This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

A Smart Power Meter to Monitor Energy

Flow in Smart Grids:
The Role of Advanced Sensing and IoT in the Electric Grid of the Future

R. Morello, Member, IEEE, C. De Capua, Member, IEEE, G. Fulco, S.C. Mukhopadhyay, Fellow, IEEE

 power grid, named microgrids, based on local demand. So

Abstract— This paper aims to describe the role of advanced energy flow has to change dynamically even its direction. As a
sensing systems in the electric grid of the future. In detail the general rule, energy must be routed from microgrids with a
project, development and experimental validation of a smart large energy amount to microgrids having an energy lack.
power meter are described in the following. The authors provide
an outline of the potentialities of the sensing systems and IoT to
Nevertheless, several factors affect this general criterion such
monitor efficiently the energy flow among nodes of electric as the intermittent production of energy from renewable
network. The described power meter uses the metrics proposed sources. In addition, the quality of the voltage and current
in the IEEE Standard 1459-2010 to analyse and process voltage signals provides further constraints to energy routing.
and current signals. Information concerning the power Consequently, the management of energy flow becomes
consumption and power quality could allow the power grid to today a really complex task, [1]. Currently, these aspects are
route efficiently the energy by means of more suitable decision
criteria. The new scenario has changed the way to exchange
not paid with sufficient attention in the grid. As a
energy in the grid. Now energy flow must be able to change its consequence, the final user sometimes has to tolerate energy
direction according to needs. Energy cannot be now routed by having low quality. The major consequences are paid by the
considering just only the criterion based on the simple shortening domestic users. Therefore today, uninterrupted energy supply
of transmission path. So, even energy coming from a far node and high quality energy are two basic and fundamental
should be preferred if it has higher quality standards. In this requirements to be guaranteed in the transmission and
view, the proposed smart power meter intends to support the
smart power grid to monitor electricity among different nodes in
distribution of electricity.
an efficient and effective way. This new concept entails new and important challenges for
researchers dealing with this field. Several issues and
Index Terms— Electric grid, smart grid, sensing systems, problems must be faced such as the development of new
smart power meter, IoT. efficient and smart sensing systems. Contextually, electric
network needs a radical renovation to be able to change
dynamically its configuration. In fact, the current architecture
I. INTRODUCTION was projected to manage only mono-directional energy flow

I N the power grid of the future, sensors and transducers will

have a significant role to monitor energy in real time
according to demand. Smart sensing systems can provide new
from the central generation plant to the final users.
Such a new scenario requires new systems which allow the
power grid to be really smart by managing the bi-directional
opportunities for automatic power measurement and data and changing flow of energy, [1]. In addition these systems
processing so to take decisions in real time. must assure interoperability between new and old equipment.
The electric network is a complex and interconnected Figure 1 shows the current scenario, where different
system commonly called grid. Growing electricity demand distributed generation plants supply their energy to users and
needs more sustainable energy generation by renewable provide the surplus to the electric network. However, energy
sources. Today, we are observing a radical transformation of production from renewable sources suffers from supply
the public electric system. For example, energy flow becomes discontinuity. Thus, the risk of blackouts and service
bidirectional due to the presence of distributed generation inefficiency increase. The smart power grid should be able to
plants. Electricity is shared among the several nodes of the prevent promptly a supply discontinuity [2]. These features
require the use of advanced and innovative sensing systems.
R. Morello, C. De Capua and G. Fulco are with the Department of So sensors must make measurements and process results in
Information Engineering, Infrastructure and Sustainable Energy (DIIES), real time to get a clear overview of power grid state in each
University Mediterranea of Reggio Calabria, Italy (e-mail: node. For instance, power meters are sensing systems which
[email protected] [email protected] [email protected])
S.C. Mukhopadhyay is with the Department of Engineering, Macquarie are able to measure power features. Depending on its purpose,
University, NSW 2109, Australia (e-mail: [email protected]) measures can include just power consumption or additional
information concerning the power quality [3]-[5].

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

Fig. 1. The current scenario of the Power Grid [1].

The general architecture of a power meter consists of a monitor the grid by performing a distributed data processing,
voltage transducer, a current transducer, an A/D converter and [8]-[14].
a processor for processing data. At the present time, most of The project, development and experimental validation of a
the commercial power meters perform measurement of active smart power meter able to monitor the power in real-time are
power. This parameter is commonly used for computing the described in the following Sections. The next Section
user power consumption. Few power meters provide describes the proposed smart power meter. Section III reports
information on the power quality. Anyway, such information the validation and experimental results. Section IV provides a
is not used by power grid for managing the energy delivery or brief description of the application to the future power grid
billing the consumptions. In addition, in literature several based on a IoT vision. The conclusions have been drawn in
definitions of reactive power exist, as a consequence the Section V.
metrics or power computing algorithms are not worldwide
univocally defined. Therefore the concept of metric is still II. THE SMART POWER METER
object of studies and research activities. The above described future vision of smart grid needs the
In this paper, the authors propose their idea of the future project and development of innovative sensing systems with
smart power grid. In detail, power meters will be integrated in specific features. The solution proposed in the present paper is
the electric grid to provide information concerning local based on a smart power meter with improved characteristics:
energy consumption and the quality of power in the several  remotely programmable and controllable;
nodes of the electric network. Such augmented information,  interoperability among several power meters;
supported by new decision criteria, will allow the power grid
 embedded data processing and decision making
to manage fault events or rapid changes in energy
algorithms;
requirements. The electric grid can be compared with the
 power quality analysis;
internet network. Therefore Internet of Things (IoT) can
 decision-based management of energy flow routing
provide new opportunities to developers during the project of
according to the power quality requirements defined by the
smart power meters. IoT can provide new criteria for sharing
final user.
data and information into the whole grid, [6], such as multi-
The hardware architecture and the soft computing
hop communication, where each sensor communicates through
algorithms are described in the following sub-Sections. A
several successive nodes. In this sight, IoT can allow sensors
remote control station has been developed in order to manage
to share information by using internet and web service
information coming from different power meters so to
architectures so to improve the grid management, [7]. In
simulate a central management station for controlling and
addition, sensing systems must cooperate and satisfy several
performing in real-time the configuration of the power
features such as to be flexible to changing conditions, be able
network. A further sub-Section describes in brief the
to monitor and predict electrical energy consumption, and to
potentialities offered by ISO/IEC/IEEE 21451 Standards.
control the grid security. So ISO/IEC/IEEE 21451 Standards
can allow smart grid to improve its efficiency by making easy A. Hardware Architecture
the interoperability among several sensing systems due to the The smart power meter architecture is based on a National
protocol standardization. In such a scenario, the modernization Instruments Single-Board RIO 9626. Two transducers allow to
of the electric network will be possible by means of power acquire the voltage and current signals, which are successively
meters geographically distributed, which can cooperate to

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

digitally converted for data processing. In detail, the power power grid on needs (see Section II.C).
meter mounts on board two additional modules: NI-9225 and
NI 9246. A 400 MHz processor with 512 MB non-volatile B. Metrics and Signal Processing Algorithms
storage and 256 MB DRAM performs the real time data The NI Single-Board RIO 9626 has been programmed by
processing. The processor is supported by a reconfigurable using the National Instruments LabVIEW software
Xilinx Spartan-6 LX45 FPGA for custom timing, inline environment. It is a graphical programming language; the
processing, and control tasks, see Figure 2 for reference. source code has been developed with the LabVIEW Real Time
Tool. In this way, the projected smart power meter is a
standalone system able to perform the previous four
operations in real-time both on-line and off-line. The code
section concerning the data processing has been entirely
developed by using all metrics suggested in the IEEE
Std.1459-2010, [21],[22]. Lastly, a Fast Fourier Transform
(FFT) algorithm allows to evaluate the harmonic content of
the voltage and current signals. All computed parameters are
reported with more detail in Table II.

TABLE II
COMPUTED PARAMETERS
Measurement
Fig. 2. The Smart Power Meter. Parameter Description Unit
PC Active Power Consumption per hour kW/h
Table I reports the main electrical and technical QC Reactive Power Consumption per hour var/h
P1 Fundamental Active Power W
specifications of the power meter. PH Harmonic Active Power W
P Total Active Power W
TABLE I Q1 Fundamental Reactive Power var
SMART POWER METER SPECIFICATIONS S Apparent Power VA
S1 Fundamental Apparent Power VA
Quantity Value SH Harmonic Apparent Power VA
DI Current Distortion Power var
Voltage Range 0-300 Vrms DV Voltage Distortion Power var
Current Range 0-20 Arms SN non-Fundamental Apparent Power VA
Peak Current 30 A N non-Active Power var
Maximum Sampling Frequency 50 kS/s PF Power Factor -
Resolution 24 bit HP Harmonic Pollution -
Temperature Operating Range -40 - +70 °C PF1 Fundamental Power Factor -
Dimension 15.4x10.3x5 cm THDV Voltage Total Harmonic Distortion -
Mass 350 g THDI Current Total Harmonic Distortion -
Supply Voltage 9-30 V k Crest Factor -
f Frequency Hz
The power meter is compliant with the specifications Vrms root mean square Voltage V
Vpk peak Voltage V
reported in the guidelines of IEC 61000-4 Standards family
V1 Fundamental Voltage V
[15]-[17], and of IEC 62052-11 and IEC 62053-21 Standards VH Harmonic Voltage V
[18], [19]. The technical specifications in Table I and the data Vrms,i root mean square Voltage of i-th V
processing algorithms allow the performed measurements to harmonic with 2<i<40
Irms root mean square Current A
meet the specifications for the range of uncertainty defined for Ipk peak Current A
metering instruments of class A [17]. I1 Fundamental Current A
The meter performs in the following order these operations: IH Harmonic Current A
root mean square Current of i-th
1. synchronous acquisition of voltage and current Irms,i
harmonic with 2<i<40
A
waveforms with a sampling rate of 5 kS/s;
2. FFT calculation of voltage and current waveforms (see The previous parameters allow the meter to provide a
Section II.B); complete overview about the energy flowing in a specific node
3. evaluation of several power and electrical parameters of the electric grid. Figure 3 shows, as an example, a section
according to embedded metrics in compliance with the of the developed code.
IEEE Std.1459-2010 [20] (see Section II.B); Data concerning voltage and current signals, frequency,
4. characterization of power quality disturbance events. power consumption and power quality is stored and made
The detected events are stored and can be used by the accessible to a remote control station for decision making
remote control station for managing and configuring the purpose.

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

from or toward a specific node depending on needs.

Fig. 3. A detail of the Source Code.

Each record includes date and time of the event, type of

disturbance (sag, swell or interruption), maximum value over
the threshold, duration of the event. In detail, the power meter Fig. 4. Screenshot of the Remote Control Panel.
compares the measurement results with user-defined
thresholds in order to characterize specific stationary and The Control Station is configured to communicate with
transient events or supply discontinuities (voltage swell, each smart meter of the grid every hour to reduce the network
voltage dip, overvoltage, undervoltage, voltage sags, micro- congestion. It is the standard time interval used to evaluate the
outages, voltage fluctuations, short and long breaks, impulsive power consumption. However, depending on needs, this time
overvoltage, over-current, blackouts, etc…) so to send a interval can be decreased or increased. To guarantee the
warning or an alert message to the remote control station if interoperability among the meters, embedded decision criteria
necessary. allow each meter to characterize the occurrence of specific
In addition, the embedded metrics allow the smart meter to faults or inefficiency conditions of the power grid. In detail,
characterize the bi-directional power flow through the node. In the meter puts constantly in comparison the measurement
detail, by considering the current sign, the meter is able to results of each parameter in Table II with user-defined
distinguish if the power is supplied by the node to the other reference values. When a threshold is overcome, the meter
ones (power production) or if it is consumed by the node alerts the Control Station. That can occur for example when
(power consumption). Such information provides important quality standards go down fixed tolerable limits, or when a
evidence about the power flow in the grid from microgrids blackout occurs, or when power consumption of the node
with a large energy amount to microgrids having an energy overcomes the power supply. Successively, the meters in the
lack. neighbouring nodes are demanded to synchronize their
measurements. Results are sent to the Remote Control Station
C. Remote Control Station for processing data. Information on power consumption and
power quality allows the grid control center to manage
The power meter has been projected in order to permit the
efficiently the energy routing by acting on actuators located in
interoperability among several meters geographically
the nodes so to configure the electric network according to
distributed in the grid. For this reason, a software-based
needs. For an instance, microgrids which supply energy with
control panel has been developed to make possible the
poor quality can be isolate, or nodes with a large power
communication with each meter. The program runs on a server
amount are connected with nodes having a power lack. All
which could simulate the control center of a smart power grid.
that can happen dynamically when network faults,
By internet network or the same electric network, the
malfunctions or disruptions occur.
control station can get access simultaneously to several meters
To improve the interoperability features, the projected smart
of the grid acquiring the computed data. The control station
meter is even able to communicate directly with the other
can even reconfigure or reprogram the single meter if
neighbouring meters so to demand power measurements or to
required. Figure 4 shows a screenshot of the control panel.
synchronize them. These features can be configured according
By means of the control panel, the remote control station
to the power grid requirements. Since the specific application
gets a clear overview of the grid state in each node.
case refers to a small-sized power grid, the control and
Information concerning the voltage and current waveforms,
communication rights have been exclusively assigned to the
power quality, stationary and transient events or supply
Remote Control Station. So the single smart meter is
discontinuities provide to the control center an instantaneous
configured to communicate only with the control center.
snapshot of the grid so to manage in real time the energy
However, when the power grid size increases, it could be
routing along specific and suitable paths. In this way,
preferable to transfer specific communication and control
information collected by the smart meter is used to configure
rights to peripheral meters so to decongest the network and to
the grid, so to manage the power flow in a bi-directional way
decentralize the grid management task.

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

D. ISO/IEC/IEEE 21451-x Standards validation, see Section III.C).

The projected power meter is compliant with the guidelines
of the IEEE 21451 Standards family so to provide a network- A. Voltage and current transducers testing
independent communication interface. This aspect becomes To check the accuracy of the two transducers, the test bench
basic when we consider that several smart transducers and configuration in Figure 5 has been used. In detail, a Calibrator
sensors will be dislocated along the power grid. Therefore, to FLUKE 5500A has been configured to test the calibration
guarantee the interoperability among the several sensing curve of each transducer. The environment temperature has
systems, the project and development of any device need been controlled and kept constant to 25 °C for the whole test.
standardization. The growing demand and interest in smart Several sinusoidal voltage and current waveforms have been
sensing systems has induced Working Groups of experts to generated with a frequency of 50 Hz and with steps of 10 V
revise the family of ISO/IEC/IEEE 21451-x Standards with and 1 A of rms amplitude, respectively, in compliance with
the joint effort of ISO/IEC/JTC1. The aim of ISO/IEC/IEEE the respective measurement ranges, see Table I.
21451-x Standards family is to provide a guideline for
projecting smart transducer interfaces and smart sensor
networks, [23]-[27]. The Standards allow users and designers
to project smart sensing systems by using different protocols,
such as eXtensible Messaging and Presence Protocol (XMPP),
TCP/IP, HTTP, and Web services so to make easy
communication among sensors and/or actuators distributed in
a wide sensing network. Transducer Electronic Data Sheets
(TEDS) are used for sensor identification and configuration
purpose. Additional Standards of ISO/IEC/IEEE 21451-x
family deal with the signal treatment. In such a scenario, the
project of sensing systems for smart grid needs specific
attention. For this reason, the remote control panel in the
Section II.C has been developed to implement a Network Fig. 5. Overview of the first test bench configuration.
Capable Application Processor (NCAP). The NCAP performs
the following functions: Results are reported in Tables III and IV.
 Transducer Interface Module (TIM) Discovery;
 Transducer Electronic Data Sheet (TEDS) Reading; TABLE III
VOLTAGE TRANSDUCER CALIBRATION CURVE (50 HZ)
 Transducer Data Reading.
Reference Value Measured Value Percentage Deviation
The TIM Discovery function allows individuating the [V] [V] %
available TIMs and automatically adding them to the list of 10 9.9978 0.0220
the installed power meters [28]. In this way, it is possible to 20 19.9957 0.0215
expand the meter network according to needs. In the TEDS of 30 29.9942 0.0193
each smart power meter are stored data concerning its 40 39.9856 0.0360
identification, its geographical location, technical 50 49.9895 0.0210
60 59.9842 0.0263
specifications, last calibration, next calibration interval.
70 69.9826 0.0248
80 79.9865 0.0168
90 89.9810 0.0211
III. VALIDATION AND EXPERIMENTAL RESULTS 100 99.9811 0.0189
110 109.976 0.0218
In this Section, the tests performed to validate the above 120 119.976 0.0200
described smart power meter are reported. Two test bench 130 129.974 0.0200
configurations have been developed to execute three different 140 139.980 0.0142
test sets. In detail, a preliminary test set has allowed us to 150 149.971 0.0193
160 159.970 0.0187
validate the measurement results provided by the voltage and
170 169.981 0.0111
current transducers (hardware testing, see Section III.A). A 180 179.973 0.0150
second test set has been performed to check the FFT algorithm 190 189.963 0.0194
so to evaluate the meter capacity to discriminate the several 200 199.973 0.0135
harmonic contributions of the voltage and current signals 210 209.973 0.0128
220 219.966 0.0154
(software testing, see Section III.B). And then finally, a further
230 229.969 0.0134
experimentation has been performed on a real application case 240 239.964 0.0150
to check the precision of the embedded metrics and the 250 249.963 0.0148
measurement accuracy of the power meter (experimental 260 259.957 0.0165

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

270 269.963 0.0137 9 450 230 229.946 0.0234

280 279.955 0.0160 10 500 230 229.935 0.0282
290 289.989 0.0037
300 299.955 0.0150
TABLE VI
FFT ALGORITHM TESTING (CURRENT WAVEFORM)
TABLE IV Reference Measured Percentage
Harmonic Frequency
CURRENT TRANSDUCER CALIBRATION CURVE (50 HZ) Value Value Deviation
Order [Hz]
[A] [A] %
Reference Value Measured Value Percentage Deviation 2 100 5 4.9985 0.030
[A] [A] % 3 150 5 4.9983 0.034
1 1.0001 0.0100 4 200 5 4.9952 0.096
2 2.0002 0.0100 5 250 5 4.9986 0.028
3 2.9992 0.0266 6 300 5 4.9985 0.030
4 3.9988 0.0300 7 350 5 4.9986 0.028
8 400 5 4.9982 0.036
5 4.9984 0.0320
9 450 5 4.9981 0.038
6 5.9983 0.0283 10 500 5 4.9979 0.042
7 6.9982 0.0257
8 7.9962 0.0475
9 8.9943 0.0633
10 9.9970 0.0300
C. Experimental Results
11 10.9939 0.0554 An additional experimental analysis has been performed by
considering a specific application case. The used test bench
configuration is showed in Figure 6.
The results show a maximum percentage deviation equal to
0.036% for the voltage calibration curve and 0.0633% for the
current calibration curve. The estimated voltage and current
offset values are 0.001 V and 0.0014 A, respectively. Such
results are compliant with the IEC requirements concerning
the electricity metering equipment so confirming the class A
for the projected power meter [17]-[19].

B. FFT and Harmonics Detection testing

The test bench in Figure 5 has been furthermore used to
check the accuracy of the harmonics detection performed by
the FFT algorithm. The Calibrator has been programmed to
generate six sinusoidal voltage and current waveforms with
different frequency values. For each waveform, the voltage
amplitude has been set equal to 230 Vrms with a current
amplitude of 5 Arms. The harmonics until the seventh order
have been considered for this test, since they are the
harmonics which occur frequently in the real case. The results
are showed in Tables V and VI for the voltage and current
waveforms, respectively. The maximum percentage deviation
obtained for the voltage waveform is equal to 0.0282% and
Fig. 6. Overview of the second test bench configuration.
equal to 0.096% for the current waveform. The values show a
good accuracy of the harmonics detection algorithm to discern
the harmonic content for each waveform. The test equipment consists of an AC Power Source Pacific
360-AMX with programmable controller, a Precision Power
TABLE V
Analyzer Yokogawa WT1800, an electric motor used as load, a
FFT ALGORITHM TESTING (VOLTAGE WAVEFORM) Hysteresis Dynamometer Magtrol HD-715-8NA with a
Reference Measured Percentage Dynamometer Controller Magtrol DSP6001.
Harmonic Frequency
Order [Hz]
Value Value Deviation The load has been supplied by applying a sinusoidal voltage
[V] [V] %
of 225 Vrms amplitude and a frequency of 50 Hz generated by
2 100 230 229.945 0.0239
3 150 230 229.959 0.0178 the power source. Voltage harmonic components until the
4 200 230 229.956 0.0191 seventh order have been added to the voltage signal in order to
5 250 230 229.951 0.0213 simulate non-sinusoidal operating conditions. Each harmonic
6 300 230 229.951 0.0213
7 350 230 229.950 0.0217 has been generated with an amplitude equal to 30% of
8 400 230 229.944 0.0243 fundamental component amplitude. The voltage and current

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

waveforms are shown in the Figures 7 and 8, respectively.

Fig. 7. Voltage waveform.

Fig. 10. Control Panel of the Remote Control Station.

Table VII reports the results of the experimental

comparison.

Fig. 8. Current Waveform. TABLE VII

EXPERIMENTAL VALIDATION RESULTS OF THE SMART POWER METER
The harmonic components of the voltage and current Percentage
waveforms are depicted in Figure 9. Parameter Reference Value Measured Value Deviation
%
PC [kW/h] - n.a. -
QC [var/h] - n.a. -
P1 [W] 530.600 527.682 0.5499
PH [W] 214.800 200.293 6.7537
P [W] 745.400 727.975 2.3376
Q1 [var] 860.600* 555.182* 35.4889*
S [VA] 1139.700 1104.911 3.0524
S1 [VA] 778.4200 765.947 1.6023
SH [VA] 347.8185 338.423 2.7012
DI [var] 507.8204 492.139 3.0879
DV [var] 533.1599 526.710 1.2097
SN [VA] 832.4532 796.338 4.3384
N [var] 860.600 831.193 3.4170
PF 0.6556 0.659 0.5186
HP 1.06 1.040 1.8867
PF1 0.6816 0.6890 1.0856
THDV 56.661 % 68.766 % 21.3639
Fig. 9. Harmonics components of the voltage and current signals. THDI 56.456 % 64.252 % 13.8089
k - 1.2045 -
f [Hz] 50.0020 50.0014 0.0011
By analysing the previous figure, it is possible to observe Vrms [V] 223.0200 222.331 0.3089
the presence of harmonic components beyond the seventh Vpk [V] 417.6800 415.8700 0.4333
order. That is the result of the distortion introduced by the V1 [V] 183.7200 183.1960 0.2852
load. The control panel displayed by the remote control station VH [V] 125.8343 125.9770 0.1134
is reported in Figure 10. The panel shows all parameters Vrms,i [V] 1) 183.72 1) 183.1960 1) 0.2852
with 1<i<8 2) 52.78 2) 52.8025 2) 0.0426
measured by the smart power meter as reported in Table II. 3) 50.77 3) 50.6235 3) 0.2885
Each measured value has been put in comparison with the 4) 50.78 4) 50.6606 4) 0.2351
value provided by the power analyser, which has been 5) 50.42 5) 50.3228 5) 0.1927
6) 51.36 6) 51.1990 6) 0.3134
considered as a reference for this experimentation.

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

7) 52.08 7) 51.9397 7) 0.2693 help to share information and data in the grid so to improve
8) 2.34 8) 2.32733 8) 0.5414
Irms [A] 5.110 4.970 2.7397
efficiency, reliability and security of the electric system [6].
Ipk [A] 6.235 6.025 3.3680 IoT aims to add value by connecting objects to internet
I1 [A] 4.237 4.1810 1.3216 network. So IoT implies that power meters can be able to
IH [A] 2.7641 2.686 2.8255 utilize internet to communicate data about their condition,
Irms,i [A] 1) 4.237 1) 4.1810 1) 1.3216 position, or other measurement parameters. Therefore smart
with 1<i<8 2) 1.615 2) 1.5485 2) 4.1176
3) 1.078 3) 1.0136 3) 5.9740 power meters will take advantage of this by using the internet
4) 1.176 4) 1.1019 4) 6.3010 network so making available data and information on the
5) 0.943 5) 0.8724 5) 7.4867 power grid with a new approach in respect to the past. Power
6) 0.941 6) 0.8688 6) 7.6726
7) 0.844 7) 0.7722 7) 8.5071
meters can even monitor constantly the power line
8) 0.495 8) 0.4260 8) 13.9393 temperature so to estimate the carrying capacity of the line.
- value not reported Such information can be used to manage dynamically the
* value obtained by using a different parameter definition
power flow amount by using suitable dynamic line rating
algorithms. Therefore, IoT can allow the smart power grid to
The analysis of the results does not allow us to make a
increase its features and services. In this new scenario, IoT
complete comparison between the two measurement systems,
promises to turn a power meter into an object which provides
since the projected smart power meter integrates a major
information about the grid and its environment. This will
number of metrics. In addition the two measurement systems
create in the next future a new way to differentiate the services
use a different definition of the reactive power, as a
of the electric network and a new source of value. This IoT
consequence the parameter Q1 is not comparable. Anyway, by
vision will improve the efficiency of the power grid and will
considering the only parameters in common, a significant
provide new opportunities. The user will be able to define and
percentage deviation has been obtained for the Harmonic
change his/her power requirements, and the smart power grid
Active Power parameter. It is due to an expected systematic
will configure itself to assure the required power quality
error caused by the voltamperometric connection of the two
specifications.
instruments. This is cause of error on the measurement of the
The main issue to be faced in the next future concerns the
harmonic current components. It is useful to observe that the
large number of power meters and sensors to be managed and
results of the FFT algorithm testing reported in the Section
maintained. Typically, an electric grid consists of 10,000 or
III.B have shown a maximum percentage deviation equal to
even 100,000 nodes. Consequently, the scalability issue
0.096%. This test has been performed by using a different test
should be resolved before considering a power grid to be
bench configuration, which has avoided the above described
smart. The collaborative signal processing among nodes is
error. Consequently, the percentage deviations of the harmonic
another important aspect. Even the querying ability is relevant
currents in Table VII are attributable to the voltamperometric
to the electric grid of the future. A node or a power meter must
method error.
be able to query an individual node or group of nodes for
getting information concerning a specific microgrid. All these
IV. TOWARDS THE GRID OF THE FUTURE: THE IOT VISION aspects have been formerly faced by the internet network. So
the main features and characteristics today requested to the
It is expected that the electric grid of future will be a
power grid are the same ones owned by the present internet
complex flow of energy and information shared among several
network. Therefore, by a IoT vision, the power grid will be
nodes. New sensing systems and services will be necessary to
able to perform its tasks: demand management, disturbances
manage a so complex distributed system, [29]. Even protocols
detection, energy flow amount management, isolation of
and standards will have an important role. The power grid can
specific microgrids, management of the energy storage,
be compared to the internet network. Each node of the grid
transport of energy from any node it is produced to nodes
can be equipped with a power meter. The resulting sensing
where it is lacking by using innovative routing algorithms. In
grid can be considered as a complex system geographically
the view of the future power grid, smart power meters could
distributed. The power meter network will have the task to
resolve all the current difficulties concerning the sensing and
monitor in real time the energy flow of a large number of
measurement issues.
nodes. Such amount of information is used to make timely
The projected smart power meter takes advantage of IoT
decisions when critical events occur. By gaining experience
concept. In fact, the current power meters are able to share
with the evolution of the internet network in the last years,
information only with the control centre of the electric
such described scenario offers a large number of research
company just to bill the user power consumption. With a
topics. The internet protocol is universal and has been widely
different approach, the proposed power meter shares
validated in the years. As a consequence, the real time control
information on consumption and power quality with the
of the power grid could take advantage of internet so to use
internet network to improve the management of the power
power meter data for taking timely decisions and configuring
grid. The power meter cannot be anymore considered as a
itself based on needs. Internet of Things (IoT) concept can
simple instrument billing consumption. By such a IoT vision,

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

information on each node of the grid is shared with the whole meter. The project and development of the proposed power
grid to increase its efficiency. Measurement data is used to bill meter have been described in the paper. Hardware architecture
consumption but at the same time to configure the power and software algorithms have been outlined. In detail, the
network based on power demands and on the power quality embedded metrics offer additional information on the energy
requirements defined by the user. In this view, the described flowing in the grid nodes in respect to the other commercial
power meter implements the IoT concept to improve the power meters. The main strengths concern:
features and the services offered by the future power grid.  the possibility to control and program remotely the
However, several further issues remain still unsolved. In meter;
detail, the new challenges include:  data are processed in real time to support the decision
 the standardization of communication protocols; making tasks;
 the improvement of the security standards;  the remote control station is able to manage
 integration of the sensing systems into existing systems simultaneously several smart meters;
to assure interoperability;  the embedded metrics offer new potential criteria to
 harmonisation of equipment standards to allow plug-and- manage efficiently the energy routing and sharing among
play and interface; several nodes by considering even the power quality
 new power flow routing algorithms and innovative features.
routing criteria; Tests and experimentation on an application case have
 management of big data coming from thousands of allowed to validate the developed prototype. Two test benches
sensing systems distributed through the grid; have been used for testing the hardware and software
 redefinition of the metrics used for billing consumptions; operations. The additional experimental results have proved
 modernisation of current electric network architecture. the compliance of the power meter with the IEC requirements.

REFERENCES
V. ACKNOWLEDGMENT
[1] Smart Grids, European Technology Platform, Vision and Strategy of
The described research activity is part of the Project Europe's Electricity Networks of the Future, European Commission
“Laboratorio RENEW-MEL” which has been funded, through Document EUR 22040, Directorate-General for Research, 2006.
[2] IEEE Smart Cities, Accessed on 10th July, 2017. [Online]. Available:
the PON Program, by the Italian Ministry of Education, http://smartgrid.ieee.org/.
Universities and Research (MIUR) and by the European [3] L. Morales-Velazquez, R. de Jesus Romero-Troncoso, G. Herrera-Ruiz,
Commission (contract no. PON03PE_000122 & D. Morinigo-Sotelo, R. Alfredo Osornio-Rios, “Smart sensor network
for power quality monitoring in electrical installations”, Measurement,
PON03PE_00012). Vol. 103, pp. 133-142, 2017.
[4] A. Cataliotti, V. Cosentino, D. Di Cara and G. Tinè, “LV Measurement
Device Placement for Load Flow Analysis in MV Smart Grids”, in IEEE
VI. CONCLUSIONS Transactions on Instrumentation and Measurement, Vol. 65, no. 5, pp.
999-1006, May 2016.
The paper proposes an innovative smart power meter to [5] M. M. Albu, M. Sănduleac and C. Stănescu, “Syncretic Use of Smart
monitor the energy flow in smart power grids. Expecting the Meters for Power Quality Monitoring in Emerging Networks”, in IEEE
power grid of the future, the authors take stock of the situation Transactions on Smart Grid, Vol. 8, no. 1, pp. 485-492, Jan. 2017.
[6] http://iot.ieee.org/
concerning the state of the current power grid. Weaknesses [7] S. Uludag, K. S. Lui, W. Ren and K. Nahrstedt, “Secure and Scalable
and strengths are discussed so to highlight the role of the Data Collection with Time Minimization in the Smart Grid”, in IEEE
advanced sensing systems in the power grid management. Transactions on Smart Grid, vol. 7, no. 1, pp. 43-54, January 2016.
[8] B. Kul, “IP-based smart sensors for energy metering and efficient
Other issues remain still unsolved before that the power grid HVAC infrastructure in buildings”, 2017 15th International Conference
can be considered really smart. IoT offers new interesting on Electrical Machines, Drives and Power Systems (ELMA), Sofia,
answers in order to face and resolve such issues. So this paper Bulgaria, pp. 258-261, 2017.
intends to solicit the debate on the role of IoT in the [9] M. Burunkaya, T. Pars, “A smart meter design and implementation
using ZigBee based Wireless Sensor Network in Smart Grid”, 2017 4th
development of the power grid of the next future. The current International Conference on Electrical and Electronic Engineering
power network needs to be updated in order to adapt itself to (ICEEE), Ankara, pp. 158-162, 2017.
the new requirements of the current power demand. As a [10] Y. Kabalci, “A survey on smart metering and smart grid
communication”, Renewable and Sustainable Energy Reviews, Vol. 57,
consequence, there is still much to be done especially in the pp. 302-318, May 2016.
matter of defining the most suitable architecture of the electric [11] L. I. Minchala-Avila, J. Armijos, D. Pesántez, Y. Zhang, “Design and
network. By means of the proposed IoT vision, the projected Implementation of a Smart Meter with Demand Response Capabilities”,
Energy Procedia, Vol. 103, pp. 195-200, December 2016.
smart power meter uses the internet network to improve the [12] K. Sharma, L. M. Saini, “Performance analysis of smart metering for
efficiency and features of the power grid. smart grid: An overview”, Renewable and Sustainable Energy Reviews,
The paper aims to propose a possible solution to the issues Vol. 49, pp. 720-735, September 2015.
[13] D. Ramírez Muñoz, D. Moro Pérez, J. Sánchez Moreno, S. Casans
concerning the sensing and measurement aspects by Berga, E. Castro Montero, “Design and experimental verification of a
discussing the potentialities of the developed smart power

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2017.2760014, IEEE
Sensors Journal

smart sensor to measure the energy and power consumption in a one- monitoring, decision-making problems and measurement uncertainty, process
phase AC line”, Measurement, Vol. 42, Issue 3, pp. 412-419, 2009. quality assurance, instrumentation reliability and calibration, energy, smart
[14] N. K. Suryadevara, S. C. Mukhopadhyay, S. Dieter, T. Kelly, S. P. S. grids, battery testing, biomedical applications and statistical signal processing,
Gill, “WSN-Based Smart Sensors and Actuator for Power Management non-invasive systems, biotechnologies and measurement, instrumentation and
in Intelligent Buildings”, in IEEE/ASME Transactions on Mechatronics, methodologies related to Healthcare. Dr. Morello is a member of the Italian
vol. 20, no. 2, pp. 564-571, April 2015. Group of Electrical, Electronic Measurements (GMEE) and IEEE.
[15] IEC 61000-4-7 International Standard, “Electromagnetic compatibility
(EMC) – Part 4-7: Testing and measurement techniques – General guide
on harmonics and interharmonics measurements and instrumentation, for Claudio De Capua (M’99) received the M.S. and the
power supply systems and equipment connected thereto”, 8 August Ph.D. degrees in Electrical Engineering from the
2002. University of Naples "Federico II", Naples, Italy. Since
[16] IEC 61000-4-15 International Standard, “Electromagnetic compatibility 2012, he is Full Professor of Electrical and Electronic
(EMC) – Part 4: Testing and measurement techniques – Section 15: Measurements at the Department of Information
Flickermeter – Functional and design specifications”, 24 August 2010. Engineering, Infrastructure and Sustainable Energy,
[17] IEC 61000-4-30 International Standard, “Electromagnetic compatibility University “Mediterranea” of Reggio Calabria. His
(EMC) – Part 4-30: Testing and measurement techniques – Power current research includes the design, realization and
quality measurement methods”, 20 February 2015. metrological performance improvement of the automatic
[18] IEC 62052-11 International Standard, “Electricity metering equipment measurement systems; web sensors and sensor data
(ac) – General requirements, tests and test conditions – Part 11: fusion; biomedical instrumentation; techniques for remote didactic laboratory;
Metering equipment”, 12 February 2003. measurement uncertainty analysis; problems of electromagnetic compatibility
[19] IEC 62053-21 International Standard, “Electricity metering equipment in measurements.
(ac) – Particular requirements – Part 21: Static meters for active energy Prof. De Capua is member of the Italian Group of Electrical and Electronic
(classes 1 and 2)”, 28 January 2003. Measurements (GMEE).
[20] IEEE 1459-2010: Definitions for the Measurement of Electric Power
Quantities Under Sinusoidal, Nonsinusoidal, Balanced, or Unbalanced
Conditions, 19 March 2010. Gaetano Fulco was born in Reggio Calabria, Italy, in
[21] A. Kumar, I.P. Singh, S.K. Sud, “Energy Efficient and Low-Cost Indoor 1990. He received from the University “Mediterranea”
Environment Monitoring System Based on the IEEE 1451 Standard”, of Reggio Calabria, Italy, a bachelor's degree in
IEEE Sensors Journal, Vol.11, Issue 10, pp.2598-2610, 2011. Electronic Engineering, in 2013, and the M.Sc. Degree
[22] J. Guevara, A. Fatecha, E. Vargas, F. Barrero, “A reconfigurable WSN (cum laude) in Electronic Engineering, in 2015. At the
node based on ISO/IEC/IEEE 21451 standard”, Proc. of IEEE present he is Ph.D. Student of Information Engineering
International Instrumentation and Measurement Technology Conference Course at the University “Mediterranea” of Reggio
(I2MTC), pp.873-877, 2014. Calabria, Italy. His main research interests are
[23] ISO/IEC/IEEE 21450: Information technology — Smart transducer development, realization and test of smart meter, power
interface for sensors and actuators — Common functions, quality, energy, smart grids.
communication protocols, and Transducer Electronic Data Sheet
(TEDS) formats, 2010.
[24] ISO/IEC/IEEE 21451-1: Information technology -- Smart transducer Subhas Mukhopadhyay (M’97, SM’02, F’11) holds a
interface for sensors and actuators - Part 1: Network Capable B.E.E. (gold medalist), M.E.E., Ph.D. (India) and
Application Processor (NCAP) information model, 2010. Doctor of Engineering (Japan). He has over 26 years of
[25] ISO/IEC/IEEE 21451-2: Information technology -- Smart transducer teaching, industrial and research experience.
interface for sensors and actuators - Part 2: Transducer to Currently he is working as a Professor of
microprocessor communication protocols and Transducer Electronic Mechanical/Electronics Engineering, Macquarie
Data Sheet (TEDS) formats, 2010. University, Australia and is Discipline Leader of the
[26] ISO/IEC/IEEE 21451-4: Information technology — Smart transducer Mechatronics Engineering Degree Programme. Before
interface for sensors and actuators — Part 4: Mixed-mode joining Macquarie he worked as Professor of Sensing
communication protocols and Transducer Electronic Data Sheet (TEDS) Technology, Massey University, New Zealand. His
formats, 2010. fields of interest include Smart Sensors and sensing technology,
[27] ISO/IEC/IEEE 21451-7: Information technology — Smart transducer instrumentation techniques, wireless sensors and network, numerical field
interface for sensors and actuators — Part 7: Transducer to radio calculation, electromagnetics etc. He has supervised over 40 postgraduate
frequency identification (RFID) systems communication protocols and students and over 100 Honours students. He has examined over 50
Transducer Electronic Data Sheet (TEDS) formats, 2011. postgraduate theses.
[28] R. Morello, C. De Capua, G. Lipari, M. Lugarà, G. Morabito, “A Smart He has published over 400 papers in different international journals and
Energy Meter for Power Grids”, 2014 IEEE International conference proceedings, written six books and thirty book chapters and edited
Instrumentation and Measurement Technology Conference (I2MTC fifteen conference proceedings. He has also edited twenty-eight books with
2014), May 12-15 2014, Montevideo, Uruguay, pp. 878-883, 2014. Springer-Verlag and seveneen journal special issues. He has organized over
[29] L. Ferrigno, R. Morello, V. Paciello, A. Pietrosanto, “Remote metering 20 international conferences as either General Chairs/co-chairs or Technical
in public networks”, Metrology and Measurement Systems, Vol. 20, Programme Chair. He has delivered 298 presentations including keynote,
Issue 4, pp. 705–714, October 2013. invited, tutorial and special lectures.
He is a Fellow of IET (UK), a Fellow of IETE (India), a Topical Editor of
Rosario Morello (M’03) was born in Reggio Calabria, IEEE Sensors journal, and an associate editor of IEEE Transactions on
Italy, in 1978. He received the M.Sc. Degree (cum Instrumentation and Measurements. He is a Distinguished Lecturer of the
laude) in Electronic Engineering and the Ph.D. Degree IEEE Sensors Council from 2017 to 2019. He chairs the IEEE IMS Technical
in Electrical and Automation Engineering from the Committee 18 on Environmental Measurements.
University “Mediterranea” of Reggio Calabria, Italy, in More details can be available at
2002 and 2006, respectively. Since 2005, he has been http://web.science.mq.edu.au/directory/listing/person.htm?id=smukhopa 
Postdoctoral Researcher of Electrical and Electronic
Measurements at the Department of Information
Engineering, Infrastructure and Sustainable Energy of
the same University. At the present he is an Assistant Professor. His main
research interests include the design and characterization of distributed and
intelligent measurement systems, wireless sensor network, environmental

1558-1748 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.