AN EXPERIMENTAL COMMUNICATION ARCHITECTURE

FOR MONITORING AND CONTROL OF SUB-GRIDS

Zoya Pourmirza

and John Brooke

School of Computer Science, The University of Manchester, Oxford Street, M13 9PL, Manchester, U.K.

Keywords: Smart Grid, Real-time Data, Communication, Network Architecture, Wireless Sensor Network, Monitoring,

Energy Efficiency, Distributed Systems.

Abstract: The Smart Grid promises to provide better monitoring and control by incorporating the communication

network over the power network. We investigate the monitoring and control of distribution sub-Grids, for

example to a local area in a city. We propose a communication architecture to be deployed into a real local

area of the sub-Grid to provide a test bed for supporting real-time data and predictive real-time system

control. This is one of the main challenges of the Smart Grid. Our system design is aimed at investigating

two key issues: firstly, energy constraints in wireless sensor networks and secondly, achieving an

appropriate balance between central and distributed control of the sub-Grid. We propose an energy efficient

distributed architecture, for control and communication and explain how it will be implemented in our

experimental test bed. Also, TinyDB, which is a query processing system for sensor networks, has been

extended to collect the real-time data from the environment, and make them accessible by the local control

unit. Finally, a visualization tool has been developed to integrate the map view of the test bed, display the

real-time data, and send an alert to the network operator for finer-grained control over the system.

1 INTRODUCTION

The Smart Grid is considered as an integration of a

power network and a communication network. The

communication network, responsible for providing

communication between different sectors in a power

network by using digital technology, has three tiers

known as wide area network, neighbourhood area

network, and home area network. It is proposed

(Ericsson, 2010) that a communication system will

be a critical tool for the operation and administrative

purposes of the future power grid. In this paper we

describe a communication network architecture for a

neighbourhood area and describe how it is being

implemented in a test bed of sensors installed in the

6.6kV sub-Grid on a University Campus. Very little

is known about the monitoring and control on this

sub-Grid level, however as the local distribution

networks cease to be purely passive but can be

controlled at multiple levels, such a communications

architecture will be essential for efficient operation.

Our work is motivated by the study of other

networked systems such as the water distribution

grids (Khan et al., 2010, Machell et al., 2010,

Stoianov et al., 2007), which integrates sensing,

computation and decision making.

Our system design is targeted towards the

solution of two key problems. The first is the energy

constraint problem of general Wireless Sensor

Network (WSN), (Abbasi and Younis, 2007,

Raghunathan et al., 2004). The second problem

addresses the weaknesses of a centralized

architecture where the entire system data are stored

at a sub-Grid database, where controls are applied.

The paper is structured as follows: in Section 2

we explain the design of our architecture, in Section

3 we justify the choice of technologies used for

architecture implementation, in section 4 we

describe how we use simulation to explore different

possibilities for deployment. In Section 5 we

demonstrate the actual experimental test bed that

will be used to examine the performance of the

architecture. It Section 6 we summarise the current

status of our work and present future plans.

2 SUB-GRID COMMUNICATION

ARCHITECTURE

Given that the predictive real-time system control

Pourmirza Z. and Brooke J..

AN EXPERIMENTAL COMMUNICATION ARCHITECTURE FOR MONITORING AND CONTROL OF SUB-GRIDS.

DOI: 10.5220/0003948600670072

In Proceedings of the 1st International Conference on Smart Grids and Green IT Systems (SMARTGREENS-2012), pages 67-72

ISBN: 978-989-8565-09-9

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

necessitates having real-time information, in this

architecture real-time data are collected, analysed,

converted to information and finally acted upon.

As shown in the figure below, we propose a

mixture of peer-to-peer (P2P) and hierarchical

architectures that integrates sensors and distributed

computation units in a unified communications

network. Notice that the Figure 1 shows only one

local area in the sub-Grid, while there exists a

number of local areas in the sub-Grid. This

architecture contains three layers of sensors for

measuring and collecting the data. Each layer of

sensors is selected according to the data type they

measure. The sensors in layer one and two are

located in the street levels, communicating

wirelessly. The sensing units in the layer three are

located at the substation level using both wireless

and wired communications. Next layer is the test and

control unit which apply some control over the

substation level. Then the database layer will store

the aggregated data received from the three sensing

layers and the test and control unit. In future we

need to figure out how much and for how long we

can store the data in a database, and how often we

need to refresh them. Finally the top level of the

architecture, showing the head of the local area, is

called the local control unit. It will apply some

control over the entire local area. The local control

unit can access the sensing units directly in

emergency situations or indirectly through the

database layer in normal scenarios for applying

some control and decision-making. Given that each

local area is likely to take the optimal decisions for

its region, which might not be the optimal solutions

for the entire system, another layer of

communication should be added over the local

control unit, in order to make each local area aware

of the state of the other local areas, and enable them

coordinate their control decisions. Thus a

publisher/subscriber system should be added on top

of this architecture that will provide asynchronous

and loosely coupled communication. Consequently,

while taking decisions, each local area not only

decides the optimal solution for its local context, but

also takes decisions that might be the optimal

solutions for the whole system.

The advantage of the proposed architecture over

the traditional one is that it is able to avoid single

point of failure, if one of the local areas goes down,

the rest can function as normal in the sub-Grid.

More importantly, energy efficiency could be

achieved through this architecture. Given that energy

for sending data is higher than energy for computing

data (Heinzelman et al., 2000), by reducing the

travelling distance and adding more computation to

the system, we may achieve an energy efficient

architecture.

Figure 1: The proposed architecture.

For this architecture a simulation should be

designed. It is supposed that initially data will be

taken from the university campus to feed the

simulation, and get the predictions of how the

system should behave. Next, the theory (simulation)

can be checked against the real world (experiments).

Finally, the expert’s feedback on this system will

result in refining the architecture and designing an

energy efficient communication network

architecture suitable for Smart Grid applications.

3 IMPLEMENTATION ISSUES

In this architecture we provide a monitoring network

that can be used by distributed control networks. The

centralized control network is the norm in the

electrical grid. Its advantage is that control can be

optimized over the whole system. However, it has

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems

problems such as information flooding, single point

of failure, less security and scalability. Migration

from centralized control to distributed control

network is also supported by Yang et al. (Yang et

al., 2011) who believe that the current central

control is no longer sufficient for the new power

network with complex operational system and

should be replaced by distributed control. In future

by monitoring the response of the system to different

control mechanisms, such as local control and

optimized global control, we can propose an optimal

control network structure.

Moreover, in this section we consider the

choice of technologies used to implement the

architecture and how we selected them in order to

address the two problems mentioned in Section 1.

We could eliminate the first problem, namely energy

constraint by sending data from the sensors via

existing electrical power lines. However information

transmission over power lines has the following

shortcomings (Galli et al., 2010): 1-Limited

bandwidth, 2-Cable breakage will cause impaired

connections, 3-Noisy channel, 4-Advanced noise

filtering and error correction mechanism is required,

5-Communications between appliances of different

vendors is a big challenge, 6-Wire line technology

needs a plug or socket to be connected to the device,

which will result in nonflexible location of the

device. Wireless technology is less reliable and

secure but it offers low cost communication with

various ranges of bandwidth. It provides

communication to geographical areas that are not on

the wired network. Nordell (2008) shows wireless

can be used either for communications in the

substation, or over longer path such as between

substations or between enterprise and the substation.

Moreover, wireless systems show resistance to

electromagnetic interference of the system in high

electric and magnetic fields in each substation. Thus

for this reason we choose wireless technology as a

means of communication.

Since we intend to integrate information

gathering and control, an intelligent monitoring

system should be implemented to keep track of

collected data by sensor nodes. Then we need to

feed the real-time data from the sensor network to

the electrical network simulator. Having real-time

data instead of historical data may help us to

perform high-fidelity simulation that will result in

more accurate predictions. In this system, unlike the

other intelligent networks, we did not use SCADA

(Supervisory Control and Data Acquisition).

SCADA (Daneels and Salter, 1999) is only able to

monitor the critical areas of a network; whereas, our

architecture goes beyond that and monitor the entire

network. Also SCADA system has slow data update

rate, which makes it inefficient for Smart Grid

applications (Qiu et al., 2011). Considering the

weaknesses of the SCADA system, the lowest layer

of this architecture designed with WSN.

For reasons of efficient utilisation of energy in

collecting data from the WSN we use cluster-based

communication. Each cluster has a cluster head,

which can be a more a powerful sensor than the

other sensors. The sensors will send their data to the

cluster head wirelessly, and the cluster head is

responsible for transmitting the data to the base

station. A survey (Abbasi and Younis, 2007) on

clustering algorithms for wireless sensor network,

also confirm our assumption that the cluster based

communication provide more efficiency and longer

life time. The survey goes on to state that in addition

to advantages of clustering such as network

scalability and energy efficiency, clustering can

localize the route set up within the clusters and

consequently decrease the size of routing table

stored at a single node. Moreover, the range of inter

cluster interactions to the Cluster Head (CH) is

restricted, and redundant message exchange is

prevented, which help in preserving communication

bandwidth. In addition, since the network topology

is stabilized with the help of clustering at sensor

level, resulting in cuts on topology maintenance

overhead. Also the sensors are only concerned about

their connection to their CH, and would not be

affected by the modifications at inter-CH levels. On

the other hand, based on application requirements,

the events monitored can be either continual or

intermittent. The continual monitoring ends up in

generating considerable amount of traffic to be

routed to the sink. In In-network data processing

similar packets from different nodes can be

aggregated in order to reduce the number of

transmissions. Since communications are more

energy consuming than computations, this technique

can save substantial amount of energy. Also, given

that the first level of sensors are using Bluetooth

communication with short range of data

transmission, a cluster head is provided to offer

longer range of communication, and be able to

transfer data from lowest level to the higher levels.

The cRIOs (compactRIO) used in the third layer

of the hierarchy are control and acquisition units.

These cRIOs can also measure various quantities

and can be programmed and used in controlling and

monitoring applications. The live data from cRIO

will be sent to the next layer of the hierarchy,

responsible for testing and controlling the substation

ANEXPERIMENTALCOMMUNICATIONARCHITECTUREFORMONITORINGANDCONTROLOFSUB-GRIDS

level. This layer is using a graphical programming

environment called LabVIEW for measuring,

testing, and controlling the system. Additionally, the

collected data from cRIOs will be transferred and

saved in the database layer, which can be accessed

by the local control unit. The local control unit,

being the highest layer of the hierarchy can query

the three sensing layers and access the real-time and

historical data.

4 SENSOR NETWORK

SIMULATION

In order to evaluate the optimal method of

implementing the WSN, TOSSIM (Levis and Lee,

2003), an existing wireless sensor network

simulator, has been selected which simulates the

nodes running on TinyOS (Levis et al., 2005)

operating system. Although TOSSIM is a more

complex simulation compared to abstract

simulations, but it enables users to take their

implementation and run it on the actual mote. This

feature motivates us to select TOSSIM over other

existing simulations. Therefore, we are able to run

and test our prototype implementation on the real

physical environment as well as laboratory-based

environment. Additionally, TinyDB (Madden et al.,

2003) which is a query processing system has been

extended to meet the needs of the Smart Grid.

TinyDB is able to query the sensors and save the

results in a database layer. The database can save

real-time data received from sensor network plus

historical data of the system. These data will then be

delivered to a higher-level unit. One difference

between TinyDB and traditional database is that

when we post a query to the sensors with TinyDB

installation we receive the real-time data instead of

receiving archived data in reply. Given that sensors

are battery-powered devices, we tend to reduce

power consumption in the system as much as

possible. A typical strategy of power-efficient in-

network processing algorithm minimizes the

communication cost by applying filtering and

aggregation inside the sensor network and provides

routing of data from a physical environment to a pc.

The reason why we select TinyDB over the other

query processing systems is that, not only it does

contain the same features of a conventional query

processing (such as join, select, project, and

aggregate information), but also it provides

functionality for minimizing power consumption by

using the acquisitional techniques. ACquisitional

Query Processing (ACQP) (Madden et al., 2005)

focuses on both conventional techniques as well as

the new features of query processing. These new

features focus on when, where, and how often data

are sampled and send to a query processing operator.

In contrast to conventional passive systems, power

consumption can be minimized by focusing on the

location and cost of data acquisition.

To extract the desired information from the

electrical network TinyDB has been extended by

adding the SmartGrid component for electrical

network applications. This extension enables us to

extract the environmental information about our

prototype electrical network and feed them to

another source of computation for the further

analysis and gaining the predictive real-time system

control. In future, this extension will be used to

produce measured data from the real environment.

5 VISUALISATION OF THE TEST

BED

In this section we will demonstrate the overlay of the

electrical network information on the map view using

Google Maps API (Application Programming

Interface). This tool enables users to click on the

desired section in the network and check the related

data.

5.1 Related Work

The work in (Haines et al., 2009) consider the map

based graphical user interface as a significant factor

for water engineers which enable them to run ‘What-

if’ simulations and visualise the results of their

simulations. They show how dynamic information

and prediction of network state can be accessed

using lightweight devices and visualised in user

friendly Google Maps based web interfaces. Also,

another study by (Stoianov et al., 2007) uses Google

map to visualize the data related to water network.

They enable their users to choose a sensor and

extract the related data from the water grid. The

Google map visualization techniques can be applied

to the other networks such as electrical grids. They

are used to retrieve the data and help the users to

better understand the network condition. The figure

2 shows how data relating to the sensors can be

overlaid on a GIS (Google Maps).

5.2 Test Bed

This real project is being implemented on the

medium voltage power network of the University of

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems

Figure 2: University of Manchester Campus.

Manchester campus. Also real equipment, real data,

and professionals in the field are provided. We are

now in the stage that this system is going live.

Considering that currently there is no information

available below 33 kV, our research is quite valuable

since it will deliver the collected data to the experts

and demonstrates what is happening in the lower

voltage network that will consequently offer better

control over the grid. This test bed only considers

eleven 6.6 kV substations that are equipped with

sixteen monitoring systems called cRIO which

collect the data from the network and process them.

The substations with one transformer are equipped

with one cRIO, whereas substations with two

transformers are supplied with two cRIOs. These

cRIOs are running at 4Hz, thus we are receiving four

samples per second, measuring three phases

voltages, currents, active powers, power factors,

voltage's spectrums (eight channels for each phase)

and current spectrums (eight channels) and

frequency. Each substation can provide wireless and

wired data communication to the data server.

In order to prevent data loss if the links go down,

data should be stored locally in the cRIO memory

and transferred once a day to the database layer.

Subsequently, the live data will be sent, from cRIO

to a pc running LabVIEW via TCP/IP or wireless

communication to test and control the substation

level. The live data will then be transferred and

stored in the database layer. Given that better

monitoring will result in more efficient control over

the system, a graphical display of the network is

provided for electrical engineers. It is helpful for

them to utilize the visualization technology in order

to track and discover the faulty part of the grid and

take remedial actions. The figure 2 illustrates the

University of Manchester campus map, where the

red pin shows the first cRIO, and the rest of the

cRIOs are link together with the red lines.

This tool enables the operators to observe the

Google map view of the test bed and schematic view

of the electrical network. The provided visualisation

tool has some control functions such as panning and

zooming. It also locates substations on the map

according to their longitude and latitude, and

demonstrates the related information. It is able to

connect to the live database and data graph to

display real-time data in order to help experts better

comprehend what is happening in the local area

network. If the measured data pass the defined

threshold the tool will send an alert, so operators can

recognize the fault very quickly and take corrective

actions. In future we will add more functionality to

the tool, and show the results of network simulations

on the map. Briefly, it is assumed that the

combination of the visualization technology and

simulation technology allow operators to better

predict the effect of course of actions by observing

generated results of simulations on the map.

ANEXPERIMENTALCOMMUNICATIONARCHITECTUREFORMONITORINGANDCONTROLOFSUB-GRIDS

6 SUMMARY

We have described how we have developed an

architecture for monitoring and controlling the local

area in the sub-Grids. We also describe how we

intend to use simulation to tune the implementation

of the architecture on a real test bed, namely a sub-

Grid covering a large university campus. The

particular interest of our work is the close

connection between design and testing of the

architecture and also that it will enable information

to be gathered from the working of a real system

operating a sub-Grid level where monitoring has not

previously been deployed.

In future, this research will focus optimising the

WSN topology and balancing between the cost of

data transmission, data computation and data

compression technique in order to bring more energy

efficiency to the system. Also, the ever increasing

volume of data gathered from the grid and the

probability of data loss in the system, necessitate

more investigations on data optimizing techniques in

databases. Finally, further work needs to be done to

establish publish-subscribe architecture to provide a

P2P architecture between local areas.

REFERENCES

Abbasi, A. A. & Younis, M. 2007. A survey on clustering

algorithms for wireless sensor networks. Computer

Communication, 30, 2826-2841.

Daneels, A. & Salter, W. 1999. WHAT IS SCADA? In:

Accelerator and Large Experimental Physics Control

Systems 1999 Trieste, Italy. 339-343.

Ericsson, G. N. 2010. Cyber Security and Power System

Communication-Essential Parts of a Smart Grid

Infrastructure. IEEE TRANSACTIONS ON POWER

DELIVERY, 25, 1501.

Galli, S., Scaglione, A. & Wang, Z. 2010. Power Line

Communications and the Smart Grid. In: First IEEE

International Conference on Smart Grid

Communications (SmartGridComm), 4-6 Oct. 303-

308.

Haines, R., Khan, K. & Brooke, J. 2009. Bringing

simulation to engineers in the field: a Web 2.0

approach. The Royal Society, 2635-2644.

Heinzelman, W. R., Chandrakasan, A. & Balakrishnan, H.

2000. Energy-Efficient Communication Protocol for

Wireless Microsensor Networks. In: Proceedings of

the 33rd Hawaii International Conference on System

Sciences-Volume 8, 820485: IEEE Computer Society,

1-10.

Khan, K., Haines, R. & Brooke, J. M. 2010. A Distributed

Computing Architecture to Support Field Engineering

in Networked Systems. In: Proceedings of the 2010

International Conference on Complex, Intelligent and

Software Intensive Systems, 1796283: IEEE

Computer Society, 54-61.

Levis, P. & Lee, N. 2003. TOSSIM: A Simulator for

TinyOS Networks. University of Berkeley. 115-148

Levis, P., Madden, S., Polastre, J., Szewczyk, R.,

Whitehouse, K., Woo, A., Gay, D., Hill, J., Welsh, M.,

Brewer, E. & Culler, D. 2005. TinyOS: An Operating

System for Sensor Networks Ambient Intelligence.

Springer Berlin Heidelberg.

Machell, J., Mounce, S. R. & Boxall, J. B. 2010. Online

modelling of water distribution systems:a UK case

study. Drinking Water Engineering and Science, 279–

294.

Madden, S., Hellerstein, J. & Hong, W. 2003. TinyDB: In-

Network Query Processing in TinyOS.

Madden, S. R., Franklin, M. J., Hellerstein, J. M. & Hong,

W. 2005. TinyDB: An Acquisitional Query Processing

System for Sensor Networks. ACM Transactions on

Database Systems, 30, 122–173.

Nordell, D. E. 2008. Communication systems for

Distribution Automation. In: Transmission and

Distribution Conference and xposition. IEEE/PES, 21-

24 April. 1-14.

Qiu, M., Gao, W., Chen, M., Niu, J. W. & Zhang, L. 2011.

Energy Efficient Security Algorithm for Power Grid

Wide Area Monitoring System. IEEE

TRANSACTIONS ON SMART GRID, 2, 715-723.

Raghunathan, V., Schurgers, C., Park, S. & Srivastava, M.

B. 2004. ENERGY EFFICIENT DESIGN OF

WIRELESS SENSOR NODES. Wireless Sensor

Networks. Kluwer Academic Publishers. 51-69

Stoianov, I., Nachman, L., Madden, S. & Tokmouline, T.

2007. PIPENET a wireless sensor network for pipeline

monitoring. In: Proceedings of the 6th international

conference on Information processing in sensor

networks, Cambridge, Massachusetts, USA. 1236396:

ACM, 264-273.

Yang, Q., Barria, J. A. & Green, T. C. 2011.

Communication Infrastructures for Distributed Control

of Power Distribution Networks. Industrial

Informatics, IEEE Transactions on, 7, 316-327.

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems