Electrical Power Systems Quality, Second Edition phần 10 pptx

1. Nameplates of transformers, motors, etc.
2. Instrumentation setups
3. Transducer and probe connections
4. Key waveform displays from instruments
5. Substations, switchgear arrangements, arrester positions, etc.
6. Dimensions of key electrical components such as cable lengths
Video cameras are similarly useful when there is moving action or ran-
dom events. For example, they may be used to help identify the loca-
tions of flashovers. Many industrial facilities will require special
permission to take photographs and may place stringent limitations on
the distribution of any photographs.
11.3.5 Oscilloscopes
An oscilloscope is valuable when performing real-time tests. Looking at
the voltage and current waveforms can provide much information
about what is happening, even without performing detailed harmonic
analysis on the waveforms. One can get the magnitudes of the voltages
and currents, look for obvious distortion, and detect any major varia-
tions in the signals.
There are numerous makes and models of oscilloscopes to choose
from. A digital oscilloscope with data storage is valuable because the
waveform can be saved and analyzed. Oscilloscopes in this category
often also have waveform analysis capability (energy calculation, spec-
trum analysis). In addition, the digital oscilloscopes can usually be
obtained with communications so that waveform data can be uploaded
to a personal computer for additional analysis with a software package.
The latest developments in oscilloscopes are hand-held instruments
with the capability to display waveforms as well as performing some
signal processing. These are quite useful for power quality investiga-
tions because they are very portable and can be operated like a volt-
ohm meter (VOM), but yield much more information. These are ideal
for initial plant surveys. A typical device is shown in Figs. 11.10 and

11.11. This particular instrument also has the capability to analyze
harmonics and permits connection with personal computers for further
data analysis and inclusion into reports as illustrated.
11.3.6 Disturbance analyzers
Disturbance analyzers and disturbance monitors form a category of
instruments that have been developed specifically for power quality
measurements. They typically can measure a wide variety of system
Power Quality Monitoring 475
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
disturbances from very short duration transient voltages to long-dura-
tion outages or undervoltages. Thresholds can be set and the instru-
ments left unattended to record disturbances over a period of time. The
information is most commonly recorded on a paper tape, but many
devices have attachments so that it can be recorded on disk as well.
There are basically two categories of these devices:
1. Conventional analyzers that summarize events with specific infor-
mation such as overvoltage and undervoltage magnitudes, sags and
surge magnitude and duration, transient magnitude and duration,
etc.
2. Graphics-based analyzers that save and print the actual waveform
along with the descriptive information which would be generated by
one of the conventional analyzers
It is often difficult to determine the characteristics of a disturbance
or a transient from the summary information available from conven-
tional disturbance analyzers. For instance, an oscillatory transient
cannot be effectively described by a peak and a duration. Therefore, it
476 Chapter Eleven

Figure 11.10 A hand-held oscillographic monitoring instrument. (Courtesy of Fluke
Corporation.)
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
is almost imperative to have the waveform capture capability of a
graphics-based disturbance analyzer for detailed analysis of a power
quality problem (Fig. 11.12). However, a simple conventional distur-
bance monitor can be valuable for initial checks at a problem location.
11.3.7 Spectrum analyzers and harmonic
analyzers
Instruments in the disturbance analyzer category have very limited
harmonic analysis capabilities. Some of the more powerful analyzers
have add-on modules that can be used for computing fast Fourier
transform (FFT) calculations to determine the lower-order harmonics.
However, any significant harmonic measurement requirements will
demand an instrument that is designed for spectral analysis or har-
monic analysis. Important capabilities for useful harmonic measure-
ments include
Power Quality Monitoring 477
Figure 11.11 Demonstrating the use of a hand-held, three-
phase power quality monitoring instrument to quickly
evaluate voltages at the mains.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
■
Capability to measure both voltage and current simultaneously so

that harmonic power flow information can be obtained.
■
Capability to measure both magnitude and phase angle of individual
harmonic components (also needed for power flow calculations).
■
Synchronization and a sampling rate fast enough to obtain accurate
measurement of harmonic components up to at least the 37th har-
monic (this requirement is a combination of a high sampling rate and
a sampling interval based on the 60-Hz fundamental).
■
Capability to characterize the statistical nature of harmonic distor-
tion levels (harmonics levels change with changing load conditions
and changing system conditions).
There are basically three categories of instruments to consider for
harmonic analysis:
1. Simple meters. It may sometimes be necessary to make a quick
check of harmonic levels at a problem location. A simple, portable
meter for this purpose is ideal. There are now several hand-held
instruments of this type on the market. Each instrument has advan-
tages and disadvantages in its operation and design. These devices
generally use microprocessor-based circuitry to perform the necessary
calculations to determine individual harmonics up to the 50th har-
monic, as well as the rms, the THD, and the telephone influence factor
(TIF). Some of these devices can calculate harmonic powers (magni-
tudes and angles) and can upload stored waveforms and calculated
data to a personal computer.
2. General-purpose spectrum analyzers. Instruments in this cate-
gory are designed to perform spectrum analysis on waveforms for a
478 Chapter Eleven
Figure 11.12 Graphics-based analyzer output.

Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
wide variety of applications. They are general signal analysis instru-
ments. The advantage of these instruments is that they have very pow-
erful capabilities for a reasonable price since they are designed for a
broader market than just power system applications. The disadvan-
tage is that they are not designed specifically for sampling power fre-
quency waveforms and, therefore, must be used carefully to assure
accurate harmonic analysis. There are a wide variety of instruments in
this category.
3. Special-purpose power system harmonic analyzers. Besides the
general-purpose spectrum analyzers just described, there are also a
number of instruments and devices that have been designed specifi-
cally for power system harmonic analysis. These are based on the FFT
with sampling rates specifically designed for determining harmonic
components in power signals. They can generally be left in the field and
include communications capability for remote monitoring.
11.3.8 Combination disturbance and
harmonic analyzers
The most recent instruments combine harmonic sampling and energy
monitoring functions with complete disturbance monitoring functions
as well. The output is graphically based, and the data are remotely
gathered over phone lines into a central database. Statistical analysis
can then be performed on the data. The data are also available for input
and manipulation into other programs such as spreadsheets and other
graphical output processors.
One example of such an instrument is shown in Fig. 11.13. This
instrument is designed for both utility and end-user applications, being

mounted in a suitable enclosure for installation outdoors on utility
poles. It monitors three-phase voltages and currents (plus neutrals)
simultaneously, which is very important for diagnosing power quality
problems. The instrument captures the raw data and saves the data in
internal storage for remote downloading. Off-line analysis is performed
with powerful software that can produce a variety of outputs such as
that shown in Fig. 11.14. The top chart shows a typical result for a volt-
age sag. Both the rms variation for the first 0.8 s and the actual wave-
form for the first 175 ms are shown. The middle chart shows a typical
wave fault capture from a capacitor-switching operation. The bottom
chart demonstrates the capability to report harmonics of a distorted
waveform. Both the actual waveform and the harmonic spectrum can
be obtained.
Another device is shown in Fig. 11.15. This is a power quality moni-
toring system designed for key utility accounts. It monitors three-phase
voltages and has the capability to capture disturbances and page power
Power Quality Monitoring 479
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
quality engineers. The engineers can then call in and hear a voice mes-
sage describing the event. It has memory for more than 30 events.
Thus, while only a few short years ago power quality monitoring was
a rare feature to be found in instruments, it is becoming much more
commonplace in commercially available equipment.
11.3.9 Flicker meters*
Over the years, many different methods for measuring flicker have been
developed. These methods range from using very simple rms meters
with flicker curves to elaborate flicker meters that use exactly tuned fil-

ters and statistical analysis to evaluate the level of voltage flicker. This
section discusses various methods available for measuring flicker.
Flicker standards. Although the United States does not currently have
a standard for flicker measurement, there are IEEE standards that
address flicker. IEEE Standards 141-1993
6
and 519-1992
7
both contain
480 Chapter Eleven
Figure 11.13 A power quality monitoring instrument capable of monitoring disturbances,
harmonics, and other steady-state phenomena on both utility systems and end-user sys-
tems. (Courtesy of Dranetz-BMI.)
*
This subsection was contributed by Jeff W. Smith and Erich W. Gunther.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Power Quality Monitoring 481
Phase C-A Voltage
RMS Variation
Trigger
Phase A Voltage
Wave Fault
Phase A Current
SS Wave
Trigger
80
85

90
95
100
105
110
115
–150
–1.5
–600
0
20
40
60
80
100
–400
–200
0
200
400
600
1.5
–0.5
0.5
0
–1
1
–100
–50
0

50
100
150
Voltage (%)Voltage (%)Voltage (pu)Current (Amps)Amps
0
0
0
0
0 1020304050
10 20 30 40 50
10 20 30 40 50 60 70
25 50 75 100 125 150 175
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Time (s)
Time (ms)
Time (ms)
Time (ms)
Harmonic
Duration
0.150 s
Min
81.38
Ave
96.77
Max
101.4
Max 1.094
Min –1.280
Fund 267.5
RMS 281.0

CF 1.772
Min –495.6
Max 498.0
THD 30.67
HRMS 86.19
TIF/IT 70249
BMI/Electrotek
Uncalibrated Data
Figure 11.14 Output from combination disturbance and har-
monic analyzer.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
flicker curves that have been used as guides for utilities to evaluate the
severity of flicker within their system. Both flicker curves, from
Standards 141 and 519, are shown in Fig. 11.16.
In other countries, a standard methodology for measuring flicker has
been established. The IEC flicker meter is the standard for measuring
flicker in Europe and other countries currently adopting IEC stan-
dards. The IEC method for flicker measurement, defined in IEC
Standard 61000-4-15
8
(formerly IEC 868), is a very comprehensive
approach to flicker measurement and is further described in “Flicker
Measurement Techniques” below. More recently, the IEEE has been
working toward adoption of the IEC flicker monitoring standards with
an additional curve to account for the differences between 230-V and
120-V systems.
Flicker measurement techniques

RMS strip charts. Historically, flicker has been measured using rms
meters, load duty cycle, and a flicker curve. If sudden rms voltage devi-
ations occurred with specified frequencies exceeding values found in
flicker curves, such as one shown in Fig. 11.16, the system was said to
have experienced flicker. A sample graph of rms voltage variations is
shown in Fig. 11.17 where large voltage deviations up to 9.0 V rms (⌬V/V
ϭ ± 8.0 percent on a 120-V base) are found. Upon comparing this to the
flicker curve in Fig. 11.16, the feeder would be experiencing flicker,
regardless of the duty cycle of the load producing the flicker, because
482 Chapter Eleven
Figure 11.15 A low-cost power quality monitor that can page power quality engi-
neers when disturbances occur.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
any sudden total change in voltage greater than 7.0 V rms results in
objectionable flicker, regardless of the frequency. The advantage to
such a method is that it is quite simple in nature and the rms data
required are rather easy to acquire. The apparent disadvantage to such
a method would be the lack of accuracy and inability to obtain the exact
frequency content of the flicker.
Fast Fourier transform. Another method that has been used to measure
flicker is to take raw samples of the actual voltage waveforms and
Power Quality Monitoring 483
IEEE 141
IEEE 519
0.01 0.1 1 10 100 1000 10000
Changes/min
⌬V/V (%)

0.1
1
10
Figure 11.16 Flicker curves from IEEE Standards 141 and 519.
Figure 11.17 RMS voltage variations.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
implement a fast Fourier transform on the demodulated signal (flicker
signal only) to extract the various frequencies and magnitudes found in
the data. These data would then be compared to a flicker curve. Although
similar to using the rms strip charts, this method more accurately quan-
tifies the data measured due to the magnitude and frequency of the
flicker being known. The downside to implementing this method is asso-
ciated with quantifying flicker levels when the flicker-producing load
contains multiple flicker signals. Some instruments compensate for this
by reporting only the dominant frequency and discarding the rest.
Flicker meters. Because of the complexity of quantifying flicker levels
that are based upon human perception, the most comprehensive
approach to measuring flicker is to use flicker meters. A flicker meter
is essentially a device that demodulates the flicker signal, weights it
according to established “flicker curves,” and performs statistical
analysis on the processed data.
Generally, these meters can be divided up into three sections. In the
first section the input waveform is demodulated, thus removing the
carrier signal. As a result of the demodulator, a dc offset and higher-fre-
quency terms (sidebands) are produced. The second section removes
these unwanted terms using filters, thus leaving only the modulating
(flicker) signal remaining. The second section also consists of filters

that weight the modulating signal according to the particular meter
specifications. The last section usually consists of a statistical analysis
of the measured flicker.
The most established method for doing this is described in IEC
Standard 61000-4-15.
8
The IEC flicker meter consists of five blocks,
which are shown in Fig. 11.18.
Block 1 is an input voltage adapter that scales the input half-cycle
rms value to an internal reference level. This allows flicker measure-
ments to be made based upon a percent ratio rather than be dependent
upon the input carrier voltage level.
Block 2 is simply a squaring demodulator that squares the input to
separate the voltage fluctuation (modulating signal) from the main
voltage signal (carrier signal), thus simulating the behavior of the
incandescent lamp.
Block 3 consists of multiple filters that serve to filter out unwanted
frequencies produced from the demodulator and also to weight the
input signal according to the incandescent lamp eye-brain response.
The basic transfer function for the weighting filter is
H(s) ϭ
и
1 ϩ s/␻
2
ᎏᎏᎏ
(
1 ϩ s/␻
3)(
1 ϩ s/␻
4)

k␻
1
s
ᎏᎏ
s
2
ϩ 2␭s ϩ␻
1
2
484 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
485
Input
transformer
Block 1 Block 2 Block 3 Block 4 Block 5
Detector and
gain control
Demodulator
with
squaring
multiplier
Input voltage
adaptor
Signal generator
for calibration
checking
dB

–3
.05
35 Hz
1
0 8.8 Hz
⌬V
V
Range
selector
Squaring
multiplier
64 level
classifier
Output
interfaces
1
st
-
order
sliding
mean
filter
0.5
1.0
2.0
5.0
10.0
20.0
Weighting filters
Squaring and

smoothing
A/D
converter
Sampling
rate
Ն 50 Hz
Programming of short and long
observation periods
Statistical evaluation of flicker level
Output
and data
display
and
recording
Square rooter
√
RMS
meter
Output 1:
Half-cycle
rms voltage
indication
Output 2:
Weighted
voltage
fluctuation
Output 3:
Range
selection
Output 4:

Short time
integration
Output 5:
Recording
1-min integrator
ƒ
Figure 11.18 Diagram of the IEC flicker meter.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
(See IEC Standard 61000-4-15 for a description of the variables used
above.)
Block 4 consists of a squaring multiplier and sliding mean filter. The
voltage signal is squared to simulate the nonlinear eye-brain response,
while the sliding mean filter averages the signal to simulate the short-
term storage effect of the brain. The output of this block is considered
to be the instantaneous flicker level. A level of 1 on the output of this
block corresponds to perceptible flicker.
Block 5 consists of a statistical analysis of the instantaneous flicker
level. The output of block 4 is divided into suitable classes, thus creat-
ing a histogram. A probability density function is created based upon
each class, and from this a cumulative distribution function can be
formed.
Flicker level evaluation can be divided into two categories, short-
term and long-term. Short-term evaluation of flicker severity P
ST
is
based upon an observation period of 10 min. This period is based upon
assessing disturbances with a short duty cycle or those that produce

continuous fluctuations. P
ST
can be found using the equation
P
ST
ϭ ͙0.031
ෆ
4P
0.1
ϩ
ෆ
0.052
ෆ
5P
1s
ϩ
ෆ
0.065
ෆ
7P
3s
ϩ
ෆ
0.28P
ෆ
10s
ϩ 0
ෆ
.08P
50

ෆ
s
ෆ
where the percentages P
0.1
, P
1s
,P
3s
,P
10s
, and P
50s
are the flicker levels
that are exceeded 0.1, 1.0, 3.0, 10.0, and 50.0 percent of the time,
respectively. These values are taken from the cumulative distribution
curve discussed previously. A P
ST
of 1.0 on the output of block 5 repre-
sents the objectionable (or irritable) limit of flicker.
For cases where the duty cycle is long or variable, such as in arc fur-
naces, or disturbances on the system that are caused by multiple loads
operating simultaneously, the need for the long-term assessment of
flicker severity arises. Therefore, the long-term flicker severity P
LT
is
derived from P
ST
using the equation
P

LT
ϭ
Ί
3
๶
where N is the number of P
ST
readings and is determined by the duty
cycle of the flicker-producing load. The purpose is to capture one duty
cycle of the fluctuating load. If the duty cycle is unknown, the recom-
mended number of P
ST
readings is 12 (2-h measurement window).
The advantage of using a single quantity, like Pst, to characterize
flicker is that it provides a basis for implementing contracts and
describing flicker levels in a much simpler manner. Figure 11.19 illus-
trates the Pst levels measured at the PCC with an arc furnace over a
Α
N
i ϭ 1
P
3
STi
ᎏ
N
486 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.

24-h period. The melt cycles when the furnace was operating can be
clearly identified by the high Pst levels. Note that Pst levels greater
than 1.0 are usually considered to be levels that might result in cus-
tomers being aware of lights flickering.
11.3.10 Smart power quality monitors
All power quality measurement instruments previously described are
designed to collect power quality data. Some instruments can send the
data over a telecommunication line to a central processing location for
analysis and interpretation. However, one common feature among
these instruments is that they do not possess the capability to locally
analyze, interpret, and determine what is happening in the power sys-
tem. They simply record and transmit data for postprocessing.
Since the conclusion of the EPRI DPQ project in Fall 1995, it was
realized that these monitors, along with the monitoring practice previ-
ously described, were inadequate. An emerging trend in power quality
monitoring practice is to collect the data, turn them into useful infor-
mation, and disseminate it to users. All these processes take place
within the instrument itself. Thus, a new breed of power quality mon-
itor was developed with integrated intelligent systems to meet this new
challenge. This type of power quality monitor is an intelligent power
quality monitor where information is directly created within the
instrument and immediately available to the users. A smart power
Power Quality Monitoring 487
0.00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00
Time
P
st
0
0.5
1

1.5
2
2.5
Figure 11.19 Flicker variations at the PCC with an arc furnace characterized by the Pst
levels for a 24-h period (March 1, 2001) (note that there is one Pst value every 10 min).
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
quality monitor allows engineers to take necessary or appropriate
actions in a timely manner. Thus, instead of acting in a reactive fash-
ion, engineers will act in a proactive fashion.
One such smart power quality monitor was developed by Electrotek
Concepts, Dranetz-BMI, EPRI, and the Tennessee Valley Authority (TVA)
(Fig. 11.20). The system features on-the-spot data analysis with rapid
information dissemination via Internet technology, e-mails, pagers, and
faxes. The system consists of data acquisition, data aggregation, commu-
nication, Web-based visualization, and enterprise management compo-
nents. The data acquisition component (DataNode) is designed to
measure the actual power system voltages, currents, and other quanti-
ties. The data aggregation, communication, Web-based visualization, and
enterprise management components are performed by a mission-specific
computer system called the InfoNode. The communication between the
data acquisition device and the InfoNode is accomplished through serial
RS-232/485/422 or Ethernet communications using industry standard
protocols (UCA MMS and Modbus). One or more data acquisition devices,
or DataNodes, can be connected to an InfoNode.
The InfoNode has its own firmware that governs the overall func-
tionality of the monitoring system. It acts as a special-purpose data-
base manager and Web server. Various special-purpose intelligent

systems are implemented within this computer system. Since it is a
Web server, any user with Internet connectivity can access the data
and its analysis results stored in its memory system. The monitoring
system supports the standard file transfer protocol (FTP). Therefore, a
database can be manually archived via FTP by simply copying the data-
base to any personal computer with connectivity to the mission-specific
computer system via network or modem. Proprietary software can be
used to archive data from a group of InfoNodes.
11.3.11 Transducer requirements
Monitoring of power quality on power systems often requires trans-
ducers to obtain acceptable voltage and current signal levels. Voltage
monitoring on secondary systems can usually be performed with direct
connections, but even these locations require current transformers
(CTs) for the current signal.
Many power quality monitoring instruments are designed for input
voltages up to 600 V rms and current inputs up to 5 A rms. Voltage and
current transducers must be selected to provide these signal levels.
Two important concerns must be addressed in selecting transducers:
1. Signal levels. Signal levels should use the full scale of the instru-
ment without distorting or clipping the desired signal.
488 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
2. Frequency response. This is particularly important for transient
and harmonic distortion monitoring, where high-frequency signals
are particularly important.
These concerns and transducer installation considerations will now be
discussed.

Signal levels. Careful consideration to sizing of voltage transducers
(VTs) and CTs is required to take advantage of the full resolution of the
instrument without clipping the measured signal. Improper sizing can
result in damage to the transducer or monitoring instrument.
Digital monitoring instruments incorporate the use of analog-to-dig-
ital (A/D) converters. These A/D boards convert the analog signal
received by the instrument from the transducers into a digital signal
for processing. To obtain the most accurate representation of the signal
being monitored, it is important to use as much of the full range of the
A/D board as possible. The noise level of a typical A/D board is approx-
imately 33 percent of the full-scale bit value (5 bits for a 16-bit A/D
board). Therefore, as a general rule, the signal that is input to the
instrument should never be less than one-eighth of the full-scale value
so that it is well above the noise level of the A/D board. This can be
accomplished by selecting the proper transducers.
Voltage transducers. VTs should be sized to prevent measured distur-
bances from inducing saturation in the VT. For transients, this gener-
ally requires that the knee point of the transducer saturation curve be
at least 200 percent of nominal system voltage.
Power Quality Monitoring 489
Signature System Architecture
Web Browsers
InfoNodes
DataNodes
Figure 11.20 A smart power quality monitoring system—it
turns data into information on the spot and makes it avail-
able over the Internet. (Courtesy of Dranetz-BMI.)
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.
Example 1. When monitoring on a 12.47-kV distribution feeder and
measuring line-to-ground, the nominal voltage across the primary of
the voltage transducer will be 7200 V rms.
A VT ratio of 60:1 will produce an output voltage on the VT of 120 V
rms (170 V peak) for a 7200-V rms input. Therefore, if the full-range
value of the instrument is 600 V rms and the instrument incorporates
a 16-bit A/D board, 13ϩ bits of the A/D board will be used.
It is always good practice to incorporate some allowance in the cal-
culations for overvoltage conditions. The steady-state voltage should
not be right at the full-scale value of the monitoring instrument. If an
overvoltage occurred, the signal would be clipped by the A/D board, and
the measurement would be useless. Allowing for a 200 percent over-
voltage is suggested. This can be accomplished by changing the input
scale on the instrument, or sizing the VT accordingly.
Current transducers. Selecting the proper transducer for currents is
more difficult. The current in any system changes more often and with
greater magnitude than the voltage. Most power quality instrument
manufacturers supply CTs with their equipment. These CTs come in a
wide range of sizes to accommodate different load levels. The CTs are
usually rated for maximum continuous load current.
The proper CT current rating and turns ratio depend on the mea-
surement objective. If fault or inrush currents are of concern, the CT
must be sized in the range of 20 to 30 times normal load current. This
will result in low resolution of the load currents and an inability to
accurately characterize load current harmonics.
If harmonics and load characterization are important, CTs should be
selected to accurately characterize load currents. This permits evalua-
tion of load response to system voltage variations and accurate calcu-
lation of load current harmonics.

Example 2. The desired current signal to the monitoring instrument
is 1 to 2 A rms. Assuming a 1-A value, the optimum CT ratio for an
average feeder current of 120 A rms is 120:1. Manufacturer’s data com-
monly list a secondary current base of 5 A to describe CT turns ratios
rather than 1 A. The primary rating for a CT with a 5-A secondary rat-
ing is calculated as follows:
CT
PRI
ϭϭϭ600
Thus, a 600:5 CT would be specified.
120 и 5
ᎏ
1
I
PRI
CT
SEC
ᎏᎏ
I
SEC
490 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Frequency response. Transducer frequency response characteristics
can be illustrated by plotting the ratio correction factor (RCF), which is
the ratio of the expected output signal (input scaled by turns ratio) to
the actual output signal, as a function of frequency.
Voltage transducers. The frequency response of a standard metering

class VT depends on the type and burden. In general, the burden
should be a very high impedance (see Figs. 11.21 and 11.22). This is
generally not a problem with most monitoring equipment available
today. Power quality monitoring instruments, digital multimeters
(DMMs), oscilloscopes, and other instruments all present a very high
impedance to the transducer. With a high impedance burden, the
response is usually adequate to at least 5 kHz. A typical RCF is plotted
in Figs. 11.21 and 11.22 for two VT burdens.
9
Some substations use capacitively coupled voltage transformers
(CCVTs) for voltage transducers. These should not be used for general
power quality monitoring. There is a low-voltage transformer in paral-
lel with the lower capacitor in the capacitive divider. This configuration
results in a circuit that is tuned to 60 Hz and will not provide accurate
representation of any higher-frequency components.
Measuring very high frequency components in the voltage requires a
capacitive divider or pure resistive divider. Figure 11.23 illustrates the
difference between a CCVT and a capacitive divider. Special-purpose
capacitor dividers can be obtained for measurements requiring accu-
rate characterization of transients up to at least 1 MHz.
Current transducers. Standard metering class CTs are generally adequate
for frequencies up to 2 kHz (phase error may start to become significant
before this).
10
For higher frequencies, window-type CTs with a high turns
ratio (doughnut, split-core, bar-type, and clamp-on) should be used.
Additional desirable attributes for CTs include
1. Large turns ratio, e.g., 2000:5 or greater.
2. Window-type CTs are preferred. Primary wound CTs (i.e., CTs in
which system current flows through a winding) may be used, pro-

vided that the number of turns is less than five.
3. Small remnant flux, e.g., ±10 percent of the core saturation value.
4. Large core area. The more steel used in the core, the better the fre-
quency response of the CT.
5. Secondary winding resistance and leakage impedance as small as
possible. As shown in Fig. 11.24, this allows more of the output sig-
nal to flow into the burden, rather than the stray capacitance and
core exciting impedance.
Power Quality Monitoring 491
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Installation considerations. Monitoring on the distribution primary
requires both voltage and current transducers. Selection of the best
combination of these transducers depends on a number of factors:
■
Monitoring location (substation, overhead, underground, etc.)
■
Space limitations
■
Ability to interrupt circuit for transducer installation
■
Need for current monitoring
Substation transducers. Usually, existing substation CTs and VTs
(except CCVTs) can be used for power quality monitoring.
Utility overhead line locations. For power quality monitoring on distribu-
tion primary circuits, it is often desirable to use a transducer that could
be installed without taking the circuit out of service. Recently, trans-
ducers for monitoring both voltage and current have been developed

that can be installed on a live line.
These devices incorporate a resistive divider-type VT and window-
type CT in a single unit. A split-core choke is clamped around the phase
492 Chapter Eleven
0.96
0.98
1
1.02
1.04
1.06
1.08
RCF
10 100 1000 10000 100000 1000000
Frequency in Hertz
Figure 11.21 Frequency response of a standard VT with 1-M⍀ burden.
10 100 1000 10000 100000 100000
0
Frequency in Hertz
0.96
0.98
1
1.02
1.04
1.06
1.08
RCF
Figure 11.22 Frequency response of a standard VT with 100-⍀ burden.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.

Any use is subject to the Terms of Use as given at the website.
conductor and is used to shunt the line current through the CT in the
insulator. This method allows the device to be installed on the crossarm
in place of the original insulator. By using the split-core choke, the
phase conductor does not have to be broken, and thus, the transducers
can be installed on a live line.
Initial tests indicated adequate frequency response for these trans-
ducers. However, field experience with these units has shown that the
frequency response, even at 60 Hz, is dependent on current magnitude,
temperature, and secondary cable length. This makes this type of
device difficult to use for accurate power quality monitoring. Care must
be exercised in matching these transducers to the instruments.
In general, all primary sites should be monitored with metering class
VTs and CTs to obtain accurate results over the required frequency spec-
trum. Installation will require a circuit outage, but convenient designs
can be developed for pole-top installations to minimize the outage.
Another option for monitoring primary sites involves monitoring at
the secondary of an unloaded distribution transformer. This will give
accurate results up to at least 3 kHz. This option does not help with the
current transducers, but it is possible to get by without the currents at
some circuit locations (e.g., end of the feeder). This option may be par-
ticularly attractive for underground circuits where the monitor can be
installed on the secondary of a pad-mounted transformer.
Power Quality Monitoring 493
V
out
V
out
V
in

V
in
Good Bad
Figure 11.23 Capacitively coupled
voltage dividers.
10 100 1000 10000 100000 100000
0
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
Equivalent circuit for determining CT frequency response
ideal
transformer
primary
winding
secondary
winding
Zp
Ze
Zs
Zb
Cs
N:5
exciting
impedance

stray
capacitance
burden
Frequency in Hertz
1/RCF
Figure 11.24 Frequency response of a window-type CT.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Primary wound CTs are available from a variety of CT manufac-
turers. Reference 2 concludes that any primary wound CT with a sin-
gle turn, or very few turns, should have a frequency response up to
10 kHz.
End-user (secondary) sites. Transducer requirements at secondary sites
are much simpler. Direct connection for the voltage is possible for
120/208- or 277/480-V rms systems. This permits full utilization of the
instrument’s frequency-response capability.
Currents can be monitored with either metering CTs (at the service
entrance, for example) or with clamp-on CTs (at locations within the
facility). Clamp-on CTs are available in a wide range of turns ratios.
The frequency range is usually published by the manufacturer.
Summary of transducer recommendations. Table 11.2 describes differ-
ent monitoring locations and the different types of transducers that are
adequate for monitoring at these locations.
Table 11.3 describes the different power quality phenomena and the
proper transducers to measure that type of power quality problem.
Tables 11.2 and 11.3 should be used in conjunction with each other to
determine the best transducer for a given application.
Summary of monitoring equipment capabilities. Figure 11.25 summa-

rizes the capabilities of the previously described metering instruments
as they relate to the various categories of power quality variations.
11.4 Assessment of Power Quality
Measurement Data
As utilities and industrial customers have expanded their power
quality monitoring systems, the data management, analysis, and
interpretation functions have become the most significant challenges
in the overall power quality monitoring effort. In addition, the shift
in the use of power quality monitoring from off-line benchmarking to
on-line operation with automatic identification of problems and con-
cerns has made the task of data management and analysis even more
critical.
There are two streams of power quality data analysis, i.e., off-line
and on-line analyses. The off-line power quality data analysis, as the
term suggests, is performed off-line at the central processing locations.
On the other hand, the on-line data analysis is performed within the
instrument itself for immediate information dissemination. Both types
of power quality data assessment are described in Secs. 11.4.1 and
11.4.2.
494 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Power Quality Monitoring 495
TABLE 11.2 VT and CT Options for Different Locations
Location VT CT
Substation Metering VTs Metering CTs
Special-purpose Relaying CTs
capacitive or resistive

dividers
Calibrated bushing taps
Overhead lines Metering VTs Metering CTs
Underground Metering VTs Metering CTs
locations Pad-mounted transformer
Special-purpose dividers
Secondary sites’ Direct connection Metering CTs
service entrance Clamp-on CTs
In facility Direct connection Clamp-on CTs
TABLE 11.3 VT and CT Requirements for Different Power Quality Variations
Concern VTs* CTs
Voltage variations Standard metering Standard metering
Harmonic levels Standard metering Window-type
Low-frequency transients Standard metering with Window-type
(switching) high-kneepoint saturation
High-frequency transients Capacitive or resistive Window-type
(lightning) dividers
*VTs are usually not required at locations below 600 V rms nominal.
Figure 11.25 Power quality measurement equipment capabilities.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
11.4.1 Off-line power quality data
assessment
Off-line power quality data assessment is carried out separately from
the monitoring instruments. Dedicated computer software is used for
this purpose. Large-scale monitoring projects with large volumes of
data to analyze often present a challenging set of requirements for
software designers and application engineers. First, the software

must integrate well with monitoring equipment and the large num-
ber of productivity tools that are currently available. The storage of
vast quantities of both disturbance and steady-state measurement
data requires an efficient and well-suited database. Data manage-
ment tools that can quickly characterize and load power quality data
must be devised, and analysis tools must be integrated with the data-
base. Automation of data management and report generation tasks
must be supported, and the design must allow for future expansion
and customizing.
The new standard format for interchanging power quality data—the
Power Quality Data Interchange Format (PQDIF)—makes sharing of
data between different types of monitoring systems much more feasi-
ble. This means that applications for data management and data
analysis can be written by third parties and measurement data from a
wide variety of monitoring systems can be accessible to these systems.
PQView (www.pqview.com) is an example of this type of third-party
application. The PQDIF standard is described in Sec. 11.6.
The off-line power quality data assessment software usually per-
forms the following functions:
■
Viewing of individual disturbance events.
■
RMS variation analysis which includes tabulations of voltage sags
and swells, magnitude-duration scatter plots based on CBEMA, ITI,
or user-specified magnitude-duration curves, and computations of a
wide range of rms indices such as SARFI, SIARFI, and CAIDI.
■
Steady-state analysis which includes trends of rms voltages, rms cur-
rents, and negative- and zero-sequence unbalances. In addition,
many software systems provide statistical analysis of various mini-

mum, average, maximum, standard deviation, count, and cumula-
tive probability levels. Statistics can be temporally aggregated and
dynamically filtered. Figures 11.26 and 11.27 show the time trend of
phase A rms voltage along with its histogram representation.
■
Harmonic analysis where users can perform voltage and current har-
monic spectra, statistical analysis of various harmonic indices, and
trending overtime.
496 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
■
Transient analysis which includes statistical analysis of maximum
voltage, transient durations, and transient frequency.
■
Standardized power quality reports (e.g. daily reports, monthly
reports, statistical performance reports, executive summaries, cus-
tomer power quality summaries).
Power Quality Monitoring 497
Samples: 1404
Minimum: 6873.0806
Average: 7284.7099
Maximum: 7600.3726
6800
6900
7000
7100
7200

7300
7400
7500
7600
7700
4/26/95 5/1/95 5/6/95 5/11/95 5/16/95 5/21/95 5/26/95 5/31/95 6/5/95
V RMS A
Figure 11.26 Time trend of an rms voltage is a standard feature in many power quality
analysis software programs.
0
20
40
60
80
100
120
Count
6870
6910
6950
6990
7030
7070
7110
7150
7190
7230
7270
7310
7350

7390
7470
7510
7550
7590
7430
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Cumulative Frequency
Samples: 1404
Minimum: 6873.0806
Average: 7284.7099
Maximum: 7600.3726
V RMS A
Figure 11.27 Histogram representation of rms voltage indicates the statistical distribu-
tion of the rms voltage magnitude.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
■

Analysis of protective device operation (identify problems).
■
Analysis of energy use.
■
Correlation of power quality levels or energy use with important
parameters (e.g., voltage sag performance versus lightning flash
density).
■
Equipment performance as a function of power quality levels (equip-
ment sensitivity reports).
11.4.2 On-line power quality data
assessment
On-line power quality data assessment analyzes data as they are cap-
tured. The analysis results are available immediately for rapid dis-
semination. Complexity in the software design requirement for on-line
assessment is usually higher than that of off-line. Most features avail-
able in off-line analysis software can also be made available in an on-
line system. One of the primary advantages of on-line data analysis is
that it can provide instant message delivery to notify users of specific
events of interest. Users can then take immediate actions upon receiv-
ing the notifications. Figure 11.28 illustrates a simple message deliv-
ered to a user reporting that a capacitor bank located upstream from a
data acquisition node called “DataNode H09_5530” was energized at
05-15-2002 at 04:56:11
A.M. The message also details the transient
characteristics such as the magnitude, frequency, and duration along
with the relative location of the capacitor bank from the data acquisi-
tion node.
Figure 11.29 shows another example of the on-line power quality
assessment. It shows the time trend of a fifth-harmonic current mag-

nitude along with its statistical distribution. The data and its analysis
are displayed on a standard Web browser. Here a user can analyze data
up to the current time. This on-line system has the capability of per-
forming a full range of transient, harmonic, and steady-state charac-
terization along with their statistical distribution analysis comparable
to that in off-line assessment analysis.
11.5 Application of Intelligent Systems
Many advanced power quality monitoring systems are equipped with
either off-line or on-line intelligent systems to evaluate disturbances
and system conditions so as to make conclusions about the cause of the
problem or even predict problems before they occur. The applications of
intelligent systems or autonomous expert systems in monitoring
instruments help engineers determine the system condition rapidly.
498 Chapter Eleven
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
This is especially important when restoring service following major dis-
turbances.
The implementation of intelligent systems within a monitoring
instrument can significantly increase the value of a monitoring appli-
cation since it can generate information rather than just collect data.
11
The intelligent systems are packaged as individual autonomous expert
system modules, where each module performs specific functions.
Examples include an expert system module that analyzes capacitor-
switching transients and determines the relative location of the capac-
itor bank, and an expert system module to determine the relative
location of the fault causing a voltage sag. Sections 11.5.1 and 11.5.2

describe the approach in designing an autonomous expert system for
power quality data assessment, and give application examples.
11.5.1 Basic design of an expert system for
monitoring applications
The development of an autonomous expert system calls for many
approaches such as signal processing and rule-based techniques along
with the knowledge-discovery approach commonly known as data min-
ing. Before the expert system module is designed, the functionalities or
objectives of the module must be clearly defined. In other words, the
designers or developers of the expert system module must have a clear
understanding about what knowledge they are trying to discover from
volumes of raw measurement data. This is very important since they will
ultimately determine the overall design of the expert system module.
Power Quality Monitoring 499
Figure 11.28 On-line data analysis can send e-mail notifications to users about the occur-
rence of specific events.
Power Quality Monitoring
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.