with Powercor Australia Ltd, a privatised
Distribution Company operating in Victoria, Australia.
Colin has had over 30 years of engineering and management experience in the Electrical
Industry in Victoria and a 12 month consulting assignment in South Sumatra.
Colin holds an Electrical Engineering Diploma from the Gippsland Institute of Advanced
Education.
His preferred job description is to have fun playing with Plant.
ABSTRACT
PJB (PT Pembangkitan Jawa-Bali), the main power utility in the Java-Bali islands, Indonesia, embarked in
2011 on an ambitious project to equip all of its important generating plants with continuous on-line partial
discharge monitoring. At that time they were facing the fact that their equipment, installed mostly in the eighties
and nineties, was approaching the period of major maintenance needs and needed a reliable tool to prioritize
effort and improve reliability. Currently, PD monitoring systems are installed on 18 turbo-generators and on 8
hydro generators, and there are more to be installed in near future. This article describes the advanced hardware
and software features of these systems, including elimination of disturbances and separation/identification of
different types of insulation defects based on synchronous, multi-channel and multi-frequency techniques. Two
case studies are also included, the first being the case of Gresik PLTU (thermal power unit) and showing the
advantage of having a fully digital system, whose settings can be easily remotely adjusted. The second is that of
Cirata Unit 7, a typical showcase of how to proceed with the assessment of the stator winding conditions, when
clear PD activities are diagnosed by the monitoring system.
KEYWORDS: Generator, Condition Based Maintenance, Stator Insulation, Partial Discharge, Monitoring.
1.
INTRODUCTION
PJB is the main power utility in the Java-Bali islands. Thanks to its installed 6793 MW, distributed in six main
power stations (fig. 1), PJB is delivering yearly 32094 GWh to the Java-Bali grid, through 150 kV and 500 kV
HV lines.
The base load generation is covered by the thermal power plants of Gresik, Paiton, Muara Karang and Muara
Tawar (fig. 2), while the 19% of the installed power is based on the hydro-power plants of Cirata and Brantas,
which cover the peak-load hours energy demand (7% of the yearly generated power).
Beside its power generation business, PJB develops and manages other diverse businesses related to power
stations in order to maximize the potential of the company. On top of that, PT PJB created business areas in
Operation and Maintenance, Engineering Procurement and Contracting, and entered a joint-venture to develop
new power stations as well as to manage plant service Operation and Maintenance business [1].
Considering the fact that most of its installed generating power was installed in the early nineties, with some
units even commissioned in the mid-eighties, PJB is continuously looking for enhancing its capabilities in
Enterprise Asset Management.
With reference to the popular survey of 1199 hydro-generators, carried out by the CIGRE Study Committee A1,
WG A1.10 in 2009 [2], it is well recognized that the principal cause of generator failure (57%) is because of
insulation damage. Considering an average life expectancy of 25-30 years for a generator, most of PJB rotating
electrical machines are approaching or are already in the critical period for major maintenance (refurbishment or
rewinding).
Aware of evidence that the peak of maintenance activity will probably be between the years of 2017 and 2020,
PJB decided in 2012 to extend its capabilities in insulation condition-based assessment by installing PD
monitoring systems on its generator fleet [3].
Since 2012, twenty one monitoring systems are operating, thirteen on turbo-generators (46% of the installed
thermal power) and eight on the generators at Cirata hydro-power plant(77% of the available hydro power).Thanks
to the positive experience with the above mentioned machine, other five system has been recently installed and
commissioned in Gresik Block 2 and Muara Karang.
This following paper presents the achievements and challenges after more than two years since the first PD
monitoring system was installed and commissioned.
2.
2.1
MAIN CONTENTS
Challenges in Partial Discharge Monitoring
Partial discharge (PD) monitoring is a widely accepted indicator of insulation condition for rotating machines.
PD measurements are specified and strongly recommended for off-line testing and on-line monitoring [4,5,6,7].
Nevertheless, extensive application of PD monitoring for Enterprise Asset Management is still a challenging
task based on the monitoring solution selected. Suggested requirements include:
x
x
x
The data acquisition and pre-processing stage must be flexible for installation to any kind of machine
(hydro generators, turbo-generators, motors) and environmental conditions (measurement noise,
distance of the sensors from the equipment under monitoring, et. al.);
The monitoring system must be capable of trending and storing significant data for long periods;
Remote access to the system must be granted to ease the access to the database;
The monitoring software package must be able to provide tools both for basic measurement and
advanced data analysis.
The introduction of fully digital systems that include the synchronous multi-channel PD measurement
technique allows sensitive measurements and accurate separation and classification of PD sources in a wide range
of rotating machines [8]. With reference to the PJB experience, the example of integrating the above-mentioned
technique within a server-based monitoring system/software is presented.
2.2
PJB represents the case of a typical utility, with a mix of different generating units, which have been installed
and commissioned at different periods.
Each monitored unit is equipped with a set of three 1.1nF capacitive couplers (one per phase), installed inside
the Insulated Phase Bus and connected to the three phases (fig. 3). The capacity of the sensors, ten times higher
than the more popular 80pF, guarantee high sensitivity to PD signals even in a frequency range lower than 1 MHz.
The sensors are connected to a three-channel data acquisition and pre-processing unit, which is based on FPGA
technology (fig.4).
Each channel is equipped with a 20 MHz high-sensitivity analogue-to-digital stage (14bits resolution
and 130dB dynamic range the best in its class), which samples pulses at 65 MSsec. The digitized signals are
then sent to real-time FPGA based system that accomplish the following functions:
x
x
x
Digitally filters the pulses within a freely-selectable bandwidth, in order to reduce measurement noise;
Populates the Phase Resolved Partial Discharge (PRPD) patterns;
Generates the 3PARD map, in order to allow separation of noise and PD [3,8,9].
The actual advantage of having a fully-digital acquisition unit is that it can be easily customized with reference
to the type of machine and installation conditions (i.e. distance of the sensors from the equipment under monitoring,
measurement noise).
The dialogue with the data acquisition and pre-processing units (usually one per machine), which include the
timing periodic for acquisition and storage of the results, is managed by the server-based monitoring software.
The monitoring software includes the following main functionalities:
x
x
x
x
The advantage of a server-based system is to have a single remote access point for several machines that can be
easily reached via the worldwide web whenever requested (fig. 5).
PLTU 1 and PLTU2, commissioned in 1981, are equipped with two twin 100 MW steam turbines
coupled to 13.2 kV - air cooled generators;
PLTU3 and PLTU4, commissioned in 1988, are equipped with two twin 200 MW steam turbines
coupled to 15.0 kV - hydrogen cooled generators.
Figure 7. Block 1 in Gresik PP
As mentioned before, the most efficient way to achieve a reliable insulation assessment of the stator insulation
is by comparing the results from similar machines. The graph below shows the combined trend from PLTU 1 and
PLTU 2
After more than one year since the commissioning, an increase of PD activity has been observed between
September and October 2013 in PLTU 1 (blue, yellow and orange trends). The trend has been zoomed in order to
investigate in detail the source of the steep increase in PD amplitude (fig. 9).
Figure 9. Combined PD data trend in PLTU 1 and 2 related to the time period Sept Oct. 2013
The increase is more evident in Phase S (yellow) and has a fluctuating behavior on all the three phases, increasing
in average since the 18th of October.
From the analysis of the recorded patterns, the cause of the increase has been identified in the appearance of a
source of external noise or disturbances that periodically appears in the PRPD patterns.
In order to restore the capability of the system to monitor PD exclusively, extra measurements have been
performed by a team of PD monitoring specialists, with the target of evaluating an adequate system fine tuning.
The results from the tests, carried out via remote connection between Germany (where the monitoring technical
support is based) and Indonesia, showed that the monitoring band-pass filter central frequency can be moved from
2 MHz to 7 MHz, allowing the system to reject the noise while preserving adequate sensitivity to PD pulses.
In order to preserve the possibility to compare of results from PLTU1 and PLTU2 and validate the settings
upgrade, the monitoring band-pass filter has been adjusted in both the machines. The results, showed in the fig.
10, clearly confirm, both for PLTU 1 and PLTU 2, that:
x
x
x
x
Resetting to
7 MHz central frequency
PLTU 1
PLTU 2
Figure 10. PRPD diagrams in PLTU 1 and PLTU 2 for two different central frequency of measurements: 2 and 7 MHz
PLTA Ciratas power house consists of two main blocks Cirata I, commissioned in 1988, and Cirata II,
commissioned in 1997. Each block is composed of 4x126 MW-16.5 kV hydro-generators with the same design
and provided by the same manufacturer. The main contribution of PLTA Cirata is to deliver energy during the
peak load hours and recover the Jawa-Bali electric system in case of black-out (Black Start-up PP).
Installation of monitoring systems on all the generators was completed at the end of 2012.
Since the first months of trending Cirata II-Unit 7, showed a steep increase of PD activity (fig.12).
As all the four generators of Cirata II have the same design and were commissioned in the same period, they
can be considered comparable. The first action taken was to compare Unit 7 with its neighbour Unit 8.
From the trend diagrams of average charge for Unit 7 and Unit 8, it is possible to understand the deviation of
Phase S-Unit 7.
To confirm the activity in phase S, pattern acquisitions have been made periodically to identify the type of source
and to confirm the phase location. In fig. 14, the results for Unit 7 and Unit 8 are presented, where each pattern is
synchronized with its relevant phase voltage.
01/03/2013
12/04/2013
28/05/2014
Unit
8
Unit
7
The next step of the evaluation aims to isolate the investigated PD source from the internal PD activity. This task
is accomplished by means of the 3PARD separation technology.
01/03/2013
12/04/2013
28/05/2014
PD
Source
Other
PD
3PARD
map
The investigated PD source, identified by the selected cluster in the 3PARD, can be associated with surface
tracking located in phase S.
Considering the other sources separately, it is possible to have an indication of the machine from the point of
view of internal discharges, which can be considered uniform in the three phases and comparable with the values
recorded in Unit 8 (twin of Unit 7).
The stator inspection carried out on September 2013 confirmed the presence of surface PD activity in six slots,
belonging to the investigated phase S.
Slot 7
Slot 44
Slot 80
Slot 99
Slot 234
Slot 252
3. CONCLUSION
Continuous PD monitoring of stator windings is proving to be an essential diagnostic tool for conditionbased maintenance of large generators operated by PJB, in order to improve their reliability and useful
life.
The main key to performing appropriate diagnosis of the state of the insulation is to accurately separate
and identify different PD sources from external noise and disturbances, even when they are concurrent
and produce similar symptoms. To achieve this, state-of-the-art synchronous multi-channel evaluation
techniques are applied through an advanced automated system.
Worldwide accessibility and possibility to cross-evaluate the monitoring data of the generators of whole
power plants, stored in central databases, allows improved integration of generators PD monitoring
within the Enterprise Asset Management system.
The usefulness of the monitoring system has already been proven by the successful correct diagnosis of
a problem in the end winding area of a large hydro generator, and subsequently fully confirmed by a
visual inspection.
REFERENCES
[1] PJB Pengbakitan Java-Bali website http://www.ptpjb.com/
[2] CIGRE Technical Brochure 392 / Survey of Hydro Generator Failures- 2009 CIGR
[3] Constant Monitoring in Electrical Rotating Machines Case Studied CEPED 2013, Bali
[4] IEC 60034-27-2, edition 1.0, On-line partial discharge measurements on the stator winding insulation of
rotating electrical machines March 2012
[5] IEEE 1434-2000, IEEE Guide to the Measurement of Partial Discharges in Rotating Machinery
[6] R. Bruetsch, M. Tari, K. Froehlich, T. Weiers and R. Vogelsang - Insulation Failure Mechanisms of Power
Generators IEEE 2008
[7] C. Hudon, M. Belec and M. Levesque Study of Slot Partial Discharges in Air-Cooled Generators IEEE
Transactions of Dielectrics and Electrical Insulation Vol.15, No. 6; December 2008
[8] L.V. Badicu, W. Koltunowicz, M. Koch and A. Piccolo - Return of experience from continuous PD
monitoring of rotating Machines;ISH2013, Seoul
[9] W. Koltunowicz, R. Plath - Synchronous Multi-channel PD Measurements - IEEE Transactions on Dielectrics
and Electrical Insulation Vol. 15, No. 6; December 2008.
)'* +$('*('$"%" +)$ *'!%" +)$ %'*,%+$('(%(" "&")
7,!&0-,-313*2'!&,,#*#130#+#,21
+(#'%#&+),-*+.'#/ +*")$)",&
$"'$#5<53DB?>93C=2
$25B5C'954
<1ECECDB91
,!&0-,-31+3*2'!&,,#*+#130#+#,21.0-4'"#,#5.-2#,2'*',*-!2'-,
'"#,2'$'!2'-, ," 1#.02'-, -$ +3*2'.*# 1-30!#1 ," 2&#'0 "'1!0'+',2'-, $0-+
#62#0,* ,-'1# ', 40'#27 -$ 11#21 &# ',"#.#,"#,2 - 1#04#01 *')# !-3.*',%
!.!'2-01 31&',% 2.1 !.!'2'4# -0 ',"3!2'4# 1#,1-01 +"# 2&# +3*2'!&,,#*
#4*32'-, $#1' *# -, +-2-01%#,#02-012&0##.&1#20,1$-0+#01-01',%*#*',#
! *# 1712#+1 5'2& !0-11 -,"',% *',)1 &0## "'$$#0#,2 +3*2'!&,,#* #4*32'-,
2#!&,'/3#1 0# .0-.-1#"
.&1# +.*'23"# 0#*2'-, "'%0+
.&1#2'+#
0#*2'-, "'%0+
,"
!&,,#* $0#/3#,!7 0#*2'-, "'%0+
&#
13!!#11$3*..*'!2'-,-$**2#!&,'/3#1'1.0#1#,2#"
P 02'* "'1!&0%#1 !0-112*) "'%,-12'!1 .22#0, 0#!-%,'2'-,
',2#0$#0#,!#13..0#11'-,
@1BD91<49C381B75%9C13?=@<5H@8IC931<@85>?=5>1
259>713?>C5AE5>35?6<?31<5<53DB931<CDB5CC3?>35>DB1D9?>9>
D85 9>CE<1D9?> )85 % F1B95DI 1>4 9DC E>CD12<5 CD?381CD93
2581F9?B=1;5%=51CEB5=5>D?>5?6D85=?CD496693E<DD1C;
9>8978F?<D175D5CD9>7$>D85?D85B81>4%=51CEB5=5>D9C
1 G?B<4G945 1335@D54 =5D8?4 6?B 9>CE<1D9?> 4917>?C9C 1>4 1
B5AE9B54@1BD?61335@D1>35D5CD9>7?6=?CD+1CC5DC
%C1B55HDB5=5<I61CD3EBB5>D@E<C5CG8?C5B9C5D9=51>4
G94D845@5>4?>D8549C381B75DI@51C13?>C5AE5>35?6<?31<
3?>49D9?>C 9> D85 9>CE<1D9?> 45653D DI@5 1>4 75?=5DBI
?<<?G9>7 D85 % C97>1< @B?@171D9?> 6B?= D85 @<135 ?6 9DC
?B979>D?D8545D53D9?>C9D5/
0D856?<<?G9>71C@53DC81F5D?
25D1;5>9>D?133?E>D
)85<?CC?69>6?B=1D9?>12?EDD8549C381B75@B?35CC9DC5<6
G9D89>3B51C9>749CD1>356B?=%?B979>
)85 9>3B51C9>7 9=@13D ?6 @E<C5 @B?@171D9?> @1D8 ?> %
G1F56?B=
)85 CE@5B@?C9D9?> 25DG55> 49665B5>D C97>1<C 57 =E<D9@<5
%C?EB353B?CCD1<;5HD5B>1<9>D5B65B5>35C
)85B54E3549>6?B=1D9?>?6D85=51CEB54%@E<C54E5D?
D85 <9=9D54 21>4G94D8 ?6 D85 3?E@<9>7 E>9D 1>4 D85 21>4
@1CC69<D5BB5CDB93D9>7D859>6?B=1D9?>D?%@E<C51@@1B5>D
381B75 F1<E5 1>4 D? D9=5 ?6 ?33EBB5>35 G9D8 B565B5>35 D?
13F?<D175@81C5?><I
)85 9>DB?4E3D9?> ?6 1 4979D1< D538>9AE5 3B51D54 >5G
?@@?BDE>9D95C D? 9=@B?F5 C5>C9D9F9DI C97>96931>35 1>4
B5@B?4E3929<9DI ?6 % =51CEB5=5>DC 2I 61B 5H35549>7 D85
31@129<9D95C?6?<45B1>1<?7%=51CEB9>7CICD5=C(?31<<54
% 69>75B@B9>DC ?B @1DD5B>C =1;5 EC5 ?6 % @81C5 1>4
")'&$(&*#"$(!& "" #&! (#&
!
!0
!
)851>1<?7%=51CEB5=5>DCCD1BD54G9D8D859>DB?4E3D9?>
?6 D85 ' 45D53D?B 9>
D85 69BCD AE1C92B?1421>4 %
9>CDBE=5>D D @1F54 D85 G1I 6?B D85 CE335CC ?6 1@@1B5>D
381B75 =51CEB5=5>DC 9> @ 9>CD514 ?6 + 1C G9D8 '+
=51CEB5=5>DC&E1C92B?1421>4%9>CDBE=5>DC81F5169H54
35>D5B6B5AE5>3I1>4D859B<?G5B1>4E@@5B3ED?666B5AE5>3I
14:ECD549>CD5@CB5CE<D9>79>121>4G94D86B?=
;JE@D?
;J ?=@1B54 D? >1BB?G21>4 9>CDBE=5>DC % @E<C5
B5C?<ED9?>G1CC97>96931>D<I9=@B?F54
)85C51>1<?7%9>CDBE=5>DCECE1<<I3?>C9CD?6121>4@1CC
69<D5B 1>4 1 @51; <5F5< 9>4931D?B C <?>7 1C D85 21>4@1CC
69<D5B 5HDB13DC % @E<C5C G85B5 C@53DB1< 45>C9DI 9C 3?>CD1>D
D85@51;?6D8569<D5BB5C@?>C59C@B?@?BD9?>1<D?D851@@1B5>D
381B75 ?6 D85 % 3EBB5>D @E<C5 )89C 6B5AE5>3I 4?=19>
@B?35CC 9C 31<<54 AE1C99>D57B1D9?> /0 )85 ?ED@ED ?6 D85
21>4@1CC69<D5BG1C3?BB5<1D54D?D8513F?<D175@81C5)85
69BCD31D1<?76?B%C?EB35B53?7>9D9?>@E2<9C8542I'
9>
4B5G851F9<I?>CE38697EB5C/
0%CE@5B9=@?C54
?> 1 !9CC1:?EC 697EB5 9C CD9<< 1 3?==?> G1I D? 49C@<1I %
2581F9?BG9D8B5C@53DD?13@81C5
!
)85 5H9CD9>7 1>1<?7 9>CDBE=5>DC =145 CD?381CD93
@B?@5BD95C ?6 % F9C92<5 2ED CD1D9CD93C 3?E<4>SD 25
1@@B?13854 )? 6E<69<< D85 45=1>4 ?6 B9C9>7 C395>D9693
9>D5B5CD =E<D9381>>5< @E<C5 85978D 1>1<IJ5BC G5B5
45F5<?@54 /0 % @E<C5C 3?E<4 81F5 255> C?BD54 1>4
3?E>D54133?B49>7D?D859B85978D@E<C585978D89CD?7B1=C
,85> =9>93?=@ED5BC 1>4 %C 253?=5 1F19<12<5 6<5H92<5
C?6DG1B521C54 CICD5=C B5@<1354 D85 =E<D9 381>>5< @E<C5
85978D 1>1<IJ5BC 6?B B5C51B38 >
D85 69BCD
3?=@ED5B9J54 % =51CEB9>7 CICD5= G1C 9>DB?4E354 D? D85
@E2<93/0CEBF5I?>51B<I45C97>C9CB5@?BD549>/01>4
1>9>45@D8DB51D=5>D?6%4979D1<41D1@B?35CC9>79C79F5>
9> /O
0 )8B55 @B9>39@<5 45C97>C ?6 4979D1<%=51CEB9>7
CICD5=C31>2549CD9>7E9C854
%81C5B5C?<F54 DB149D9?>1< 45C97> G9D8 1>1<?7 21>4
@1CC69<D5B3?=29>54G9D8@51;45D53D?B?B@E<C5C81@9>7
1=@<9695BC<?G3?>F5BD5B1>43?=@ED5B
)9=5B5C?<F54G9D8G94521>4@B51=@<9695B4979D1<CD?B175
?C39<<?C3?@5?B61CD3?>F5BD5B1>43?=@ED5B
)9=5B5C?<F54 G9D8 G94521>4 @B51=@<9695B 61CD
3?>F5BD5BB51<D9=5@B5@B?35CC9>71>43?=@ED5B
)85 =19> 14F1>D175 ?6 1 4979D1< @81C5B5C?<F54 =51CEB9>7
CICD5=9CD8531@129<9DID?B53?B41@@1B5>D381B751=@<9DE45
% 1>4 13 F?<D175 @81C5 1>7<5 ?6 5138 C9>7<5 %
@E<C5 )89C 41D1 9C EC54 D? 1@@<I CD1D9CD93 5F1<E1D9?>
=5D8?4C 57 D? 75>5B1D5 ?B 89CD?7B1=C G9D8
B5C@53D D? % 1=@<9DE45 % B5@5D9D9?> B1D5 1>4 13 F?<D175
@81C597EB5
'%30# 1>4%'%@1DD5B>
0
>CD514 ?6 21>4<9=9D9>7 69<D5BC CE38 CICD5=C EC5 G94521>4
@B51=@<9695BC 1>4 3?==5B391<<I 1F19<12<5 4979D9J5BC G9D8
(CC1=@<9>7B1D51>41>1@@B?@B91D51>1<?721>4G94D8
J ?B =?B5 )85 C1=@<54 41D1 6B?= D85 4979D9J5B 9C
DB1>C65BB54 D? D85 3?=@ED5B 6?B @?CD@B?35CC9>7 )85 =1:?B
2?DD<5>53;C?6D89CCICD5=1B5D85<9=9D5441D1DB1>C65BC@554
1>4D85<9=9D54B5DB9775BB1D5?6D854979D9J5B
335<5B1D54 @B?7B5CC 9> C?6DG1B5 1>4 81B4G1B5 ?6 4979D1<
D538>9AE5C =145 9D @?CC92<5 D? 3?=29>5 3?>D9>E?EC
B53?B49>7?6%41D1G9D8G94521>4%=51CEB5=5>DC
,94521>4 1>4 E<DB1G94521>4 % =51CEB5=5>DC B5AE9B5
8978 C1=@<9>7 B1D5C 6?B 4979D9J1D9?> > 3?>C5AE5>35 D85
D9=56B1=5 6?B 3?>D9>E?EC 41D1 13AE9C9D9?> 9C <9=9D54 2I D85
C9J5 ?6 D85 4979D9J5B =5=?BI )85B56?B5 4979D1< G94521>4
% CICD5=C 4979D9J5 @E<C5C ?><I G85> D85 9>@ED C97>1<
5H3554C D85 DB9775B D8B5C8?<4 %B?35CC9>7 @5B6?B=1>35 ?6
3?EBC5 9C <9=9D54 2I D85 B5DB9775B B1D5 1>4 D85 CD?B175
31@129<9DI
531EC5 D9=5 9> 25DG55> 3?>C53ED9F5 % @E<C5C 3?>D19>C
>?B5<5F1>D9>6?B=1D9?>DB9775B9>7?>D85@E<C59DC5<6C55=C
D?25D851@@B?@B91D5C?<ED9?>D?@B5F5>D6B?=4979D9J9>7D89C
@5B9?42EDDB9775B9>7G9<<G?B;G5<<?><I6?B8978C97>1<D?
>?9C5 B1D9? (#' > B51< CICD5=C 8?G5F5B 9>D5B65B5>35
31> 31EC5 1 49<5==1 6 D85 D8B5C8?<4 9C C5D D? 8978 F1<E5C
D85 4979D9J9>7 CICD5= G9<< ?><I DB9775B ?> <1B75 @E<C5C >
3?>DB1CD 1D <?G D8B5C8?<4 9>D5B65B5>35 31> DB9775B D85
CICD5=5@5>49>7?>D85B5<1D9?>C89@?6%1>4@E<C5DI@5
>?9C5 B5@5D9D9?> B1D5 <?G D8B5C8?<4 31> <514 D? 4979D9J5
=19><I>?9C59>CD514?6%@E<C5C
C <?>7 1C % @E<C5C 1B5 =51CEB54 3<?C5 D? D859B ?B979>
1>4 3?>C5AE5>D<I G9D8?ED C97>96931>D 49CD?BD9?> E<DB1
G94521>4 G1F56?B= % 1>1<IC5C 1B5 B5<1D54 D? D85 @8IC93C
?6 D85 49C381B75 @B?35CC /
O
0 9665B5>D 49C381B75C 57
3?B?>1F?9449C381B75CEB613549C381B75C8?G49665B5>35C
9> % @E<C5 G1F56?B= ?><I F9C92<5 G9D8 G94521>4
=51CEB9>7CICD5=CD<?G=51CEB9>76B5AE5>395C8?G5F5B
@E<C5C 6B?= 49665B5>D 49C381B75 @B?35CC5C 253?=5
9>49CD9>7E9C812<5 1C D859B C@53DB1< 45>C9D95C 1B5 CD9<<
3?>CD1>D"51CEB5=5>DC?61@@1B5>D381B75F1<E5133?B49>7
D? 1B5 21C54 ?> 3?>CD1>D C@53DB1< 45>C9DI D?
@5B6?B=3?BB53DAE1C99>D57B1D9?>?6D85%@E<C53EBB5>D
>31C5?6+5AE9@=5>DG9D83?=@<5HDB1>C65B6E>3D9?>
<9;5 57 DB1>C6?B=5BC ?B B?D1D9>7 =1389>5C D85 %
G1F56?B=9CC97>96931>D<I16653D542I@E<C5@B?@171D9?>1>4
3?>C5AE5>D<I D85 G1F56?B= <??C5 1<< 9>6?B=1D9?> 12?ED D85
DI@5?6D85%C?EB35
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
)'* +$('*('$"%" +)$ *'!%" +)$ %'*,%+$('(%(" "&")
!
!
)8514F1>3544979D1<%=51CEB9>7CICD5=%CICD5=G1C
45C97>54D??F5B3?=5D85<9=9D1D9?>C=5>D9?>549>C53D9?>
)89C CICD5= 9C 31@12<5 D? @5B6?B= 3?>D9>E?EC 1>4
CI>38B?>?EC =E<D9381>>5< % =51CEB5=5>DC 9> B51<D9=5
/
0
!
)85 C9=@<96954 45C97> ?6 D85 %CICD5= 9C @B5C5>D54 9>
97EB5
'%30# 4F1>354B51<D9=54979D1<%=51CEB9>7CICD5=
)85 CICD5= 9C =?4E<1B 1>4 3?>C9CDC ?6 ?>5 ?B =?B5
13AE9C9D9?> E>9DC 1 6925B ?@D93 2EC3?>DB?<<5B 1>4 1 %
9>3<E49>7=51CEB9>7C?6DG1B5)85%C97>1<C1B569<D5B54
1=@<96954 1>4 4979D9J54 9> B51<D9=5 @9@5<9>5
1B389D53DEB5 ,9D8 1> 1=@<9DE45 AE1>D9J1D9?> ?6
29D
1>4 1 C1=@<9>7 B1D5 ?6
"(C 1 CE2C1=@<5 D9=5
133EB13I ?6 >C 9C 13895F54 6?B CI>38B?>?EC =E<D9
381>>5< =51CEB5=5>DC )85 AE1C99>D57B1D9?> 9C
@5B6?B=54 2I 1 4979D1< 21>4@1CC 69<D5B )85 35>D5B
6B5AE5>3I 1>4 D85 21>4G94D8 31> 25 6B55<I 38?C5> 9> D85
B1>75 6B?= 43 E@ D? "J C5<53D12<5 69<D5B 21>4G94D8
6B?= ;J E@ D? "J D? 1F?94 49CDEB21>35C 1>4 D?
B5138 1> ?@D9=1< (#' 5F5> E>45B >?9CI ?>C9D5
3?>49D9?>C?B13@81C5CI>38B?>9J1D9?>1CG5<<1C6?B13
G1F56?B= 49C@<1I 1>4 1=@<9DE45 B5149>7 1 C53?>4
3?>F5BD5B4979D9J5CD85F?<D1753EBF5G9D8
29D;(C
<<41D113AE9C9D9?>1>4@B5@B?35CC9>71B5@5B6?B=549>D85
13AE9C9D9?> E>9D D? 7E1B1>D55 ?@D9=1< @5B6?B=1>35 9> C@554
1>4C97>1<AE1<9DI6?EB@1B1=5D5BF53D?B6?B5138C9>7<5%
@E<C5 9C 945>D96954 /"
%
(0 G85B5 " 9C D85 13AE9C9D9?>
E>9D >E=25B % D85 1@@1B5>D 381B75 F1<E5 D85 13 F?<D175
@81C5 1>7<5 1>4 ( D85 12C?<ED5 D9=5 > @1B1<<5< D85
9>CD1>D1>5?ECF1<E5?6D8513F?<D1759CC1=@<545F5BI
C
<<@1B1=5D5BC31>25CD?B546?B?66<9>5@B?35CC9>7G9D89>D85
B5@<1I =?45 ?B 1<D5B>1D9F5<I 5H@?BD54 6?B @?CD@B?35CC9>7
G9D8"1D!12>?C39<<?C3?@5=?45 "JG9D8C1=@<9>7
B1D5?6
"(C1>41)21C54C@53DBE=F95G5BD?1>1<IJ5
9>@ED C97>1<C G9D8 6E<< 21>4G94D8 1>4 D? <?31D5 % 2I D9=5
4?=19>B56<53D?=5DBI1B51<C?9>3<E45413813AE9C9D9?>E>9D
3?>D19>C 1> 9>D5B>1< D5CD 75>5B1D?BD?5>12<53<?C54<??@C5<6
D5CD9>7
56?B5 =E<D9381>>5< % =51CEB9>7 5AE9@=5>D 2531=5
1F19<12<5 D8B55@81C5 1@@1B1DEC G5B5 D5CD54 2I EC9>7 1
=E<D9@<5H5BD?3?>>53DD853?E@<9>745F935C1D5138@81C5
?>5 16D5B 1>?D85B D? D85 % =51CEB9>7 9>CDBE=5>D
569>9D5<I D85 EC5 ?6 1 =E<D9@<5H5B 9>8929DC C9=E<D1>5?EC
% =51CEB5=5>DC ?=@1B9C?> ?6 % =51CEB5=5>D
B5CE<DC D1;5> G9D8 1 =E<D9@<5H5B 9C D85B56?B5 >535CC1B9<I
21C54 ?> 49665B5>D D9=5 C57=5>DC 5C945C D85
49C14F1>D175D?C@<9DD851F19<12<5D5CD9>7D9=59>25DG55>
D85 D8B55 @81C5C B5CE<D9>7 9> <5CC % 41D1 6?B CD1D9CD931<
1>1<IC9C =E<D9@<5H5B21C54 % =51CEB5=5>DC CE665B
5C@5391<<I6B?=3B?CCD1<;
(5F5B1< 3?E@<9>7 =5381>9C=C 71<F1>93 31@139D9F5
9>4E3D9F5 5<53DB?=17>5D93 695<4 31EC5 3B?CCD1<; 9>
25DG55> @81C5C 9> D8B55@81C5 @?G5B 1@@1B1DEC <9;5
75>5B1D?BC ?B DB1>C6?B=5BC % 3B?CCD1<; CE@5B9=@?C5C
@81C5B5C?<F54 % =51CEB5=5>DC C? =1;5C % @1DD5B>
9>D5B@B5D1D9?>C?=5D9=5CF5BI496693E<D?>C5AE5>D<I%
3B?CCD1<;G1C=19><IB571B4541CC?EB35?69>D5B65B5>35O
G85>=51CEB54G9D81=E<D9@<5H5B
> 3?>DB1CD D85 B5CE<DC ?2D19>54 G9D8?ED =E<D9@<5H5B
1@@<I9>7 D85 CI>38B?>?EC =E<D9381>>5< CICD5= C8?G 3<51B
5F945>35 D81D 3B?CCD1<; 56653DC 3?>D19> 9=@?BD1>D 9>6?B=1D9?>
?> % @E<C5 ?B =?B5 3?==?> ?> C97>1< @B?@171D9?>
?@5>9>7>5GCDB1978D6?BG1B4=5D8?4CD?B5=?F549CDEB21>35C
1>4 =?B5?F5B D? 3<51B<I 49CD9>7E9C8 25DG55> =E<D9@<5 %
C?EB35C E5 D? 49665B5>D C9D5C ?6 ?B979> %
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
'%30#
853;?6D85%=51CEB9>7CICD5=@5B6?B=1>35O@B?@171D9?>?631<92B1D9?>@E<C51<?>7 ;=?6
;+-!%312<5<9>5
31@139D?BC3?>>53D54D?D85D5B=9>1<C?675>5B1D?BC>31C5?6
% =51CEB5=5>DC ?> <?>7 D8B55@81C5 C9>7<5<9>5 312<5
CICD5=C 3B?CC2?>49>7 <9>;C 5>12<5 49CDB92ED54 9>4E3D9F5
D8B55@81C5%45D53D9?>6?B?><9>5=51CEB5=5>DC/
0
'%30# "E<D9381>>5<%41D1@B?35CC9>7G9D8%13AE9C9D9?>E>9DC
%' F9CE1<9J5C D85 B5<1D9?>C 1=?>7 1=@<9DE45C ?6 1
C9>7<5 % @E<C5 9> ?>5 @81C5 1>4 9DC 3B?CCD1<; 75>5B1D54
C97>1<C9>D85?D85BDG?@81C5C%'3?>CDBE3D9?>2ID8B55
CD5@C9CC8?G>9>97EB5
)85 69BCD CD5@ 97EB5 9C D? 31<3E<1D5 D85 <?71B9D8= ?6
D85 12C?<ED5 F1<E5 #
%
?6 1<< D8B55 @E<C5C ?6 1 %
DB9@<5)85>5138@E<C59CDB1>C6?B=549>D?1F53D?BB5<1D54D?
9DC@81C5?6?B979>97EB5
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
)'* +$('*('$"%" +)$ *'!%" +)$ %'*,%+$('(%(" "&")
9>1<<I 1449D9?> ?6 D85 D8B55 F53D?BC B5CE<DC 9> 1 C9>7<5 4?D
@?C9D9?>B5@B5C5>D9>7?>5%DB9@<54?D9>97EB5
'%30# ?>CDBE3D9?>?6%'
)85%'81C35BD19>@B?@5BD95C
)85 4?D @?C9D9?> 45@5>4C ?><I ?> D85 <9>51B 1>4
45D5B=9>9CD93 DB1>C65B 6E>3D9?> 25DG55> @E<C5 C?EB35 1>4
45D53D9?>C9D5
)854?D@?C9D9?>9C9>45@5>45>D6B?=@E<C5=17>9DE451>4
13F?<D175@81C5OEC56E<1<C?6?B43%D5CDC
%E<C5C?6?>5@81C5G9D8CI==5DB933B?CCD1<;D?2?D8?D85B
@81C5C G9<< 25 3?>35>DB1D54 ?> ?>5 ?6 D85 %' 1H9C
DI@931< 6?B 31<92B1D9?> ?B 5H1=@<5 96 ?>5 @E<C5 %
45D53D541D@81C5!9C89785BD81>9DCCI==5DB933B?CCD1<;
B53?B4541D@81C5C!
1>4!97EB5%%%
%D85>D854?D9C49C@<1I545H13D<I?>D851H9C!)85
4?D49CD1>356B?=%'?B979>9>3B51C5CG9D89>3B51C9>7
B1D9?D?9DC3B?CCD1<;@E<C5C
6 D85 3?E@<9>7 D? D85@81C5C!1>4!9C>?DCI==5DB93
57 % % % D85> D85 @?C9D9?> ?6 D85 4?D 9C
C896D54D?G1B4D851H9C?6D85C53?>489785CD@E<C5
%E<C5 DB9@<5C G9D8 5AE1< =17>9DE45 3?==?> =?45
9>D5B65B5>35 G9<< 25 3?>35>DB1D54 3<?C5 D? D85 %'
?B979>%%%
)85 DB1>C65B 6E>3D9?> 25DG55> @E<C5 C?EB35 1>4 45D53D9?>
@<1359C75>5B1<<I9>F1B91>D8?G5F5BCE@5B9=@?C54>?9C5
1>4 F1B91>35 ?6 D85 49C381B75 G1F56?B= G9<< 3B51D5
3<ECD5BC9>CD514?6C9>7<54?DC/
0
I B5@5D9D9?> ?6 D89C @B?354EB5 6?B 1 <1B75 >E=25B ?6 %
@E<C5C%C?EB35CG9D89>D85D5CD?2:53D1CG5<<1C?ED5B>?9C5
1@@51B 1C 3<51B<I 49CD9>7E9C812<5 3?>35>DB1D9?> ?6 4?DC
3<ECD5B9>1%'
0
)851BB9F1<D9=5?61%DB9@<545@5>4C?>D85@B?@171D9?>
@1D86B?=%?B979>D?D85D5B=9>1<C?6D8B55@81C51@@1B1DEC
/
0
6 D85 % C?EB35 9C 61B 6B?= D85 45D53D9?> C9D5 @E<C5
1=@<9DE45C ?6 1 % DB9@<5 D5>4 D? 5AE1< 25DG55> D85 @81C5C
1>4G9<<C8?G3?==?>=?45@B?@171D9?>?>C5AE5>D<I61B
49CD1>D%C?EB35CG9<<2549C@<1I543<?C5D?D85?B979>?6D85
%')89C<9=9DCD8531@129<9DID?49CD9>7E9C825DG55>CE38
C?EB35C EBD85B=?B5 49CD1>D @B?@171D9?> G9<< 41=@ 8978
6B5AE5>3I3?=@?>5>DC?6D85%C97>1<C<?G@1CC56653D1>4
C? B5AE9B5C 1 <?G =51CEB5=5>D 6B5AE5>3I D? =19>D19>
'%30#
+9CE1<9J1D9?>C57=5>DC?6%)'6?BC9H49665B5>D1BB9F1<?B45BC
?6D85%DB9@<5@E<C5C
)85 F53D?B >?B=
9>3B51C5C G9D8 9>3B51C9>7 45<1I 25DG55>
D85 69BCD 1>4 D85 <1CD @E<C5 ?6 1 % DB9@<5 , B5@B5C5>DC D85
?66C5D1>7<5?6D85@81C5G85B5D8569BCD@E<C5?6D85%DB9@<5
9C45D53D54
1>4(1B5
=51CEB54 D9=5 ?6 D85 69BCD C53?>4 1>4 D89B4 =51CEB54 @E<C5C
G9D89>1%DB9@<5
97EB5 C8?GC D85 F9CE1<9J1D9?> ?6 49665B5>D D9=5
49665B5>35C 25DG55> D85 % @E<C5C 9> 31C5 D81D 69BCD @E<C5 9C
45D53D54 9> @81C5 !
D85 C53?>4 9> ! 1>4 D85 <1CD ?>5 9>
@81C5! ?B 9>CD1>35 96 D85 D9=5 49665B5>35 25DG55> 69BCD
DG? @E<C5C 9C F5BI C=1<< D85> D85 B5CE<D9>7 4?D G?E<4 25
49C@<1I54 25DG55> D85 1H9C G85B5 D85C5 @E<C5C 1B5 45D53D54
97EB5 $> D85 ?D85B 81>4 96 D85 <1CD DG? @E<C5C ?33EB
1<=?CD C9=E<D1>5?EC<I D85 B5CE<D9>7 4?D G?E<4 25 49C@<1I54
?>D851H9CG85B5D8569BCD@E<C59C45D53D5497EB5>
D85 31C5 D81D 1<< D8B55 @E<C5C 1B5 45D53D54 1D 1<=?CD D85 C1=5
D9=5D85F53D?B<5>7D8G?E<425J5B?
((
G893849C@<1ICD854?D1DD85?B979>?6D854917B1=97EB5
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
'%30# )5CD39B3E9DG9D8DG?%C?EB35C1>4
!
0
)85 381>>5< 6B5AE5>3I B5<1D9?> 4917B1= 3?BB5<1D5C
CI>38B?>?EC=E<D9C@53DB1<%=51CEB5=5>DD?C5@1B1D5%
6B?= 9>D5B65B5>35C > 3?>DB1CD D? %' 1>4 %)'
=5D8?4C ' 5F1<E1D9?> 4?5C >?D >535CC1B9<I B5AE9B5
D8B559>45@5>45>D13AE9C9D9?>E>9DC/
0
"E<D9C@53DB1< % =51CEB5=5>DC 1B5 21C54 ?> @E<C5
C@53DB1 3?BB5<1D9?> 2I 1@@<I9>7 D8B55 49665B5>D 21>4@1CC
69<D5BC 49665B5>D 35>D5B 6B5AE5>395C 1>4 21>4G94D8
C9=E<D1>5?EC<I )85 D8B55 21>4@1CC 69<D5BC 31> 25 DE>54 D?
1B51C G9D8 <?G 9>D5B65B5>35 <5F5< > 75>5B1< D85 21>4@1CC
69<D5B >?
C8?E<4 25 DE>54 D? 1 <?G 35>D5B 6B5AE5>3I D?
5>12<53?>6?B=9DID? 1>4@1CC>?1>41B5
DE>54D?89785B35>D5B6B5AE5>395CG85B5@E<C5@B?@171D9?>
56653DC31EC541<B514I49CD9>7E9C812<549665B5>35C9>C@53DB1<
B5C@?>C5C ?6 % C97>1< I @B?@5B 38?935 ?6 D85 21>4 @1CC
35>D5B 6B5AE5>395C 9D 253?=5C @?CC92<5 D? @5B6?B= %
=51CEB5=5>DC 133?B49>7 D? D85 CD1>41B4C 1>4 9> D85 C1=5
=?=5>DD?B5=?F5@B13D931<<I1<<CE@5B9=@?C549>D5B65B5>35
' 3?BB5<1D5C D85 ?ED@ED ?6 D85 D8B55 21>4@1CC 69<D5BC
5H13D<I<9;5%'4?5CG9D8D85@E<C51=@<9DE45C?61%
DB9@<5
)? 45=?>CDB1D5 D85 56653D9F5>5CC ?6 ' =5D8?4 1
C9=@<5 + 39B3E9D G1C 38?C5> )89C 39B3E9D 3?>C9CD54 ?6 1
3?E@<9>7 31@139D?B 1 3?E@<9>7 45F935 DG? >554<5
D?@<1>5 1BB1>75=5>DC 1>4 1>4 1> 9>4E3D1>35
97EB5 )85 13 D5CD F?<D175 G1C B19C54 E>D9< 3?B?>1
49C381B75CG5B575>5B1D541D1>497EB5
'%30# (E@5B9=@?C54%@1DD5B>C1D1>4
'%30# '?6>?9C51>4%C6B?=1>4
97EB5C
1>4
45=?>CDB1D5 D85 31@129<9DI ?6 D85 =E<D9
C@53DB1< % 5F1<E1D9?> )85 CE@5B9=@?C54 % @1DD5B>C 1B5
3<51B<IC5@1B1D54
$><9>5 % =51CEB5=5>D 1D 35>D5B 6B5AE5>3I ?6
"J
D9C?2F9?ECD81D%@1DD5B>C?6CE388978C9=9<1B9DI31> 1>46B5AE5>3I21>4G94D8?6
"JG1C@5B6?B=54?>;+
81B4<I 25 C5@1B1D54 I 9>DB?4E39>7 D85 9>4E3D1>35 ! D85 DEB2?75>5B1D?B 97EB5
C8?GC D85 % @1DD5B>C %'% 9>
% @E<C5 G1F56?B= 1D G1C 381>754 )85 % 45D53D9?> D8B55 @81C5C !
! 1>4 ! 97EB5
C8?GC %' 1>4
CICD5= 3?>>53D54 D? D85 3?E@<9>7 45F935 G1C DE>54 D? %)' 4917B1=C G85B5 65G 3<51B<I 49CD9>7E9C812<5 3<ECD5BC
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
)'* +$('*('$"%" +)$ *'!%" +)$ %'*,%+$('(%(" "&")
'%30#
%@1DD5B>1D?><I
'%30#
%@1DD5B>1D?><I
'%30#
%'E@@5B?>51>4%)'<?G5B?>5
'%30#
'%30#
%'%@1DD5B>C9>D8B55@81C5C?61DEB2?75>5B1D?B
!
)85 16D5B9>CD1<<1D9?> D5CD ?6 D85 ;=
;+ -!%
312<5CICD5=C55C53D9?>71F5D85?@@?BDE>9DID?=51CEB5
% ?> 312<5 1335CC?B95C C9=E<D1>5?EC<I )? 3853; D85
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
@5B6?B=1>35 ?6 D85 9>CD1<<54 % =51CEB9>7 CICD5= 5HD5B>1<
% G1C 75>5B1D54 1D 1 G9B5 1DD13854 D? D85 + 5<53DB?45 ?6
D85D5CDD5B=9>1<)85%'%@1DD5B>C45D53D541D3?>C53ED9F5
:?9>DC
1>41B5C8?G>9>97EB5
'%30#
:?9>DC
)85%'%@1DD5B>CC8?G5HD5B>1<%1CG5<<1C9>D5B65B5>35
6B?= D85 B5C?>1>35 3?>F5BD5B EC54 D? 75>5B1D5 13 D5CD F?<D175
)85 9>D5B65B5>35 6B?= D85 B5C?>1>35 3?>F5BD5B 1@@51B54 1C
6?EB F5BD931< <9>5C 31EC54 2I CG9D389>7 ?6 D85 C?<94CD1D5
3?>F5BD5B G9D8?ED 1>I :9DD5B 9> 13 F?<D175 @81C5 2ED G9D8
F1BI9>7 1=@<9DE45 )85 49CD1>35 25DG55> DG? :?9>DC G1C
1@@B?H9=1D5<I
;= C? @E<C5C >55454 C D? @B?@171D5 6B?=
:?9>DD?:?9>D
%?G5B312<5C13DC<9;51<?CCIDB1>C=9CC9?><9>531EC9>7>?D?><I
<5>7D845@5>49>7 45<1I 2ED 1<C? 41=@9>7 <?G@1CC 2581F9?B
49C@5BC9?> 1>4 @1BD<I B56<53D9?> 1D 5138 :?9>D 97EB5 )89C
5H@<19>CD853?=@<5H%'@B5C5>D549>97EB5
C8?G9>7>?D
?><I41=@9>7?65HD5B>1<%1>4B5C?>1>353?>F5BD5B9>D5B65B5>35
9>DG?C5@1B12<53<ECD5BC
1>42ED1<C?CE@5B9=@?C543<ECD5BC
31EC54 2I =E<D9@<5 B56<53D9?> #5F5BD85<5CC C5@1B1D9?> ?6 %'%
@1DD5B>CG1C13895F5497EB5
'%30#
335<5B1D54 @B?7B5CC 9> C?6DG1B5 1>4 81B4G1B5 ?6 4979D1<
D538>9AE5C=145@?CC92<5D?3?=29>53?>D9>E?ECB53?B49>7
?6%41D1G9D8G94521>4%=51CEB5=5>DC
(I>38B?>?EC=E<D9381>>5<%=51CEB5=5>DC9>B51<D9=5
31> 25 @5B6?B=54 G9D8 14F1>354 4979D1< % =51CEB9>7
CICD5= D85 CI>38B?>9C= ?6 % 13AE9C9D9?> E>9DC ?6
=E<D9381>>5<CICD5=9C9>1B1>75?6 >C?><I
(I>38B?>?EC%=51CEB5=5>DC1B5?67B51D9=@?BD1>356?B
C9=E<D1>5?EC 45D53D9?> 1>4 <?31D9?> ?6 % 9> + 312<5C
1335CC?B95C4EB9>716D5B9>CD1<<1D9?>495<53DB93D5CDC
%' ' 1>4 %)' 5F1<E1D9?> =5D8?4C 5>12<5
3<51B 49CD9>3D9?> ?6 =E<D9@<5 % C?EB35C 1C G5<< 1C
CE@@B5CC9?> ?6 5HD5B>1< 9>D5B65B5>35C 1<< D85C5 =5D8?4C
C8?G49665B5>D@E<C5DI@5C?EB35C9>C5@1B12<53<ECD5BC1>4
51383<ECD5B9CC5<53D12<56?BC8?G9>7D85%'%@1DD5B>?6
?>5C9>7<5%C?EB35?><I
(5@1B1D9?> ?6 % C?EB35C F1CD<I 9=@B?F5C % @1DD5B>
1>1<IC9C2IB5=?F9>71>ICE@5B9=@?C54@1DD5B>1>41<<%
5F1<E1D9?> =5D8?4C G8938 B5<I ?> C9>7<5 C?EB35
3?BB5<1D9?> <9;5 D9=5?66<978D ?B @E<C5C5AE5>35 1>1<IC9C
>?G253?=5EC56E<5F5>E>45B3B9D931<3?>49D9?>C
'%30#
%'45B9F546B?=%'%@1DD5B>C
)85 1ED8?BC G?E<4 <9;5 D? D81>; "B =1>E5< "B
$2B1<93 "B 1<;?> 1>4 B '5D8=595B 6?B D859B
3?>DB92ED9?>9>@B5@1B1D9?>?6D89C@1@5B
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
)'* +$('*('$"%" +)$ *'!%" +)$ %'*,%+$('(%(" "&")
/
0 ."9>7(9B<1C5;1B1>1>4(8?9Q(9=E<1D9?>?6 %1BD91<
9C381B75%B?@171D9?>9>%?G5B#5DG?B;R
)B1>C
%?G5B
5<9F5BI+?< @@
/0 & (81?J85> 1>4 ( 9B<1C5;1B1> Q)85 (DE4I ?6 % %B?@171D9?>
%85>?=5>?>9>%?G5B#5DG?B;R)B1>C%?G5B5<9F5BI+?<
@@
O
/0 K>971>4.#'1?%1BD91<9C381B75C9> <53DB931<
%?G5B
@@1B1DEC+
/
0 ' ,
Q'53?7>9D9?> ?6 49C381B75CR <53DB1 #?
@@
O
/0 ' 1BD>9;1C 1>4 !5F9 Q C9=@<5 @E<C5 85978D 1>1<IJ5B 6?B @1BD91<
49C381B75B1D5=51CEB5=5>DCR)B1>C>CDBE="51C+?<
@@
O
/0 ECD9> 1>4 ' 1=5C Q$><9>5 4979D1< 3?=@ED5B CICD5= 6?B
=51CEB5=5>DC?6@1BD91<49C381B75C9>9>CE<1D9?>CDBE3DEB5R<53DB
>CE<"17+?<
#?
/0 ,1B4Q979D1<D538>9AE5C6?B@1BD91<49C381B75=51CEB5=5>DCR
)B1>C%?G5B5<9F5BI+?<@@
O
/0 (89=(?B1781>1>4,(95GQ979D1<C97>1<@B?35CC9>71@@<954D?
D85 45D53D9?> ?6 @1BD91< 49C381B75 > ?F5BF95GR <53DB >CE<
"17+?<
#?@@O
/0 (89= (?B1781> 1>4 , (95G Q@@<931D9?> ?6 4979D1< C97>1<
@B?35CC9>7 1@@<954 D? D85 45D53D9?> ?6 @1BD91< 49C381B75 %1BD
$@D9=9J54 3?>F5BC9?>R <53DB >CE< "17 +?<
#?
@@
O
/
0 (89= (?B1781> 1>4 , (95G Q5D53D9?> ?6 % ED9<9J9>7 4979D1<
C97>1< @B?35CC9>7 =5D8?4C %1BD $@5><??@ >?9C5 B54E3D9?>R
<53DB>CE<"17+?<
#?
@@O
/
/
0 ,,59CC5>25B71B94'%<1D8 '5D8=595B1>4, 1<;>5BQ$>
C9D5 % 45D53D9?> 1D 3B?CC2?>49>7 <9>;C ?6 + 312<5CR '
(5CC9?>C%1B9CB1>35%1@5B
/
0 52BEI>5 1>4 $ !5C19>D Q =5D8?4 D? <?31D5 1 % C?EB35 9> 1
G9>49>7 2I @B?35CC9>7 C97>1<C =51CEB54 1D 9DC D5B=9>1<CR
D8 (
1>71<?B5>491@1@5B>?
/
0 $2B1<93 , 1<;>5B 1>4 ' %<1D8 Q(5@1B1D9?> 1>4 9>49F94E1<
5F1<E1D9?> ?6 % C?EB35C 9> CD1D?B G9>49>7 ?6 B?D1D9>7 =1389>5C 2I
CI>38B?>?EC =E<D9381>>5< =51CEB5=5>DR
D8 ( !:E2<:1>1
(<?F5>91%1@5B)
/ 0 ( (381@5B $2B1<93 ' %<1D8 1>4 , 1<;>5B Q(I>38B?>?EC =E<D9
D5B=9>1< ?>C9D5 %=51CEB5=5>DC ?> @?G5B DB1>C6?B=5BC G9D8 1>
5>81>354 D9=5 49665B5>D91< 5F1<E1D9?> =5D8?4R
D8 ( !:E2<:1>1
(<?F5>91%1@5B)
/
0 ' %<1D8 Q"E<D9381>>5< % =51CEB5=5>DCR
D8 (59:9>789>1
%1@5B
/0 '5D8=595B " BL75B B15D75 ' %<1D8 , ?<DE>?G93J
$2B1<93 1>4 , 1<;>5B QH@5B95>35C 9> ?>C9D5 @1BD91< 49C381B75
=51CEB5=5>DC 1>4 @B?C@53DC 6?B % =?>9D?B9>7R 9> ?>49D9?>
"?>9D?B9>7917>?C9C" 59:9>789>1%1@5B"
-(!'#!& -*23,-5'!8 G1C 2?B> 9> <9G935
%?<1>49>
5B5359F54D85"(%81>4
(3457B55C9>5<53DB931<5>79>55B9>76B?=D85
,1BC1G *>9F5BC9DI ?6 )538>?<?7I 9>
85G1CG9D8>CD9DED5?6%?G5B9>%?<1>4
1C 1 B5C51B38 C395>D9CD 9> D85 978 +?<D175
5@1BD=5>D B?=
D? 85 G1C G9D8
( D1<I G85B5 85 G1C =19><I 9>F?<F54 9>
+D5CD9>71>44917>?CD93C?6+5AE9@=5>D>
85 :?9>54 $"'$# G85B5 85 9C
9>F?<F549>=?>9D?B9>7?6+5AE9@=5>D59C=5=25B?6',
1>4'4F9C?BI,?B;9>7B?E@
R>CE<1D9>71C5CR1>49DC,C
59C1ED8?B?6C5F5B1<9>D5B>1D9?>1<B5@?BDC
-,*" *2& G1C 2?B> 9> 5B<9> 5B=1>I 9>
5B5359F54D85"(1>4%8457B55C9>
5<53DB931< 5>79>55B9>7 6B?= D85 5B<9> *>9F5BC9DI
?6 )538>?<?7I 9>
1>4
B5C@53D9F5<I
B?=
D?
85 G1C G9D8 D85 5B<9>
*>9F5BC9DI ?6 )538>?<?7I 9> 5B=1>I 1C 1
B5C51B38 C395>D9CD 9> D85 978 +?<D175
5@1BD=5>D B?=
D?
85 G1C G9D8
(D1<IG85B585G1C=19><I9>F?<F549>D85
69BCD
;+ @B5AE1<96931D9?> D5CDC ?6 -!%
9>CE<1D54 312<5 CICD5=C B?=
D? 85
G1C 8514 ?6 D85 8978 F?<D175 <12 ?6 % 5B<9> > 85 :?9>54
$"'$# G85B5 85 9C 9>F?<F54 9> =?>9D?B9>7 ?6 + 5AE9@=5>D 5 9C
=5=25B ?6
5B=1> =9BB?B 3?==9DD55 ?6 )
1>4 ?6
'
G?B;9>7 7B?E@C 1>4 )C 5 9C 1ED8?B ?6 C5F5B1<
9>D5B>1D9?>1<B5@?BDC
Authorized licensed use limited to: Ronald Plath. Downloaded on December 22, 2008 at 17:21 from IEEE Xplore. Restrictions apply.
A-26
Per.Christensen@nktcables.dk
insulation condition is necessary. This information can be
derived by monitoring certain diagnostic parameters during the
operation of the equipment. Consequently, continuous
monitoring is an essential tool for proper maintenance
management to guarantee the high level of asset reliability [2].
The monitoring system not only provides a set of excellent
components but also the know-how in data evaluation and long
term partnership, supporting any utilities decision over the
complete asset life cycle.
The concept of applied continuous monitoring system is
presented in Fig. 1. The signals from different sensors
measuring partial discharges, distributed temperature, oil
pressure in terminations and sheath voltage limiters are
acquired by multi-channel data acquisition units.
I.
HV/EHV
XLPE
cables,
partial
INTRODUCTION
917
Figure 4.
Figure 2.
PD acquisition system
Figure 3.
918
Figure 7.
Permanent
Periodic
Partial Discharge
for 1 every 1 h
every 2- 3 sec
once every 1 h
SVL status
--
once every 8 h
During the permanent mode, the data are acquired every 23 seconds, compared with threshold values and in real time
displayed in graphical user interface. In case those data are
within normal margins, they will be colored in green. If the
values exceed thresholds for "warning" or "alarm" they are
colored in yellow or red accordingly.
Periodic measurements are initiated in equidistant time
spans, e.g. every hour. Duration of the periodic measurement
is normally 1 min. During this time span all mentioned scalar
values are calculated and PRPD (phase resolved PD) and
3PARD diagrams are acquired. These data are saved for later
post-processing and trend visualization. Unscheduled periodic
measurements are triggered in case of one or more measured
quantities exceeding the threshold level.
PD activity is displayed as PRPD for each phase/sensor
and for each separated PD source, respectively. Trend
diagrams of statistical parameters such as PD magnitude,
frequency of occurrence of PD pulses etc. are available.
Suitable filter options enable the user to select the data display
according to his specific interest. The user can set limits which
cause warning or alarm messages when exceeded. The
measured values are continuously compared with signal
levels. The measured quantities are colored according to their
value related to pre-seated warning or alarm threshold levels.
This means, if the detected PD level on any channel on one
assed exceeds a configurable threshold, the corresponding
value will be drawn in red (Fig. 8).
The SVL status is verified within the Server Software by
FFT-based spectral analysis of the signal injected to the XB
link loop impedance from PD acquisition unit. A statistical
model of normal SVL behavior is used for reference. The
model is constructed based on the SVL data collected during
starting limited time period of cable system operation with
different load conditions. SVL status analysis is based on
statistical comparison parameters between the model and
current SVL measurements which clearly distinguish normal
operation of SVL from short circuited and open loop SVL.
919
System
element
Maintenance scheme
Activity to be
performed
Periodicity
visual check
yearly
owner
functionality check
every 3 years
owner &
system provider
updates
every 3 years
system provider
data evaluation
periodic reports
system provider
expert consultancy
in case of PD
event
system provider
Hardare
Sofware
Responsibility
CONCLUSIONS
[2]
[3]
920