An Analytical Method for the Reliability

Evaluation of Wind Energy Systems

A. Ehsani, M. Fotuhi, Senior Member, IEEE, A. Abbaspour, and A. M. Ranjbar

71max Maximum efficiency.

Abstract-Wind generation is one of the most successful
sources of renewable energy for the production of electrical VCut-in Cut-in velocity.
energy. The technical characteristics of wind generation make Cut-out velocity.
existing conventional generation models not directly applicable.
This paper presents an analytical method for the reliability
VRating Rated velocity.
assessment of the flat-rated wind turbine generation systems.
An example of flat rating is that of the MOD-2, a second- V Mean wind velocity.
generation class of wind turbines. The power-velocity Mean energy velocity.
characteristic of the flat-rated wind turbines is employed in this VE
paper to model the operating behavior of the installed wind n Number of wind velocity observations.
turbine generators. For wind-power potential estimation, the V Wind velocity.
Weibull distribution model is used. The performance of the Reference velocity.
developed method is demonstrated with computational results. Vr
Hr Reference height.
Index Terms-Reliability evaluation, Weibull probability
distribution model, wind turbine generation. P(X) Probability of X.
F(X) Gamma function of X.
I. NOMENCLATURE DG Dispersed generation.
Failure rate in failure/year. WTG Wind Turbine Generation
,u Repair rate in repair/year.
FOR Forced outage rate. II. INTRODUCTION
MTTF Mean time to failure. pUBLIC environmental concerns with electrical energy
MTTR Mean time to repair. derived from fossil resources have created an increased
MTBF Mean time between failures. interest in the development and use of alternative sources.
L Load demand. The technologies for the generation of electrical energy from
L max Maximum demand. renewable energy sources have evolved in recent years from
small experimental plants to a real option for the utilities to
L min Minimum demand.
produce significant amounts of electric power. Wind
LOLP Loss of load probability. generation is one of the most successful sources of renewable
LOLE Loss of load expectation. energy for the production of electrical energy. Many utilities
LOEE Loss of energy expectation. throughout the world are considering using wind energy as a
ptotal Total power of wind stream. substitute for conventional generation due to its huge
potential and minimal pollution [ 1 ] - [2].
mh Mass flow rate of wind stream.
One of the most promising applications of wind energy is
Vi Velocity of wind stream. in remote, windy places which have weak, autonomous power
p Air mass density. systems. There are several Iranian regions meeting these
A Area of wind stream. conditions. As the number of wind power installations
p Maximum power. connected into existing electric power networks grows rapidly
max
worldwide, there is a need to study these energy sources and
17 Efficiency.
assess their effects on the system. To evaluate the connection
of wind energy sources to the electric supply system and
A. Ehsani is with the Department of Technology, Engineering and Sciences, particularly to distribution networks, it is necessary to
Islamic Azad University, Shahr-e-Rey Branch, Yadegar-e-Emam Academic evaluate how the reliability and production of the system may
Complex, Tehran, Iran (e-mail: a ehsanina
M. Fotuhi, A. Abbaspour, and A. M. Ranjbar are with the Department of be affected. However, the technical characteristics of wind
Electrical Engineering, Sharif University of Technology, Tehran, Iran (e-mail: generation make existing conventional generation models not
fotuhi(d,sharifedu, abaspour(dshlarifedu, ranjbar(osharifedtu).

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.
 2

directly applicable [2]-[3]. A. Generation Model

From a reliability modeling and evaluation viewpoint, The basic generating unit parameter used in static capacity
dispersed generation can generally be grouped into two main evaluation is the probability of finding the unit on forced
types: those which have an output dependent on a variable outage at some distant time in the future. This probability is
energy source (e.g. wind, solar) and cannot be prescheduled defined as the unit unavailability, and historically in power
and those that are not so dependent (e.g. hydro, gas, and system applications it is known as the unit forced outage rate
diesel) and could be prescheduled. The latter type can be (FOR),
modeled using conventional generation approaches and their
contribution to the system supply is only dependent on need Unavailability (FOR)= + (1)
A+
and the availability of the units themselves. The former,
however, are much more difficult to deal with because their Availability= p (2)
contribution to the system supply is also dependent on the A+p
primary source of energy being available as well as need and The concepts of availability and unavailability as
unit availability [4]. Several excellent textbooks and illustrated in equations (1) and (2) are associated with the
references that provide a detailed description of the modeling simple two-state model shown in Fig. 3. This model is
and evaluation efforts are available [1]-[15]. This paper will directly applicable to a generating unit which is either
not attempt to focus on these references, but will use their operating or forced out of service.
approach and philosophy to develop a new method.
The paper has the following structure. A brief introduction A/ 1
to reliability evaluation of generation systems is provided in Un up Un~&w
section III; section IV describes concepts and characteristics
of wind turbine generation; numerical examples are presented Fig. 3. Two-state model for a generating unit.
in section V; and finally, conclusions are presented in section
VI. The two-state model of Fig. 3 is the situation in which X
and ,u are constant values. In this case
111. RELIABILITY EVALUATION OF GENERATION SYSTEMS
The determination of the required amount of system Mean time to failure= MTTF = (3)
generating capacity to ensure an adequate supply is an
important aspect of electric power system planning and Mean time to repair= MTTR =
(4)
operation. The basic approach to evaluating the adequacy of a Mean time between failures= MTBF = MTTF + MTTR (5)
particular generation configuration consists of three parts as
shown in Fig. 1. This can be illustrated using the representation shown in Fig.
4.

Time

Fig. 4. Average history of a generating unit.

Fig. 1. Conceptual tasks in generating capacity reliability evaluation.
The generation and load models shown in Fig. 1 are
combined to form the appropriate risk model. The calculated
indices do not normally include network constraints. The
system representation in a conventional study is shown in Fig.
The generation model is sometimes known as a capacity
outage probability table. As the name suggests, it is a simple
array of capacity levels and the associated probabilities of
existence. If all the units in the system are identical, the
capacity outage probability table can be easily obtained using

I
2. the binomial distribution. Otherwise, the units can be
Total Total
combined using basic probability concepts and this approach
ys tem * sYStem can be extended to a simple but powerful recursive technique
geneiaion load in which units are added sequentially to produce the final
model. (Table I)
Fig. 2. Conventional system model.

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.
 3

TABLE I IV. WIND TURBINE GENERATION

CAPACITY OUTAGE PROBABILITY TABLE FOR A GENERATION SYSTEM WITH
Two NON-IDENTICAL UNITS A. Wind Power
The total power of a wind stream is equal to the rate of the
Stae i n it I n 2
lrt Capacity ill ( apacit iotit Sta'te proabillity,
1 LIp Ip CI r
f2 (,, I= UR-(J]R 1) *(I =......R)
, ) incoming kinetic energy of that stream, or
Up Dow L
(cc.1 (1 .OR OI (FJR) Ptoa 21 2 (6)
(.I (),
t RHH/
Ptotal 2 V
4 D)ownr Downl (., C(2 f- K,m, RM,
The mass flow rate is given by the continuity equation
B. Load Model and Risk Indices iii= pAV. (7)I
The generation system model illustrated in section A can Thus
be convolved with an appropriate load model to produce
system risk indices. One way to partially assess how a Ptotal = 2 pA V. (8)
particular generation system satisfies a variable demand is to
determine several load states based on the load duration Thus the total power of a wind stream is directly
curve; a simple load model that is used quite extensively. This proportional to its density, area, and the cube of its velocity. It
curve uses individual hourly load values, and the area under it can be proved that the total power cannot all be converted to

represents the energy required in the given period. Fig. 5 mechanical power. There is a maximum power Pmax which
shows a load duration curve with a maximum demand can be obtained from the wind
of Lm, a minimum of Lmi and that indicates the demand
spends t hours per year above L. Pmax 27 pA V3 (9)
4
The ideal, or maximum, theoretical efficiency 17ma (also
Ln,ax I
called the power coefficient) of a wind turbine is the ratio of
the maximum power obtained from the wind to the total
power of the wind, or
p
7max
max = 0.5926 (1o)
i I 00.~~~~
ltotal
In other words, a wind turbine is capable of converting no
Time duratson (ou) more than 60 percent of the total power of a wind to useful
Fig. 5. Load duration curve shows the amount of time that the demand is above power.
any value during the year.
Since a wind-turbine wheel cannot be completely closed,
This curve can be used in conjunction with the capacity and because of spillage and other effects, practical turbines
outage probability table to obtain the expected number of achieve some 50 to 70 percent of the ideal efficiency. The real

hours in the specified period (i.e., a week, a month or a year) efficiency q is the ratio of actual to total power
in which the load will exceed the available capacity. Fig. 5
shows a typical system load-capacity relationship where the P=1iP
max
=-I7pA
2
V.31 (II)
load model is shown as a continuous curve for a period of
8760 hours (one year). Where q varies between 30 and 40 percent for real turbines.
A number of probabilistic criteria and indices are available
for generation system risk studies. These include, but are not B. Flat-rated Wind Turbine Generation
limited to those which are given in Table II [16]. Severe fluctuations in power are undesirable. They pose
power-oscillation problems on the grid and severe strains on
TABLE II the wind turbine hardware. It is, therefore, more cost-effective
PROBABILISTIC CRITERIA AND INDICES FOR GENERATION SYSTEM RISK
STUDIES
to design a wind turbine to produce rated power at less than
the maximum prevailing wind velocity, and to maintain a
dciX Dtefinition constant output at all wind speeds above rating. This is called
flat rating. (Fig. 6)
LOLP probability that mthe load ,wi . h

.a-ttv
availble ixemaLig
_

LOLEA: IThe. av ae numLbeTr of hourf wbi, lh the load is Because of a severe loss in efficiency and power at low
illpectd tL exceed the a..Ji.ibl g..1....JinH1 capa1c,itv wind velocities, a wind turbine is designed to come into
LOEE TIe expetl1ed enerav t at jw 1 11 ot sLIuppIied dute to operation at a minimum wind speed, called the cut-in
tIhose oCDca s1o1n Ih en th.e load exCeeds tih e ailab1le
velocity. To protect the turbine wheel against damage at very
.cWfKrai I1L' ____'t
high wind velocities, it is designed to stop operation at a cut-

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.
 ,
4

out velocity. Thus the wind turbine operates with variable where
load over a narrow range between the cut-in and the rated K =1.09 + 0.2V (15)
velocities and at constant power between the rated and cot-out r
velocities and ceases operation above the cut-out velocity V
C (16)
[17]. 1
F(1+ I)
K
r
I. F= gamma function
p
A height above ground of 9.1 m is usually used as a
reference elevation for which wind velocities are listed. The
velocity to be used in determining the power of a wind
turbine, however, is that at the hub of the turbine wheel. The
ratio of wind velocity V at a height H to the reference velocity
V at reference height H is given by a wind-shear model
r r
as
4 - C -
Cut-mi Rai'ig CIA-ut ax
V H (17)
Velocily V H
r r)
Fig. 6. Power-velocity characteristics of a flat-rated wind turbine. where
C. Wind Speed Model
Wind-power potential can be estimated from the mean or =ao(I ~~loglogV)or (18)
a
0t
V (8

wind velocity V, which is based on measurements over a

period of time. It is given by At elevations H, other than the reference H = 9.1 m, the
r
n Weibull distribution model is used, but the parameters
Vi K and C are modified to K and C , given by
r r H H
v=i 1 (12)
n K
K r (19)
n
where E V. is the sum of all velocity observations and n is
i=1
the number of these observations. °[ Hr ) °]
Power plant sizing would be underestimated if it were rated and
at the mean wind velocity. A more representative figure =0~~ aH
would be based on the mean energy velocity VE, which CH (20)
because of the power dependence on the cube of the wind
velocity, is given by where
n
( log Cr
3 cx =ac 1- (21)
H 0 log V0
it (13)
E

r
n

For estimation purposes, the Weibull distribution model

-,
0.2
(22)
has been found useful and appropriate for wind-turbine
performance analysis by many investigators. This model gives
the probability that the wind velocity is greater than a selected Z =0.4mandV =67.1m/s
0 0
value V for a locality where the mean wind velocity V is
D. WTG Modelingfor Reliability Evaluation
known. It is given at reference height H = 9.1 m, where V.
r i As described in part B, the output of a wind turbine
and Vare in m/s, by generation unit is a function of the wind speed. When a WTG
K is in the normal operating condition, it can be represented by
-(V/C ) r a three-state model. Upl, Up2 and Down are three states,
P(V. > V)= e r
(14) which represent variable, constant and zero outputs,

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.
 5

respectively, in terms of wind speeds. A WTG can also suffer the convolution of capacity outage probability tables.
a forced outage, which can be represented by Up and Down Consequently it is not possible to calculate a generation
states. In order to consider the joint effects of both wind speed output capacity table by convolving the output table of each
and the forced outage, a WTG can be represented by the wind turbine. Determination of the output characteristics of a
three-state model shown in Fig. 7 and Table III. A wind farm wind farm for reliability analysis requires the simultaneous
usually consists of many units and therefore the specified consideration of all wind turbines [4].
wind velocity is assumed to be the same for all the units in the
farm. The power output of a wind farm is the summation of Upi Up2
the output of all the available units [7].
The total output of a generation station is obtained by
appropriately aggregating the outputs of each individual
generator. The conventional approach to aggregate these
outputs using probabilistic analytical techniques is the
convolution of their individual capacity outage probability
tables. This approach is applicable to non-intermittent sources
such as gas and diesel, but is not valid for wind generators.
The problem is that the outputs of units such as wind turbines
are dependent on a common source, e.g. the wind. Therefore Fig. 7. Three-state Markov model of a flat-rated wind turbine generation system.
there is dependence between the different generation output
states, whereas independence is an underlying assumption in
TABL:
FLAT-RATED WIND TURBINE GENERATION S THREE STATE REPRESENTATION

LU nit state State prubability [xpected output power

* / ~ ~i
/t'1? /

7 t/XO '1:
#t
i..qUEMrfEl jl lV

p. X.
W2. - [.Nf w < V Xi-)Ni¢ .)
ji` r t~ rI
I. ~ R P T;
_ < p, V: .u
0.0000285/ kilometer x 2 kilometers x 8759 hours
V. NUMERICAL EXAMPLES = 29.96 minutes
A. Reliability of Combined WTG and T&D Network * Total interruption time as seen by customers on
Fig. 8 shows a five-segment, 20kV radial feeder, a segment A = 89.96 minutes
simplified diagram of the type of circuit typical in many
Iranian rural/suburban areas. Each segment serves 200 150 C
customers with an average peak coincident load of 5kVA
each. Each segment is composed of two kilometers of
overhead line, assumed here to have uniform expected failure Substation
150 k E
rate per kilometer of once every ten years, with an average A E D
expected repair time of 2.5 hours, for an average of 15
minutes of outage per year per kilometer (for a failure ewe dak nir N
probability of 0.0000285). The source of power at the E
substation, itself, is expected to be out of service an average of
one hour per year [14].
The total time customers on the first segment are expected Fig. 8. Simple model of a suburban primary distribution feeder. Fuses are
assumed to function as needed, with no likelihood of failure. Outage minutes are
to be without power is computed as: shown on each segment, per year, rounded to the nearest minute.
* Time each year when the substation is out of service
= 60 minutes Similarly, the expected out-of-service times for the other
* Time during remainder of year when line is out of three sections shown in Fig. 8 can be computed.
service Suppose that a flat-rated wind turbine generation unit, with
an outage rate of 300 were placed on segment D, as shown in

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.
 6

Fig. 8. The wind pattern that follows the Weibull distribution unit's capability of 855kVA, as well as those 18.65 minutes
model has a reference mean velocity 1Om/s, measured at the when the unit will be unavailable. Fig. 9 shows the load
reference height above ground of 9. 1m at 1 standard duration curve for segment D. Load exceeds 435kVA and
855kVA during 8033 and 1288 hours per year, or 91.7% and
atmosphere pressure and 20 C temperature. Consider a 14.7% oftime, respectively. Thus,
turbine with a hub 50m above ground that has cut-in, rated, * Time when segment D will be out of service itself
and cut-out velocities of 4.97, 8.84, and 16.29m/s, 30 minutes
respectively, all at the reference height above ground of 9. tim. * Time during outage of substation, segment A, or
1960m 2cross-sectional segment B when WTG unit will be out of service, too
The turbine wheel has a area (50m =18.65 minutes
diameter). The wind turbine and generator efficiencies are
* Time during remaining 101.35 minutes when load
38% and 9400, respectively.
will be greater than 855kVA = 14.7% of 101.35
Using (14) to (22), the velocities at a height 50m above
ground and their probabilities are computed as: minutes = 14.9 minutes
* Time during remaining 86.45 minutes when the load
V = 15.11 will be greater than 435kVA and the WTG unit will
v 1 711 1 be in UpI state
Cut - in S 0.050567 91.7 14.7 x 86.45minutes
0.050567 + 0.794036 100 - 14.7
V = 13.36 = 4.67 minutes
Rating s
* Total interruption time as seen by customers on
V =24.61m segment D = 68.2 minutes
C;ut - outs
P(V> VCut - in Thus, the WTG unit will cut the expected out-of-service
in) 0.975366 time for the 200 customers on segment D from 150 minutes to
68.2 minutes, an improvement of about an hour and one-half
P V>V jj 0.923235
( Rating) per year, even though it cannot serve the peak load on the
segment.
Pt > C oV
Cut- out )
0.104641 .!

The results are shown in Table IV.

1im
TABLE IV
THREE STATE CAPACITY OUTAGE PROBABILITY TABLE OF WTG UNIT

Unit State ELpected output 435 6 -1- --

st 1I abiltAn
ptob

i~2
01'5056 4 3544 `L
8 5478I5 I
==== iI _.
Down J 5 i)3 7 1288 8033 8760
a-

The WTG unit has just enough capacity to support the
customers on segment D (200 x 5kVA = IMVA) unassisted by
the system. When segment D is not damaged or disabled,
itself, but power to it is unavailable due to outage at
substation or one of the segments upstream, service could be
provided to these 200 customers by this WTG unit. Such
conditions are expected to occur 120 minutes per year (the
other 30 of its expected 150 minutes per year, the segment is
failed). The WTG unit should be available about 84.5% of
this time, or 101.35 minutes, during this period, leaving only
18.65 minutes during the time each year when the feed of
power from the feeder and the WTG unit, itself, are both
expected to be out.
The WTG unit is incapable of serving the peak load on the
segment at any time when it exceeds 855kVA. Thus, a failure
condition for segment D now includes any of those expected
120 minutes per year where power flow through segment B is
unavailable and the load on segment D exceeds the WTG
Tinm IufiDnQiour)
Fig. 9. Load duration curve for segment D. The segment achieves its peak load
of lOOOkVA only 7 hours per year. Load exceeds 435kVA and 855kVA during
8033 and 1288 hours per year, or 91.7% and 14.7% of time, respectively.
B. Wind Farm Model
Consider a wind farm consisting of the same WTG units as
the unit described in part A. Without the loss of generality, it
is assumed that there are only two units in the farm. This
assumption can easily be removed allowing more units to be
considered.
The state probability analysis of the WTG units is shown in
Table V. As discussed in part D of section IV, the outputs of
units are dependent on the wind. Consequently it is not
possible to calculate the generation output capacity table by
convolving the output table of each unit.

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.
 7

[12] R. Billinton and G. Bai, "Generating capacity adequacy associated

with wind energy," IEEE Trans. Energy Conversion, vol. 19, pp.
TABLE V 641-646, Sep. 2004.
EXAMPLE STATE PROBABILITY ANALYSIS OF Two WTG UNITS [13] W. E. Klein and V. R. Lali, "Model-OA wind turbine generator:
failure modes & effects analysis," in Proc. 1990 Annual Reliability
State Wind velocit Unit 1 Unit 2 State Expeted and Maintainability Symp., pp. 337-340.
probability output
[14] H. L. Willis and W. G. Scott, Distributed Power Generation, New
(Wat%
I York: Marcel Dekker, 2000, pp. 409-447.
p

I)Dwn OA01471
XF tI : - 0 435442
Down Up OA0j47l
4 Dlon Down. 0,00t46
5 Up 47108
6 VLii XYrI.Iw Up t)i~on O .023106
7 Do)wn
Down
Up
Down
0.0 23106
OAlilll07!1.1;;5
fVti 126p 1)O-Wzup .14 62-l
I1,h .OICII 'ut
IP DowX n. 4522z
ruiu Do)wn ;up O.0t'4522
11l.L Down Down OJ00140
VI. CONCLUSION
An analytical probabilistic model for the evaluation of the
reliability of wind energy systems has been presented. The
model considers the stochastic nature of wind, the failure and
repair rates of the WTG units, and the flat rated wind turbine
output characteristics in detail. The model can be easily
integrated into existing reliability evaluation models for
electric power networks. The performance of the developed
model has been demonstrated with computational results.
VII. REFERENCES
[1] P. Wang and R. Billinton, "Reliability benefit analysis of adding
WTG to a distribution system," IEEE Trans. Energy Conversion,
vol. 16, pp. 134-139, Jun. 2001.
[2] F. Castro Sayas and R. N. Allan, "Generation availability assessment
of wind farms," IEE Proc. Generation, Transmission and
Distribution, vol. 143, pp. 507-518, Sep. 1996.
[3] E. S. Gavanidou, A. G. Bakirtzis, and P. S. Dokopoulos, "A
probabilistic method for the evaluation of the performance and the
reliability of wind-diesel energy systems," IEEE Trans. Energy
Conversion, vol. 8, pp. 197-206, Jun. 1993.
[4] N. Jenkins, R. Allan, P. Crossley, D. Kirschen, and G. Strbac,
Embedded Generation, London: The Institution of Electrical
Engineers, 2000, pp. 189-226.
[5] A. G. Bakirtzis, "A probabilistic method for the evaluation of the
reliability of stand alone wind energy systems," IEEE Trans. Energy
Conversion, vol. 7, pp. 99-106, Mar. 1992.
[6] J. R. Ubeda and M. A. R. Rodriguez Garcia, "Reliability and
production assessment of wind energy production connected to the
electric network supply," IEE Proc. Generation, Transmission and
Distribution, vol. 146, pp. 169-175, Mar. 1999.
[7] P. Wang and R. Billinton, "Time-sequential simulation technique for
rural distribution system reliability cost/worth evaluation including
wind generation as alternative supply," IEE Proc. Generation,
Transmission and Distribution, vol. 148, pp. 355-360, Jul. 2001.
[8] R. Billinton and R. Karki, "Maintaining supply reliability of small
isolated power systems using renewable energy," IEE Proc.
Generation, Transmission and Distribution, vol. 148, pp. 530-534,
Nov. 2001.
[9] R. Karki and R. Billinton, "Reliability/cost implications of PV and
wind energy utilization in small isolated power systems," IEEE
Trans. Energy Conversion, vol. 16, pp. 368-373, Dec. 2001.
[10] R. Billinton, Bagen, and Y. Cui, "Reliability evaluation of small
stand-alone wind energy conversion systems using a time series
simulation model," IEE Proc. Generation, Transmission and
Distribution, vol. 150, pp. 96-100, Jan. 2003.
[11] R. Karki and R. Billinton, "Cost-effective wind energy utilization for
reliable power supply," IEEE Trans. Energy Conversion, vol. 19,
pp. 435-440, Jun. 2004.
[15] R. Billinton and Bagen, "A sequential simulation method for the
435442
generating capacity adequacy evaluation of small stand-alone wind
O. energy conversion systems," in Proc. 2002 IEEE Canadian
Conference on Electrical & Computer Engineering., pp. 72-77.
l0Q57 [16] R. Billinton and R. N. Allan, Reliability Evaluation of Power
854795 Systems, 2nd ed., New York and London: Plenum Press, 1996, pp.
854795 18-77.
[17] M. M. El-Wakil, Powerplant Technology, Singapore: McGraw-Hill,
O
1988, pp. 589-626.
O
VIII. BIOGRAPHIES
Arash Ehsani was born in Iran in 1973. He received his
B.Sc. and M.Sc. degrees in electrical engineering from
Isfahan Univesity of Technology and Sharif University of
Technology in 1996 and 1998, respectively. Since 1998,
he has been cooperating with National Dispatching
Department of Iran. Currently, he is researching his
doctoral thesis in the area of electric power systems at
Sharif University of Technology and is teaching at Azad
University, Shahr-e-Rey branch, Tehran, Iran.
Mahmud Fotuhi-Firuzabad (M'97, SM'99) was born
in Yazd, Iran. He received his B.Sc. (Hons.) and M.Sc.
degrees in electrical engineering from Sharif University
of Technology in 1986 and the University of Tehran in
1989, respectively and obtained M.Sc. and Ph.D. degrees
in electrical engineering from the University of
Saskatchewan in 1993 and 1997, respectively. He
worked as a postdoctoral fellow in the Department of
Electrical Engineering, University of Saskatchewan from
Jan. 1998 to Sept. 2000 and from Sept. 2001 to Sept.
2002 where he conducted research in the area of power system reliability. He
worked as an assistant professor in the same department from Sept. 2000 to Sept.
2001. Currently, he is an associate professor of electrical engineering department
at Sharif University of Technology. His main research interests are power system
reliability evaluation, unit commitment and power system operation, power
system economics, application of probabilistic techniques to power system
analysis and computer applications in power system.
Ali Abbaspour-Tehrani-Fard was born in Tehran, Iran in
1950. He received his B.Sc. degree (Hons.) in electrical
engineering from Amirkabir University of Technology and
his M.Sc. degree in nuclear engineering from the University
of Tehran. He obtained a Master degree in nuclear fuel
management from M.I.T. and a Ph.D. degree in nuclear
engineering from the University of California, Berkeley.
Currently, he is an assistant professor of electrical
engineering department at Sharif University of Technology
and is a member of Iran's Parliament. His research interests are in renewable
energy resources, power plant technology and nuclear engineering.
Ali Mohammad Ranjbar was born in Boroujerd, Iran in
1943. He received his M.Sc. degree in electromechanical
engineering from Technical College of Tehran University
in 1967, and Ph.D. degree in electrical engineering from
Imperial College of London University in 1975. Since
1975, he has been teaching at Sharif University of
Technology, Tehran, Iran and since 1980, he has been
Deputy Minister of Energy in Iran. He has also been the
director of Niroo Research Institute (N.R.I.), since 1996.
Currently, he is a professor of electrical engineering
department at Sharif University of Technology. He has
published 2 books and more than 100 research papers and technical reports. His
main research interests are in the area of electric power system protection and
operation and electrical machines. Dr. Ranjbar is a member of Iran's Academy
of Sciences and a member of CIGRE.

Authorized licensed use limited to: Malaviya National Institute of Technology Jaipur. Downloaded on October 18,2023 at 06:23:42 UTC from IEEE Xplore. Restrictions apply.