[ieee 2013 10th international conference & expo on emerging technologies for a smarter world...
Post on 16-Feb-2017
220 Views
Preview:
TRANSCRIPT
1
Novel Smart Grid and SCADA System
Interdependency Networks for Future’s Clean,
Sustainable and Green Energy
Pravin Chopade1, Dr. Serap Karagol
2, Dr. Marwan Bikdash, Dr. Ibraheem Kateeb, Dr. Numan Dogan
Computational Science and Engineering Department-COE, CST-SOT, ECEN-COE
North Carolina A&T State University, Greensboro, NC, USA
pvchopad@ncat.edu, serap.karagol@omu.edu.tr, bikdash@ncat.edu, kateeb@ncat.edu, dogan@ncat.edu.
(1 Ph.D. candidate, NCA&TSU, USA and Associate Professor, Bharati Vidyapeeth Deemed University College of Engineering, Pune, India, 2 Assistant
Professor, Electrical and Electronics Engineering, Ondokuz Mayis University, Samsun, Turkey)
Abstract— A novel approach of interdependency modeling of
Smart Grid and Supervisory Control and Data Acquisition
(SCADA) networks is introduced. Methodological approach for
vulnerability reduction considering both structural and
functional vulnerability is introduced in this paper. The analysis
of cyber-attacks on Smart Grid and SCADA network is
discussed. Paper discusses importance and need of clean,
sustainable and green energy for the future. Paper also analyzes
interdependent Smart Grid and SCADA network under severe
emergency situations. The contribution of this paper provides
novel network with integrated infrastructures for active and
survivable network operation. Thus novel Smart Grid and
SCADA interdependency network will act as future’s clean,
sustainable and Green energy model.
Keywords- Smart Grid, SCADA, Vulnerability, Emergency,
Interdependency modeling, Green Energy.
I. INTRODUCTION
Past historic blackouts highlight the vulnerability of the
Smart electric grid infrastructures and their interdependencies.
The large geographic extension of power failures effects is
related to the high interconnectivity of power grid
transmission and distribution infrastructures and the multiple
interdependencies existing between these infrastructures and
Supervisory Control and Data Acquisition (SCADA) i.e. the
information infrastructures supporting the control, the
monitoring, the maintenance and the exploitation of power
supply systems. The blackout is an example of the potential
ramifications of a failure or attack on the Smart Power grid
and SCADA systems. SCADA systems and their components
can be found in a number of national infrastructures including
the water, oil, and gas industries. SCADA systems are
computer controlled devices that perform and relay physical
changes in infrastructure systems to technical operators. They
are capable of monitoring millions of data points
simultaneously, and can therefore be manipulated by a cyber-
attack [1]. An adversary thus can penetrate the electrical
power grid, or other control system, with little more than a
laptop and an Internet connection. This is a major threat [2].
The source of vulnerability includes natural disasters,
equipment failures, human errors, or deliberate sabotage and
attacks. Increased electricity demand and transmission
bottlenecks make the complex power system even more
vulnerable and, with triggering disruption at opportune time,
topple it over to blackouts [3], [4].
Clearly there is a need to analyze and model critical
infrastructures in the presence of interdependencies in order:
i) To understand how such interdependencies may
contribute to the occurrence of large outages and
blackouts and
ii) To develop architectural solutions those are well
suited to improve the dependability and resilience of
power grid infrastructures.
In this paper we aim to address these objectives focusing
attention on two interdependent infrastructures: the Smart
Power Grid infrastructure and SCADA i.e. the information
infrastructures supporting management, control and
maintenance functionality.
Section II addresses cyber-attacks on smart grid and
SCADA network. Section III discuses need of clean,
sustainable and green energy for future. In Section IV, we
discuss novel Smart Grid and SCADA network modeling.
Section V analyses interdependent Smart Grid and SCADA
network under severe emergencies.
II. CYBER-ATTACKS ON SMART GRID AND SCADA
NETWORKS
Smart Grid design and deployment must take into account
the current cyber vulnerabilities in the legacy of power grid.
The known vulnerabilities in the existing legacy power grid
should continue to be addressed and mitigated in concert with
the implementation of Smart Grid technologies [5]. Resistance
to attack is one of the seven principle characteristics of the
Smart Grid vision [6]. However, implementation of a Smart
Grid that is resistant to attack is particularly difficult for
several reasons. The Smart Grid deployment will increase the
This work is supported by The Defense Threat Reduction Agency (DTRA) and Pennsylvania State University under contract DTRA01-03-D-
0010/0020 and sub-contract S03-34.
978-1-4799-2546-9/13/$31.00 ©2013 IEEE
2
complexity of the existing system and will include the addition
of many new communication paths. The increased complexity
and expanded communication paths can easily lead to an
increase in vulnerability to cyber-attack [3]. The size (millions
of nodes) of a fully implemented Smart Grid and an
unpredictable intelligent adversary make it difficult to
anticipate how attacks may be manifested [7], [8]. Smart Grid
technology that has known vulnerabilities has already been
deployed in some parts of the current power grid [9].
Furthermore, the goal of “resistance to attack” is in
competition with some of the other desired characteristics of
the Smart Grid, e.g. the goal of “optimizing assets and
operating efficiently”. The desire to minimize costs and to
provide services tend to take priority over the desire for
security in the face of a threat that is not well understood.
Chronology of reported cyber-attacks on electric and other
utilities including data obtained by Motter [10] are given
below and other sources, reports some incidents and attacks
that affected electricity and other critical utilities during the
last decade. It is noteworthy that due to the high sensitivity of
such data, a complete historical data base of these events is not
publicly available (and probably it will never be in the future).
Nevertheless, these examples show that the threat is real and it
will be increasing due to market deregulation and the
increased complexity and openness of the future SCADA
architectures for power infrastructures. Actually, several
working groups and initiatives dedicated to the analysis and
assessment of security related vulnerabilities and threats in the
context of power system infrastructures and the proposal of
appropriate solutions to mitigate them have been created
recently.
Chronology of reported cyber-attacks on SPGN and SCADA
network [10].
1994: Salt River Project: A water facility in Arizona
was breached by a cyber-attack. The hacker
trespassed in critical areas that could have caused
significant damage.
1997: A teenager remotely disabled part of the public
switching network in Massachusetts, which shutdown
telephone service to 600 customers.
2000: A disgruntled employee of an Australian
company used his laptop car computer to remotely
hack into the controls of a sewage treatment system,
which caused 264,000 gallons of raw sewage to be
released into public waterways of Australia over a
period of two months. This caused marine life to die
and creek water to turn black, producing an
unbearable stench to nearby residents, among other
impacts.
2000: On October 13, the Control System of Ertan
Hydro Station received unexpected signals, and then
they reduced generation 900 MW within 7 seconds,
almost causing Sichuan power system to collapse.
2001: Hackers attacked the California Independent
System Operator managing the electricity supply of
California. The Los Angeles Times reported that the
cyber hackers “got close” to disrupting power flow
during the California rolling blackouts in May 2001.
2001: October 1, many Fault Recorders
dysfunctioning were caused by a Timer Logical
Bomb, this type of device had been installed on 146
sets in China.
2003: The SQL Slammer worm infected and disabled
internal systems at nuclear power plant in Ohio,
Safety was never compromised, but a safety
parameter display system and the plant process
control computer were knocked off-line by the cyber
worm for several hours.
2003: On December 30, several viruses were found in
the control systems of 3 HVDC convert stations
(Longquan, Zhengping, Ercheng), which transfer
total 6000 MW from Three Gorge to East and South
of China.
2006: May 18/17/15, Malware Infection LEAKS
Japanese Power Plant Data. A malware infection has
being blamed for the leak of sensitive Japanese
power plant information onto the Internet. The
information included key facility location and
operation procedures for the Chubu Electric Power
Company’s thermal power plant in Owase, Mie
Prefecture; some employee data were also
compromised. A sub-contractors use of file sharing
software is suspected to have caused the malware
infection.
Securing the assets of electric power delivery systems,
from the control center to the substation, to the feeders and
even to customer meters, requires an end-to-end security
infrastructure that protects the myriad of communication
assets (control center-based) SCADA, RTUs (Remote
Terminal Units), PLCs (Programmable logic controllers),
power meters, digital relays, and bay controls) used to operate,
monitor, and control power flow and measurement.
III. NEED OF CLEAN, SUSTAINABLE AND GREEN ENERGY
FOR THE FUTURE
Today, an electricity disruption such as a blackout can have
a domino effect- a series of failures that can affect banking,
communications, traffic, and security. This is a particular
threat in the winter, when homeowners can be left without
heat. A smarter grid will add resiliency to our electric power
system and make it better prepared to address emergencies
such as severe storms, earthquakes, large solar flares, and
terrorist attacks. Because of its two-way interactive capacity,
the Smart Grid will allow for automatic rerouting when
equipment fails or outages occur. This will minimize outages
and minimize the effects when they do happen [11].
When a power outage occurs, Smart Grid technologies will
detect and isolate the outages, containing them before they
become large-scale blackouts. The new technologies will also
help ensure that electricity recovery resumes quickly and
strategically after emergency- routing electricity to emergency
services first, for example. In addition, the Smart Grid will
3
take greater advantage of customer-owned power generators to
produce power when it is not available from utilities. By
combining these “distributed generation” resources, a
community could keep its health center, police department,
traffic lights, phone system, and grocery store operating
during emergencies [12].
In addition, the Smart Grid is a way to address an aging
energy infrastructure that needs to be upgraded or replaced.
It’s a way to address energy efficiency, to bring increased
awareness to consumers about the connection between
electricity use and the environment. And it’s a way to bring
increased national security to our energy system- drawing on
greater amounts of home-grown electricity that is more
resistant to natural disasters and attack [13].
The Smart Grid is not just about utilities and
technologies; it is about giving you the information
and tools you need to make choices about your
energy use. If you already manage activities such as
personal banking from your home computer, imagine
managing your electricity in a similar way. A smarter
grid will enable an unprecedented level of consumer
participation.
For example, you will no longer have to wait for your
monthly statement to know how much electricity you
use. With a smarter grid, you can have a clear and
timely picture of it. “Smart meters”, and other
mechanisms, will allow you to see how much
electricity you use, when you use it, and its cost.
Combined with real-time pricing, this will allow you
to save money by using less power when electricity is
most expensive.
While the potential benefits of the Smart Grid are
usually discussed in terms of economics, national
security, and renewable energy goals, the Smart Grid
has the potential to help you save money by helping
you to manage your electricity use and choose the
best times to purchase electricity. And you can save
even more by generating your own power.
IV. NOVEL SMART GRID AND SCADA NETWORK
MODELING
The complex network theory has been successfully applied
in the analysis of various technological networks. As the
power grid increases in size and complexity, it is becoming
more important to understand the emergent behaviors that can
take place in the system. We try to apply complex network
theory to smart grid and SCADA network analysis. Both these
interdependent networks can often be represented in a useful
way as networks; the structure (topology) of networks is
mathematically described in terms of graphs, i.e., sets of
vertices (nodes) and edges (links). For a smart electric power
grid, the vertices can be power plants, stations and power
users, and the edges power lines [14] and for SCADA
networks, Master Control Station (MCS), Remote Terminal
Units (RTUs), Intelligent Electronics Devices (IEDs),
Programmable Logic Controllers (PLC) can be described as
nodes and information and communication lines can be
modeled by links (as adopted in this work), or vice versa [13],
[15]. Consider an example of Smart Grid and SCADA system
shown in Fig. 1, where the electrical needs of a node in the
SCADA network can be supplied by one or more nodes in the
Smart Grid. Fig. 1 shows extracted graph topology (Structural
Topology) of Smart Grid, SCADA network and their
interconnections.
Figure 1. The interdependent Smart Grid and SCADA network.
A novel or methodological approaches to comprehensively
analyze the vulnerability of interdependent infrastructures are
required, two types of vulnerability are considered:
Structural vulnerability and
Functional vulnerability
For structural vulnerability, infrastructures topologies are the
only information while operating regimes of different
infrastructures are further taken into consideration to analyze
functional vulnerability. The vulnerability analysis process of
interdependent infrastructures can be seen in Fig. 2. From the
figure, the first step is to extract the topology of each
infrastructure, i.e. what are described by nodes and what are
modeled by links. When infrastructure topologies have been
extracted, their operating regimes can be further considered
for analysis on functional vulnerability. However, no matter it
is the analysis on structural vulnerability or functional
vulnerability, the most important thing is to model
interdependences between two infrastructures. There are
several types of interdependences between infrastructures.
Different scholars have different view on the classification [7].
A. Structural vulnerability Analysis
Vulnerability is related to attacks and can be described as
the decrease of system efficiency after an attack. To analyze
structural vulnerability, infrastructure topologies are only
considered and the most important thing is to determine what
are used to describe structural efficiency. There are many
definitions on structural efficiency, such as average shortest
distance, network diameter, and cluster efficiency, but they all
have some limitation. Usually, the average reciprocal shortest
path lengths of networks are used to measure the structural
efficiency and it is generally accepted [8], [16].
4
Figure 2. The vulnerability analysis process of interdependent infrastructures
Smart Grid and SCADA network.
The topology is represented as a graph },{ EVG with N
nodes, }{ ivV is the set of vertices and E is the set of edges,
denote ),( ji vvd by the shortest path lengths connecting two
nodes in the network or it indicates the minimum number of
edges that one crosses to traverse from node i to node j , then
the structural efficiency )(GX of one infrastructure can be
defined as follow:
(1) 1
.
1)( ∑
∈∈ , SP GjGi ijSP dNNGX
Where PN is the number of resource nodes (such as
generator nodes in the smart grid network and RTU nodes in
the SCADA network) and SN is the number of load nodes
(nodes offer services to other systems). This distance matrix
ijd has the same dimensions as the adjacency matrix A .
The adjacency matrix A of size NN be written as
(2) vertex to vertex from edge no is thereif0
vertex to vertex from edgean is thereif1
ji
jiAij
edges. parallel by connecteddirectly are and if kjikAij
When two nodes are not connected at all, or become
disconnected due to attacks, their shortest path length
),( ji vvd becomes infinite, and then ),(
1
ji vvd is zero. If
)(GX is large, it is indicated that the network is well
connected and has high efficiency.
In order to fully understand the structure of a power grid,
one needs to know not only its topology, but also the structure
that results from the physical properties that govern flow. To
understand the electrical structure of a given smart grid we
need a measure of electrical distance. Electrical distance does
not perfectly represent all of the ways in which components in
a grid are connected; it is a useful starting point for structural
analysis. The electrical distance, is the absolute value of the
inverse of the system admittance matrix given by Eq. (3).
(3) 11
YYEd
Taking electrical distance into consideration the structural
efficiency )(GX is given by,
(4) 1
.
1)( ∑
∈∈ , SP GjGi ijSP ZNNGEX
(5) .
1)( ∑
∈∈ , SP GjGi
ijSP
YNN
GEX
Where, )(GEX gives electrical structural efficiency. Zij is
the absolute value of the series impedance of the shortest
electrical path between buses i and j.
We consider the power from a generator to be accessible to
a consumer if there is a path of transmission lines between the
two. In practice, the existence of a connection between two
substations does not always imply that power can be
transferred across it as there may be capacity or other
constrains present. In addition, another important thing is to
model the structural interdependence. For analysis on
structural vulnerability, when one smart grid node is attacked,
all SCADA nodes connected by this power node will be
deleted. Similarly, when a SCADA node is attacked, the
corresponding load-based power generators will be also
removed. This will change network topologies. The structural
efficiency can be calculated and structural vulnerability can be
further analyzed.
B. Functional Vulnerability Analysis
To analyze the functional vulnerability, operating regimes
of different infrastructures should be further considered.
While the connectedness of the smart grid allows for the
transmission of power over large distances, it also implies that
local disturbances propagate over the whole grid. The failure
of a power line due to lightning strike or short-circuit leads to
the overloading of parallel and nearby lines. Power lines are
guarded by automatic devices that take them out of service
when the voltage on them is too high. Generating substations
are designed to switch off if their power cannot be transmitted;
this protective measure has the unwanted effect of diminishing
5
power for all consumers. Another possible consequence of
power line failure is the incapacitation of transmission
substations, possibly causing that the power from generators
cannot reach distribution substations and ultimately
consumers.
In order to account for such functional vulnerability of the
system we need to consider power network dynamics. The
power network dynamics are coupled by its network Eq. (6).
(6) .VYI bus
The bus admittance of system is given by
(7) /1 ijijijij jBGZY
In the system, there is a complex power injected into the thi
bus is given by [17]
Ni
jQPIVS iiiii
,.........3,2,1
(8) *
Real or active power coming out of bus i :
(9) )sin()cos(
1
N
jjiijjiijjii BGVVP
Reactive power coming out of bus i :
(10) )cos()sin(
1
N
jjiijjiijjii BGVVQ
Where ,iP iQ and ,iV are the active power, reactive power
and complex voltage at bus ,i respectively. )( ji is the
difference in angles of the voltage phasors at two buses i and
.j
For a medium length or a long transmission line .RX For
ease of calculation we can thus ignore .R After simplification
active and reactive power flow equations are given by
(11) )sin(.
)sin(. jiij
jijiijjiij
X
VVBVVP
))cos((. jijiiijij VVVBQ
(12) ))cos((
ij
jijiiij
X
VVVQ
Power balance equation is given by [18]
(13)
11
11
Loss
N
iD
N
iG
Loss
N
iD
N
iG
QQQ
PPP
ii
ii
Under different vulnerability or attacks conditions above
mentioned system parameters changes which affect the power
flow in the system and sometimes it leads to complete system
collapse or blackout situation.
V. ANALYSIS OF INTERDEPENDENT SMART GRID AND
SCADA NETWORKS UNDER SEVERE EMERGENCIES
The increased complexity and expanded communication
paths can easily lead to an increase in vulnerability to cyber-
attack [5,13]. The size (millions of nodes) of a fully
implemented smart grid and an unpredictable intelligent
adversary make it difficult to anticipate how attacks may be
manifested [14]. Smart Grid technology that has known
vulnerabilities has already been deployed in some parts of the
current power grid [9, 14].
We modeled IEEE-30 bus [19] combined smart grid with
Microgrid as spinning reserve capacity and SCADA network
using MATLAB [20]. Some of the details of this model are
given in Table I.
Table I. Configuration of IEEE-30 bus Smart Grid and SCADA Network.
Smart Grid network SCADA Network
No of Buses or
Nodes
30 No of RTUs
(Nodes)
25
No of lines or
branches
41 No of IEDs
(Nodes)
30
No of Generator
Buses
06
At Bus 1, 2,
5, 8, 11, 13
PLC 06
Slack Bus Bus 1 MTU 01
We tested Interdependent network performance for structural
and functional vulnerability. Under failure conditions i.e. the
case of structural vulnerability case shown in Fig. 3 which
shows average distance under random vertex failures of
combined Smart Grid and SCADA network. The functional
behavior is shown in Fig. 4. Fig. 4 shows the analysis of
interdependency network with Microgrid as spinning reserve.
Figure 3. Average distance under random vertex failures of interdependent
Smart Grid and SCADA network.
When 5 Units with 710 MW tripped down without
interdependent and Microgrid network frequency drops to
59.886 Hz within 5.8 seconds. If there is no immediate
spinning reserve capacity available from Microgrid then
complete grid may reach to failure or blackout stage but with
0 0.2 0.4 0.6 0.8 10
1
2
3
4
5
fv-fraction of removed vertices
lv-a
vera
ge d
ista
nce
Random vertex failures of combined Smart Grid and SCADA network
6
interdependent and Microgrid network frequency excursion
arrests at 59.950 within 0.7 seconds and complete system
survive. Smart Grid and SCADA interdependency network
combined with Microgrid can reduce Grid congestion and thus
it will help to avoid complete system failure or blackouts
situations.
Figure 4. Simulation test results for Novel Smart Grid and SCADA System
Interdependency network.
VI. CONCLUSIONS
Smart grid technologies will enable higher percentages of
centralized and distributed renewable generation to be
integrated into the grid efficiently and reliably, so they can
become significant contributors to our overall energy
platform. Renewable power can become as mainstream as coal
is today, thereby reducing carbon emissions, natural resource
depletion, and dependence on foreign oil -ultimately helping
us improve our energy security. Structural and functional
vulnerability analysis can be used to analyze the vulnerability
of interdependent infrastructures. Our results indicate that
interdependent smart grid and SCADA network is more
vulnerable. Microgrid can share large portion of the load and it
will reduce pressure of main power grid. Thus it will provide
better economical solution. Novel network will reduce carbon
footprint which will provide efficient and sustainable energy
solutions.
ACKNOWLEDGMENT
The authors gratefully acknowledge The Defense Threat Reduction Agency (DTRA) and Pennsylvania State University for their support and finance for this Project.
The Authors of this paper are greatly thankful to the Management of Bharati Vidyapeeth Pune, India, Bharati Vidyapeeth Deemed University Pune, Dr. Anand R. Bhalerao, Principal and Dean, Bharati Vidyapeeth Deemed University
College of Engineering, Pune, India, for their support and constant inspiration.
REFERENCES
[1] C. Ten; C. Liu; G. Manimaran, “Vulnerability Assessment of Cybersecurity for SCADA Systems,” Power Systems, IEEE Transactions on , vol.23, no.4, pp.1836,1846, Nov. 2008 doi: 10.1109/TPWRS. 2008.2002298
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652578
& isnumber=4652575
[2] Tai-hoon Kim, “Securing Communication of SCADA Components in Smart Grid Environment”, International Journal of Systems Applications, Engineering & Development, Issue 2, vol. 5, 2011, pp. 135-142.
[3] A. Murray, and T. Grubesic, “Critical Infrastructure-Reliability and Vulnerability”, Advances in Spatial Science, Springer Publications, 2007, ISBN 978-3-540-68055-0.
[4] F. R. Spellman and R.M. Bieber, Energy Infrastructure Protection and Homeland Security, Government Institution (GI), The SCARECROW Press, Inc, 2010, ISBN 978-1-60590-678-2.
[5] NISTIR 7628, U. S. Department of Commerce, “Guidelines for Smart Grid Cyber Security: Vol.1, Smart Grid Cyber Security Strategy, Architecture, and High-Level Requirements”, The Smart Grid Interoperability Panel- Cyber Security Working Group, August 2010.
[6] U.S. Department of Energy Office of Electricity Delivery and Energy Reliability, “Study ofSecurity Attributes of Smart Grid Systems - Current Cyber Security Issues”, April 2009.
[7] P. Pederson, D. Dudenhoeffer, S. Hartley, M. Permann, “Critical Infrastructure Interdependency Modeling: A Survey of U.S. and International Research”, INL Report, U.S. Department of Energy
National Laboratory and Battelle Energy Alliance, August 2006.
[8] S. Sun, Z. Liu, Z. Chen, Z. Yuan, “Error and attack tolerance of evolving networks with local preferential attachment”, Physica A 373, 2007, pp. 851-860, Elsevier, doi:10.1016/j.physa.2006.05.049.
[9] “Smart Grid Legislative and Regulatory Policies and Case Studies”, U.S. Energy Information Administration, U.S. Department of Energy, Washington, DC 20585, December 2011.
[10] A. E. Motter, “Cascade Control and Defense in Complex Networks”, Phys. Rev. Lett. 93, 098701, 2004.
[11] T. G. Lewis, Critical Infrastructure Protection in Homeland Security Defending a Networked Nation, A John Wiley & Sons, Inc., Publications, 2006, ISBN-13: 978-0-471-78628-3.
[12] Cisco Smart Grid Security Solutions, 2009.
[13] Clark W. Gellings, The Smart Grid: Enabling Energy Efficiency and Demand Response, CRC Press, 2009, ISBN-10: 0-88173-623-6.
[14] T. Flick, J. Morehouse and C. Veltsos, Securing the Smart Grid: Next Generation Power Grid Securiy, Elsevier Publications, 2011, ISBN 978-1-59749-570-7.
[15] Ronald L. Krutz, Securing SCADA Systems, Wiley Publishing, Inc., 2006, ISBN-13: 978-0-7645-9787-9.
[16] Charles J. Kim, Obinna B. Obah, “Vulnerability Assessment of Power Grid Using Graph Topological Indices”, International Journal of Emerging Electric Power Systems, vol. 8, Issue 6, 2007, Article 4, pp. 1-17.
[17] Leonard L. Grigsby, Electric Power Engineering Handbook, Second Edition, CRC Press, 2006.
[18] Yoshihide Hase, Handbook of Power System Engineering, John Wiley & Sons Ltd, 2007, ISBN-13: 978-0-470-02742-4.
[19] University of Washington, power systems test case archive. http://www.ee.washington.edu/research/pstca.
[20] http://www.mathworks.com/
top related