Download Validation of the Measurement Characteristics in an Instrument for

energies
Article
Validation of the Measurement Characteristics
in an Instrument for Power Quality
Estimation—A Case Study †
Romuald Masnicki
Marine Electrical Engineering Faculty, Gdynia Maritime University, 81-225 Gdynia, Poland;
[email protected]; Tel.: +48-58-5586-666
† This is an expanded version of a conference paper: Masnicki, R. Validation of the measurement algorithms in
instrument for power quality estimation. In Proceedings of the IEEE 16th International Conference on
Environment and Electrical Engineering (EEEIC), Florence, Italy, 7–10 June 2016.
Academic Editor: Rodolfo Araneo
Received: 15 February 2017; Accepted: 6 April 2017; Published: 15 April 2017
Abstract: An acceptable quality of electrical energy is seen today as an important component of
ecology. Several instruments for estimating the quality of electrical power have been elaborated.
Each supplier assures that the instrument meets the applicable standards and that the uncertainty
of the measurement results obtained using the instrument does not exceed the established levels.
The accuracy of the measurement results depends on a couple of things, e.g., the correctness of the
measurement algorithms implemented in the instrument and the quality of its calibration. In this
paper, the basic features of an “estimator/analyzer” (E/A) instrument, as well as the calibration
methods of the instrument, the verification of its measurement algorithms, and also the obtained
exemplary results, are shown. The proposal of the strategy of the reliable validation of embedded
measurement algorithms for the identification of parameters characterizing electrical power quality
in the power grid is discussed.
Keywords: electrical power quality estimation; measurement algorithm; measurement track
examination; validation procedures
1. Introduction
Ecology is the study of environmental systems, and is not only interested in the individual
components of nature, but also in how the parts interact. Industrial ecology is a discipline that has
recently been developed, particularly in Europe. Industrial ecology is an interdisciplinary framework
for designing and operating industrial systems as living systems interdependent from natural systems.
Its aim, among others, is to follow the energy use throughout the industrial processes, in order to
improve the efficiency of energy producing. Many ecological and environmental impacts are the side
products of electrical power generation, delivery, and consumer systems [1].
The electrical power system can be regarded as a large network system consisting of electrical
power generators, voltage transformers, connecting cables, and power consumers [2]. The environmental
impact of the production and consumption of electrical energy is becoming more significant because
modern society is using an increasing amount of electrical power [3]. This power is usually produced
at power plants that convert some other kind of energy into electrical power: conventional, nuclear,
and renewable energy sources (RES). Each system has pros and cons, but many of them may pose
a threat to the environment. Additionally, the common use of electronic power converters affects
the deterioration of electrical power quality. This lower quality of power places a greater burden on
the environment [4].
Energies 2017, 10, 536; doi:10.3390/en10040536
www.mdpi.com/journal/energies
Energies 2017, 10, 536
2 of 16
The electrical power quality is a set of parameters describing the properties of the process of
provision energy to the user under operating conditions, determining the continuity of the supply
(power outage), and characterizing the supply voltage (its value and frequency, shape of waveform,
and asymmetry in the multi-phase systems) [5]. The voltage waveforms in the electrical network,
often thought of as composed of pure sinusoidal signals, can contain a wide variety of disturbances [6].
The main phenomena related to power quality include:
•
•
•
•
•
Harmonic and other derogations from the expected frequency of the alternating voltage,
Voltage fluctuations, particularly those causing flickering lights,
Voltage dips and short interruptions,
Asymmetry within the three-phase system,
Surges, having the characteristics of high frequency.
A set of standards has been introduced to cover this field. They define the types and sizes of
disturbance, and also the tolerance of various types of equipment to the possible disturbances that
may be encountered. The principal standards in this field are the series of IEC 61000 [7], norms EN
50160 [8] and IEEE 1159 [9].
Electrical energy is a product, and like all other products, should meet the relevant quality
requirements [10]. A proper operation of electrical equipment requires that the supply voltage only
varies within a specified range around the nominal value [11,12]. The presence of disturbances in
the electrical network (deterioration of the electrical power quality) is associated with a decrease
in the energy production and conversion efficiency [13]. On the other hand, disturbances in the
supply voltage mean that the components consuming electrical energy may reach an excessive usage
level. It is important to note that the production technologies of the equipment for the generation of
electrical energy from renewable sources are not fully “green”. The RES devices also introduce some
disturbances to the electrical network.
To minimize the influence of these disturbances, some protectors or noise filters are used [4,14].
A proper choice of equipment can be achieved after the determination of parameters of electrical
disturbances as a result of the assessment of the electrical power quality.
Commercially available instruments for electrical power quality estimations also measure other
parameters characterizing the power quality. For example, the:
•
•
•
•
•
•
Effective value of current and voltage,
Active, reactive, and apparent power and energy,
Frequency,
Power factors: cosφ, tgφ, or λ,
rms value of given harmonics for voltage and current,
Total harmonic distortion (THD) coefficients for voltage and current.
In [15], the results of an interesting experiment relating to the designation of power quality
parameters were shown. Ten instruments used in the experiment (Table 1), owned by various
institutions, measured the same set of parameters describing the electrical power quality. “Own design”
means the prototype instrument, developed in the scientific institution and taking part in the
experiment. The measurements included the network frequency, voltage rms values, total harmonic
distortion, rms values of the dominant harmonics, and also the voltage fluctuation parameters:
short and long term flicker severity. The measurements were carried out at bus-bars of the medium
voltage 6 kV, by means of voltage transformers.
Energies 2017, 10, 536
3 of 16
Table 1. List of instruments used in the measurements [15].
Energies 2017, 10, 536
No.
Type of Instrument
Owner
3 of 16
1
Panensa Table
MEF 1. List of instruments
Technical University
Radom, Radom,
used in the of
measurements
[15]. Poland
2
Memobox 686
Technical University of Radom, Radom, Poland
Type
of Instrument
3 No. Own
design
Technical University ofOwner
Lodz, Lodz, Poland
Panensa
MEF
Technical
UniversityInstitute,
of Radom,Gdansk,
Radom, Poland
4 1 Own
design
Electrotechnical
Poland
2
Memobox
686
Technical
University
of
Radom,
Radom,
Poland
5
Memobox 800
Semicon, Warszawa, Poland
3
Own design
Technical University of Lodz, Lodz, Poland
Power recorder 1650 University of Mining and Metallurgy, Krakow, Poland
6
4
Own design
Electrotechnical Institute, Gdansk, Poland
7
Own design
University of Mining and Metallurgy, Krakow, Poland
5
Memobox 800
Semicon, Warszawa, Poland
8 6 Own
design
Institute of Power Engineering, Katowice, Poland
Power recorder 1650 University of Mining and Metallurgy, Krakow, Poland
9 7 Memobox
686
Energopomiar-Elektryka,
Gliwice,
Poland
Own design
University
of Mining and Metallurgy,
Krakow,
Poland
10 8 Own
GdyniaofMaritime
Academy,
Gdynia,
Poland
Owndesign
design
Institute
Power Engineering,
Katowice,
Poland
9
10
Memobox 686
Own design
Energopomiar-Elektryka, Gliwice, Poland
Gdynia Maritime Academy, Gdynia, Poland
A similar set of ten instruments was used for the following measurements [16]. A schematic
diagram of
measurement
site is shown
Figure
1. The
measurement
points[16].
were
on the
A the
similar
set of ten instruments
wasinused
for the
following
measurements
A located
schematic
secondary
side
of
a
supply
transformer,
at
the
substation
denoted
as
GSTS-30
kV
(phase-to-phase
diagram of the measurement site is shown in Figure 1. The measurement points were located on the
secondary
side=of
supply
the substation
denoted
as GSTS-30
(phase-to-phase
voltage
level Urms
30akV).
Tentransformer,
measuring at
instruments
were
connected
via thekV
voltage
transformers.
voltage
level Urms =used
30 kV).inTen
measuring
instruments
connected
via the voltage
transformers.
All
instruments
both
experiments
werewere
declared
as complying
with
the terms of
All instruments
used in standards
both experiments
were declared
as complying
with
of the and
the respective
IEC 61000-4-15
[17], according
to the
statements
of the
theterms
designers
respective IEC 61000-4-15 standards [17], according to the statements of the designers and
manufacturers. Each of the two measurements was carried out by independent experiments. They were
manufacturers. Each of the two measurements was carried out by independent experiments. They
held at the same point of the supply network at the same time, for each experiment. Although the
were held at the same point of the supply network at the same time, for each experiment. Although
instruments
measured the same voltage signals at the same time, their indications differed significantly,
the instruments measured the same voltage signals at the same time, their indications differed
independently
forindependently
each experiment.
significantly,
for each experiment.
Figure 1. Schematic diagram of the measurement site in the electric steelworks power supply
system [16].
Figure 1. Schematic diagram of the measurement site in the electric steelworks power supply system [16].
In [18],
the analysis
of possible
reasonsfor
forthe
the different
different indications
of of
instruments
observed
in in
In [18],
the analysis
of possible
reasons
indications
instruments
observed
the aforementioned
experimentsisispresented.
presented. As
error,
the authors
of the of
paper
the
the aforementioned
experiments
Asa source
a source
error,
the authors
the indicate
paper indicate
human factor
factor (erroneous
(erroneous set-up
parameters)
or design
drawbacks
of the
the human
set-upofofthe
theinstrument’s
instrument’s
parameters)
or design
drawbacks
of the
manufacturer’s equipment. After an estimation of the relations between the results obtained during
manufacturer’s equipment. After an estimation of the relations between the results obtained during
the measurements, they formulate the simplified linear formulas showing the dependencies between
the measurements, they formulate the simplified linear formulas showing the dependencies between
the obtained results and bring the result sets from all of the instruments to a common course, within
the obtained
results and bring the result sets from all of the instruments to a common course, within the
the accepted range of deviations. They concluded that the reason for the deviations in the
accepted
range
of deviations.
They
concluded
reason
for the deviations
in the
measurements
measurements
results could
have
originatedthat
in the
variable
amplification
and could
be due
to the
results
could
have
originated
in
variable
amplification
and
could
be
due
to
the
constant
offset of the
constant offset of the measurement characteristics of the individual instruments.
measurement characteristics of the individual instruments.
Energies 2017, 10, 536
4 of 16
The description of effects related to the deviations of the characteristics of the measurement
channel (i.e., lost performance or improper calibration) of various devices, causing the deterioration of
the operation performance, can be found, e.g., in [19,20].
The vital issue for millions of customers, connected with the calibration quality of energy meters,
is presented in [21]. The paper concerns research focused on the accurate assessment of the commonly
used static energy meters. In some specific load conditions, their indications can differ by more
than 580% when compared to the conventional (electromechanical) energy meters. The research was
performed in accordance with the relevant standards of test setups. The research included the energy
meters equipped with a variety of current sensors. However, the question, “What is the cause of an
error in the processing characteristics of the energy meters?”, remains unanswered. In the context of
this investigation, the necessity to measure the quality of the electrical energy is becoming increasingly
important, because the number of non-linear and variable-energy loads connected to the electrical
network is constantly growing.
The descriptions of the methods and systems for electrical power quality assessment, in their
various aspects, can be found, e.g., in [22–27].
In [28–30], issues concerning the configuration and operation of the measurement track of the
“estimator/analyzer” (E/A) instrument for the designation of electrical power quality coefficients,
especially for ship power network estimations, were considered. Additionally, considerations
connected with investigating the characteristics of the E/A measurement tracks are described in [31–33].
Further detailed analyses are based on the case study concerning the designed and constructed
instrument, in the area under consideration.
The author of the present paper uses the term “calibration”, which can be defined as: “a test during
which known values of measures are applied to the device input and corresponding output readings
are recorded under specified conditions”. A related notion of “validation”, which in engineering means
confirming that a product or service meets the needs of its users, is used in this paper. An overview of
the selected validation methods can be found in [34].
For the user of a specific instrument, it is difficult to assess the correctness of the instruments’
indications during the measurement operation. They can only relay on the declarations of the
instrument manufacturer.
This work is focused on the method of reliable validation of the E/A instrument characteristics.
The main goal of this paper is to show the solution used in the developed instrument. The paper deals
with an additional configuration, necessary to implement the validation procedures of the algorithms
applied in the E/A instrument.
In Section 2, the measurement functions, as well as the basic measurement algorithms,
implemented in the instrument are in short given. Section 3 describes the calibration procedures
applied during the examination of the instrument. The constraints of the procedures used are
mentioned. Section 4 presents the proposal of a reliable method for the validation of measurement
algorithms implemented in the instrument measurement functions as embedded processor software.
This work is summarized in Section 5.
2. Measurement Functions of the E/A
In this section, the hardware configuration of the E/A and its measurement functions are in short
presented. The configuration of the instrument is shown in Figure 2.
The voltages from the electrical network (3- or 4-wires) and phase currents from generating sets
(up to three) are delivered to the input terminals of the signal conditioning units. The configuration
of the signal conditioning units applied in the voltage channels is shown in Figure 3. The voltage
dividers (VD) adapt the level of the network voltages to the value acceptable in the next parts of
the measurement channels. The galvanic separation (GS) block ensures a safe operation for other
circuits of the E/A. The low-pass filter is switched between 10 and 100 kHz, depending on the actual
measurement function. The voltage follower (VF) provides the adequate output resistance of this part
Energies 2017, 10, 536
5 of 16
of the measurement channel. The output signal is fed to the input of the analog-to-digital converter
5 of 16
(ADC). The network currents are measured using the current probe type LFR 1/15 (Power Electronic
Measurements Ltd.,
Ltd., Nottingham,
Nottingham, UK)
UK) (based
(based on
on the
the Rogowski
Rogowski coil).
coil). The
The configuration
configuration of
of the
the signal
signal
Measurements
conditioning
units
applied
in
the
current
channels
is
similar
to
the
signal
conditioning
unit
used
in the
conditioning units applied in the current channels is similar to the signal conditioning unit used
in
voltage
channels,
but
in
the
current
channels,
the
VD
blocks
are
missing.
The
current
probe
signals
are
the voltage channels, but in the current channels, the VD blocks are missing. The current probe signals
directly
fed fed
to the
inputs
of the
are
directly
to the
inputs
of GS
the blocks.
GS blocks.
The
designation
of
coefficients
describing the
the electrical
electrical power
power quality
quality is
is carried
carried out
out using
using aa
The designation of coefficients describing
ADSP-TS201 TigerSHARC
TigerSHARC (Analog
(Analog Devices,
Devices, Norwood,
Norwood, MA,
MA, USA)
USA) digital
digital signal
signal processor
processor (DSP)
(DSP)
ADSP-TS201
embedded
on
the
DSP
Subsystem
Module
(DSM;
STSL
stands
for
“Single
TigerSHARC
with
Logic”)
embedded on the DSP Subsystem Module (DSM; STSL stands for “Single TigerSHARC with Logic”)
(Kaztek Systems,
Acton, MA,
MA, USA)
The configuration
configuration of
of the
the DSM
DSM STSL
STSL module
(Kaztek
Systems, Acton,
USA) [35]
[35] (Figure
(Figure 2).
2). The
module
resources is
is shown
shown in
in Figure
Figure 4.
There is
is also
also aa field-programmable
field-programmable gate
gate array
array (FPGA)
(FPGA) Xilinx
Xilinx Spartan-3
Spartan-3
resources
4. There
(Xilinx, Inc.,
Inc., San
San Jose,
Jose, CA,
CA, USA)
USA) which
which is
is used
used for
for the
the data
data exchange
exchange between
between the
the DSP
DSP and
and other
other systems
systems
(Xilinx,
of
the
instrument.
For
the
communication
of
DSM
devices
with
other
E/A
circuits,
user-defined
FPGA
of the instrument. For the communication of DSM devices with other E/A circuits, user-defined FPGA
input/output (I/O)
(I/O)lines
lineswere
wereused.
used. The
Theuser
userinterface
interface(Figure
(Figure2),
2), designed
designed for
for the
the control
control and
and
input/output
selection
of
individual
measurement
functions
of
the
instrument,
is
supported
by
the
general
purpose
selection of individual measurement functions of the instrument, is supported by the general purpose
processor (GPP)
(GPP) LPC3250
LPC3250 (ARM926EJ-S
(ARM926EJ-S core)
core) (NXP
(NXP Semiconductors
Semiconductors Netherlands
Netherlands B.V.,
B.V., Eindhoven,
Eindhoven,
processor
The
Netherlands).
This
processor
also
controls
the
indication
and
registration
of
the
measurement
data.
The Netherlands). This processor also controls the indication and registration of the measurement data.
Energies 2017, 10, 536
Bus bars
of Main
switchboard
Estimator/analyzer of power quality
L1
L2
L3
N
Signal
conditioning
unit
I G11
I G13
I G21
I G23
DSM STSL Module
ADC
AD7656
FPGA
Spartan 3
Xilinx
DSP
TigerSHARC
ADC
Signal
conditioning
unit
I G31
I G33
AD7656
RAM
Flash
memory
GPP
USB
Ethernet
keyboard
LCD
display
Figure 2. The hardware structure of instrument [30]. ADC: analog-to-digital converter; DSM STSL:
Figure 2. The hardware structure of instrument [30]. ADC: analog-to-digital converter; DSM
DSP subsystem module; FPGA: field-programmable gate array; DSP: digital signal processor; RAM:
STSL: DSP subsystem module; FPGA: field-programmable gate array; DSP: digital signal processor;
random access memory; GPP: general purpose processor; USB: universal serial bus; LCD: liquid
RAM: random access memory; GPP: general purpose processor; USB: universal serial bus; LCD:
crystal display.
liquid crystal display.
In Figure 5, the main operations carried out in the measurement track of the instrument,
In Figure
the main
operations
carried out
in differ
the measurement
trackrate
of the
connected
with5,data
processing,
are shown.
They
in the sampling
of instrument,
the electricalconnected
network
with
data
processing,
are
shown.
They
differ
in
the
sampling
rate
of
the
electrical
network
signals,
signals, as well as in a set of measured quantities and therefore in the set of designated
coefficients.
as well
as in a set
of measured
quantities
and therefore sampled
in the set of
The
electrical
network
signals
are simultaneously
in designated
all voltagecoefficients.
channels atThe
theelectrical
rate of
network
signals
are
simultaneously
sampled
in
all
voltage
channels
at
the
rate
of
210
kS/s
all
210 kS/s or in all of the voltage and current channels at the rate of 25 kS/s, dependingoroninthe
of
the
voltage
and
current
channels
at
the
rate
of
25
kS/s,
depending
on
the
performed
measuring
performed measuring function of the E/A.
function of the E/A.
Energies 2017, 10, 536
6 of 16
Energies 2017, 10, 536
Energies 2017, 10, 536
6 of 16
6 of 16
Energies 2017, 10, 536
6 of 16
Figure 3. The configuration of the signal conditioning units applied in the voltage channels.
Figure
3. The
configuration
ofofthe
unitsapplied
applied
voltage
channels.
Figure
3. The
configuration
thesignal
signal conditioning
conditioning units
in in
thethe
voltage
channels.
Figure 3. The configuration of the signal conditioning units applied in the voltage channels.
Figure 4. The block diagram of the DSM STSL module [35].
Figure 4. The block diagram of the DSM STSL module [35].
Figure 4. The block diagram of the DSM STSL module [35].
Figure 4. The block diagram of the DSM STSL module [35].
Sampling at 210 kS/s
Sampling at 210 kS/s
Operating options
Operating options
Operating options
Sampling at 210 kS/s
Estimator
Estimator
Estimator
Designation and
Registration
Designation
Registration
Designation
and
registration
ofand Registration
of raw
registration
raw
registration
of of
ofofraw
selected
samples
selected
samples
selected
samples
coefficients
coefficients
coefficients
of electrical
of electrical
of
electrical
power
quality
power
quality
power quality
Sampling at 25 kS/s
Sampling at 25 kS/s
Sampling at 25 kS/s
Analyzer
Analyzer
Analyzer
Designation and
Registration
Designation
and
Designation
and
Registration
registration
of Registration
of raw
registration
of raw
registration
of raw
selectedofof
samples
selected
samples
selected
samples
coefficients
coefficients
coefficients
of
electrical
of electrical
of electrical
power
quality
power
quality
power quality
Figure
Mainoperations
operations performed
performed ininthe
instrument.
Figure
5. 5.
Main
the
instrument.
Figure
5. 5.Main
performedininthe
the
instrument.
Figure
Mainoperations
operations performed
instrument.
The
algorithms
of
the
digital
data
processing
(DDP)
used
to
determine
the the
valuevalue
of theof the
The algorithms of the digital data processing (DDP) used to
determine
individual
parameters
of
the
electrical
power
quality
are
based
on
the
common
application
theof the
The algorithms
ofof
the
data
processing
(DDP)
used
the value
The
algorithms
of the
digital
data processing
(DDP)
used
to determine
the value
ofofthe
individual
individual
parameters
thedigital
electrical
power
quality
are
based
onto
thedetermine
common
application
of the
discrete
wavelet
transform
(DWT),
fast
Fourier
transform
(FFT),
discrete
Fourier
transform
(DFT),
orof the
individual
parameters
of
the
electrical
power
quality
are
based
on
the
common
application
discrete
wavelet
transform
(DWT),
fast
Fourier
transform
(FFT),
discrete
Fourier
transform
(DFT),
or
parameterschirp
of the
electrical(CZT),
power
quality are
on the common
application
of the
the wavelet
discrete wavelet
Z-transform
performed
in abased
complementary
way [30,36,37].
Namely,
discrete
wavelet
transform
(DWT),
fast Fourier
transform (FFT),
discrete
Fourier
transform
(DFT),
or
chirp
Z-transform
(CZT),
performed
in
a
complementary
way
[30,36,37].
Namely,
the
wavelet
transform coefficients
(DWT), fast
Fourier
transform
(FFT), discrete
Fourier
transform
(DFT),
chirp
Z-transform
after
respective
levels of wavelet
decomposition
are used
as the input
dataorfor
various
chirp
Z-transform
(CZT), performed
in a complementary
wayused
[30,36,37].
Namely,
wavelet
coefficients
after
levels ofthe
wavelet
decomposition
as the
input
datathe
for
various
procedures
to calculate
rms[30,36,37].
values
and coefficients
in
thewavelet
Fourier
series,
in combination
(CZT),
performed
inemployed
arespective
complementary
way
Namely,are
the
coefficients
after
respective
coefficients after respective levels of wavelet decomposition are used as the input data for various
procedures employed to calculate the rms values and coefficients in the Fourier series, in combination
levelsprocedures
of wavelet
decomposition
as the
input
data for
various
employed to
employed
to calculateare
theused
rms values
and
coefficients
in the
Fourierprocedures
series, in combination
calculate the rms values and coefficients in the Fourier series, in combination with the transient
monitoring algorithm. The main components of the measurement algorithms are shown in Figure 6.
These operations are carried out in the DSP (Figure 2).
Energies 2017, 10, 536
7 of 16
Energies 2017,with
10, 536
the transient monitoring algorithm. The main components of the measurement algorithms are
shown in Figure 6. These operations are carried out in the DSP (Figure 2).
Energies 2017, 10, 536
7 of 16
7 of 16
Components above
with the transient monitoring algorithm. The main components
of the measurement algorithms are
50th harmonic
shown in Figure 6. These operations are carried out in the DSP (Figure 2).
Range up to 9 kHz
Rms value
calculation
Components above
50th harmonic
FFT
Range up to 9 kHz
Wavelet
decomposition Rms value
calculation
Transient detection
and assessment
Wavelet
decomposition
DFT/CZT
FFT
Spectrum up to
50th harmonic
DFT/CZT
Figure 6. The diagram of operations performed in the measurement algorithms [30].
Figure 6. The diagram of operations
performed
measurement
algorithms [30].
Transient
detection in the
Spectrum
up to
and assessment
3. Procedures for Assessing the Characteristics
of the E/A
50th harmonic
3. Procedures To
forevaluate
Assessing
the
Characteristics
ofofthe
E/A
measurement
the in
whole
measurement
track, the
E/A was tested
Figurethe
6. The
diagram
of characteristics
operations performed
the measurement
algorithms
[30].
in the laboratory as well as in the ship environment for several different test signals.
To evaluate
measurement
characteristics
theE/A
whole
measurement
track,
the
3. Procedures
for Assessing
the
Characteristics
oftrack
the
Thethe
configuration
of each
E/A
measurementof
can
be illustrated
in simplified
form,
as E/A
shownwas tested
in the laboratory
as main
in the
ship
foracquisition
several different
test
signals.
in Figureas
7. well
The two
parts
of environment
the track, the data
(DA) and the
DDP,
determine the
To evaluate the measurement characteristics of the whole measurement track, the E/A was tested
correctness of the
measurement
result. An evaluation
of the
measurementincharacteristics
of the as shown
The configuration
of
each
E/A
measurement
track
can
be
illustrated
simplified
form,
in the laboratory as well as in the ship environment for several different test signals.
designed instrument was made considering these two parts of the track.
in Figure 7. The
main parts
ofE/A
themeasurement
track, the track
datacan
acquisition
(DA)
and the
DDP,
determine the
Thetwo
configuration
of each
be illustrated
in simplified
form,
as shown
in Figure
The two main parts
of
the track,
the data acquisition
and the DDP, determine
the
correctness
of the7.measurement
result.
An evaluation
of the
measurement
characteristics
of the
Digital
Data(DA)
Processing
Data
Acquisition
correctness of the measurement result. An evaluation of the measurement characteristics of the
designed instrument was made considering these two parts of the track.
designed instrument was made
Signal considering these two parts of the track.
In
conditioning
ADC
Data Acquisition
unit
DSP
UI
Digital Data Processing
Figure 7. The
simplified block diagram of the digital measurement track.
Signal
In
conditioning
DSP
UI
ADC
unit in the DA part of the track is carried by the instantaneous values of
Measurement information
the analog signals, and the conversion of these signals is performed systemically (in analog track).
Figure 7. The simplified block diagram of the digital measurement track.
The course
of characteristics
of the DA
may
change, of
because
this part
of the measurement
Figure
7. The simplified
block
diagram
the digital
measurement
track. track is
susceptible to the influence of many external impacts. The calibration of the DA part is based on the
Measurement information in the DA part of the track is carried by the instantaneous values of
determination of its actual characteristics of the signal conversion.
the analog signals, and the conversion of these signals is performed systemically (in analog track).
In the DDP
part, the information
a form
digital
representation
thethe
signals
from the DA, values of
Measurement
information
in the DAinpart
ofofthe
track
is carriedofby
instantaneous
The course of characteristics of the DA may change, because this part of the measurement track is
after
they have
been sampled
and converted
in signals
the ADC,is
is performed
presented by the
stream of digital
the analogsusceptible
signals,
and
conversion
of these
(inondata.
analog
track).
to
the the
influence
of many external
impacts.
The calibration
of thesystemically
DA part is based
the
The processing of data in this part of the track is performed in the software. The characteristics of the
determination
of its actual
characteristics
of the
signal conversion.
The course
of
characteristics
of
the
DA
may
change,
because
this
part
of
the
measurement
track
is
DDP, describing the data processing in the software algorithms, are substantially unchanging. The
In
the
DDP
part,
the
information
in
a
form
of
digital
representation
of
the
signals
from
the
DA,
susceptiblevalidation
to the influence
many external
The
calibration
of the
DA part
is based on the
of the DDPof
characteristics
meansimpacts.
checking the
correctness
(as well
as further
correction)
after they have been sampled and converted in the ADC, is presented by the stream of digital data.
of theof
measurement
algorithms
performed
in this
part of
the track.
determination
its
actual
characteristics
of
the
signal
conversion.
The processing of data in this part of the track is performed in the software. The characteristics of the
In theDDP,
DDP
part, the
inina the
form
of digital
representation
of the
signals The
from the DA,
describing
theinformation
data processing
software
algorithms,
are substantially
unchanging.
3.1. The Calibration of DA Part of Measurement Channel
of the
DDP characteristics
means in
checking
the correctness
(as well
further
correction)
after they validation
have been
sampled
and converted
the ADC,
is presented
byasthe
stream
of digital data.
calibration of the DAperformed
part is easy
to perform.
It track.
relys on determining the relationship
of theThe
measurement
in this
of the
The processing
of data in algorithms
this part of the track
is part
performed
in the software. The characteristics of
between the output signal and the input signal (under stabilized conditions) over the range of the
the DDP, describing
thecalibration
data
processing
the
software
algorithms,
are substantially
unchanging.
3.1.
Thesignal.
Calibration
of
DA Part
of the
Measurement
Channel
input
The
of
inputinanalog
circuits, starting
from the voltage
dividers (for three
The validation
of
the
DDP
characteristics
correctness
well
as
further
voltage
channels)
and
the Rogowski means
coils (forchecking
six currentthe
channels)
to the (as
ADC’s
(Figure
2), wascorrection)
The calibration of the DA part is easy to perform. It relys on determining the relationship
performed using the pattern
sources of the alternating
current. In Figure 8, the exemplary
of the measurement
this(under
partvoltage
of
theand
track.
between the algorithms
output signal performed
and the inputin
signal
stabilized
conditions) over the range of the
characteristics obtained as a result of the calibration of one of the current probes used, together with
input signal. The calibration of the input analog circuits, starting from the voltage dividers (for three
a designation
of itsPart
trendofline,
are shown. Figure
9 contains the deviation of the trend line from the
3.1. The Calibration
of DA
Measurement
Channel
voltage channels)
and the
Rogowski coils (for
six current channels) to the ADC’s (Figure 2), was
probe characteristics. The absolute value of this deviation (∆approx) slightly exceeds the value of 0.8 A.
performed using the pattern sources of the alternating voltage and current. In Figure 8, the exemplary
The calibration
ofobtained
the DAaspart
is easy
perform.ofItone
relys
oncurrent
determining
the together
relationship
characteristics
a result
of thetocalibration
of the
probes used,
with between
designation
its input
trend line,
are shown.
9 contains
the deviation
of the
lineoffrom
the outputa signal
andofthe
signal
(underFigure
stabilized
conditions)
over
thetrend
range
thethe
input signal.
probe characteristics.
Theanalog
absolutecircuits,
value of this
deviation
(∆approx
) slightly
exceeds
the value(for
of 0.8
A. voltage
The calibration
of the input
starting
from
the
voltage
dividers
three
channels) and the Rogowski coils (for six current channels) to the ADC’s (Figure 2), was performed
using the pattern sources of the alternating voltage and current. In Figure 8, the exemplary characteristics
obtained as a result of the calibration of one of the current probes used, together with a designation of
its trend line, are shown. Figure 9 contains the deviation of the trend line from the probe characteristics.
The absolute value of this deviation (∆approx ) slightly exceeds the value of 0.8 A. Taking into account the
range of the current probe LFR 1/15 (peak current to 3000 A), it is an acceptable value in the application
under consideration.
Energies 2017, 10, 536
8 of 16
Energies 2017, 10, 536
8 of 16
Energies
2017, 10,
536account the range of the current probe LFR 1/15 (peak current to 3000 A), it is an8 of 16
Taking
into
Taking intovalue
account
range of under
the current
probe LFR 1/15 (peak current to 3000 A), it is an
acceptable
in thethe
application
consideration.
acceptable
value
Energies 2017,
10, in
536the application under consideration.
8 of 16
Taking into account the range of the current probe LFR 1/15 (peak current to 3000 A), it is an
acceptable value in the application under consideration.
Figure 8. The exemplary calibration report of current probe No. 1.
Figure 8. The exemplary calibration report of current probe No. 1.
Figure 8. The exemplary calibration report of current probe No. 1.
Figure 8. The exemplary calibration report of current probe No. 1.
Figure 9. The deviation between the trend line and the current probe characteristics No. 1.
Figure 9. The deviation between the trend line and the current probe characteristics No. 1.
Figure
9. The deviation
between
trend line
and
the current
probe characteristics
The
equations
of the trend
lines, the
obtained
in the
calibration
procedures
of the DA No.
part1.of the
The equations
the trend
lines, used
obtained
in the
procedures
of digital
the DAdata
part in
of the
measurement
track of
(Figure
8), were
in the
DDPcalibration
part for the
scaling of
measurement
track
(Figure
8),
were
used
in
the
DDP
part
for
the
scaling
of
digital
data
in
theof the
Figure
9.
The
deviation
between
the
trend
line
and
the
current
probe
characteristics
No.
1.
corresponding
channels.
The equations of the trend lines, obtained in the calibration procedures of the DA part
corresponding
channels.
Current
research
on
the
configuration
of
the
E/A
is
carried
out
in
a
way
which
leads
to
the
measurement track (Figure 8), were used in the DDP part for the scaling of digital data in the
The
equations
of the
trend
lines, obtained
in
theis calibration
procedures
ofthe
themeasurement
DA part
of the
Current
research
on
theto
configuration
theself-calibration
E/A
carriedofout
in
a part
way of
which
leads
to the
provision
ofchannels.
its
input circuits
the system
forofthe
the
DA
corresponding
measurement
track
(Figure
8), DA
were
used
in self-calibration
the DDP
part
for
theDA
scaling
data
provision
its input
circuits
the
system
forobtained
the
of the
part
ofofthe
measurement
track.
Theof
DDP
processing
oftothe
data,
for the
standard
test
signal
Sdigital
t, allows
oneinto the
Current
researchchannels.
on the configuration of the E/A is carried out in a way which leads to the
corresponding
track.
Thethe
DDP
processing
of the DA and
data,correct
obtained
for the
signal
St, allows one
to
designate
actual
DA characteristics
the DA
datastandard
obtainedtest
for the
measurement
signal
provision
of
its
input
circuits
to the
the configuration
system
for
the
self-calibration
of out
thefor
DA
of
the measurement
Current
on
of along
the DA
E/A
is the
carried
in
aofpart
way
which
leads
to the
designate
the
DA characteristics
andresult,
correct
data
obtained
the
measurement
signal
S
m, as well
asactual
toresearch
determine
the corrected
with
estimation
its uncertainty.
The
track.detailed
DDP
processing
of
theto
DA
obtained
for
thethe
standard
signal
S
,measurement
allows
one to
input
circuits
the data,
system
for
the
self-calibration
of the test
DA
of the
SmThe
, provision
as well
asof
toitsdetermine
the
corrected
result,
along
with
ofpart
its[38].
uncertainty.
The
description
of the
procedures
and
the
calibration
circuit
isestimation
shown,
e.g.,
in
Ittis
assumed
designate
theadditional
actual
DAoperations
characteristics
correct
the DA
data
obtained
forenable
thein
measurement
signal
track.
The DDP
processing
of theand
DA
data,
obtained
for
the
test
signal
SItt, is
allows
one toSm ,
detailed
description
of
the procedures
and
the
circuit
isstandard
shown,
e.g.,
[38].determination
assumed
that
the
performed
in
thecalibration
DDP
(shown
in
Figure
10)
the
designate
the actual
characteristics
and
the
DA
data
obtained
foruncertainty.
thethe
measurement
signal
as well
as
to
determine
theDA
corrected
result,in
along
with
the
estimation
of its
The detailed
that
the
additional
performed
thecorrect
DDP
(shown
in
Figure
10)
enable
determination
of
the
uncertainty
ofoperations
results.
S
m
,
as
well
as
to
determine
the
corrected
result,
along
with
the
estimation
of
its
uncertainty.
Thethe
of
the
uncertainty
of
the
results.
description of the procedures and the calibration circuit is shown, e.g., in [38]. It is assumed that
Sm of the procedures and the calibration circuit is shown, e.g., in [38]. It is assumed
detailed
description
additional operations performed
in the DDP (shown
inDDP
Figure 10) enable
the determination of the
Corrected
Data
DA
Sm
that the additional
performed in
the DDP (shown
in Figure
10) enable the determination
result
correction
Corrected
DDP
Data
uncertainty
of the results.operations
DA
St
of the uncertainty
of the results.
result
correction
St
Real
Uncertainty
Uncertainty
characteristics
estimation
of result
Test Sm
Real
Uncertainty
Uncertainty
Corrected
DDP
Data
DA
signal
characteristics
estimation
of result
Test
result
correction
signal
St
Figure 10. The idea of the implementation of the auto-calibration procedures in the E/A operations.
Real
Uncertainty
Uncertainty
Figure 10. The idea
of the auto-calibration
E/A operations.
characteristics
estimationprocedures in
of the
result
Testof the implementation
signal
Figure 10. The idea of the implementation of the auto-calibration procedures in the E/A operations.
Figure 10. The idea of the implementation of the auto-calibration procedures in the E/A operations.
Energies 2017, 10, 536
9 of 16
Energies
2017, 10, 536of Characreristics of Measurement Channel
3.2. The
Designation
9 of 16
The
proper
calibration
of the instrument’s
input Channel
circuits seems to be a basic requirement. Probably,
3.2. The
Designation
of Characreristics
of Measurement
for most The
of the
effects
described
in
[15,16],
it
can
be circuits
assigned
to the
change
the conversion
proper calibration of the instrument’s input
seems
to be
a basicofrequirement.
characteristics
of most
the DA
section
the devices
useditin
the
the accuracy
Probably, for
of the
effectsof
described
in [15,16],
can
beexperiment.
assigned to theIndependently,
change of the conversion
of thecharacteristics
designationofofthe
a particular
quality
coefficient
also depends
on the correctness
of the
DA section power
of the devices
used
in the experiment.
Independently,
the accuracy
algorithms
applied inofthe
DDP part
of the
measurement
channels.
goal
of the performed
of the designation
a particular
power
quality
coefficient also
depends The
on the
correctness
of the
algorithms
applied
in the
part of the
measurement
Theingoal
the performed
experiments
is, most
of all,
theDDP
examination
of the
algorithmschannels.
embedded
the of
software
executed in
experiments
is, most
of all,
the examination
of the
algorithms
embedded inconfigurations
the software executed
in in
the DSP.
The research
was
carried
out using the
four
basic arrangement
presented
the11.
DSP.
research was
carried out
the fourin
basic
arrangement
configurations
in
Figure
TheThe
estimation
procedures
areusing
performed
applications
running
on a PC.presented
The distinction
Figure 11. The estimation procedures are performed in applications running on a PC. The distinction
of the DA and DDP parts in the measurement channel allows one to apply various test signals and
of the DA and DDP parts in the measurement channel allows one to apply various test signals and
calibration techniques for the estimation of the measurement features of the E/A instrument.
calibration techniques for the estimation of the measurement features of the E/A instrument.
(a)
(b)
(c)
(d)
Figure 11. Experimental setups for the verification of measurement algorithms: (a) The preliminary
Figure 11. Experimental setups for the verification of measurement algorithms: (a) The preliminary
assessment of algorithms in the development kit; (b) The estimation of algorithms with the use of
assessment of algorithms in the development kit; (b) The estimation of algorithms with the use of
pattern analog signals; (c) The estimation with the use of the raw samples of signal and coefficients
pattern
analog signals; (c) The estimation with the use of the raw samples of signal and coefficients
designated in the E/A; and (d) The estimation using the reference instrument.
designated in the E/A; and (d) The estimation using the reference instrument.
The experimental layout in Figure 11a figuratively shows the preliminary tests of the E/A
instrument
algorithms.
The software
designing,
well as the initial
verification
of the measurement
The
experimental
layout
in Figure
11a as
figuratively
shows
the preliminary
tests of the
algorithms
implemented
in
software
of
the
E/A,
were
carried
out
using
the
evaluation
board
EVALE/A instrument algorithms. The software designing, as well as the initial verification
of the
TS201S-EZKIT Analog Devices (Analog Devices, Norwood, MA, USA) (Figure 12) [39], equipped
measurement algorithms implemented in software of the E/A, were carried out using the evaluation
with two TigerSHARC processors, the same as the DSP used in the E/A instrument (Figure 2). Due
board EVAL-TS201S-EZKIT Analog Devices (Analog Devices, Norwood, MA, USA) (Figure 12) [39],
equipped with two TigerSHARC processors, the same as the DSP used in the E/A instrument (Figure 2).
Energies 2017, 10, 536
10 of 16
Due to the different configurations of the peripheral circuits of the DSP processors on the evaluation
Energies 2017, 10, 536
10 of 16
board and in the E/A instrument, the verification only included the selected portions of the
measurement
functions.
However,
the
verification
ofofthe
foronmeasuring
the
individual
to the different
configurations
of the
peripheral
circuits
the algorithms
DSP processors
the evaluation
board
and
in the E/A
instrument,
thefully
verification
thestandard
selected portions
of the measurement
factors of
energy
quality
has been
carriedonly
out included
using the
sets of digital
data. These studies
functions.to
However,
verification of
algorithms
for measuring
thedata
individual
of energy
were important
start thetheinstrument,
inthe
spite
of a limited
set of test
whichfactors
cannot
cover various
quality
has
been
fully
carried
out
using
the
standard
sets
of
digital
data.
These
studies
were
important
measurement conditions. As a result, the software can be embedded in the DSP on the E/A instrument.
to start the instrument, in spite of a limited set of test data which cannot cover various measurement
Figure 11b shows the calibration setup using the standard analog test signals. Just as in the
conditions. As a result, the software can be embedded in the DSP on the E/A instrument.
previous layout,
of testing
signals issetup
limited,
calibration
to in
both
Figurethe
11bset
shows
the calibration
usingbut
thethe
standard
analogprocedure
test signals.refers
Just as
the parts of
the channel:
thelayout,
DA and
DDP.
As signals
a standard
signal
the 3-phase
generator
type 6590
previous
thethe
set of
testing
is limited,
butsource,
the calibration
procedure
refers toChroma
both parts
of the channel:
the DASource
and the was
DDP.used
As a standard
signal
source,
thesource
3-phasewas
generator
Chroma
type others,
Programmable
AC Power
[40]. The
same
signal
also used,
among
6590 Programmable
AC Power
used [40]. The
same in
signal
source
was
also
used, among results
in the setup
shown in Figure
11b,c.Source
In thewas
experimental
setup
Figure
11b,
the
measurement
others, in the setup shown in Figure 11b,c. In the experimental setup in Figure 11b, the measurement
from the E/A are assessed in a PC-based estimation system, taking into account the features of the test
results from the E/A are assessed in a PC-based estimation system, taking into account the features
signal from
the Chroma generator.
of the test signal from the Chroma generator.
Figure 12. The block diagram of evaluation board EVAL-TS201S-EZKIT [39]. CPLD: complex
Figure programmable
12.
The block
diagram
evaluation converter;
board EVAL-TS201S-EZKIT
[39].
CPLD:
logic device;
DAC: of
digital-to-analog
ADC: analog-to-digital converter;
complex
programmable
logic device;
DAC: digital-to-analog
converter;
analog-to-digital
JTAG:
join test action group;
LEDs: light-emitting
diodes; PBs: push buttons;
PLL: ADC:
phase-locked
loop;
FLAGs:
processor’s
status
register;
IRQs: interrupt
SDRAM:diodes;
synchronous
converter;
JTAG:
join test
action
group;
LEDs: requests;
light-emitting
PBs:dynamic
push random
buttons; PLL:
access memory.
phase-locked
loop; FLAGs: processor’s status register; IRQs: interrupt requests; SDRAM:
synchronous dynamic random access memory.
Another way of verifying the measurement characteristics of the E/A was achieved using the
registered raw samples of the signals. In the experimental setup shown in Figure 11c, for the
stationary
test of
signals,
the digital
of the
raw samples ofof
these
Another
way
verifying
therepresentations
measurement
characteristics
thesignals
E/A were
wasregistered.
achieved using
For the same
signals,of
after
of instrument operation
(Figure 5),
particular
the registered
rawtest
samples
thechanging
signals.theInoption
the experimental
setup shown
in the
Figure
11c, for the
coefficients are registered. In the estimation arrangement, supported by the PC computer with
stationary test signals, the digital representations of the raw samples of these signals were registered.
running software package (independently tested using pattern data), the registered raw samples are
For the processed
same testand
signals,
after changing the option of instrument operation (Figure 5), the particular
then the calculated coefficients are compared with the coefficients obtained from the
coefficients
are registered.
In the estimation arrangement, supported by the PC computer with running
examined
E/A instrument.
software package
tested using
data), thetriangle
registered
raw
samples are
processed
Table 2 (independently
contains the measurement
resultspattern
of the three-phase
signal
(experimental
setup
as the
in Figure
11b), generated
usingare
a Chroma
type with
6590 Programmable
ACobtained
Power Source
and then
calculated
coefficients
compared
the coefficients
from(Chroma
the examined
ATE Inc., Taipei Hsien, Taiwan) (declared values of Urms, U1 and THD for standard triangle signal
E/A instrument.
DST017 from its waveforms library) [40], combined with the results obtained from the E/A instrument
Table 2 contains the measurement results of the three-phase triangle signal (experimental setup as
and as a result of the raw samples processing in the PC system (experimental setup as in Figure 11c).
in Figure 11b), generated using a Chroma type 6590 Programmable AC Power Source (Chroma ATE
Inc., Taipei Hsien, Taiwan) (declared values of Urms , U1 and THD for standard triangle signal DST017
from its waveforms library) [40], combined with the results obtained from the E/A instrument and as
a result of the raw samples processing in the PC system (experimental setup as in Figure 11c).
Energies 2017, 10, 536
11 of 16
Table 2. The results obtained from the E/A and in the PC system for the standard signal DST017 from
Energies
2017, 10,
536
11 of 16
the
Chroma
Programmable
AC Power Source.
Table 2. The results obtained from the E/A and in the PC system for the standard signal DST017 from
Device
Size
Unit
L1
L2
L3
the Chroma Programmable AC Power Source.
Device
E/A
E/A
PC
PC
DST017
DST017
Urms
USize
1
Urms
THD
U1
Urms
THD
UU
1 rms
THD
U1
THD
Urms
UU
1 rms
U1
THD
THD
(V)
220.75
220.56
219.70
Unit
(V)
(V)
(%)
(V)
(V)
(%)
(V)
(V)
(%)
(V)
(%)
(V)
(V)
(V)
(V)
(%)
(%)
L1
219.13
220.75
12.00
219.13
220.80
12.00
219.23
220.80
12.18
219.23
12.18
220.00
220.00
218.42
218.42
12.04
12.04
L2
218.93
220.56
10.24
218.93
220.020
10.24
219.05
220.020
12.23
219.05
12.23
220.00
220.00
218.42
218.42
12.04
12.04
L3
218.10
219.70
11.97
218.10
220.24
11.97
218.68
220.24
12.14
218.68
12.14
220.00
220.00
218.42
218.42
12.04
12.04
The experimental
layout
shown
(as well
wellasasininFigure
Figure
11c)
is free
from
the signal
The experimental
layout
shownininFigure
Figure 11d
11d (as
11c)
is free
from
the signal
limitations.
TheThe
testtest
signals
were
same time,
time,totothe
theinputs
inputs
examined
limitations.
signals
wereprovided,
provided, at
at the
the same
of of
thethe
examined
E/A E/A
instrument
and to
thetoreference
instrument.
The reference
instrument
was based
on theon
PCIthe
eXtensions
instrument
and
the reference
instrument.
The reference
instrument
was based
PCI
eXtensions
for
Instrumentation
(PXI)
platform
(National
Instruments
Corporation,
Austin,
TX,
USA)
for Instrumentation (PXI) platform (National Instruments Corporation, Austin, TX, USA) equipped
with suitable and
peripheries
and the
virtual instrument
designed
for measuring
the power
with equipped
suitable peripheries
the virtual
instrument
designed
for measuring
the power
quality
quality
parameters
(PCI
stands
for
Peripheral
Component
Interconnect—personal
computer
bus). 13,
parameters (PCI stands for Peripheral Component Interconnect—personal computer bus). In Figure
In Figure 13, the exemplary results of calibration of the phase voltage channels are presented.
the exemplary results of calibration of the phase voltage channels are presented. The differences
The differences between the rms values, measured by the tested instrument and also the PXI system,
between the rms values, measured by the tested instrument and also the PXI system, are shown.
are shown.
Figure 13. The calibration results of the voltage measurement channel of the instrument.
Figure 13. The calibration results of the voltage measurement channel of the instrument.
Table 3 contains the absolute errors ∆ representing the deviations between the rms values of the
phase voltages,
measured
by the E/A
instrument
and the PXI
obtained
during the
the calibration
Table
3 contains
the absolute
errors
∆ representing
thesystem,
deviations
between
rms values of
of the instrument
for differentby
frequencies
the network voltage,
wellsystem,
as the relative
(percentage)
the phase
voltages, measured
the E/Aofinstrument
and theasPXI
obtained
during the
errors
δ,
adopting
the
value
indicated
by
the
PXI
device
as
the
actual
value
of
the
voltage,
calibration of the instrument for different frequencies of the network voltage, as well as the relative
respectively. The values and course of the errors indicate that their main source were the systematic
(percentage) errors δ, adopting the value indicated by the PXI device as the actual value of the voltage,
effects resulting from the incorrect calibration of the processing characteristics of the measurement
respectively. The values and course of the errors indicate that their main source were the systematic
channels. The random influences cause the dispersion of errors ∆ as a function of frequency (for L2
effects
resulting
from the
ofAssuming
the processing
characteristics
of the
measurement
and
L3 channels)
not incorrect
exceeding calibration
∆max = 0.2 V.
the uniform
distribution
of errors,
the
channels.
The
random
influences
cause
the
dispersion
of
errors
∆
as
a
function
of
frequency
(for L2 and
standard uncertainty can be calculated using Formula (1):
L3 channels) not exceeding ∆max = 0.2 V. Assuming
the uniform distribution of errors, the standard
Δ
(1)
uncertainty can be calculated using Formula=(1): ≅ 0.12
√3
∆max
u= √ ∼
= 0.12 V
3
(1)
Energies 2017, 10, 536
Energies 2017, 10, 536
12 of 16
Table 3. The voltage deviations with reference to the network frequency.
12 of 16
Table 3. The voltage deviations with reference to the network frequency.
f (Hz)
Size
Unit
L1
L2
L3
f (Hz)
45
45
50
50
55
55
Size
U
E/A
UPXI
E/A
U
U∆PXI
δ∆
δ
UE/A
U
UPXI
E/A
U∆PXI
δ∆
Unit
(V)
(V)
(V)
(V)
(V)
(%)
(V)
(%)
(V)
(V)
(V)
(V)
(V)
(%)
(V)
L1
221.4
221.4
220.9
0.5
220.9
0.23
0.5
0.23
221.4
220.9
221.4
0.5
220.9
0.23
0.5
L2
220.7
220.7
220.4
0.3
220.4
0.14
0.3
0.14
220.9
220.4
220.9
0.5
220.4
0.23
0.5
L3
220.3
220.3
220.8
−
0.5
220.8
−−0.5
0.23
−0.23
220.5
220.8
220.5
−
0.3
220.8
−−0.3
0.14
UE/A
δ
U
E/A
UPXI
U∆PXI
δ
∆
δ
(V)
(%)
(V)
(V)
(V)
(V)
(%)
(V)
(%)
221.4
0.23
220.9
221.4
0.5
220.9
0.23
0.5
0.23
220.8
0.23
220.4
220.8
0.4
220.4
0.19
0.4
0.19
220.3
−0.14
220.8
220.3
−
0.5
220.8
−0.23
−0.5
−0.23
In sum, all of the applied testing methods for the assessment of the measurement track
In sum, all
the applied
methods
for theThe
assessment
of the of
measurement
track
characteristics
areofreliable
for thetesting
specified
instrument.
obtained results
the experimental
characteristics
reliable
for the
specified instrument.
The obtained
the of
experimental
verification
areare
positive
within
an acceptable
tolerance range.
However,results
on theofbasis
the results,
verification
are about
positive
tolerance range.
However,
on the basis
of the results,
no
conclusions
thewithin
degreeanofacceptable
possible modification
algorithms
to improve
the measurement
no conclusions
accuracy
can be about
drawn.the degree of possible modification algorithms to improve the measurement
accuracy
Whatcan
is be
thedrawn.
reason for the phenomena described in [15,16]? Can these cases relate to the
What is the
reason
the phenomena
in is[15,16]?
these cases
to the
characteristics
of the
E/A for
instrument?
Perhaps described
the problem
that theCan
instrument
underrelate
test and
characteristics
of the operate
E/A instrument?
Perhaps
the problem
the instrument
under
and the
testing
arrangement
on different
data sets.
Because is
thethat
individual
coefficients
aretest
calculated
testingthe
arrangement
on different
data sets.
Because
the individual
coefficients
are calculated
using
samples of operate
signals from
the electrical
network,
collected
in time windows
of approximately
using
the
of signals
the differ.
electrical
collected in time
windows
approximately
200
ms,
thesamples
sets of the
source from
data can
Fornetwork,
specific measurement
conditions,
theofsmall
inaccuracy
200the
ms,
the sets of the
source data
differ.
For specific
measurement
conditions,
of
measurement
algorithms
may can
result
in a large
difference
of indication
comparedthe
to small
other
inaccuracy of So,
the next
measurement
algorithms
result in
large
of indication
compared
to
instruments.
to the quality
of the may
calibration
ofathe
DAdifference
part of the
instrument,
the proper
other instruments.
So, next to
the quality
of the
of the DA
of the
instrument,
validation
of the algorithms
performed
in the
DDPcalibration
part (the difficult
taskpart
to do)
is important
for the
proper validation
of the algorithms
performed
in the DDP part (the difficult task to do) is important
quality
of the measurement
functions
of the instrument.
for the quality of the measurement functions of the instrument.
4. The Concept of Validation of Measurement Algorithms
4. The
Concept is
ofthe
Validation
Measurement
Validation
techniqueofused
to compareAlgorithms
the results to be validated with the results obtained
through
other numerical
methods
validated.
words, with
one technique
been
Validation
is the technique
usedpreviously
to compare
the resultsIntoother
be validated
the results has
obtained
validated
beforenumerical
it is used as
a reference
to validate
the second
method.
through other
methods
previously
validated.
In other
words, one technique has been
The
idea
of
the
reliable
validation
of
measurement
algorithms,
applicable
for the E/A instrument
validated before it is used as a reference to validate the second method.
(and The
probably
other devices
as of
well),
relies on algorithms,
the assumption
that for
thethe
dataset
used for
idea of for
the reliable
validation
measurement
applicable
E/A instrument
the
coefficients
the relies
examined
the
asused
the for
onethe
used
in the
(andcalculation
probably forofother
devices asinwell),
on theinstrument
assumption is
that
thesame
dataset
calculation
reference
system.
of coefficients in the examined instrument is the same as the one used in the reference system.
This assumption
assumption is
is met
met ififthe
themeasurement
measurementdata,
data,processed
processedininthe
the
DSP
designation
DSP
forfor
thethe
designation
of
of
coefficients,
are
stored
and
further
used
in
the
reference
system
to
calculate
the
corresponding
coefficients, are stored and further used in the reference system to calculate
coefficients (Figure 14).
Test
signals
DDP
DA
Collection of selected
samples & calculated
coefficients
Estimation
Figure 14.
Figure
14. The
The experimental
experimental setup
setup for
for the
the verification
verification of
of the
the measurement
measurement algorithms
algorithms using
using registered
registered
samples and
and the
the corresponding
corresponding coefficients,
coefficients, also
also in
in addition
addition to
to the
the PC-based
PC-based estimation
estimation system.
system.
samples
Figure 4 shows the resources of the DSM STSL module applied in the E/A instrument (Figure 2).
The module is additionally equipped with 128 MB of the synchronous dynamic random access
memory (SDRAM). This memory was not used in the E/A operations, because of the sufficient
Energies 2017, 10, 536
13 of 16
Figure 4 shows the resources of the DSM STSL module applied in the E/A instrument (Figure 2).
The module is additionally equipped with 128 MB of the synchronous dynamic random access memory
(SDRAM). This memory was not used in the E/A operations, because of the sufficient capacity of the
internal memory of the DSP (24 Mb) for the calculation of the adopted coefficients of the electrical
power quality.
The proposed idea of measurement algorithms validation assumes the use of the SDRAM memory
(in Figure 4 marked in red ellipse) for storing the selected samples from the ADC, but only those
which were used by the DSP for the coefficients calculation, together with the designated coefficients.
The capacity of the SDRAM enables one to accumulate the data from several processing cycles,
corresponding to measurement windows of approximately 200 ms. For example, for the “estimator”
operating option (Figure 5), for memorizing the data collected from the samples from three voltage
channels, converted in the ADC at the rate of 210 kS/s, no more than 256 kB of memory is needed for
one measurement window.
To perform the validation of embedded algorithms, the original source data, together with the
DSP processing results, are sent to the external system. In this way, it is possible to assess the real
characteristics of data processing, as well as those corresponding to the DA part as embedded in the
DSP software, in the DDP part of the instrument. The implementation of the proposed method in the
E/A instrument is not connected with any hardware modifications. Only software changes are needed
to obtain the fixed goal.
The reference system stores the set of data received from the instrument and uses them for the
independent off-line calculation of respective coefficients, resulting from the sequentially selected
measurement functions of the E/A. The system is under development in the LabVIEW environment.
The estimation part of the reference system compares the values of the coefficients designated in the
E/A with the results obtained off-line in the PC estimation system. The simple criterion (2) is used for
the validation of the examined algorithm implemented in the DSP:
∆i ≥ CiM − CiE
(2)
where: CiM —the value of coefficient calculated in the E/A, CiE —the value of the coefficient obtained
in the estimation operation in the reference system, ∆i —the admissible error specified for the given
algorithm under test, and i—the coefficient identifier.
To measure the strength of the relation between the set of CiM and CiE coefficients, obtained in
the E/A instrument and in the PC estimation system for many measurement windows, the Pearson
coefficient [34] will be used.
The results, shown in [15,16], indicate that other methods of verification of the sophisticated
algorithms, even using reliable examination systems, are not always sufficiently correct.
The imperfection of the measurement algorithms is often the result of the rounding used,
simplifications, and approximations, and their optimization can help to improve the properties of the
measuring instrument.
More detailed descriptions of some other aspects concerning data processing in the measurement
track of the E/A instrument are analyzed in [28,29,31–33].
5. Conclusions
Nowadays, the exploitation of electrical appliances has a significant impact on our natural
environment. The monitoring of electrical power quality is an important part of its protection.
The main problem presented in this paper deals with the calibration procedures of the
measurement channel and the validation of the measurement algorithms. A case study-based
consideration concerning the performed calibration procedures of the E/A is presented, along with the
obtained exemplary results. Another approach for the verification of the measurement characteristics,
especially applied to the software part of the instrument, is proposed. The concept of the validation of
Energies 2017, 10, 536
14 of 16
the measurement algorithms embedded in the software of the instrument for the designation of the
coefficients of electrical power quality, as well as its application using the resources available onboard
of the DSP module of the elaborated E/A instrument, are discussed.
The development of self-calibration procedures and their implementation to the A/D functions,
in combination with the validation of the measurement algorithms, can improve the quality of the
measurement results. The use of the proposed solutions in the considered instrument certainly cannot
solve all of the problems concerning accuracy of the measurement results, but it allows an evaluation
of the data processing algorithms implemented in the built-in DSP software and the identification of
errors affecting the deteriorations of the instrument’s accuracy.
This research is ongoing and the results will be presented in future papers.
Acknowledgments: The cost of publishing in open access is covered by the author’s employer (Gdynia
Maritime University).
Conflicts of Interest: The author declares no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Arrillaga, J.; Bradley, D.A.; Bodger, P.S. Power System Harmonics; John Wiley & Sons: New York, NY, USA,
1985; p. 664.
Caciotta, M.; Leccese, F.; Trifirò, T. From power quality to perceived power quality. In Proceedings of
the IASTED International Conference on Energy and Power Systems EPS 2006, Chiang Mai, Thailand,
29–31 March 2006; Volume 526–119, pp. 94–102.
Caciotta, M.; Leccese, F.; Trifirò, T. Curve-fitting-algorithm (CFA) as power quality basic algorithm.
In Proceedings of the XVIII International Measurement Confederation (IMEKO) World Congress, Metrology
for a Sustainable Development, Rio de Janeiro, Brazil, 17–22 September 2006.
Fuchs, E.F.; Masoum, M.A.S. Power Quality in Power Systems and Electrical Machines, 1st ed.; Academic Press:
Cambridge, MA, USA, 2008; p. 638.
Arrillaga, J.; Watson, N.R.; Chen, S. Power System Quality Assessment; John Wiley & Sons: New York, NY, USA,
2000; p. 300.
Caciotta, M.; Grossoni, M.; Leccese, F. Power quality measurements in telecommunication
exchanges. In Proceedings of the International Telecommunications Energy Conference, INTELEC 2008,
San Diego, CA, USA, 14–18 September 2008.
International Electrotechnical Commission (IEC). IEC 61000-4-30:2015. Electromagnetic Compatibility
(EMC)—Part 4-30: Testing and Measurement Techniques—Power Quality Measurement Methods; IEC:
Genova, Switzerland, 2015.
EN 50160, Voltage Characteristics of Electricity Supplied by Public Distribution Systems. Available online:
http://www2.schneider-electric.com/library/SCHNEIDER_ELECTRIC/SE_LOCAL/APS/204836_1312/
DraftStandard0026rev2-DraftEN501602005-05.pdf (accessed on 10 March 2017).
IEEE 1159-2009—IEEE Recommended Practice for Monitoring Electric Power Quality; IEEE Standards Association:
Piscataway, NJ, USA, 2009.
Leccese, F. A first analysis of perceived power quality for domestic customers. In Proceedings of the 12th
IMEKO TC1 & TC7 Joint Symposium on Man Science & Measurement, Annecy, France, 3–5 September 2008;
pp. 246–253.
Fuller, J.F.; Fuchs, E.F.; Roesler, D.J. Influence of harmonics on power system distribution protection.
IEEE Trans. Power Deliv. 1988, 3, 546–557. [CrossRef]
Dugan, R.C.; McDermott, T.E. Operating conflicts for distributed generation on distribution systems.
In Proceedings of the IEEE Industry Applications Society (IAS) 2001 Rural Electric Power Conference
Record, Little Rock, AR, USA, 29 April–1 May 2001.
Mindykowski, J. Assessment of Electric Power Quality in Ship Systems Fitted with Converter Subsystems; Press of
Shipbuilding and Shipping Ltd.: Gdańsk, Poland, 2003; p. 276.
Leccese, F. Study and characterization of a new protection system against surges and over voltages for
domestic telecommunication networks. In Proceedings of the International Telecommunications Energy
Conference INTELEC 2007, Rome, Italy, 30 September–4 October 2007; pp. 363–368.
Energies 2017, 10, 536
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
15 of 16
Bień, A.; Firlit, A.; Folga, P.; Gallus, E.; Hanzelka, Z.; Hartman, M.; Hashad, M.; Iwaszkiewicz, J.; Jarocha, R.;
Mieński, R.; et al. Voltage fluctuation measurement—Experiment in the industrial environment. Electr. Power
Qual. Util. EPQU 2001, 7, 9–18.
Szlosek, M.; Piekarz, M.; Hanzelka, Z.; Bien, A.; Piatek, M.; Loziak, W.; Pietrucha, R.; Hashad, M.; Wolski, L.;
Olczykowski, Z.; et al. Comparative tests of flickermeters. In Proceedings of the 17th International
Conference on Electricity Distribution, Barcelona, Spain, 12–15 May 2003.
IEC. IEC 61000-4-15:2010. Electromagnetic Compatibility (EMC)—Part 4-15: Testing and Measurement
Techniques—Flickermeter—Functional and Design Specifications; IEC: Genova, Switzerland, 2010.
Hashad, M.; Hanzelka, Z.; Hartman, M.; Bien, A. The hypothesis for the wrong measurements results
obtained during the flickermeter comparative test. In Proceedings of the 6th International Conference
Electrical Power Quality and Utilisation, Cracow, Poland, 19–21 September 2001.
Hoon, Y.; Radzi, M.A.M.; Hassan, M.K.; Mailah, N.F. DC-link capacitor voltage regulation for three-phase
three-level inverter-based shunt active power filter with inverted error deviation control. Energies 2016,
9, 533. [CrossRef]
Ok, Y.; Lee, J.; Choi, J. Analysis and solution for operations of overcurrent relay in wind power system.
Energies 2016, 9, 458. [CrossRef]
Leferink, F.; Keyer, C.; Melentjev, A. Static energy meter errors caused by conducted electromagnetic
interference. IEEE Electromagn. Compat. Mag. 2016, 5, 49–55. [CrossRef]
Caciotta, M.; Giarnetti, S.; Leccese, F.; Pedruzzi, E. Curve fitting algorithm FPGA implementation.
In Proceedings of the 10th International Conference on Environment and Electrical Engineering—EEEIC2011,
Rome, Italy, 8–11 May 2011.
Caciotta, M.; Giarnetti, S.; Leccese, F.; Trinca, D. Development of an USB Data acquisition system for
power quality and smart metering applications. In Proceedings of the 11th International Conference on
Environment and Electrical Engineering—EEEIC2012, Venice, Italy, 18–25 May 2012.
Di Pasquale, S.; Giarnetti, S.; Leccese, F.; Trinca, D.; Caciotta, M. A new platform for high accuracy power
quality measurements: The forensic point of view. In Proceedings of the 20th International Measurement
Confederation (IMEKO) Technical Committee 4 (TC4) International Symposium and 18th International
Workshop on ADC Modelling and Testing, Benevento, Italy, 15–17 September 2014; pp. 404–409.
Di Pasquale, S.; Giarnetti, S.; Leccese, F.; Trinca, D.; Cagnetti, M.; Caciotta, M. A Distributed Web-Based
System for Temporal and Spatial Power Quality Analysis. In Power Quality Issues in Distributed Generation;
In Tech: Rijeka, Croatia, 2015. [CrossRef]
Giarnetti, S.; Leccese, F.; Caciotta, M. Non recursive multiharmonic least-square fitting for frequency
estimation for grid frequency estimation. Measurement 2015, 66, 229–237. [CrossRef]
Giarnetti, S.; Leccese, F.; Caciotta, M. Non recursive Nonlinear Least Squares for periodic signal fitting.
Measurement 2017, 103, 208–216. [CrossRef]
Masnicki, R. The fluency of data flow in the instrument for measurement and registration of parameters of
the electrical power network. In Proceedings of the 15 IEEE International Conference on Environment and
Electrical Engineering, Rome, Italy, 10–13 June 2015; pp. 2040–2044.
Masnicki, R.; Mindykowski, J. Coordination of operations in registration channel of data from electrical
power system. Measurement 2017, 99, 68–77. [CrossRef]
Mindykowski, J.; Tarasiuk, T.; Masnicki, R.; Szweda, M.; Gorniak, M. Universal estimator/analyzer of
electrical power quality—Part II—Design and technical implementation. In Scientific Papers of the Gdynia
Maritime University; Publishing House of Gdynia Maritime University: Gdynia, Poland, 2010; pp. 21–36.
(In Polish)
Masnicki, R. Validation of the measurement algorithms in instrument for power quality estimation.
In Proceedings of the 16 IEEE International Conference on Environment and Electrical Engineering,
Florence, Italy, 6–8 June 2016; pp. 2519–2523.
Masnicki, R.; Mindykowski, J. Examination of the instrument for power quality estimation—Case study.
In Proceedings of the 21st IMEKO TC4 International Symposium and 19th International Workshop on
ADC Modelling and Testing “Understanding the World through Electrical and Electronic Measurement”,
Budapest, Hungary, 7–9 September 2016; pp. 240–244.
Energies 2017, 10, 536
33.
34.
35.
36.
37.
38.
39.
40.
16 of 16
Masnicki, R. Verification of algorithms in measuring instrument to assess the quality of electrical energy.
In Scientific Papers of the Faculty of Electrical and Control Engineering of Gdansk University of Technology;
Politechnika Gdańska: Gdansk, Poland, 2016; Volume 49, pp. 63–67. (In Polish)
Jauregui, R.; Silva, F. Numerical Validation Methods. In Numerical Analysis—Theory and Application; In Tech:
Rijeka, Croatia, 2011. Available online: http://www.intechopen.com/books/numerical-analysis-theoryand-application/numerical-validation-methods (accessed on 30 March 2017).
Kaztek Systems. Available online: http://www.kaztek.com/ (accessed on 7 April 2017).
Tarasiuk, T. Estimator-analyser of power quality: Part I—Methods and algorithms. Measurement 2011, 44,
238–247. [CrossRef]
Tarasiuk, T. Estimator-analyser of power quality: Part II—Hardware and research results. Measurement 2011,
44, 248–258. [CrossRef]
Masnicki, R. The improvement of measurement accuracy in microprocessor instruments. In Proceedings
of the 4th International Workshop, Compatibility in Power Electronics, CPE 2005, Gdynia, Poland,
1–3 June 2005.
Analog Devices. ADSP-TS201S EZ-KIT Lite, Evaluation System Manual. Available online: http://www.
analog.com/media/en/dsp-documentation/evaluation-kit-manuals/ (accessed on 7 April 2017).
Chroma ATE Inc. Chroma Programmable AC Power Source 6560/6590. In User’s Manual. Available online:
http://www.chromaate.com/product/6500_series_Programmable_AC_Source.htm (accessed on 7 April
2017).
© 2017 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).