1302

IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012

Impact of WAMS Malfunction on Power System

Reliability Assessment
Farrokh Aminifar, Member, IEEE, Mahmud Fotuhi-Firuzabad, Senior Member, IEEE,
Mohammad Shahidehpour, Fellow, IEEE, and Amir Safdarian, Student Member, IEEE

AbstractMonitoring/control infrastructures are often assumed to be fully reliable in power system reliability studies.
However, recent investigations on blackouts have revealed the
crucial impacts of monitoring/control system malfunctions. This
paper addresses the impact of situational awareness and controllability on power system reliability assessment. A methodology is
proposed to simulate a situation in which a limitation of either
or both monitoring and control functions could spread the consequence of power system events throughout the grid. It is assumed
that the monitoring/control infrastructure is based on a wide-area
measurement system (WAMS). While the proposed methodology
is applicable to a variety of strategies for grid operations, certain
assumptions are made for the simulation purposes. The Monte
Carlo simulation is applied and a scenario reduction technique is
considered for overcoming computational burdens. The performance of the proposed approach is simulated and analyzed on
9-bus and the IEEE 57-bus systems.

Set of observable lines.

Set of scenarios comprising monitoring/control
and power system contingencies.
Set of scenarios for power system contingencies.
Set of scenarios for WAMS contingencies.
B. Parameters
Reactance of line .
Element of network incidence matrix, 1 when bus
is the sending bus of line ,
when bus is
the receiving bus of line , 0 otherwise.

Index TermsComposite system reliability assessment, observability analysis, phasor measurement unit (PMU), wide-area measurement system (WAMS).

Generating power of unit in the base scenario.

,

Maximum and minimum capacity limits of unit .

Power demand at bus .

NOMENCLATURE

Capacity of line .

A. Indices and Sets

Binary number representing state of line in

scenario .

Unit index.

Binary number representing state of unit in

scenario .

Line index.
Bus index.

C. Functions and Variables

Scenario index.
Scenario index for power system.

Expected demand not served at bus .

Scenario index for WAMS.

Expected demand not served of the entire system.

Set of buses.

Objective function in OLS of scenario .

Set of buses with controllable units and loads.

Load shed at bus

Set of units connected to bus .

Power flow of line in scenario .

Set of lines.

Voltage phase angle of bus

in scenario .
in scenario .

Power redispatch of unit in scenario .

Manuscript received August 21, 2011; revised November 27, 2011; accepted
January 02, 2012. Date of publication March 16, 2012; date of current version
August 20, 2012. This work was supported by the Iran National Science Foundation and is acknowledged by the first two authors. Paper no. TSG-00382-2011.
F. Aminifar, M. Fotuhi-Firuzabad, and A. Safdarian are with the Center
of Excellence in Power System Control and Management, Electrical Engineering Department, Sharif University of Technology, Tehran, Iran (e-mail:
[email protected]; [email protected]; [email protected]).
M. Shahidehpour is with the Electrical and Computer Engineering Department, Illinois Institute of Technology, Chicago, IL 60616 USA (e-mail: ms@iit.
edu).
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TSG.2012.2183397

I. INTRODUCTION

NVESTIGATIONS conducted following the occurrences

of large events in power grids have revealed the indispensible role of critical information at control centers and the
impact of situational awareness on managing power system operations. In many occasions, the impact of outages could have
been prevented or significantly reduced if system operators
had a more complete knowledge on the corresponding state of
stressed power systems [1].

1949-3053/$31.00 2012 IEEE

AMINIFAR et al.: IMPACT OF WAMS MALFUNCTION ON POWER SYSTEM RELIABILITY ASSESSMENT

Synchrophasor measurement technology (SMT) was recognized as a promising alternative for enhancing the situational
awareness. It provides an unprecedented insight on the realtime state of power systems with a wide-area coverage [1][3].
Phasor measurement units (PMUs) are the building block of
SMT [3]. SMT, which is also referred to as wide-area measurement system (WAMS), is presently accounted for monitoring
and control of static and dynamic power systems.
In the literature, few studies focused on the role of monitoring/control infrastructures in the probabilistic power system
reliability assessment. Reference [4] reported a probabilistic
case study on the overall reliability of control center facilities in
Ontario Hydro. The paper considered failures in control center
equipment; however, it did not study malfunctions associated
with the supervisory control and data acquisition (SCADA)
system and their impacts on power system performance. A
joint SCADA and power system model was considered in [5]
in which the impact of SCADA failures on power system load
curtailments was analyzed. The idea was valuable; however,
the approach was not elaborated on its details. The paper
presented limitations on controllability but did not incorporate
observability deficits in power system analyses.
The WAMS applications were examined in [6] using simple
examples. The probability of successful operations was calculated based on the availability of WAMS elements. However,
the probabilistic power system analyses were preliminary. Reference [7] conducted the calculation of multiple reliability indices for a regional network of WAMS. The outages of PMUs
and communication links were considered, while those of other
key components such as measuring transformers or transmission lines were excluded. The authors extended their works in
[8] by presenting a quantitative reliability evaluation method for
the communication network of WAMS and the overall WAMS
from a hardware reliability perspective.
In [9], the observability of WAMS-based power networks was
analyzed from a probabilistic viewpoint. A probabilistic observability index was introduced for buses associated with a PMU
placement configuration. The average of bus indices represented
a system index with an insight on the overall network observability. The proposed model in [9] was extended in [10]; however, both studies were devised with the intention of PMU placement in power systems. References [6][10] studied the WAMS
performance as a single system; however, they did not investigate the impact of WAMS failures on power system performance analyses.
In this paper, the proposed methodology considers WAMS as
a viable monitoring/control infrastructure in electric power systems. This assumption is due to straightforward observability
rules embedded in the WAMS model. The WAMS application
to the enhancement of power system operation is discussed
and requirements for the joint reliability assessment of WAMS
and power system infrastructures are identified. As reliability
analyses assess the performance of static power systems, the
WAMS capability in monitoring power system dynamics is
not modeled here. This feature makes the proposed framework
applicable in the SCADA-based power systems where the
supply-demand balance and transmission system loading are
just monitored. A set of assumptions in the proposed study
represent the complexity of the problem which is, in turn, due
to miscellaneous infrastructures deployed across the world.

1303

Fig. 1. Actual power system and mimic diagram.

However, such assumptions would be tailored for other WAMS

cases. For instance, WAMS is assumed in the paper to be the
key entity in charge of monitoring and control functions for
power system reliability studies. However, other options for
the monitoring and control in power systems are discussed and
simulated as well.
Next, the proposed methodology which incorporates WAMS
malfunctions in the power system reliability evaluation is
devised. The outline of the algorithm is followed by the step
by step procedure for the implementation of the proposed
approach and the detailed formulations of the algorithm. Numerical analyses are then conducted on a 9-bus system and the
IEEE 57-bus test system and the results are analyzed further
based on the assumptions presented throughout the paper.
II. PROPOSED CONCEPTS
Contingency analysis in the reliability evaluation of power
systems assumes a complete observability over the system faults
and operation status. It also supposes specific remedial actions
and applies load shedding as the last resort at any load points requested. These procedures imply a reliable monitoring and control system, which could be somewhat an unrealistic assumption.
In Fig. 1, a two-layer power system model is demonstrated.
The lower layer implies the actual power system. The upper
layer shows the mimic diagram in a control center. The WAMS
is the interface between these two layers. When power system
contingencies occur within an unobservable region, the operator would not be directly informed on the event and might
only be able to monitor consequential changes in observable
power system portions. Remedial actions applied to observable
regions of power systems could shift the power system state
where the unobservable region might experience further sever
conditions. Such ill-conditioned actions could trigger cascading
outages which might remain unknown to system operators. A
realistic reliability assessment methodology should consider the
entire modeling of the actual power system layer. Such an effort
would be presented in the next section.

1304

IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012

Fig. 2. Proposed algorithm for joint reliability evaluation of power system and WAMS.

We assume in the proposed approach that the protection

system is perfect and focus our attention on WAMS failures.
Such malfunctions could be accommodated in the extended
version of the proposed reliability analyses.
III. PROPOSED METHODOLOGY
This section presents the proposed methodology for assessing
the impact of WAMS malfunctions on the power system reliability assessment. The flowchart of the algorithm is depicted in
Fig. 2. The procedure devised in this figure is as follows.
Step 1: The first step is to determine a base case OPF [11].
Step 2: The outer loop iterates on WAMS scenarios. The initial set of scenarios is truncated to a tractable level
[12]. The composite power system reliability evaluation is conducted for each scenario.
Step 3: The inner loop considers power system scenarios in
which the scenario reduction is applied again.
Step 4: An observability analysis accounts for forced outages in monitoring and power system levels [13],
[14]. In this analysis, the PMU measurement of a

given line current offers an indirect voltage measurement associated with the line other end. The
observability rules are summarized as: i) a PMU
located at a given bus would make itself and adjacent buses observable; ii) a transmission line
with observable terminals is observable since its
current and power flow could be calculated. The
addition of PMUs and the provision of redundant communication channels would enhance the
power system observability. Note as well that for
SCADA-equipped power systems, this step merely
represents the state estimation (SE). Assuming
WAMS for both monitoring and control, this step
will also identify uncontrollable parameters. In the
event of any WAMS failures, the communication
path between the control center and system buses
could become partially unavailable. Accordingly,
certain system variables would become uncontrollable. However, if the WAMS is only responsible for
the system monitoring, we assume that the system
is fully controllable for the sake of this study.

AMINIFAR et al.: IMPACT OF WAMS MALFUNCTION ON POWER SYSTEM RELIABILITY ASSESSMENT

Step 5: With the event of violations in observable regions,

the system operator will prescribe remedial actions
for mitigating them. These actions are simulated by
an optimal load shedding (OLS) optimization algorithm which will be outlined in this section.
Step 6: This step simulates the role of protection system in
tripping out overloaded lines. Any violations in the
unobservable region is checked at this step. In the
case, the violated lines will be tripped by the protection system.
Step 7: The isolated buses are identified, their generating
units are shut down, and the respective loads are curtailed. Since the system switches to a new operating
point, the algorithm returns to Step 5 to check other
violations.
Step 8: Reliability indices are updated based on load curtailments in scenario and the probability of occurrence
of this scenario.

1305

Fig. 3. Single-line diagram of the nine-bus system.

TABLE I
GENERATING UNIT DATA

A. Formulation
The OLS problem, in Step 5, is formulated in (1)(8).
(1)
Subject to
(2)

(3)
(4)
(5)
(6)
(7)
(8)

TABLE II
SYSTEM DEMAND DISTRIBUTION

TABLE III
TRANSMISSION SYSTEM DATA

Here, the joint scenario set for power system and WAMS comand
as
prises scenario sets
(9)
and
is an ordered pair of scenarios for
where
WAMS and power system.
In each scenario, the objective is to minimize the load shedding in (1). The line flow calculation (2) has a binary parameter
where 0 stands for the state of lines on outage making the
power flow of such lines equal to 0. In (3), the bus-level power
balance is considered and the units on outage are excluded by in. The load shedding varicorporating the binary parameter
is considered in the power balance equation to make
able
the problem feasible. The set of constraints (4) compels observable line flows to be within limits. The range of possible redispatch associated with controllable and uncontrollable generating units are determined by (5) and (7), respectively. Sets
of constraints (6) and (8) declare the range of deployable load
shedding associated with controllable and uncontrollable demands.
If WAMS is merely in charge of monitoring and the system
control is managed through other media such as telephone lines,

the proposed formulation will be modified slightly. Equation

(1)(3) will remain since they model the power flow problem
in the given scenario. Also, as the system WAMS malfunctions
still restrict the operator insight, the flow limits are met just in
observable regions. So the set of constraints (4) will remain. In
, and
contrast, (5)(6) will be formed for all buses, i.e.,
(7)(8) will be removed.
IV. CASE STUDY AND DISCUSSIONS: A NINE-BUS SYSTEM
In this section, a nine-bus system is studied [9], [10], [13].
The nine-bus system with the base case power flow solution is
depicted in Fig. 3. The data are given in Tables IIII.
As shown in Fig. 3, the WAMS network allocated for the
system monitoring/control consists of three PMUs installed at
buses 4, 6, and 8. These devices make the network fully observable, while buses 5, 7, 9 have redundant PMU measurements.

1306

IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012

TABLE IV
WAMS COMPONENTS RELIABILITY DATA

TABLE V
CASE 1: RELIABILITY INDICES

The reliability data of WAMS components are given in Table IV.

For each phasor calculation, three-phase measurements are necessary [3]. So a PMU requires three CTs in the operating state
for calculating a current phasor and the associated availability
. The requirement for the voltage phasor
would be
is the same. The following cases are studied:
Case 1: Power system contingencies with reliable WAMS.
Case 2: Few joint contingencies of power system and
WAMS.
Case 3: All contingencies of power system and WAMS.
A. Case 1
This case is the comparison benchmark to quantify the impact of WAMS malfunctions on the power system reliability indices. Table V gives the reliability indices for load points and
the power system. Here, the largest load point index is at bus 9.
to the system index is not well
The contribution of
to the system demand (see
correlated with the ratio of
is much larger than
Table II). In addition,
while
. These observations are justified by load
shedding procedure associated with contingencies.
B. Case 2
This case simulates a set of joint contingencies in power
system and WAMS.
1) Contingency 1: PMU at bus 4 and line 14 (or generating unit at bus 1) are on outage. This contingency simulates
the frequency instability and potential power blackout condition. In this contingency, the system generation capacity is 160
MW and, at least, 70 MW load shedding is required. Due to the
WAMS malfunction, the system curtailable load is just 60 MW
at bus 7. Accordingly, the demand exceeds the generation and
the system frequency drops which will lead to a blackout. In
practice, the automatic under-frequency protection could prevent such widespread impact of the disturbance. The protection
model will be incorporated in the follow-up version of the proposed reliability analyses.
2) Contingency 2: PMU at bus 6 and line 14 are on outage.
This contingency simulates the impact of WAMS availability on
load shedding. From the power system viewpoint, this contingency is similar to Contingency 1. However, WAMS failure will
influence other components. In Contingency 2, the load point at
bus 9 and its linking transmission lines are controllable and observable, respectively. So the 70 MW loss of generation is balanced by a load shedding at bus 9.

Fig. 4. Case 2, Contingency 3: load flow following line 67 outage. (a) Precascading outage. (b) Postcascading outage.

3) Contingency 3: PMU at bus 8 and line 67 are on outage.

This contingency simulates cascading outages and bus isolation.
The observability analysis shows buses 2, 7, 8 and lines 28,
78, 89 as unobservable. Fig. 4(a) shows the power flow solution once line 67 is on outage. Line 78 with a capacity of
55 MW is the only path to supply the 60 MW load at bus 7. So
line 78 is overloaded while it is not observable in the control
center. Consequently, the protection system trips line 78. The
cascading event leaves bus 7 isolated in which the load will be
curtailed. The curtailment of 60 MW load requires a reduction
in generation dispatch of unit 1 by 120 MW. Fig. 4(b) shows
the line flow in post-cascading outage. Based on the common
reliability evaluation, the solution of this contingency would be
to relieve the line 78 overload regardless of its
un-observation in the control center.
4) Contingency 4: The PMU at bus 8 and generating unit at
bus 2 are on outage. This contingency simulates the situation
that the faulted system remains healthy. Following the outage
of generating unit 2, the system slack unit at bus 3 compensates
the generation loss and the new power flow solution does not
violate transmission limits. Table VI summarizes the load shedding results in Case 2.
C. Case 3
In this case, all contingencies associated with both power
system and WAMS network are considered including four
states analyzed in the previous case. Table VII gives the load
points and system reliability indices. This case is comparable
with Case 1 in which the WAMS network was assumed to be
fully reliable. In Table VII, WAMS contingencies deteriorate

AMINIFAR et al.: IMPACT OF WAMS MALFUNCTION ON POWER SYSTEM RELIABILITY ASSESSMENT

TABLE VI
CASE 2: LOAD CURTAILMENTS OF CONTINGENCIES 14

1307

TABLE IX
RELIABILITY INDICES FOR VARIOUS PLACEMENT SCHEMES OF 4 PMUS

TABLE VII
COMPARISON OF RELIABILITY INDICES IN CASES 1 AND 3
TABLE X
AVAILABILITY OF GENERATING UNITS

TABLE VIII
RELIABILITY INDICES WITH A RELIABLE WAMS

the system reliability. The load point reliability index varies in

large interval ranging from 2.82 to 20.78%.
D. Impact of Key Parameters
1) Availability of WAMS Component Data: Two situations
in Case 3 are examined here in which the unavailability of
WAMS (see Table IV) is multiplied by a factor of 0.1 or 10.
The former simulates an extremely reliable WAMS network.
Table VIII shows the results which are compared with those of
Case 3. The less reliable WAMS could deteriorate the power
system reliability. Table VIII shows that load point indices vary
in large intervals which is due to the OLS feasibility and its
multiple solutions.
2) Deployment of Additional PMUs: In general, improving
the component availability could be very costly or even infeasible. A more practical approach would be to incorporate redundant components [18]. In WAMS system, additional PMUs can
be installed in a given bus to enhance the system observability
in the event of either power system or WAMS outages. Accordingly, we consider 1 more PMU for the small nine-bus system.
The additional PMU could be located at buses 1, 2, 3, 5, 7, and 9.
So there are 6 solutions for the placement of 4 PMUs. Table IX
presents the breakdown of reliability index for the base and new
placement schemes. The improvement varies for different PMU
schemes.
In Table IX, adding a PMU at bus 9 is the most effective
solution since the system bulk load is connected to this bus.
The second solution refers to an additional PMU at bus 1 which
consists of the system largest generating unit. The next effective
solution is the placement of PMUs at buses 4, 6, 7, and 8. In this
because
solution, the major improvement occurs in
bus 7 is equipped with the additional PMU and the chance of
its isolation originated by line overloads significantly declines.

Hence, buses with large units and/or loads are better candidates for the PMU installation. However, if the value of lost
load (VOLL) or synonymously interrupted energy assessment
rate (IEAR) is considered in the power system reliability assessment, the buses serving economical or critical loads are suited
for enhancing the observability.
V. CASE STUDY AND DISCUSSION: IEEE 57-BUS SYSTEM
In this section, the proposed reliability evaluation methodology is tested on the IEEE 57-bus system [9], [10], [13]. The
system has 80 transmission lines, 7 generating units with 1975
MW generation capacity, and 42 bulk loads with 1250 MW demand. The data for the base case power flow solution are available at [19]. Generating units data are taken from [19] and their
availabilities are given in Table X. Line capacities are presented
in Table XI and their availability data are taken from [9].
Fig. 5 depicts the IEEE 57-bus system and 17 PMUs for the
WAMS network. The reliability data considered for the WAMS
components are the same as those given in Table IV. The numerical simulations consist of the following cases:
Case 1: Power system contingencies with a reliable
WAMS.
Case 2: All contingencies of power system and WAMS.
Case 3: Same as Case 2 but the controllability is fully reliable.
Table XII shows the system reliability index associated with
the three case studies. The CPU time for the calculation of Case
2 was about 30 minutes on a Core i7 1.60-GHz processor and 4
GB of RAM.
in Case 1 represents the system reliability based on
the conventional evaluation analysis.
in Case 2 shows
a 41% increase as compared with that of Case 1. So incorporating the malfunction of WAMS components, as expected, provides a more comprehensive system reliability index. In Case 3,
it is assumed that WAMS is just in charge of system monitoring
and control actions are rendered through another media which
is not affected by WAMS failures. As expected, in this case, the
is greater than that of Case
system reliability index

1308

IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012

TABLE XI
CAPACITY OF TRANSMISSION LINES

TABLE XII
SYSTEM

TABLE XIII
MAJOR LOAD POINT RELIABILITY INDICES IN CASE 2

1. This is due to incorporating the observability deficits associated with WAMS malfunctions. On the contrary, comparison
of Cases 2 and 3 reveals that accounting controllability deficits
would further deteriorate the system reliability.
The following discussion is limited to Case 2 which is the
major case study. Due to the large dimensions, it is not possible
to present the results at all load points. So the major load point
buses, with demands greater than 25 MW, are adopted and their
respective reliability indices are detailed in Table XIII. For the
reliability index is employed since
sake of comparison,
load shedding amount directly influences this index. As shown
of all buses, except bus 12, are in a same
in Table XIII,
order of magnitude. The reason can be declared through analysis of the measurement robustness associated with buses 8, 9,
and 12 as those serve largest loads. As depicted in Fig. 5, bus 8
is redundantly observable through two paths, namely PMUs at
buses 6 and 9. Also, bus 9 is equipped with its own PMU and
its observability is independent on the network topology. However, bus 12 is neither furnished with a PMU nor does it have
redundant observability sources. So the probability of observability [10] of this bus is significantly lower than those of buses
8 and 9.
As a cascading event, the concurrent outages of PMU at bus
9 and line 117 is considered. Here, buses 9, 10, 11, 12, and 55
and all linking lines are unobservable. So the overloaded line
1216 will trip. Thereafter, the overloaded line 1213 will trip.
The cascading event is ended when lines 89, 910, 911, 912,
and 1012 trip.
VI. CONCLUSIONS

Fig. 5. IEEE 57-bus network with the given PMUs.

This paper has enabled a methodology for incorporating

WAMS malfunctions in power system reliability assessment. It
was assumed that WAMS is in charge of the monitoring and/or
control activities.
Based on numerical studies, the proposed method is able to
model the situation in which the WAMS network suffers failures and an event in power systems would consequently spread
out. Such events, although might be unlikely, could significantly

AMINIFAR et al.: IMPACT OF WAMS MALFUNCTION ON POWER SYSTEM RELIABILITY ASSESSMENT

contribute to load shedding and simulate major changes in reliability assessments. These events are usually overlooked in
conventional reliability evaluation studies. Numerical analyses
show that buses with large generating units and loads should be
candidates for PMU installations. It was deduced that, due to
very long execution times, a few shortcuts should be devised
to speed up the process. Scenario reduction was the approach
adopted here while other alternatives such as parallel processing
or running over pre-selected contingencies would also be possible. These studies underlined the indisputable role of information infrastructure on the power system adequacy evaluation.
Other infrastructures such as the natural gas network and water
resources could influence the problem as well and the joint analysis of all these systems should be considered in future studies.

REFERENCES
[1] Real-time application of synchrophasors for improving reliability,
North American Electric Reliability Corporation, 2010.
[2] V. Terzija, G. Valverde, C. Deyu, P. Regulski, V. Madani, J. Fitch, S.
Skok, M. M. Begovic, and A. Phadke, Wide-area monitoring, protection, and control of future electric power networks, Proc. IEEE, vol.
99, no. 1, pp. 8093, Jan. 2011.
[3] G. Phadke and J. S. Thorp, Synchronized Phasor Measurements and
Their Applications. New York: Springer, 2008.
[4] L. Wang, P. P. Gelberger, and N. Ramani, Reliability assessment of
the operational functions of a power system control center, Proc. IET
Prob. Methods Appl. Electr. Power Syst., pp. 229234, Jul. 1991.
[5] A. G. Bruce, Reliability analysis of electric utility SCADA systems,
IEEE Trans. Power Syst., vol. 13, no. 3, pp. 844849, Aug. 1998.
[6] M. Zima, M. Larsson, P. Korba, C. Rehtanz, and G. Andersson,
Design aspects for wide-area monitoring and control systems, Proc.
IEEE, vol. 93, no. 5, pp. 970996, May 2005.
[7] Y. Wang, W. Li, J. Lu, and H. Liu, Evaluating multiple reliability
indices of regional networks in wide area measurement system, Electr.
Power Syst. Res., vol. 73, no. 10, pp. 13531359, Oct. 2009.
[8] Y. Wang, W. Li, and J. Lu, Reliability analysis of wide-area measurement system, IEEE Trans. Power Del., vol. 25, no. 3, pp. 14831491,
Jul. 2010.
[9] F. Aminifar, M. Fotuhi-Firuzabad, M. Shahidehpour, and A. Khodaei,
Observability enhancement by optimal PMU placement considering
random power system outages, Energy Syst., vol. 2, no. 1, pp. 4565,
Mar. 2011.
[10] F. Aminifar, M. Fotuhi-Firuzabad, M. Shahidehpour, and A. Khodaei,
Probabilistic multistage PMU placement in electric power systems,
IEEE Trans. Power Del., vol. 26, no. 2, pp. 841849, Apr. 2011.
[11] A. J. Wood and B. F. Wollenberg, Power Generation, Operation, and
Control, 2nd ed. New York: Wiley, 1996.
[12] J. Dupacov, N. Grwe-Kuska, and W. Rmisch, Scenario reduction
in stochastic programming: An approach using probability metrics,
Math. Program., vol. A 95, pp. 493511, 2003.
[13] F. Aminifar, A. Khodaei, M. Fotuhi-Firuzabad, and M. Shahidehpour,
Contingency-Constrained PMU placement in power networks, IEEE
Trans. Power Syst., vol. 25, no. 1, pp. 516523, Feb. 2010.
[14] F. Aminifar, C. Lucas, A. Khodaei, and M. Fotuhi-Firuzabad, Optimal
placement of phasor measurement units using immunity genetic algorithm, IEEE Trans. Power Del., vol. 24, no. 3, pp. 10141020, Jul.
2009.
[15] F. Aminifar, S. Bagheri-Shouraki, M. Fotuhi-Firuzabad, and M.
Shahidehpour, Reliability modeling of PMUs using fuzzy sets,
IEEE Trans. Power Del., vol. 25, no. 4, pp. 23842391, Nov. 2010.
[16] M. J. Rice and G. T. Heydt, The measurement outage table and state
estimation, IEEE Trans. Power Syst., vol. 23, no. 2, pp. 353360, May
2008.

1309

[17] Performance requirements task team (PRTT), performance requirements, Part II, Targeted applications: State estimation, Eastern Interconnection Phasor Project, 2005.
[18] H. Pham, Handbook of Reliability Engineering. New York: Springer,
2003.
[19] MATPOWER [Online]. Available: http://www.pserc.cornell.edu/matpower
Farrokh Aminifar (S07M11) received the M.Sc.
and Ph.D. degrees in electrical engineering from
Sharif University of Technology, Tehran, Iran, in
2007 and 2010, respectively.
He worked with the Electrical and Computer
Engineering Department at Illinois Institute of
Technology, Chicago, as a Research Associate.
Currently, he is a Postdoctoral Fellow with the
Electrical Engineering Department Sharif University
of Technology, Iran. His research interests include
wide-area measurement system, reliability modeling
and assessment, and smart grid technologies.
Dr. Aminifar is the Guest Editor of a Special Issue on Microgrids of the IEEE
TRANSACTIONS ON SMART GRID. He received the IEEE Best Ph.D. Dissertation
Award from Iran Section for his research on the probabilistic schemes for the
placment of phasor measurement units.

Mahmud Fotuhi-Firuzabad (SM99) received the

B.Sc. degree in electrical engineering from Sharif
University of Technology, Tehran, Iran, in 1986, the
M.Sc. degree in electrical engineering from Tehran
University, Tehran, in 1989, respectively, and the
M.Sc. and Ph.D. degrees in electrical engineering
from the University of Saskatchewan, Saskatoon,
Canada, in 1993 and 1997, respectively.
Currently, he is a Professor and Head of the Department of Electrical Engineering, Sharif University
of Technology. He is also an Honorary Professor in
the Universiti Teknologi Mara (UiTM), Shah Alam, Malaysia.
Dr. Fotuhi-Firuzabad is a Member of the Center of Excellence in Power
System Control Management and Control. He serves as an Editor of the IEEE
TRANSACTIONS ON SMART GRID.

Mohammad Shahidehpour (F01) is Carl Bodine

Chair Professor and Director of Robert W. Galvin
Center for Electricity Innovation at the Illinois
Institute of Technology, Chicago. He is an Honorary
Professor at Sharif University of Technology and
the North China Electric Power University. He is the
recipient of the 2009 Honorary Doctorate from the
Polytechnic University of Bucharest.

Amir Safdarian (S11) obtained the B.S. (Hon.)

degree in electrical engineering from University of
Tehran, Tehran, Iran, in 2008 and the M.S. (Hon.)
degree in electrical engineering from Sharif University of Technology, Tehran, in 2010. He continued
his studies there, where he is currently working
toward the Ph.D. degree.
His research interests include reliability, vulnerability, and operation of power systems.