Wireless Sensor Networks for Smart Micro-Grids

Abdellatif Elmouatamid, Elkamoun Najib, Lakrami Fatima, Bakhouya

Mohamed

To cite this version:

Abdellatif Elmouatamid, Elkamoun Najib, Lakrami Fatima, Bakhouya Mohamed. Wireless Sensor
Networks for Smart Micro-Grids. Conference, May 2018, El Jadida, Morocco. �hal-04607723�

HAL Id: hal-04607723

https://hal.science/hal-04607723v1
Submitted on 11 Jun 2024

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.

Open licence - etalab

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/329178819

Wireless Sensor Networks for Smart MicroGrids

Conference Paper · May 2018

CITATIONS READS

0 933

1 author:

Najib EL KAMOUN
Université Chouaib Doukkali
207 PUBLICATIONS 1,200 CITATIONS

SEE PROFILE

All content following this page was uploaded by Najib EL KAMOUN on 25 November 2018.

The user has requested enhancement of the downloaded file.

 Wireless Sensor Networks for Smart Micro-
Grids
A. Elmouatamid 1, 2, N. Elkamoun1, F. Lakrami1, M. Bakhouya2
1
Chouaib Doukkali University
Faculty of Sciences, STIC Lab, El Jadida, Morocco
elmouatamid13@gmail.com, elkamoun.n@ucd.ac.ma, fatima.lakrami@gmail.com
2
International University of Rabat
Faculty of Computing and Logistics, TICLab, Sala El Jadida, Morocco
{abdellatif.elmouatamid, mohamed.bakhouya}@uir.ac.ma

Abstract— Smart micro-grid has emerged as a new infrastructure mid to high data rates (e.g., voice, video…). However, up till
to further enhance existing power grid through the integration of now, there is only ZigBee that could meet WSN requirements
renewable energy sources and communication technologies. In for micro-grid. In fact, ZigBee was deployed (e.g. Xbee) in
fact, smart micro-grid infrastructure is characterized by several applications that does not require high data rates but
renewable energy integration, automatic metering
does require low cost and low power. This proprietary system
infrastructure, distribution automation and complete monitoring
and control of the entire power grid. The monitoring is based on was designed because there were no standards that meet these
a set of sensors, which are used for collecting data production, requirements. The ZigBee is not pushing a technology, rather
data consumption and weather conditions. These data should be it is providing a standardized base set of solutions for WSNs
transmitted by a wireless network to a main location or sink for [1, 2]. This work provides first an introduction to WSNs and
visualization and analysis. For this reasons, it is important to their applications in the smart-grid. A performance study of
develop a self-configured wireless sensor network able to handle different communication topologies using ZigBee technology
efficiently the massive amount of data. This paper investigates the was then conducted and results are reported to show their
ZigBee network in the context of micro grid by comparing two efficiency in SMG.
routing topologies.
Keywords- Wireless Sensor Networks, Smart Micro-Grids, Data II. RELATED WORK
Collection, ZigBee, Routing Topologies.
Recent research work in WSN focused mainly on the
I. INTRODUCTION development of a wireless communication with low-cost and
Smart micro-grid (SMG) communication infrastructure is very low power protocols communication. The aim is to
a set of Wireless Sensor Networks (WSNs) that collect extend the life of the network while improving their bandwidth
different data (e.g., power production, consumption, weather management and processing speed. For instance, authors in
conditions…). These data are analyzed for controlling the [15] presented different factors (e.g., energy efficiency), which
power flow and balancing the power Demand/Response in need to be considered for designing WSN. They studied
order to avoid the blackout problem. The data are transmitted several constraints, such as topology change, scalability and
by different sensors’ nodes to a base station named Sink. In power consumption in order to build a WSN with high
fact, the sink acts like an interface between the users and the flexibility and reliability for almost all applications’
network and the nodes sense their environment (e.g., power requirements. Authors in [16] discussed and analyzed various
production, temperature) and route sensed data from one node wireless communication standards, cyber security issues and
to another until it reach the sink node (see Figure 1). solutions for WSN and explored topology control for cyber
Sensor nodes can wirelessly communicate with other its security especially for Smart Grid.
neighboring nodes via radio signals; it is equipped with radio A smart grid communications system is presented in [3, 4,
transceivers, sensing/computing devices, and a power supplier 13, 14], authors explored three major systems, namely the
component. However, their efficient deployment is an smart infrastructure system, the smart management system,
important issue that needs to be considered. In fact, after their and the smart protection system. For the smart protection
deployment, they have to self-configure and self-organize in system, authors explored various failure protection
order to form an appropriate communication infrastructure. mechanisms, which improve the reliability and the flexibility
These sensors are battery powered devices with limited of the Smart Grid. For the smart infrastructure system, authors
resources (i.e., processing speed, communication bandwidth explored the smart information subsystem, the smart energy
and storage capacity), and therefore, these characteristics need subsystem and the smart communication subsystem. For the
to be taken into consideration when developing this type of smart management system, they have explored various
wireless communication network, especially in the context of management objectives, such as improving energy efficiency,
micro-grid, which requires extensive communication among profiling demand, maximizing utility, reducing cost, and
its nodes. controlling emission. They have also explored various
Since only a small size of data that need to be transmitted management methods to achieve these objectives. In [5], a
sensors does not need a high bandwidth, low latency and very statistical characterization of wireless channel in different
low energy consumption are highly required. There are a electric-power-system environments has been presented. WSN
multitude of standards like WiFi and Bluetooth that address design and deployment challenges on wireless communication
reliability are identified and experimentally studied. The communication technologies (e.g. IEEE 802.11 based wireless
impacts of electric-power-system environments on low-power LAN, IEEE 802.16 based WiMAX, 3G/4G cellular, ZigBee
wireless communications have been also revealed and the based on IEEE 802.15, IEEE 802.20). Moreover, challenges
potential applications of WSNs in smart grid are summarized. related to each wireless communication technologies have
Authors in [6] have proposed a various smart grid applications, been briefly discussed.
which are achieved through standardized wireless

Fig. 1. Wireless sensors network achitecture applied to the smart grid

[10], we can manage the power flow in order to minimize the

III. WSN FOR SMART MICRO-GRID consumption from the grid [11]. Several works were also
WSNs have been successfully applied in various carried out to minimize the consumption [12] by controlling
application domains [1, 2, 7, 8, 9], such as in transportation, active equipment inside the building.
healthcare, and environment monitoring. In the context of The smart infrastructure system applied to manage the
SMG, WSN could be deployed to measure physical quantities power sources flow can be divided into three subsystem [3]: i)
(e.g. current, voltage, frequency). When sensors detect the smart energy subsystem for power generation, ii) a smart
event being monitored, the event is reported to one of the base information subsystem for information metering and
stations, which then takes appropriate action as illustrated in measurement (Sensor, Smart meter) as well as information
Figure 2. In fact, a microgrid is a localized grouping of management for data analyze and control decision, and iii) a
electricity generations, loads and energy storages. In the smart communication subsystem for wireless data transfer
normal operation, it is connected to a traditional power grid. In between sensors nodes and the information management
the traditional power grid, the electricity is generated by the subsystem. In other words, the smart information subsystem is
hydraulic generator, fossil or nuclear energy then moved by responsible for advanced information monitoring, metering,
transmission and distribution grid to users. In a microgrid the and management in the context of the SMG. The smart
users can generate locally a low power electricity using communication subsystem is responsible for communication
renewable energy sources, such as solar panels, wind turbines connectivity and information transmission among all SMG
and fuel cells. Through the application of control strategies devices and applications.

Fig. 2. Micro-grid infrastructure

In this work we focus mainly on the smart communication topologies as depicted in Figure 3. For example in a start
subsystem by investigating different ZigBee topologies in the topology, a single base station can send and receive the data to
context of SMG. In fact, three main topologies could be a number of remote nodes. The remote nodes are not permitted
deployed as a communication medium, start, mesh, and hybrid to send messages to each other (Figure 3-a). The advantage of
this type of network is the simplicity and the ability to keep the system. Unlike these abovementioned topologies, a hybrid
remote node’s power consumption to a minimum. It also between the star and mesh topologies could provide a robust
allows low latency communications between the remote node communications network, while maintaining the ability to
and the base station. Alike star topology, mesh topology is keep the wireless sensor nodes power consumption to a
known as multi-hop communications in which data is minimum. The sensor nodes with lowest power are not enabled
transmitted from one node to another in the network, which is as data forwarder, while other nodes are enabled with multi-
within its radio transmission range. (Figure 3-b). This network hop capability, allowing them to forward the data from the low
topology has the advantage of redundancy and the range of the power nodes to other farther nodes in the network. In the rest
network is not necessarily limited by the range between single of this work, we investigate these three topologies by
nodes; it can simply be extended by adding more nodes to the conducting simulations in different context.

(a) (b)
Fig. 3. Micro-grid communication topologies, a) star, b) mesh

IV. SIMULATION RESULTS different devices perform the routing in the network. Routing
of data is decentralized in mesh network. A mesh network is
In this section, a wireless sensor network system was
in this way self-healing so that if a node fails another route is
designed using OPNET Modeler Academic Edition 17.5 as a
used for the delivery.
simulation tool. Different ZigBee topologies are created and
For conducting performance evaluation, different
evaluated. Using the ZigBee standard, a performance analysis
scenarios have been considered. The first scenario studies the
of the two topologies has been conducted. The different
behavior of a ZigBee network when the coordinator fails. We
network topologies built up by ZigBee devices are star, and
use two coordinators each fails for a period of time to figure
mesh network. In the star topology a coordinator is responsible
out the behavior of network nodes accordingly. The network
for the network. All other devices are end devices and
contains two coordinators and 24 routers and terminals. Each
communicate directly with the coordinator. This topology is
router and terminal in the scenario has its PAN ID set to Auto-
suitable for time critical applications and for networks with a
Assigned. The coordinators respectively have all their PAN
centralized device. In mesh network, the coordinators are still
IDs at 1 and 2. Some of the nodes should join one of the two
responsible for the network initialization and maintenance,
coordinators (a given node can join one of the two coordinators
while the routers are used to extend the network. A mesh
(Coordinator or Coordinator_0).
network shall allow full peer-to-peer communication and the

Fig. 4: Scenario of coordinator fails in a ZigBee netwok with two coordinators

Two minutes of simulation, the first coordinator fails. It coordinator fails, the nodes connected to this PAN should
will remain unsuccessful until the fourth minute, and it will leave and join the second coordinator. When the second
recover and restore the network again. At eight minutes, the coordinator fails, all nodes must join the first coordinator.
second coordinator will fail. It will remain unsuccessful until There are two main results registered with this network
the second minute. The expected behavior initially is that model. The network structure can be better visualized via the
about half of the nodes will join one of the two coordinators output report or visualization functions (Figure 5.a). The blue
and the rest will join the other coordinator. When the first curve is the first coordinator (PAN 1), while the red is the
second coordinator (PAN 2). At first, the two PANs have the fails at eight minutes, the load increases from PAN 1, while it
same loads (PAN 1 has more than PAN 2 due to a little more drops to zero for PAN 2. Figure 5.b presents the PAN
nodes of assembly of this network). After 2 minutes, the PAN affiliation, in these two graphs, we can see that when the
1 load drops to zero while the load increases in the PAN 2. coordinators have failed they have a negative PAN ID (i.e., ID
These loads remain constant even after the first coordinator is equal -1), which is the code indicating that they are not
recovered (i.e., after 4 minutes). When the second coordinator currently connected to the network.

(a) (b)
Fig. 5: a) Global MAC load par PAN, b) PAN Affiliation for the two coordinators

We can also see that 15 nodes are connected to PAN 1, this network model. Firstly, Figure 7.a shows the numbers of
while 11 are connected to PAN 2 with a balance in the number Hops for a router (Mesh and star routing). This graph shows
of packets sent for each node. At the beginning, the nodes the number of network traffic hops from Router 1 to reach its
connect by the automatic routing to the different coordinator destination. Which is a hops for the mesh routing (blue curve),
and each time a coordinator fails, the nodes change the PAN and two hops for the star routing topology. Figure 7.b shows
ID with as result a concentration of the traffic on the the End to End Delay (e2e delay) for the two topologies. The
coordinator which is running. curve in blue indicates the e2e delay for the mesh routing
scenario while the curve in red indicates the e2e delay for star
routing. The e2e delay of mesh routing is lower, because the
routing process finds more efficient paths than star routing.
The delay for the mesh routing varies between 0.009s and
0.016s while the delay for star routing varies between 0.017s
and 0.021s.
(a) Finally, Figure 7.c shows the Global Load MAC per PAN. The
curve in blue represents the global load for the mesh routing
scenario while the second is the global load for star routing.
The load for mesh routing is lower due to fewer hops for
application traffic, which results in less overall traffic seen at
the MAC layer. Also note that it has a very small peak load for
(b) mesh routing near the beginning of the simulation that is not
seen for star routing. This is due to the routing of messages
broadcast at that time.
In summary, the mesh topology is more adaptive to the
sensor network thanks to the set of topology stabilization
performances with less load compared to the star topology,
(c) less exchange of packets for routing with a number hops equal
Fig. 6: a) Two coordinators are connected to the network, b) coordinator 1
to 1 in order to reach the coordinator. We always consider the
fails, c) coordinator 2 fails energy constraints for the sensors, the listening of the channel,
the exchange of the frames of the routing protocol, the
The second scenario compares the performance of a sensor retransmission in case of error and the number of nodes to
network for a mesh topology based on the routing performance reach the coordinator are energy constraints for the sensors and
in a ZigBee network, the same topology will be analyzed later a network that degrades these parameters makes it possible to
but with a star topology. There are two results registered for have a longer duration of a WSN.
 (a) (b)

(c)
Fig. 7: Comparison: Mesh and Star Routing, a) Numbers of Hops for a router, b) Global End-to-end Delay, c) Global Load MAC per PAN

V. CONCLUSION AND PERSPECTIVES [6] Parikh, Palak P., Mitalkumar G. Kanabar, and Tarlochan S. Sidhu.
"Opportunities and challenges of wireless communication technologies
The deployment of SMG technology needs a smart for smart grid applications." Power and Energy Society General
infrastructure system for managing the power production and Meeting, 2010 IEEE. IEEE, 2010.
consumption flows. This infrastructure is based on a [7] Akyildiz, Ian F., et al. "Wireless sensor networks: a survey." Computer
networks 38.4 (2002): 393-422.
distributed collection of sensor nodes situated at various places
[8] Bharathidasan, A. R. C. H. A. N. A., V. Anand, and S. Ponduru. "Sensor
for measurement and communication of dedicated parameters Networks: An Overview, Department of Computer Science, University
such as current, voltage, and frequency. WSNs are effective of California." DAVIS, CA 95616 (2001).
solutions for energy management system in SMG, it will make [9] Yick, Jennifer, Biswanath Mukherjee, and Dipak Ghosal. "Wireless
the system cost and power effective, it is a cost effective sensor network survey." Computer networks 52.12 (2008): 2292-2330.
solution for measurement, monitoring and controlling SMG. [10] A. El mouatamid, Y. Nait Malek, R. Ouladsine, M. Bakhouya, N.
In this paper, we investigated two WSN topologies based on Elkamoun, K. Zine-Dine, M. Khaidar, R. Abid Towards a
ZigBee standard. Preliminary simulation results showed that Demand/Response Control Approach for Micro-grid Systems, in
‘CoDIT’2018, Grece.
the mesh topology outperforms the star topology, mainly it
[11] Kaplan, Stan Mark, Fred Sissine, and Thecapitol Net. Smart Grid:
provides a better reliability, quality of service, packet loss, Modernizing electric power transmission and distribution; Energy
energy consumption and delay. Extensive simulations will be independence, Storage and security; Energy independence and security
conducted together with experiments to further investigate act of 2007 (EISA); Improving electrical grid efficiency,
these topologies in real-sitting scenarios communication, reliability, and resiliency; integrating new and
renewable energy sources. The Capitol Net Inc, 2009.
REFERENCES [12] M. Bakhouya, Y. NaitMalek, A. Elmouatamid, F. Lachhab, A. Berouine,
[1] Kashif Dar, Mohamed Bakhouya, Jaafar Gaber, Maxime Wack, Pascal S. Boulmrharj, R. Ouladsine, V. Felix, K. Zine-dine, M. Khaidar, N.
Lorenz: Wireless communication technologies for ITS applications Elkamoune, Towards a Data-Driven Platform using IoT and Big Data
[Topics in Automotive Networking]. IEEE Communications Technologies for Energy Efficient Buildings, in CloudTech,2017, pp. 1-
Magazine 48(5): 156-162 (2010) 5, Rabat, Morocco.
[2] Yalgashev O, Bakhouya M, Nait-Sidi-Moh A, Gaber J. Wireless sensor [13] Sharma, Priya, and Gitanjali Pandove. "A Review Article on Wireless
networks: basics and fundamentals. the book Cyber-Physical System Sensor Network in Smart Grid." International Journal of Advanced
Design with Sensor Networking Technologies, ISBN. 2016:978-1 Research in Computer Science 8.5 (2017).
[3] Fang, X.; Misra, S.; Xu, G.; Yang, D. ‘’Smart grid—The new and [14] Vançin, Sercan, and Ebubekir Erdem. "Design and simulation of
improved power grid: A survey’’. IEEE Commun. Surv. Tutor. 2012, 14, wireless sensor network topologies using the ZigBee
944–980. standard." International Journal of Computer Networks and Applications
(IJCNA) 2.3 (2015): 135-143.
[4] Feng, Z.; Yuexia, Z. ‘’Study on smart grid communications system based
on new generation wireless technology’’. In Proceedings of the IEEE [15] Matin, M. A., and M. M. Islam. "Overview of wireless sensor
International Conference on Electronics, Communications and Control network." Wireless Sensor Networks-Technology and Protocols.
(ICECC), Ningbo, China, 9–11 September 2011; pp. 1673–1678. InTech, 2012.
[5] Gungor, Vehbi C., Bin Lu, and Gerhard P. Hancke. "Opportunities and [16] Chhaya, Lipi, et al. "Wireless Sensor Network Based Smart Grid
challenges of wireless sensor networks in smart grid." IEEE transactions Communications: Cyber Attacks, Intrusion Detection System and
on industrial electronics 57.10 (2010): 3557-3564. Topology Control." Electronics 6.1 (2017): 5.

View publication stats