Power Transmission Network Vulnerability Analysis
Power Transmission Network Vulnerability Analysis
[Link] (0123456789().,-volV)(0123456789().,-volV)
TECHNICAL PAPER
Abstract
The functionality of electric power network are continuously challenged by dynamic and diverse disruptions. The vul-
nerability analysis in the context of resiliency, using complex networks has significantly modified in the past few years,
enlighten valuable ideas to examine the structural, topological and constructional behavior of a power transmission
network. This network having similarity with graph theoretic approach in accordance with complex graph theory. Vul-
nerability is assessed in this paper by evaluating betweenness and netability for topological analysis of vulnerability.
Undesirable high impact low occurrence events result in disruption of a large power network as well as to the whole grid.
Framework of resiliency in this paper is done in conceptual long-term basis and vulnerability assessment approach.
Simulation carried out in this paper to realize the assessment of vulnerability using netability. The considered test system is
standard IEEE 14 bus system. One FACTS device, namely STATCOM is employed to the system to show the
improvement in the steady state bus voltage.
1 Introduction outage of the power grid all through the Country (Zang
et al. 2019; Xue and Xiao 2013). It is a very crucial time
Electrical power system is the most indispensable infras- for all the researcher to identify the critical lines which are
tructure of modern society. Large outage or blackout in the very important to the uninterrupted supply of power flow.
power system is not only affects the consumers but also Critical loads should be identified and the essential mea-
affects the socio-economic structure of a developing sures should be taken to supply the power at that node.
country. Resiliency of a power network is capability of a Now a days power transmission network is getting more
network to regain its original conditions after some unde- complicated and complex as various type of equipment are
sirable events. A conceptual framework of resiliency is incorporated in the system. In consonance with the elec-
presented by analyzing vulnerability in electrical power trical power transmission network, the change in the flow
systems. In high impact low frequency events, which part or losses of power in the network is specified as vulnera-
of the network will be more harden to enhance the resi- bility. The range or the durability of that loss by damaging
liency of the system, can be determined by vulnerability one node or one line is being calculated as vulnerability. In
analysis. Power blackout over a huge region may cause other words, it may be representing as the impact of the
major blackouts indulging severity, measured either in
& Dipanjan Bose terms of lost of power or economic loss. Complex network
dbose535@[Link] theory is applied in this analysis from graph theoretic
[Link] approach. Analysis of vulnerability performed by consid-
Chandan Kumar Chanda ering topological architecture (Bompard et al. 2011; Beyza
ckc_math@[Link] et al. 2018; Kinney et al. 2005; Holmgren 2007; Rout et al.
[Link] 2016; Chen et al. 2007; Mei et al. 2011; Wei et al. 2018;
Abhijit Chakrabarti Wu et al. 2017; Chen et al. 2012; Sun 2005; Bompard et al.
achakrabarti58@[Link]; 2008; Rout et al. 2015; Guohua et al. 2008) of complex
[Link]
network theory.
1
Department of Electrical Engineering, Indian Institute of
Engineering Science and Technology, Shibpur,
Howrah 711103, India
123
Microsystem Technologies
2 Power system implemented as a complex be one of the indicators of criticality of the nodes in the
network power networks.
The electrical power network can be implemented by graph 2.4 Distance and efficiency
theoretic approach. Fundamental graph theory is backbone
of complex network theory. The power network is pre- The characteristics of a line along a path generally depend
sented in an undirected graph, where the number of ver- on its impedances and the power flows through the path.
tices/nodes represents by and the number of edges/lines The economical and nominal loss situation may possible if
represents by N and L are the total number of nodes and the impedances through the path is minimal. So, distance is
edges respectively. The connection between two nodes can dependent on the above said properties of the power net-
be shown by adjacency matrix, whose order is N 9 N. The work. If the impedance in the line is fixed the costs changes
entry of the adjacency matrix is either zero or one. with the change in power flow and if the power flow is
fixed then the costs changes with the change in
2.1 Geodesic distance impedances.
Zij ¼ zii þ zjj 2zij ð4Þ
In an un-weighted or weighted graph, the lines joining the
nodes are called length of a path. A minimum length path is where, the bus impedance matrix is denoted by Z. The
considered to be the shortest path or Geodesic path. Geo- geodesic distance is replaced by electrical distance when
desic distance, dij , is the length of a geodesic path con- approaching extended topological method.
necting through ith bus to jth bus of the power network or dij ¼ Zij ð5Þ
graph.
The minimum path walked to reach from ith node to jth
2.2 Average shortest path length node through alternative sequential nodes is described by
the distance. Efficiency of the power network associated
The mean shortest path length L is described as the average with grid is calculated as follows: (Chowdhury et al. 2015)
of geodesic lengths over all couples of nodes: 1 X N
1
E¼ ð6Þ
1 X N NðN 1Þ i;j6¼1 dij
L¼ dij ð1Þ
NðN 1Þ i;j¼1
where number of nodes denoted by, N and shortest path for
The number of lines connecting the couple of nodes in node i to j is denoted by dij .
an undirected and un-weighted graph is considered to be its
length. 2.5 Electrical centrality and its measurement
where weights of the lines joining i and j represents by wij . The idea from graph theory, it is proven that, if an edge
In a network the connectivity of a node depicts by the counts in more number of paths, then that component
degree of node. Higher connectivity nodes are more rele- becomes more critical for the transmission in power net-
vant than other nodes. Therefore, the degree of node could work. This may be employed as a justification to the
importance of an edge in terms of its betweenness.
123
Microsystem Technologies
The number of shortest paths connecting the two non- harmonic mean of efficiencies of all the links the path
adjacent buses is the main criterion for transferring the passed by. In all of the paths connecting nodes i and j, the
energy from one node to another. Thus the vulnerability one with the maximal efficiency is called the most efficient
measures as the number of shortest paths passes through path, and its efficiency is denoted as E. If there’s no path
that element is defined as betweenness centrality of that between two nodes i and j, we have dij = 0.
element. The global efficiency of the network is given by
Node betweenness of a node v can be expressed by the (Chowdhury et al. 2015):
following formula: X N
1 1
X
N X
N E¼ ð9Þ
rij ðvÞ NðN 1Þ i;j6¼1 dij
BðvÞ ¼ ; i 6¼ j 6¼ voV ð7Þ
i j
rij
Vulnerability of a line is being calculated by removing
where rij ðvÞ is the number of geodesics from node i to j the line from the power network and is formulated as
through node v and rij is the total number of geodesics below:
between i and j. E EðlÞ
Similarly, critical lines in the network may be calculated Ve ðlÞ ¼ ð10Þ
E
from edge betweenness. Line betweenness centrality can be
obtained by Eq. (8): So, the relative decrement in performance of the power
network is being measured by damaging that line. Global
X
N X
N
rij ðlÞ
BðlÞ ¼ ; i 6¼ joV; lOE ð8Þ efficiency of the network being E without any damage and
i j
rij EðlÞ is the efficiency after removing the line l of that net-
work, then vulnerability of that line l can be expressed as:
where rij ðlÞ is the number of geodesics from node i to j
through line l. V ¼ max½Ve ðlÞ ð11Þ
A component in power transmission network has a Greater the value in vulnerability, the risk in that part of
specific value of betweenness depending upon the config- the network is more. Considering all of the above methods
uration of the network. When these values are little bit there are some drawbacks in the system. Firstly, geodesics
higher than other, its indicates that a higher number of only consider shortest path but power flow does not
shortest paths pass through the component focusing the maintain it. Secondly, there is no need to consider all the
higher criticality of the component. Thus, betweenness nodes, power only can flow from generation end to load
value being used as the identifier for ranking of the com- end and lastly, the transfer of power is different for dif-
ponents for a specified network as the critical component. ferent network which is not employed calculating global
efficiency.
2.7 Electrical betweenness
2.9 Vulnerability using netability
In conventional topological approach, the sum of the
probability for a vertex or an edge to belong to a randomly Netability, used as a tool for finding power transmission
selected geodesic path linking any other pair of vertices is capacity, power flow limit, power transmission distribution
defined as betweenness. It is a more useful method for factor(PTDF) and also the performance of the power
measuring the magnitude of critical load on the given node/ network.
edge in the network. Network connectivity is a most Cij is denoted by the injection of power from bus i to j,
important part of betweenness. It can be considered as a when the first line of all the paths connecting the generation
local metric in order to describe the degree centrality and node i and the load node j reaches its limit called as power
corresponding criticality of components. transmission capability. It is given by
Plmax
2.8 Vulnerability using global efficiency Cij ¼ min l ð12Þ
l2L fij
A general measure for vulnerability analysis of power where Plij is the transmission limit of line l and fijl is the
transmission network is the so-called efficiency index, and PTDF of line l of the path joining generation node i to load
it is also being described as the transmitting efficiency of node j. This is the difference of power for line l injecting at
the power network. The transmitting efficiency of the link
generation bus i and withdrawn at load bus j. fijl reflects the
connecting nodes i and j is denoted by E, and the initial
transmitting efficiency is 1 for every link in the network. difference in the entries fil and fjl of the PTDF matrix, it is
The efficiency of a path between nodes i and j is the formulated as
123
Microsystem Technologies
123
Microsystem Technologies
123
Microsystem Technologies
123
Microsystem Technologies
40
35
30
20
15
10
0
0 2 4 6 8 10 12 14 16 18 20
Line number
Table 3 Critical lines identified from graph The line vulnerabilities are calculated using Eqs. 12 and
17 and the comparable graphs are resulted as shown below
Line no. Connecting nodes Vulnerability in (%)
in Fig. 4:
3 2–3 17.53 It is observed that netability metric which includes the
6 3–4 19.35 electrical properties of extended topological approach
7 4–5 18.91 gives enhanced results of vulnerability analysis. The criti-
8 4–7 17.11 cal lines can be identified as listed below in Table 3.
11 6–11 18.83 From the above it is noticed that the line number 14 is
14 7–8 29.2 highly vulnerable line in the given network. One of critical
15 7–9 17.3 lines i.e. line no. 3 connects two generation buses. And
1000
800
Node betweenness
600
400
200
0
0 2 4 6 8 10 12 14
Node Number
123
Microsystem Technologies
Table 4 Comparison of voltage magnitudes of IEEE 14 bus system which justifies the applicability of the developed model i.e.
without and with STATCOM netability model. The analysis identifies critical lines in the
Bus No. Voltage Magnitude (p.u.) power grid which is very important in maintaining the
network security.
Without STATCOM With STATCOM
Fig. 6 Comparison of voltages Voltage magnitude comparison of IEEE14 bus test system with or without STATCOM
1.1
of IEEE 14 bus system with and
without STATCOM without STATCOM
with STATCOM
1.08
1.06
Voltage Magniutude (p.u.)
1.04
1.02
0.98
0 2 4 6 8 10 12 14
Bus Number
123
Microsystem Technologies
well as adopting mitigating measure for improvement of electric utility deregulation and restructuring and power tech-
resiliency of the network along with its steady state nologies. IEEE, pp 1554–1558
Holmgren ÅJ (2007) A Framework for Vulnerability Assessment of
characteristics. Electric Power Systems. In: Murray AT, Grubesic TH (eds)
Critical Infrastructure. Springer, Berlin
Kahnamouei AS, Bolandi TG, Haghifam MR (2017) The conceptual
framework of resilience and its measurement approaches in
References electrical power systems
Kinney R, Crucitti P, Albert R, Latora V (2005) Modeling cascading
failures in the North American power grid. Eur Phys J B
Albert R, Albert I, Nakarado GL (2004) Structural vulnerability of the Condens Matter Complex Syst 46(1):101–107
North American power grid. Phys Rev E 69(2):025103 Mei S, Zhang X, Cao M (2011) Vulnerability assessment of static
Beyza J, Yusta J, Correa G, Ruiz H (2018) Vulnerability assessment power grids. In: Power grid complexity. Springer, Berlin,
of a large electrical grid by new graph theory approach. IEEE pp 208–228
Lat Am Trans 16(2):527–535 Rout GK, Chanda CK, Chowdhury T (2015) A novel method based
Bompard E, Masera M, Napoli R, Xue F (2008) Assessment of on complex network theory approach for location of distributed
structural vulnerability for power grids by network performance generators in radial distribution systems. In: Michael Faraday
based on complex networks. In: International workshop on IET international summit
critical information infrastructures security. Springer, Berlin, Rout GK, Chowdhury T, Chanda CK (2016) Betweenness as a tool of
pp 144–154 vulnerability analysis of power system. J Inst Eng (India) Ser B
Bompard E, Wu D, Xue F (2011) Structural vulnerability of power 97(4):463–468
systems: a topological approach. Electr Power Syst Res Sun K (2005) Complex networks theory: A new method of research in
81(7):1334–1340 power grid. In: 2005 IEEE/PES transmission and distribution
Chen X, Sun K, Cao Y, Wang S (2007) Identification of vulnerable conference & exposition: Asia and Pacific. IEEE, pp 1–6
lines in power grid based on complex network theory. In: 2007 Wei X, Gao S, Huang T, Bompard E, Pi R, Wang T (2018) Complex
IEEE power engineering society general meeting. IEEE, pp 1–6 network-based cascading faults graph for the analysis of
Chen G, Zhao J, Dong ZY, Weller SR (2012) Complex network transmission network vulnerability. IEEE Trans Ind Inf
theory based power grid vulnerability assessment from past to 15(3):1265–1276
future Wu H, Peng M, Zhang H, Zhu L, Che H, Liu Z (2017) Node
Chowdhury T, Rout GK, Chanda CK (2015) Netability analysis of vulnerability assessment for distribution network based on
critical lines of power grid based on betweenness approach. In: complex network theory. Complex Syst Complex Sci
2015 international conference on energy, power and environ- 14:38–45. [Link]
ment: towards sustainable growth (ICEPE). IEEE, pp 1–6 Xue Y, Xiao S (2013) Generalized congestion of power systems:
Chowdhury T, Chanda CK, Chakrabarti A (2017) Resiliency insights from the massive blackouts in India. J Modern Power
improvement for a part of south Indian power transmission Syst Clean Energy 1(2):91–100
network. In: 2017 Australasian universities power engineering Yang Y, Tang W, Liu Y, Xin Y, Wu Q (2018) Quantitative resilience
conference (AUPEC). IEEE, pp 1–5 assessment for power transmission systems under typhoon
Dehghanian P, Aslan S, Dehghanian P (2017) Quantifying power weather. IEEE Access 6:40747–40756
system resiliency improvement using network reconfiguration. Zang T, Gao S, Huang T, Wei X, Wang T (2019) Complex network-
In: 2017 IEEE 60th international midwest symposium on circuits based transmission network vulnerability assessment using
and systems (MWSCAS). IEEE, pp 1364–1367 adjacent graphs. IEEE Syst J. [Link]
Fenu G, Pau PL (2015) Evaluating complex network indices for 2019.2934317
vulnerability analysis of a territorial power grid. J Ambient Intell
Humaniz Comput 6(3):297–306
Guohua Z, Ce W, Jianhua Z, Jingyan Y, Yin Z, Manyin D (2008) Publisher’s Note Springer Nature remains neutral with regard to
Vulnerability assessment of bulk power grid based on complex jurisdictional claims in published maps and institutional affiliations.
network theory. In: 2008 third international conference on
123