Content uploaded by Hermes Jose Loschi
Author content
All content in this area was uploaded by Hermes Jose Loschi on Dec 16, 2019
Content may be subject to copyright.
Waseem El SAYED, Hermes LOSCHI, Robert SMOLENSKI, Piotr LEZYNSKI, Choon LONG LOK
Uniwersytet Zielonogórski, Instytut Inżynierii Elektrycznej
Performance Evaluation of the Effect of Power Converters
Modulation on Power line Communication
Abstract. The increase in the use of power converters in a smart grid system causes the presence of high-level of conducted electromagnetic
interference. Consequently, this leads to a miscommunication problem between the smart grid elements. This paper focuses on the influence of two
power converter modulation techniques: deterministic and random modulation, on the performance of the Power Line Communication (PLC) signal.
The paper presented a practical implementation of the system and discussed the results for different operating scenarios.
Streszczenie. Stosowanie przekształtników energoelektronicznych w systemach Smart Grid prowadzi do wysokiego poziomu zaburzeń
elektromagnetycznych mogących zakłócać komunikację między elementami systemu. W artykule zbadano wpływ zaburzeń elektromagnetycznych
generowanych przez przekształtniki energoelektroniczne na komunikację PLC. Przedstawione badania eksperymentalne wykonano dla
deterministycznych i losowych technik modulacji stosowanych w przekształtnikach i różnych wariantów pracy układów komunikacyjnych.(Ocena
wpływu zaburzeń elektromagnetycznych generowanych przez przekształtniki energoelektroniczne na komunikację PLC.)
Keywords: Power Line Communication (PLC), Electromagnetic interference (EMI), Deterministic modulation, Random modulation.
Słowa kluczowe: Komunikacja po liniach elektroenergetycznych, Zaburzenia elektromagnetyczne, Modulacja deterministyczne,
Modulacje losowe
Introduction
In the last decade, the use of renewable energy has
gained a meditative interest in a lot of applications,
especially, in the home applications, traction, and industrial
systems. Consequently, this claims growing concerns on
the reliability of the communications between the smart grid
elements. The Power Line Communication (PLC) is
considered as one of the most used techniques for the
communication between the smart grid elements, as it uses
the existing power cables in the system to provide data
transmission capabilities. The use of the electrical cable
infrastructure results in lowering the cost and providing high
data transition rate, however, a lot of problems could
appear in the communication due to the presence of
harmonics and conducted Electromagnetic Interference
(EMI) in the power cables [1].
The use of several power electronic converters
connected through PLC increases the probability of
appearance of data transmission errors, caused by the
generated EMI in the system, due to the high switching
frequency used for the control of the converter operation [2].
The most traditional control approach is pulse width
modulation (PWM). In the case of three or more phase
systems, named “polyphase systems", the industry-
standard today is the so-called space vector modulation, in
which switching commands for all three phases are
generated in a coordinated fashion [3]–[6]. Two approaches
became more popular; one is the optimization the Power
Electronic Interfaces thought their switch modulation
strategies. Deterministic PWM waveforms (“programmed
switching”), and other is an alternative in the form of
randomized modulation. The concept of frequency
modulation techniques is based on the modulating the
original constant clock frequency to spread the energy of
every single harmonic into the well-defined frequencies,
thus reducing the peak amplitude of EMI at harmonic
frequencies. There is no continues noise spectrum. The
strategy with programmed switching frequency is using a
variable switching frequency, obtained by the modulation of
a base value, i.e., carrier function using control signals and
having a spectrum with lower peak amplitude than the
constant frequency square signal. This strategy is also
known as spread spectrum frequency modulation (SSFM).
The investigations of such techniques applied to EMI
reduction of digital systems is a subject of significant
concern, inclusive opening a new area of research for the
modulation techniques with randomized modulation [7]–
[13].
The core difference of these approaches is that the
effect of the randomization in attenuating the discrete
spectrum, equal the spread spectrum frequency modulation
(SSFM) approach, but also introduce a continuous
spectrum. One possible problem in converters with
randomized modulation is the increase of “ripple” (i.e.,
deviation around the nominal waveform) when compared
with programmed switching. This particularly bothersome is
in the low-frequency range (“drift”)[14]–[17].
The effect of different modulation techniques on the
communication has been studied before in [18][19], it shows
great influence on RS-232 communication standard.
This paper focuses on studying the influence of
converter modulation techniques on PLC performance.
Moreover, the effect of the distance between the source of
EMI and the sending device has studied as well. The paper
is divided as following, Section 1 introduce the PLC
standards and regulation. The difference between the
deterministic and random modulation is presented in
section 2, section 3 present the hardware connection of the
used setup and section 4 introduce the results from the
setup, conclusion of the work is given in section 5
respectively.
PLC Standardization and Regulations
The PLC works by modulating carrier signal and adding
it to the main power signal. it operates through two main
frequency bands: Narrowband (NB-PLC) and broadband
PLC (BB-PLC). The NB-PLC typically works at a range of 3
kHz to 500 kHz and BB-PLC works at a range of 1.8MHz to
250MHz [1]. The NB-PLC is working in the smart metering
application, traction and battery charging systems, however,
the BB-PLC is used in high-speed application like the
internet. Therefore, this paper concerns only on NB-PLC.
As shown in Table 1, there are three main regulations
are used, the first is the European Norm (EN) 50065, it was
first established by CENELEC in 1992 [1]. It consists of four
frequency bands, which are commonly referred to as
CENELEC-A (3 – 90.6 kHz), CENELEC-B (95–125 kHz),
CENELEC-C (125–140 kHz) and CENELEC-D (140–148.5
kHz) respectively. The second is the U.S. Federal
Communications Commission (FCC), the PLC is used in the
range between 9 and 490 kHz band, this regulation is not
used for smart meter applications. The last regulation is
established by the Japanese Association of Radio
Industries and Businesses (ARIB) in the form of Standard
T84, which allow the use of PLC in the 10–450 kHz band.
Table 1, Standard and Regulation of PLC [1].
Region
Standard/
Regulation
Type
Frequency in
(kHz)
Europe
EN 50065
3–148.5 kHz
CENELEC A
3 to 90.6
CENELEC B
95 to 125
CENELEC C
25 to 140
CENELEC D
140 to 148.5
USA
FCC
9 – 490 kHz
Japan
ARIB STD T-84
10–450 kHz
A lot of industries start developing of PLC solution
based on the regulation in Table1, G3-PLC has been
developed the G3-PLC Alliance and The industry
specifications PRIME (Power line Related Intelligent
Metering Evolution) by the PRIME Alliance. Fig.1 shows the
developed industrial solutions and the standards that they
support [1].
Fig.1. Industrial solutions and operating frequency ranges [1].
Random Modulation Vs Deterministic modulation in the
spectrum from
In standard PWM switching strategy with the
programmed switching frequency, switching harmonics
usually occur at fixed and well-defined frequencies and are
thus named “discrete harmonics.”
According to [18], [20] and [21], the concept of
randomness into standard PWM strategy is to spread the
harmonic power which exists at well-defined frequencies
(discrete harmonics) over a wide range of frequencies so
that no harmonic of significant magnitude exists. As a
result, discrete harmonics are significantly reduced, and the
harmonic power is spread over the spectrum as “noise”
(continuous spectrum) of insignificant magnitude.
The strategy of most randomized modulation is based
on schemes in which successive randomizations of the
switching pulse train (or of the periodic segments of this
pulse train) are statistically independent and governed by
invariant probabilistic rules [17],[20]–[22] Therefore, the
randomized modulation strategy must enable precise
control of the time-domain performance of randomized
switching, in addition to spectral shaping in the frequency
domain. Randomized modulation presented an inherent
invariant deterministic and probabilistic structure.
Fig.2 shows the frequency spectrum average detector,
reading from EMI receiver for one Switched Mode Power
supplies (SMPSs) using deterministic and random
modulation, work with a switching frequency of 28 kHz. The
reading was taken in in accordance with the EN 55022
standard.
Hardware setup of the system
The system work using two Texas Instrument PLC Kits,
they are connected to the same PC as shown in Fig.4, both
kits are configured to work on the PRIME mode which
works in the frequency range CENELEC-A standard. Fig.3
shows the frequency spectrum of the PLC communication
without connecting any source of conducted EMI noise, it
was noticed that the frequency range is between 42 and
90.6 kHz which is the frequency used by the PRIME
industrial solution respectively.
Fig.2. Frequency Spectrum of Deterministic and Random
modulations.
Fig.3. Frequency spectrum for the PLC using PRIME.
The communication was established using a cable of
42m length, divided into 5 sockets ever 10m to study the
effect of the distance in the PLC status. The Line
Impedance Stabilization Network (LISN) was used to isolate
the conducted EMI noise coming from the grid and is
connected to the receiving terminal of the cable as shown in
Fig.4. Two Switched Mode Power supplies (SMPSs) are
used as a source of conducted EMI on the circuit, they work
on 28 kHz switching frequency. The converters were
modified to work using both deterministic and random
modulation, The used switched mode power converters are
shown in Fig.5 as the electromagnetic interference sources.
Fig.4. PLC Kits.
Fig.5. The used SMPS as EMI Source.
Fig.6 shows the connection diagram for the hardware
setup. The place of converters changes according to the
needed scenario respectively.
Fig.6. Connection diagram of the system.
Results and discussion
The study was done on 1373 data packages; each
package is 50bit frame of data. In the normal case, the
average waiting time between each package and the other
is 0.4sec, the used modulation for transferring of the PLC
signal is Deferential Binary Phase Shift Keying (DBPSK).
1) Deterministic Vs Random Modulation
The first scenario is connecting the two converters at the
sending point (0m from the sending PLC), both converters
work on a output voltage of 12V and current of 1A. Fig.7
shows the difference between the frequency spectrum of
the two converters connected together at the stated
operating points in case of deterministic and random
modulation, the results shows that spreading the frequency
in case of random modulation has a greater influence on
the data transition rate than the deterministic modulation as
the amplitude of the hall frequency spectrum increase, this
is confirmed in Fig.8. Fig.8 shows the box and whisker plot
for the data transition packages in case of deterministic and
random modulation, the mean value of the waiting time in
case of deterministic modulation is 0.7121 sec, while for
random modulation, it is 0.836sec, this means that the
deterministic modulation shows less effect on the data
transition flow rate. However, the median value is almost
the same in all cases, it is equal to 0.4 sec.
2) Distance influence
The second scenario is changing the place of the EMI
Source. Fig.9 shows the box and whisker plot for the data
transition of 1373 package in case of connecting 2
converters using Deterministic modulation at different
distance from the sending terminal, it was noticed that the
EMI source has the maximum effect on data transition rate
when the two converters are connected at the terminal
ends, as in those cases the EMI source is connected
directly at the same points of the sending and receiving
PLC Kites, the mean value of the waiting time in sending
and receiving terminal reach 0.712sec and 0.55sec. On the
other hand, as we go far from the terminals, the effect of the
EMI on the PLC performance decreases, the minimum
effect of EMI source appears on the middle of the cable, at
20m from the sending terminal and 22m from the receiving
terminal, the mean value of the waiting time in for each
package, in this case, is 0.401sec.
Fig.7. Frequency spectrum of two SMPSs in case of deterministic
and random modulations.
Fig.8. Box and whisker plot for Normal case, Deterministic and
random modulation cases at 0m.
PLC Sending Kit
PLC Receiving Kit
SMPS1
SMPS2
Resistive Loads
Resistive Loads
Fig.9. Box and whisker plot in case of 2 converters using
Deterministic modulation and in different distance from the sending
end.
Fig.10 shows the effect of using the random modulation
in converters on the PLC performance at different distances
at the same number of packages and bit frame, the same
performance was noticed as in Deterministic modulation, as
the EMI source is closer to the terminals, the greater the
effect on the data transition rate. As shown in Fig.10,
Random modulation has a greater effect on the data
transition rate than the deterministic modulation, the mean
value of the waiting time between each package reach
0.8326sec at 0m and 0.75 sec at 42m, this is due to the
concept of randomness which based on is spreading the
harmonic power over a wide range of frequencies, so that
no harmonic of significant magnitude exists. However, it
causes increase on the amplitude of whole spectrum,
respectively, causing higher effect on the PLC signal,
especially, when the PLC signal and the converters
switching frequency and its harmonics lays between the
same frequency range as shown in Fig.3 and Fig.4, the
effect change according to the switching frequency and its
harmonics.
Fig.11 shows the average bps in case of connecting of
two converters in different distance from the sending
terminal, with deterministic and random modulation. The
curves show that the maximum average bps is in the middle
for both deterministic and random modulation, it reaches
992bps in case of deterministic modulation and 725 bps in
case of random modulation, the minimum values were in
the terminals. On other hands, the ratio between the
average bps for deterministic and random modulation in all
points is almost equal to1.4, this means that in case of
smart meter systems, the number of the meters connected
to the data concentrator should be decreased by a factor of
0.7.
As a result, the probability of data transition errors will be
maximum at the terminals and reach its minimum at the
center for both deterministic and random modulations as
shown in Fig.12. In case of deterministic modulation, the
probability of error reach 19.88% and 15% in case of
connecting the converters directly to the sending and
receiving ends, however, it reaches 7.5% in the center.
Also, in case of random modulation, the probability of error
reach 34.1% and 31.1% in case of connecting the
converters directly to the sending and receiving ends,
however, it reaches 21.4% in the center.
Fig.13 shows the total waiting time for sending all the
packages in case of connecting of two converters in
different distance from the sending terminal, with
deterministic and random modulation, it is opposite to the
average bps curves.
Fig.10. Box and whisker plot in case of 2 converters using Random
modulation and in different distance from the sending end.
Fig.11. Average bps in case of Deterministic and Random
modulation.
Fig.12. The Probability of data transition error in the case of
deterministic and random modulation.
0
200
400
600
800
1000
1200
010 20 30 40 50
Average (bps)
Distance (m)
Avg bps in 2D Avg bps in 2R
0
10
20
30
40
0 10.2 20 30 41.5
Error in %
Distance (m)
Probability of Error in Determenistic
Probability of error in Random
Fig.13. Total waiting time in case of Deterministic and Random
modulation.
Conclusion
This paper focuses on the influence of SMPS as the
source of conducted EMI on the PLC communication signal
from different aspects. A comparison between the effect of
the deterministic and random modulation is presented.
When compared with deterministic switching strategies
(“programmed switching”), the effect of randomization
generates a reduction of the discrete spectrum and
introduction of a continuous spectrum. Therefore, we can
assume that the main benefit from randomized switching
strategies is better utilization of the available harmonic
content of waveforms at the power supply/equipment
interface. Consequently, the deterministic modulation has
less effect in PLC signal than the random modulation at 28
kHz switching frequency of the converters, but not in all
cases, only at this switching frequency, the effect change
according to the switching frequency and its harmonics.
On the other hand, the distance between the source of
the EMI and the sending or the receiving devices plays an
important role. As this situation of switching frequency and
operating conditions, the highest probability of data
transition error appears when the source of the EMI is
connected directly to the sending and the receiving devices,
in both random and deterministic modulation. The lowest
probability of data transition error appears in the middle.
The investigations of such effect on the PLC
communication open the gate for a lot of discussions in the
future, such as studying the effect of changing the switching
frequency of the converter on the results. Also, studying the
effect of changing the way of PLC data modulation
(DBPSK, DQPSK, and D8PSK) on the results. Moreover,
trying different types of converters and loads which may
give interesting results.
Acknowledgment
This paper is part of a project that has received funding from
the European Union’s Horizon 2020 research and innovation
programme under the Marie Sklodowska-Curie grant
agreement No 812391 – SCENT.
References
[1] C. Cano, A. Pittolo, D. Malone, L. Lampe, A. M. Tonello, and
A. Dabak, “State-of-the-art in Power Line Communications :
from the Applications to the Medium,” IEEE J. Sel. Areas
Commun., vol. 8716, no. c, pp. 1–19, 2016.
[2] Y. Lai, S. Member, and Y. Chang, “Novel Random-Switching
PWM Technique With Constant Sampling Frequency and
Constant Inductor Average Current for Digitally Controlled
Converter,” IEEE Trans. Ind. Electron., vol. 60, no. 8, pp.
3126–3135, 2013.
[3] M. H. Hedayati, S. Member, V. John, and S. Member, “Filter
Configuration and PWM Method For Single Phase Inverters
with Reduced Conducted EMI Noise,” IEEE Trans. Ind. Appl.,
vol. 51, no. 4, pp. 3236–3243, 2014.
[4] G. Krishnan, “A Neoteric Method Based on PWM ON PWM
Scheme with Buck Converter for Torque Ripple Minimization
in BLDC Drive,” in International Conference on Magnetics,
Machines & Drives (AICERA-2014 iCMMD) A, 2014, no. 1,
pp. 1–6.
[5] S. K. Sahoo, A. Ramulu, S. Batta, and S. Duggal,
“Performance Analysis and Simulation of Th ree Phase
Voltage Source Inverter using basic PWM Techniques,” in Th
ird International Conference on Sustainable Energy and
Intelligent System (seiscon 2012), 2012, pp. 1–7.
[6] S. K. Vlfv et al., “Voltage source multilevel inverter voltage
quality comparison under multicarrier sinusoidal PWM and
space vector PWM of two delta voltages,” in International
Multi-Conference on Engineering, Computer and Information
Sciences (SIBIRCON), 2017, pp. 439–444.
[7] F. Lin and D. Y. Chen, “Reduction of Power Supply EM1
Emission by Switching Frequency Modulation,” IEEE Trans.
Power Electron., vol. 9, no. 1, pp. 132–137, 1993.
[8] J. Michel, C. Neron, and R. Luciolcs, “A Frequency
Modulated,” in Twelfth Annual IEEE International ASIC/SOC
Conference, 1999, pp. 362–365.
[9] Z. Shen, S. Member, D. Jiang, S. Member, T. Zou, and S.
Member, “Dual-Segment Three-phase PMSM with Dual
Inverters for Leakage Current and Common-mode EMI
Reduction,” IEEE Trans. Power Electron., vol. PP, no. c, p. 1,
2018.
[10] H. S. A, H. Kobayashi, T. Suzuki, and T. Iijima, “Spread-
Spectrum Clocking in Switching Regulators to Reduce EM1,”
in Proceedings. IEEE Asia-Pacific Conference on ASIC,
2002, pp. 141–144.
[11] K. K. Tse, R. W. Ng, H. S. Chung, and S. Y. R. Hui, “An
Evaluation of the Spectral Characteristics of Switching
Converters With Chaotic Carrier-Frequency Modulation,”
IEEE Trans. Ind. Electron., vol. 50, no. 1, pp. 171–182, 2003.
[12] H. Guo, Z. Li, and B. Zhang, “Yrs,” in 2009 International
Conference on Communications, Circuits and Systems,
2009, pp. 693–696.
[13] T. P. Inverters, C. K. Lee, S. Y. R. Hui, S. Member, H. S.
Chung, and Y. Shrivastava, “A Randomized Voltage Vector
Switching Scheme for,” IEEE Trans. POWER Electron., vol.
17, no. 1, pp. 94–100, 2002.
[14] H. Lev-ari, S. Member, and A. M. Stankovic, “Randomized
Modulation in Power Electronic Converters,” vol. 90, no. 5,
2002.
[15] K. Hardin, R. A. Oglesbee, and F. Fisher, “Investigation Into
the Interference Potential of Spread-Spectrum Clock
Generation to Broadband Digital Communications,” IEEE
Trans. Electromagn. Compat., vol. 45, no. 1, pp. 10–21,
2003.
[16] K. K. Tse, H. S.-H. Chung, and H. C. So, “Analysis and
Spectral Characteristics of a Spread-Spectrum Technique for
Conducted EMI Suppression,” IEEE Trans. POWER
Electron., vol. 15, no. 2, pp. 399–410, 2000.
[17] K. B. Hardin, J. T. Fessler, and D. R. Bush, “Spread
Spectrum Clock Generation for the Reduction of Radiated
Emissions,” in Proceedings of IEEE Symposium on
Electromagnetic Compatibility, 1994, pp. 227–231.
[18] J. Bojarski, R. Smolenski, P. Lezynski, and Z. Sadowski,
“Diophantine equation based model of data transmission
errors caused by interference generated by DC-DC
converters with deterministic modulation,” Bull. POLISH
Acad. Sci. Tech. Sci., vol. 64, no. 3, pp. 575–580, 2016.
[19] R. Smolenski, J. Bojarski, A. Kempski, and P. Lezynski,
“Time-Domain-Based Assessment of Data Transmission
Error Probability in Smart Grids With Electromagnetic
Interference,” IEEE Trans. Ind. Electron., vol. 61, no. 4, pp.
1882–1890, 2014.
[20] V. Adrian, J. S. Chang, B. Gwee, and S. Tedjaseputro,
“Spectral Analysis of Randomized Switching Frequency
Modulation Scheme with a Triangular Distribution for DC-DC
Converters,” in 2009 International Conference on Computing,
Engineering and Information, 2009, pp. 0–3.
[21] K. Cui, V. Adrian, Y. Sun, B. Gwee, and J. S. Chang, “A Low-
Harmonics Low-Noise Randomized Modulation Scheme for
0
200
400
600
800
1000
1200
1400
020 40 60
Total Awaiting Time in Sec
Distance (m)
Total Time in 2R Total Time in 2D
Multi-Phase DC-DC Converters,” in 15th IEEE International
New Circuits and Systems Conference (NEWCAS), 2017, pp.
165–168.
[22] S. D. Converters, K. Cui, V. Adrian, B. Gwee, S. Member,
and J. S. Chang, “A Noise-Shaped Randomized Modulation
for,” IEEE Trans. CIRCUITS Syst. Regul. Pap., pp. 1–12,
2017.
Authors: mgr inż.Waseem El Sayed, Uniwersytet Zielonogórski,
Instytut Inżynierii Elektrycznej, ul. Licealna 9, 65-417 Zielona Góra,
E-mail: waseem.elsayed@ieee.org;
mgr inż.Hermes Loschi, Uniwersytet Zielonogórski, Instytut
Inżynierii Elektrycznej, ul. Licealna 9, 65-417 Zielona Góra, E-mail:
eng.hermes.loschi@ieee.org;
prof. dr hab. inż. Robert Smolenski,Uniwersytet Zielonogórski,
Instytut Inżynierii Elektrycznej, ul. Licealna 9, 65-417 Zielona Góra,
E-mail: r.smolenski@iee.uz.zgora.pl;
dr hab. inż. P. Lezynski, Uniwersytet Zielonogórski, Instytut
Inżynierii Elektrycznej, ul. Licealna 9, 65-417 Zielona Góra, E-mail:
p.lezynski@iee.uz.zgora.pl;
mgr inż. Choon Long. Lok, Uniwersytet Zielonogórski, Instytut
Inżynierii Elektrycznej, ul. Licealna 9, 65-417 Zielona Góra, E-mail:
marshall.lok@gmail.com;