Cyber-interdependency in Smart Energy Systems

Razgar Ebrahimy

and Zoya Pourmirza

School of Computing Science, Newcastle University, Newcastle upon Tyne, U.K.

School of Electrical and Electronic Engineering, Newcastle University, Newcastle upon Tyne, U.K.

{razgar.ebrahimy, zoya.pourmirza}@ncl.ac.uk

Keywords:

Smart Energy Systems, ICT Architecture, Complex Interdependencies.

Abstract:

Critical infrastructures are highly interdependent due to the services they receive and provide to one another.

These interdependencies include physical, logical, geographical and cyber. Most of these interdependencies

have been studied extensively apart from the cyber interdependency which is the main focus of this paper.

Critical infrastructures have cyber interdependency when the state of a physical infrastructure (energy, trans-

port, water, waste, etc.) depends on the information transmitted through the information infrastructure. The

communication network is the backbone of smart energy systems and is responsible for transmission of data

from all sub-systems in both directions. Due to the complexity and combination of many sub-systems that

form smart energy systems, data is generated at different levels within such systems. The data at each layer

is different and has its own cyber characteristics. Knowing these characteristics and interdependencies at each

layer provides the foundation for designing an appropriate ICT architecture that ﬁts that segment (layer) rather

than having an ICT architecture that is generic and designed for all layers but susceptible to risks. This paper

outlines a new approach by focusing only on the cyber interdependencies in smart energy systems and how

they effect the Smart Grid.

1 INTRODUCTION

Critical infrastructures are highly interdependent due

to the services they receive and provide to one an-

other. These interdependencies vary from physical,

geographical and logical to cyber interdependencies

(Rinaldi et al., 2001). These interdependencies could

be regarded as providing both risks and opportunities

to overall infrastructure system operation. In gen-

eral, infrastructure systems such as power, transport

and water are legacy systems that have been designed

and developed during the last century and are still

being developed (Ebrahimy, 2014). The availability

of communication networks and digital connectivity

(Oughton et al., 2016) to power systems and the ben-

eﬁts of data transmission have led the power sector to

adopt these new technologies and integrate them into

its operation in the form of Supervisory Control and

Data Acquisition (SCADA) systems, which are used

for controlling and monitoring.

Whilst the integration of communication network

with power grid has great beneﬁts, it is also suscep-

tible to risks and failures due to its interdependency

and reliance on external services. For instance an

unplanned shut-down of a power station in Italy in

2003 led to failure of a communication node and the

SCADA system (Buldyrev et al., 2010). This event

resulted in further failures, causing a cascading fail-

ure in the system.

The arrival of the Smart Grid and its integration

with the legacy power grid has opened new avenues

to explore. This exploration must cover the expansion

of already complex power systems as they become

smarter and more reliant on communication technolo-

gies at the same time as integrating with the legacy

systems. This all requires new approaches in deﬁning

the interdependencies and relationships with external

services.

Smart Grid is the integration of traditional elec-

trical power grid systems with information and com-

munication technologies (ICT) (Aloula et al., 2012).

This integration improves the efﬁciency and avail-

ability of the power system, empowers the utility

providers in terms of how to operate their systems

based on consumer demands, while constantly moni-

toring and checking the demand and supply using ICT

infrastructure.

In this research we identiﬁed the lack of discus-

sion on cyber interdependency of the Smart Grid.

Since the state of the Smart Grid depends on the

data transmitted through ICT architecture, the explo-

ration of this cyber-dependencywithin the Smart Grid

Ebrahimy, R. and Pourmirza, Z.

Cyber-interdependency in Smart Energy Systems.

DOI: 10.5220/0006262805290537

In Proceedings of the 3rd International Conference on Information Systems Security and Privacy (ICISSP 2017), pages 529-537

ISBN: 978-989-758-209-7

529

makes this research timely, and critical. Unlike the

existing grid, the Smart Grid is comprised of distri-

bution, intelligence, operation, system optimization,

data decentralization and high dependency on com-

munication. Smart Grid components are interdepen-

dent with bidirectional interactions which rely on pre-

deﬁned and correct operation of other dependant sys-

tems. Due to the complexity and collection of sub-

systems in Smart Grid it is important to note that

the data dependency at each layer of Smart Grid has

different implications on the operation of the entire

system. Reliable and real-time information gathering

from each layer of the Smart Grid is critical in order

to minimize the impact of failures and to optimize the

power ﬂow operation in the power system.

The shift towards Smart Grid as referred to by

some as an energy transition period (Jefferson, 2008)

and moving away from centralized and fossil based

systems to a distribution conﬁguration that make use

of low carbon alternatives highlights new opportuni-

ties and challenges that could arise. One of the key

challenges that this paper addresses is the cyber inter-

dependency in smart energy systems and the need for

novel, adaptable and resilient ICT architectures that

fulﬁls future interdependent systems.

The remainder of this paper is organized as fol-

lows: Section 2 addresses the related work, Section

3 is about the importance of communication in Smart

Grid and exploring the cyber dependency. Section 4

addresses different layers of the systems where data is

generated and Section 5 analyzes the cyber character-

istics of the information. Section 6 is about the differ-

entiation of the ICT architecture and ICT infrastruc-

ture, Section 7 explains the essential requirements of

Communication and Information handling. Sections

8 and 9 address cyber risks and conclusions respec-

tively.

2 RELATED WORK

There has been intensive research on the Smart Grid

from various angles by many researchers. There are

many papers which focus on new developments in in-

formation and communication (Wissner, 2011), (Wu

et al., 2011) and some focusing on consumers and

how to involve them in active operation of power con-

sumption management by introducing technical op-

eration systems or providing economic incentives to

facilitate their demand (Wolsink, 2012) while others

focus on new challenges of dealing with risks and un-

certainties in Smart Grid (Zio and Aven, 2011)(Mo-

moh, 2012). There has also been an attempt to focus

on cascading failures in interdependent Smart Grid

networks and designing cascade resilient networks in

Smart Grid using optimum allocation of interdepen-

dencies (Rahnamay-Naeini, 2016). As well as focus-

ing solely on electricity there is some research which

focuses on a complete set of intelligent management

energy sources (Orecchini and Santiangeli, 2011) and

some emphasis on market integration (Nielsen et al.,

2011).

Surveys such as those conducted by (Wang et al.,

2011), (Yan et al., 2013) and (Safdar et al., 2013) have

studied the communication infrastructures of Smart

Grids, the importance of such communication infras-

tructures, their challenges and requirements, and the

available communication technology. However, there

is lack of research on the deign of an ICT architec-

ture, cyber Interdependency within the Smart Grid,

and how this cyber interdependency affect the design

on an ICT architecture for the smart power grids.

3 SMART GRID AND

COMMUNICATION

The communication system is the key part of Smart

Grid for data transmission (Laverty et al., 2010). As

mentioned in Section 1, Smart Grid has been devel-

oped to utilize the transmitted data to deliver a bet-

ter, more reliable, cost effective,secure and optimized

service to consumers (Gungor et al., 2011). An Smart

Grid system consists of collection of different types

of data from consumer usage data to transmission and

substation sensors data and control unit data.

Cyber interdependency in infrastructure (energy,

transport, water, waste, etc.) as deﬁned by (Rinaldi

et al., 2001), is when a state of a physical infrastruc-

ture depends on the information transmitted through

the information infrastructure. Figure 1 is a represen-

tation of an optimized energy system that is the in-

tegration of smart energy providers, smart consumers

and advanced management systems with a communi-

cation network being an integral part of it.

The direction of data ﬂow and the purpose of

transmitted data from each system or sub component

is very important. As can be seen from Figure 1 the

type and criticality of the data depends on the trans-

mission direction and the data generation source. For

instance any data generated from the energy network

is either load data or information about the status of

the system which is transmitted to advanced manage-

ment system.

However if the information is from the manage-

ment system then the data is considered critical since

it sends control data and the availability of such data is

important for overall operability of the system. Again

ICISSP 2017 - 3rd International Conference on Information Systems Security and Privacy

530

depending on the destination of the control data, the

criticality varies; for instance if the control data is

to send a signal to increase generation capacity or

to switch load on different lines then it is regarded

as critical whereas if the signal is to the consumer to

manage their load, based on the demand and response

management system, then it is not classed as critical

since this is only a suggestion by a utility provider and

there is no obligation for consumers to comply with

it. In order to understand the data interaction at each

layer of the system and the nature of the relationship

and criticality we have structured and categorized data

sources in three different layers derived from Figure

1 in section 4.

4 DIMENSION OF CYBER

DEPENDENCY

Smart Grid data originates from several sources and

sensors such as advanced metering systems, phasor

measurement units, intelligent relays and remote ter-

minal units. The advanced metering system is a two-

way communication network consisting of smart me-

ter, computer hardware and software, monitoring sys-

tem and data management that enable the collection

and redistribution of data between meters and utility

providers.

A large and broad spectrum of sources that gen-

erate data and increase communication could lead to

increasing vulnerability of Smart Grid with many po-

tential points of failure and attacks to be exploited.

In smart energy systems, similar to Smart Grid,

the ﬁrst layer of data dependency is between the

consumers (residential/commercial) and local control

centres via a communication network. At this layer

consumer usage and activity is reported back to a lo-

cal or central control centre. Since the communica-

tion is two way, it is possible for the operator to send

some signals back to consumers to suggest reduction

in load or limiting their access to the services. If we

call this the ﬁrst layer interaction, then the absence

of this information at this level is not very critical for

overall operation but instead its availability helps in

demand-side response and would give economic ben-

eﬁts to consumers.

The second layer of data dependency is between

transmission substation, distribution substation, load

serving entity and energy management system (EMS)

which are part of modern smart energy provision.

An EMS is fed from sources such RTUs, PLCs and

smart relays that take system status information and

load at different nodes via the communication net-

work and then feeds the information into a state es-

timator to perform optimal power ﬂow, contingency

analysis and detect bad and false data received to op-

erate the optimal power system. An EMS is also ca-

pable of receiving the transmitted data from the power

network sensor readings of substation to transmission

linens’ and distributed and renewable energy sources

to monitor the network.

The generated data at this layer is critical for the

operation of the system since it has direct inﬂuence

on how a system should react in terms of whether to

increase/decrease capacity, switch the load or discon-

nect a node from the grid.

The third layer of data dependency in smart en-

ergy systems is the data generated from interdepen-

dent systems which are not directly part of the Smart

Grid architecture but whose inputs have direct inﬂu-

ence and implications on how these systems operate.

For instance accurate weather data could be used to

forecast more realistic energy demand and prepare for

sudden changes. Access to historical data about usage

and weather data enables advanced advisory systems

to provide more accurate and realistic predictions of

energy demand in conjunction with current demand

data. Although at this layer the availability of such

data cannot be regarded as critical, the addition would

help in better monitoring and operation of the system.

The data interdependencyin smart energy systems

is tight and cohesive and this is why there is a need for

better understanding of layer dependencies to enable

accurate measurements, performance analysis and en-

sure security of supply.

5 CYBER CHARACTERISTICS

OF THE DATA

Within each interdependent layer the characteristics

of the data and communication networks differ. For

instance, depending on what data is transmitted or

where it originates, the implications caused by such

data on smart energy systems could vary. Figure 2

is a representation of the data that each has a source,

meaning that in large complex systems such as Smart

Grid there are many sources which generate informa-

tion. Secondly each piece of information has a data

type and ﬁnally the characteristics of the data which

could be one or the collection of the set shown in Fig-

ure 2. Below are the data characteristics of the in-

formation generated, that depending on the layer and

source would posses the characteristics shown in the

following subsections:

Cyber-interdependency in Smart Energy Systems

531

Communication Network

RTUs

Optimal Power Flow

Emergency and

Contingency

False Data Detection

State

Estimator

System

Operator

Advanced Management System

Energy Network

Communication Network

Wired, Wireless, Powerline, etc.

RTUs

Energy storage

RTU

Communication Network

Control

System

Monitoring

Metering

Advisory and Control

System,

Demand Side Response

Data Flow

Two way power flow

Electricity Line

Sensor Network / Smart Meters

Smart Energy Provider

Electric Vehicles

Vehicles to Grid

Figure 1: Energy Systems and Data Flow.

5.1 Communication Technologies

As mentioned before, a two-way communication sys-

tem which includes both wired and wireless is es-

sential for Smart Grid and smart energy systems as

a whole. However based on the layer of transmis-

sion, the communication could be different. For in-

stance, the ﬂow of information from sensors and elec-

trical appliances to smart meters could be via pow-

erline or wireless communications such as ZigBee,

Zwave, and others (Luan et al., 2010). The communi-

cation from smart meters to utility providers could be

accomplished through mobile network or the Internet.

Depending on the type of implemented technology on

each layer or on each component the interactions and

points of failures are different.

5.2 Margin of Error

Sometimes it is acceptable to have delay in data trans-

mission from smart meters to utility providers, pro-

viding it is only the consumption data. The resolution

of data in this case is usually not the same for each

provider or similar to other countries. Usually the

data is collected at discrete time slots of equal length.

For instance, depending on the provider the resolu-

tion of data could vary from 1 slot a day to 48 (half

hourly) (Yang et al., 2014) per day. In cases where the

sensors transmit critical data such as PMUs or RTUs

where the data triggers automotive decisions the mar-

gin of delay and resolution is much smaller. There-

fore the source of the data is key for determining the

acceptable scale of errors.

5.3 Security Measures

Security is a major concern in Smart Grid due to two-

way communication in the system as modern power

systems rely on information infrastructure to oper-

ate. Legacy power systems are protected by Supervi-

sory Control and Data Acquisition (SCADA), Energy

Management Systems (EMS) and Real Time Oper-

ICISSP 2017 - 3rd International Conference on Information Systems Security and Privacy

532

Data Characteristics

Data Sources

Data Type

Communication Protocol

Margin of Error

Security Measures

Layer Position

Data Sh aring

Smart Meters

RTUs & PMUs

Asset He al th Monitoring

Operational data

External data (weather, historical, etc.)

Load u sage measurement

Low level data

Information data

Control a nd management

Shareable data

Figure 2: Data interdependency.

ating Systems (RTUS) to ensure optimal operability.

To ensure the operability of the Smart Grid informa-

tion infrastructure, the network and data management

tools need to be secure. This is why the IEC62351

protocol was developed with the focus on authenti-

cation of data transfer (ANSI C12.22), authenticated

access, intrusion detection and eavesdropping. Nev-

ertheless there are still some concerns with bad data

injection in Smart Grid systems that can not be iden-

tiﬁed by detection or state estimator (Huang et al.,

2013).

5.4 Data Types

Generally there are two types of data in smart energy

systems: status data and control data. Status data are

generated from sensors in transmission, distribution

and smart meters, etc whereas control data are gener-

ated from RTUs and management systems to trigger

an event.

5.5 Resource and Status Sharing

Distributed energy is becoming an important part of

the Smart Grid, speciﬁcally when the consumer feeds

energy back to the system using solar or storage bat-

teries. In traditional power systems in the UK, Na-

tional Grid is responsible for generating electricity

based on forecast ﬁgures and contingency measure-

ments. However when consumers generate their own

electricity then to track the overall generation the

electricity suppliers need to have access to such data

from the consumer side. This is where consumers

generate data of their usage and generation, each tar-

geting the same or different operators. This type of

resource sharing in terms of data beneﬁts both utility

and grid operators.

5.6 Information Infrastructure & Big

Data

Due to the complexity and integration of different sys-

tems, the amount of data that smart energy systems

generate is very large scale. This is one reason that

data analysis is becoming a core part of any smart sys-

tem to enable the advisory systems in better decision

making and optimal operation of the entire system.

6 ICT ARCHITECTURE VERSUS

ICT INFRASTRUCTURE

The term infrastructure used to refer to ”the underly-

ing foundation or basic framework (as of a system or

organization)” (MA, 1993). In 1997, the term gained

more importance and advanced meaning based on the

work of PCCIP (US Presidents Commission on Criti-

cal Infrastructure Protection). In this new era the term

infrastructurewas widely used to mean the framework

of interdependent networks and systems comprising

identiﬁable industries, institutions (including people

and procedures),and distribution capabilities thatpro-

vide a reliable ﬂow of products and services essential

to the defence and economic security of the United

Cyber-interdependency in Smart Energy Systems

533

States, the smooth functioning of governments at all

levels, and society as a whole. This section deﬁnes

and delimits the term infrastructure in the domain of

Communication and Information Technology (ICT).

Additionally, it differentiates between ICT infrastruc-

ture and ICT architecture, terms which are commonly

misused by researchers in a number of areas of study.

Next, we will discuss how ICT architecture can inﬂu-

ence the cyber dependency of the Smart Grid.

An ICT infrastructure describes the fundamental

components such as monitoring devices, computers,

controllers, software, middleware, storage, and data

communication media and technologies, that offer

services such as access, transmission, storage, mon-

itoring, control, conversion of data to useful informa-

tion, analysis, and ﬁnally taking action based on anal-

ysed data in a system or organization.

ICT infrastructure resembles the foundation or in-

frastructure of a building that can support a num-

ber of different building architectural styles such as

Achaemenid, Ancient Chinese, Victorian, or Roman.

Although different buildings might have varied archi-

tectural styles, their infrastructures are similar. Like-

wise, based on similar ICT infrastructure we can de-

sign and develop a number of different ICT archi-

tectures, corresponding to diverse organizational re-

quirements.

Conversely, an ICT Architecture refers to the de-

sign or style in which components and services of

the ICT infrastructure interact. In other words, ICT

infrastructure is about the components or building

blocks of a system that support the ICT architecture.

ICT architecture design should be based on estab-

lished architectural principles and requirements of the

organization. Such architecture should not only of-

fer, but facilitate, enhance, and sustain services re-

quired by the organization. Without a proper ICT

architecture design, the cyber system of an organi-

zation will not be future proof, as it may not incor-

porate upcoming functionalities, ideas, and changes

to the system. To summarise, while ICT infrastruc-

ture is almost the same for all levels of Smart Grids,

the ICT architecture varies according to a number of

parameters, which will be discussed in the next sec-

tion. Accordingly, ICT infrastructure does not affect

the cyber-interdependency of the Smart Grid as much

as the ICT architectures does. This also highlights the

importance of rigorous considerations of an ICT ar-

chitecture in the design phase, as it can isolate or cas-

cade failures in such cyber-interdependent systems.

7 COMMUNICATION AND

INFORMATION HANDLING

REQUIREMENTS

This section identiﬁes the ICT (Information and Com-

munication Technology) requirements of the Smart

Grid, and discusses how some of these requirements

can inﬂuence the cyber-interdependency of the power

grid. As mentioned earlier, in order to design a

sustainable and futureproof architecture, the design

team should ﬁrst identify the requirements of such

architecture. Since there is no one ICT architecture

that can ﬁt all the different cases (Pourmirza, 2015),

the requirements and design decisions identiﬁed in

this section can help ICT developers to improve and

modify their designs. The ICT requirements for the

Smart Grid are classiﬁed into three categories, namely

functional, non-functional, and architectural require-

ments. Functional requirements are deﬁned as the

concrete functionalities of a system, non-functional

requirements are described as the qualitative charac-

teristics of a system which address the performance

concerns (Committee and Board, 1998), and architec-

tural requirements are deﬁned as the design decisions

which relate to the ICT architecture itself (Rohjans

et al., 2012). Amongst these three categories, require-

ments deﬁned under the architectural requirements or

design decisions affect the degree of cyber interde-

pendency of the power grid the most. Proper consid-

eration of some of these parameters is critical in the

design of an ICT architecture for the Smart Grid, as

they can directly or indirectly cascade or isolate the

fault in the system, and prevent the rest of the grid

from being adversely affected. Figure 3 presents a

number of requirements in each category.

The state of the Smart Grid depends on the out-

put of its ICT architecture. Some of the parameters

that highlight the cyber interdependency of the Smart

Grid are as follows: distributed, layered, component-

based, and loosely coupled. These parameters can af-

fect ﬂexibility and adaptiveness of the system in case

of failure or when the system is stressed. One of the

main requirements of an ICT architecture is to be a

distributed system (Pourmirza and Brooke, 2013). A

centralized ICT system has a single point of failure;

by moving from centralized to distributed systems we

can prevent this problem. Fault tolerance is one of the

principles in distributed systems (Emmerich, ); it is

the ability to deal with the reliability of the system in

the event of a fault, and it also enables the system to

progress despite the presence of failure in the system.

It is usually implemented by circumventing a single

point of failure in the network. A distributed system

can be represented by a hierarchical architecture. Hi-

ICISSP 2017 - 3rd International Conference on Information Systems Security and Privacy

534

Smart Grid ICT requirements and characteristics

Functional requirements

System qualitative

characteristics

/ Non-

functional requirements

Design decision/

architectural requirement

Sense

Configuration

Distributed

Transmit

Quality of service

Layered

Store

Dependency

Component-based

Monitor

ICT energy management

Loosely coupled

Control

Consideration of big data

Data collection

Convert data to knowledge

ICT constraints and other issues

Data analysis

Analyse

Resiliency

2-way communication

Take action

Reliability

Alarm handling

Environmental

Fault tolerant

Security

Manageability

Information coherence

Scalability

Confidentiality

Integrity

Availability

Non-repudiation

Accountability

Authenticity

Figure 3: ICT Smart Grid requirements.

erarchical systems have a tree structure such that each

node is connected to several other nodes (leaves) and

each of these nodes is connected to a number of differ-

ent nodes, and so on. A hierarchical network is a type

of network wherein processing and control function-

alities are performed at different levels (Fed, 1996).

These functionalities can be on top of, or below, each

other or else they can be at the same level. Hierar-

chical systems are to some degree interdependent and

partially fault tolerant. Although a failure at the leaf

level may not cause huge disturbance to the whole

system, their root can be considered as a single point

of failure.

The other requirement of an ICT architecture is

to be a layered system (Pourmirza, 2015). Hierarchi-

cal relations can also be represented through use of

layers. As deﬁned by the National Communications

System Technology and Standards Division in the US

(Emmerich, ), layering in a communication system is

referred to as a group of related functions that are per-

formed in a given level in a hierarchy of related func-

tions. Layering addresses how different sections of

the ICT of the Smart Grid are connected, and how the

information passes between each section. A strictly

layered hierarchical architecture is an interdependent

system, if one layer fails, the whole architecture will

fail. This is because an entity in each layer can only

interact with an entity in its own layer, or with the

layer directly below, because the upper layer asks for

services or data from layers below.

The next requirement of an ICT architecture is to

be a component-based system (Pourmirza, 2015). A

component has been deﬁned by the National Com-

munications System Technology and Standards Divi-

sion (Fed, 1996) as a part of a system that is essential

for the operation of a bigger system and is a direct

sub-division of the system to which it belongs. Some

of the components that result from a sub-division of

the ICT architecture are: smart meters, WSNs, moni-

toring devices in the substations, databases and visu-

alization tools. Accordingly, some of these compo-

nents can also be sub-divided into other components.

Cyber-interdependency in Smart Energy Systems

535

For example, monitoring devices in the substations

can be componentized into other components, namely

cRIOs, data storage, a control unit and a router. Since

these components are interdependent, failure to pass

correct or updated data to another component may

cause a fault in the system. Therefore, what happens

to one layer or component can directly and indirectly

affect the rest of the system, and ﬁnally impact the

operation of the whole power grid.

The other main requirement for the design of the

an ICT architecture is loose coupling which falls un-

der the architectural requirement category. Infrastruc-

tures can be tightly or loosely coupled. Tight cou-

pling addresses the strong dependency of one layer or

component on another one in a cyber system. Cas-

cading failures are usually caused by tightly coupled

systems, as disturbances in one section can affect the

other sections to which they are tightly coupled. In

cyber infrastructure, loose coupling allows for ﬂex-

ibility of design of the ICT architecture. In a lay-

ered ICT architecture a loosely coupled system has

a little or no knowledge about the internal details of

the other layers and communication between layers

is based on abstractions. This will affect their oper-

ational performance, and ﬂexibility to deal with fail-

ures and changes in the system.

Overall, ICT requirements such as distributed,

layered, component-based, and loosely coupled can

inﬂuence the degree of cyber-dependencies of the

Smart Grid.

8 CYBER RISKS

Due to the nature, complexity and interdependency

of the Smart Grid it is essential to consider security

objectives in its communication systems. These ob-

jectives are conﬁdentiality, integrity and availability

which are required for control management and mon-

itoring and must operate within acceptable risk level

(Zhang et al., 2010). Since the emergence of dif-

ferent Smart Grid technologies, standards and tech-

niques have been developed by the community. How-

ever there is still a lack of a widely accepted protocol

which includes all the elements of Smart Grid such as

smart meters, smart devices and household appliances

and the integration of renewable energy. One of the

reasons for the lack of full deployment of Smart Grid

is regional policy making and the deﬁnition of Smart

Grid in that region or country. Nowadays power sys-

tems are much more susceptible to failures and black-

outs due to increase in connectivity to information in-

frastructure than when power systems were operating

in isolation. This means the whole system needs to be

more resilient and secure by implementing new secu-

rity measurements and in most cases there is a need

to change the ICT architecture that ﬁts the system and

satisﬁes the new requirements.

9 CONCLUSION

Smart Grid and smart energy systems are interdepen-

dent and highly dependent on communication net-

works. In addition there is high dependencyon the in-

formation and data generated by these systems which

are critical for the operation of dependent systems.

This paper has presented a top down approach to rep-

resent the data and communication interdependencies

between legacy power systems, Smart Grid and smart

energy systems individually and as a whole. It has

also shown how important it is to design and develop

an ICT architecture that facilitates such systems using

the existing ICT infrastructure.

Additionally, it has demonstrated how a number

of design decisions while developing an ICT archi-

tecture affects the cyber interdependency of the Smart

Grids. Having an ICT architecture that ﬁts the re-

quirement of each discrete system enables the over-

all systems to be more reliable, resilient and secure

against any threats and failures.

ACKNOWLEDGEMENTS

The research reported in this paper was partially sup-

ported by of the STRATA; Layers for Structuring

Trustworthy Ambient Systems funded by the Engi-

neering and Physical Sciences Research Council un-

der Programme Grant EP/N023641/1.

The authors also would like to thanks Cliff Jones

for his feedbacks and guidance on this paper.

REFERENCES

(1996). Telecommunications: Glossary of telecommunica-

tion terms.

Aloula, F., Al-Alia, A., Al-Dalkya, R., Al-Mardinia, M.,

and El-Hajj, W. (2012). Smart grid security: threats,

vulnerabilities and solutions. International Journal of

Smart Grid and Clean Energy, 1(1):1–6.

Buldyrev, S. V., Parshani, R., Paul, G., Stanley, H. E., and

Havlin, S. (2010). Catastrophic cascade of failures

in interdependent networks. Nature, 464(7291):1025–

1028.

Committee, I. C. S. S. E. S. and Board, I.-S. S. (1998).

Ieee recommended practice for software requirements

ICISSP 2017 - 3rd International Conference on Information Systems Security and Privacy

536

speciﬁcations. Institute of Electrical and Electronics

Engineers.

Ebrahimy, R. (2014). Investigating SCADA failures in in-

terdependent critical infrastructure systems. CoRR,

abs/1404.7565.

Emmerich, W. Distributed system principles.

Gungor, V. C., Sahin, D., Kocak, T., Ergut, S., Buccella,

C., Cecati, C., and Hancke, G. P. (2011). Smart grid

technologies: Communication technologies and stan-

dards. IEEE Transactions on Industrial Informatics,

7(4):529–539.

Huang, Y., Esmalifalak, M., Nguyen, H., Zheng, R., Han,

Z., Li, H., and Song, L. (2013). Bad data injection

in smart grid: attack and defense mechanisms. IEEE

Communications Magazine, 51(1):27–33.

Jefferson, M. (2008). Accelerating the transition to sus-

tainable energy systems. Energy Policy, 36(11):4116–

4125.

Laverty, D. M., Morrow, D. J., Best, R., and Crossley, P. A.

(2010). Telecommunications for smart grid: Backhaul

solutions for the distribution network. In IEEE PES

General Meeting, pages 1–6.

Luan, W., Sharp, D., and Lancashire, S. (2010). Smart grid

communication network capacity planning for power

utilities. In IEEE PES T D 2010, pages 1–4.

MA (1993). MerriamWebsters Collegiate Dictionary, vol-

ume 10th ed. Springﬁeld.

Momoh, J. (2012). Smart grid: fundamentals of design and

analysis, volume 63. John Wiley & Sons.

Nielsen, S., Sorknæs, P., and Østergaard, P. A. (2011). Elec-

tricity market auction settings in a future danish elec-

tricity system with a high penetration of renewable

energy sources–a comparison of marginal pricing and

pay-as-bid. Energy, 36(7):4434–4444.

Orecchini, F. and Santiangeli, A. (2011). Beyond smart

grids–the need of intelligent energy networks for a

higher global efﬁciency through energy vectors inte-

gration. International Journal of hydrogen energy,

36(13):8126–8133.

Oughton, E. J., TRAN, M., JONES, C. B., and

EBRAHIMY, R. (2016). 9 digital communica-

tions and information systems. The Future of Na-

tional Infrastructure: A System-of-Systems Approach,

1995(2000):181.

Pourmirza, Z. (2015). An ICT architecture for the neigh-

bourhood area network in the Smart Grid. PhD thesis,

University of Manchester.

Pourmirza, Z. and Brooke, J. M. (2013). A realistic ict

network design and implementation in the neighbour-

hood area of the smart grid. Smart Grid and Renew-

able Energy, 4(6):436–448.

Rahnamay-Naeini, M. (2016). Designing cascade-resilient

interdependent networks by optimum allocation of

interdependencies. In 2016 International Confer-

ence on Computing, Networking and Communications

(ICNC), pages 1–7.

Rinaldi, S. M., Peerenboom, J. P., and Kelly, T. K. (2001).

Identifying, understanding, and analyzing critical in-

frastructure interdependencies. IEEE Control Sys-

tems, 21(6):11–25.

Rohjans, S., Dnekas, C., and Uslar, M. (2012). Require-

ments for smart grid ict-architectures. In 2012 3rd

IEEE PES Innovative Smart Grid Technologies Eu-

rope (ISGT Europe), pages 1–8.

Safdar, S., Hamdaoui, B., Cotilla-Sanchez, E., and Guizani,

M. (2013). A survey on communication infrastructure

for micro-grids. In 2013 9th International Wireless

Communications and Mobile Computing Conference

(IWCMC), pages 545–550.

Wang, W., Xu, Y., and Khanna, M. (2011). A survey on the

communication architectures in smart grid. Computer

Networks, 55(15):3604–3629.

Wissner, M. (2011). The smart grid–a saucerful of secrets?

Applied Energy, 88(7):2509–2518.

Wolsink, M. (2012). The research agenda on social accep-

tance of distributed generation in smart grids: Renew-

able as common pool resources. Renewable and Sus-

tainable Energy Reviews, 16(1):822–835.

Wu, Y.-n., Chen, J., and Liu, L.-r. (2011). Construc-

tion of china’s smart grid information system anal-

ysis. Renewable and Sustainable Energy Reviews,

15(9):4236–4241.

Yan, Y., Qian, Y., Sharif, H., and Tipper, D. (2013). A sur-

vey on smart grid communication infrastructures: Mo-

tivations, requirements and challenges. IEEE Commu-

nications Surveys Tutorials, 15(1):5–20.

Yang, L., Xue, H., and Li, F. (2014). Privacy-preserving

data sharing in smart grid systems. In Smart Grid

Communications (SmartGridComm), 2014 IEEE In-

ternational Conference on, pages 878–883.

Zhang, R., Zhao, Z., and Chen, X. (2010). An overall re-

liability and security assessment architecture for elec-

tric power communication network in smart grid. In

Power System Technology (POWERCON), 2010 Inter-

national Conference on, pages 1–6.

Zio, E. and Aven, T. (2011). Uncertainties in smart grids

behavior and modeling: What are the risks and vul-

nerabilities? how to analyze them? Energy Policy,

39(10):6308–6320.

Cyber-interdependency in Smart Energy Systems

537