energies

Article
Dynamic Modeling of Wind Turbines Based on
Estimated Wind Speed under Turbulent Conditions
Ahmed G. Abo-Khalil 1,2, *, Saeed Alyami 1 , Khairy Sayed 3 and Ayman Alhejji 4
1 Electrical Engineering Department, Majmaah University, Al-Majmaah 11952, Saudi Arabia;
s.alyami@mu.edu.sa
2 Electrical Engineering Department, Assiut University, Assiut 71515, Egypt
3 Electrical Engineering Department, Sohag University, Sohag 82524, Egypt; khairyfathy@yahoo.com
4 Electrical and Electronics Engineering Technology Department, Yanbu Industrial College,
Yanbu Al Bahr 46452, Saudi Arabia; alhejjia@rcyci.edu.sa
* Correspondence: a.abokhalil@mu.edu.sa




Received: 19 April 2019; Accepted: 15 May 2019; Published: 18 May 2019


Abstract: Large-scale wind turbines with a large blade radius rotates under fluctuating conditions
depending on the blade position. The wind speed is maximum in the highest point when the
blade in the upward position and minimum in the lowest point when the blade in the downward
position. The spatial distribution of wind speed, which is known as the wind shear, leads to periodic
fluctuations in the turbine rotor, which causes fluctuations in the generator output voltage and power.
In addition, the turbine torque is affected by other factors such as tower shadow and turbine inertia.
The space between the blade and tower, the tower diameter, and the blade diameter are very critical
design factors that should be considered to reduce the output power fluctuations of a wind turbine
generator. To model realistic characteristics while considering the critical factors of a wind turbine
system, a wind turbine model is implemented using a squirrel-cage induction motor. Since the
wind speed is the most important factor in modeling the aerodynamics of wind turbine, an accurate
measurement or estimation is essential to have a valid model. This paper estimates the average wind
speed, instead of measuring, from the generator power and rotating speed and models the turbine’s
aerodynamics, including tower shadow and wind shear components, without having to measure the
wind speed at any height. The proposed algorithm overcomes the errors of measuring wind speed in
single or multiple locations by estimating the wind speed with estimation error less than 2%.

Keywords: Doubly-fed Induction Generator (DFIG); wind speed estimation; tower shadow; wind shear

1. Introduction
Currently, electrical power is the main energy source in homes, industries, and workplaces.
Industrial development and population growth have led to increased power consumption in the last
three decades. Natural resources such as gas, petroleum, and coal are quickly depleting because of
their growing use in power plants, vehicles, and industries. This serious problem motivates countries
to find alternative energy sources through inexhaustible natural resources. One solution is found in
clean, sustainable, inexhaustible, environmentally friendly, and renewable sources such as solar energy,
wind energy, and tidal ocean currents. Wind energy has attracted more interest for several reasons and
has increasingly become the most harvested renewable energy source [1]. Increased interest in wind
turbines to generate electrical power entails studying and modeling the steady state and dynamic
behavior of the wind turbine in laboratory conditions to prevent possible problems during installation
and later use. In addition, over the last few decades, new control algorithms have been established to
improve the performances of wind-generation systems. Before applying a new algorithm, it must be

Energies 2019, 12, 1907; doi:10.3390/en12101907 www.mdpi.com/journal/energies

Energies 2019, 12, 1907 2 of 25

validated in the laboratory by simulating the entire system to verify the performance, control, and
disturbances. Developing and improving the characteristics of a wind turbine simulator will help reach
this objective and reflect the dynamic behavior of a wind turbine in a controlled environment [1–3].
Several wind turbine emulators have been proposed and studied in-depth. Some studies have
investigated direct current (DC) motors to model the dynamic characteristics because of their robust
characteristics such as output torque and input current using armature current control [4,5]. These
emulators, however, entail a bulky and expensive DC motor, which requires frequent maintenance
given its commutators and brushes. With regard to AC motor emulators, squirrel-cage induction
motors (SCIM) are less costly and more reliable and robust [6]. Several research articles feature simple
experiment platforms utilizing SCIM to model the characteristics of wind turbines. However, these
emulators have simple wind turbine models that neglect the dynamics of wind turbine components,
an important factor in wind energy systems [7,8]. A scaled-down emulator operated by a programmable
wind speed model has been developed in several studies to investigate wind turbine performance
during turbulences and stochastic wind speed changes [9].
Nevertheless, these emulators disregard several dynamic performance characteristics related to
tower shadow, wind turbulence, and shear. These characteristics clearly influence the power quality
and stability of wind-energy conversion systems (WECS) [10,11]. Specifically, fluctuations due to
tower shadow and wind shear affect the WECS stability when the fluctuation frequency equals the
natural frequency of the system [12]. In addition, all these studies used the measured wind speed,
leading to inaccurate wind turbine dynamic characteristics and misleading results. Similarly, using
anemometers in measuring wind speed leads to several problems in calibration and measurement
accuracy and higher initial WECS costs. Therefore, using a wind speed estimator that is based on
turbine characteristics instead of mechanical anemometers is desirable. In the literature, different
techniques have been developed to estimate the effective wind speed without using an anemometer.
These techniques can be categorized into two approaches. The first is the polynomial-based method
for wind speed estimation, which uses a power equation as a function of the power coefficient and
tip-speed ratio. This method utilizes root-finding algorithms to solve the polynomial with a high order
of power coefficient for accurate estimation. However, the method is time-consuming and therefore
impractical. Second, artificial intelligence and machine-learning methods, such as adaptive neuro-fuzzy
inference systems [13,14], multilayer perceptron neural networks [15], and support vector regressions
(SVR) [16,17], have been widely implemented to estimate effective wind speed. These methods help
map the relation between electrical measured quantities, such as turbine power and rotational speed,
and wind speed using numerous samples, and then use this online map to estimate the wind speed
and the optimum wind speed, which produce the maximum turbine power point.
In this framework, a wind turbine modeling based on the estimated wind speed is presented
by simulating both the static and dynamic characteristics using a SCIM. The tower shadow and
wind shear are modeled as a function of the turbine parameters and continuously varying wind
speed. A continuously varying wind speed profile composed of a sum of harmonics with wide-range
frequencies to present the stochastic nature is developed. A particle swarm optimization–support
vector regression (PSO-SVR) approach is employed for online wind speed estimation and generator
speed adjustment for maximum power point tracking. Because the SVR performance alone severely
depends on the tuning of its parameters, the PSO algorithm is used, which enables a fast online-based
approach with high-parameter estimation accuracy. All dynamic features in the wind turbine are
experimentally implemented to validate the proposed model shown in Figure 1.

2. Mathematical Model of Wind Energy System

Figure 1 illustrates the structure of the wind power generation system, which exploits a SCIM
to emulate the wind turbine characteristics. The details of the wind emulator components, the DFG
modeling, and the generator controller are discussed in the next sections.
 Energies 2017, 10, x FOR PEER REVIEW 3 of 26

Energies 2017,12,
Energies2019, 10,1907
x FOR PEER REVIEW 33ofof2526

Figure 1. Basic configuration of the proposed wind-energy conversion system (WECS).

2.1. Wind Turbine Model

Figure 1. Basic configuration of the proposed wind-energy conversion system (WECS).
Figure 1.power
The extracted Basic configuration
from the wind of the proposed
at any windwind-energy conversion
speed and turbine system (WECS).
rotational speed is given by
2.1. Wind Turbine Model
[18]
2.1. The
Windextracted
Turbine Model
power from the wind at any wind speed and turbine rotational speed is given
1at any wind
by [18]The extracted power from the P
t =
wind ρπ R 2υ 3Cspeed
p (β , λ ) .
and turbine rotational speed is given(1)by
[18] 2 ρπR2 υ3 Cp (β, λ).
Pt =
1
(1)
2
Equation (1)
Equation (1) indicates
indicates that
that the
the power
power 1coefficient
coefficient is influenced by the TSR and the blade pitch
2 3 is influenced by the TSR and the blade pitch
angle, which is always below the
angle, which is always below the ratedt wind
P
rated =
wind ρπ R
speed.υInC p ( βenergy
wind , λ ) . systems,
energy systems,when
whenthe thepitch angle(1)
pitchangle isis
keptconstant,
constant,the
thepower
powercaptured
capturedwill willbe
2 speed. In wind
beaafunction
functiononly
onlyin inthe
theturbine
turbinerotational
rotationalspeed,
speed,maximum
maximum
kept
at a Equation
certain (1)
speed. indicates
Hence, to that the
maximize power
the coefficient
captured is
power, λ
influenced
should by the
be
at a certain speed. Hence, to maximize the captured power, λ should be kept at its optimum value TSR
kept atand
its the blade
optimum pitch
value
λλangle,, andwhich
thenisthe
always belowpower
captured the rated wind speed.computed
is technically In wind energyfrom systems,
the windwhen
opt , and then the captured power is technically computed from the wind speed and the blade radius.
opt speedtheandpitch
theangle
blade is
kept
When
radius. constant,
theWhen
windthethe
speed power
wind captured
is above
speedthe will be
rated speed,
is above a function
the pitch
the rated only
speed,angle in the
adjusts
the pitch turbine rotational
to reduce
angle speed,
thetoimpact
adjusts maximum
reduceofthe
excessive
impact
at excessive
wind
of a certain
speed onspeed. Hence,
the speed
wind ontothe
generated maximize
power. Here,
generated β captured
thepower. power,
is keptHere,
constant λ simplicity.
for
β is kept should befor
constant kept at its optimum value
simplicity.
λopt Figure
, and then the captured
Figure 22 presents
presents the power
the turbine is
turbine blade’stechnically
blade’s power computed
power variation
variation with from
with windthe wind
wind speed
speed andspeed and thespeed.
and rotational
rotational blade
speed.
The
The maximum
radius. When power
maximum power
the output
windoutput
speed occurs
occursatataathe
is above particular
particular rotational
rotational
rated speed, speed.
speed.
the pitch angle adjusts to reduce the impact
of excessive wind speed on the generated power. Here, β is kept constant for simplicity.
Figure 2 presents the turbine blade’s power variation with wind speed and rotational speed.
The maximum power output occurs at a particular rotational υ =12 m/s
speed.
υ =11 m/s
υ =10 m/s υ =12 m/s

υ =9 m/s υ =11 m/s

υ =8 m/sυ =10 m/s
υ =7 m/s
υ =6 m/s υ =9 m/s
υ =5 m/s υ =8 m/s
υ =7 m/s
υ =6 m/s
Figure2.2. Wind
Figure Windturbine υ =5 m/scurves:
turbinecharacteristic
characteristic curves: Wind
Windturbine
turbinepower
power(Watt)
(Watt)vs.
vs.rotational
rotationalspeed
speed(rpm).
(rpm).

2.2.
2.2. Wind
WindShear
Shearand
andTower
TowerShadow
ShadowModelModel
Different
Figure 2.wind
Different windspeeds
Wind turbine
speedsatcharacteristic
different altitudes
at differentcurves: and
Wind
altitudes the
and passing
turbine of the
power
the passing ofblade
(Watt) in front of speed
thevs.blade
rotational the tower
in front the result
of(rpm).
tower
in periodic torque pulsations in wind turbines.
result in periodic torque pulsations in wind turbines.
2.2. Wind Shear and Tower Shadow Model
Different wind speeds at different altitudes and the passing of the blade in front of the tower
result in periodic torque pulsations in wind turbines.
Energies 2019, 12, 1907 4 of 25

Based on the height, the wind speed changes as follows [19,20]:

!α
υ2 h
= 2 . (2)
υ1 h1

Table 1 summarizes the geographical features and represents a general wind shear index [1].

Table 1. The empirical wind shear index [1].

Terrain Empirical Wind Shear Exponent

Smooth, hard ground, lake, or ocean 0.1
Smooth, level, grass-covered 0.14
Tall row crops, low bushes with few trees 0.2
Many trees, occasional buildings 0.24

As a result of height-dependent wind speeds, the turbine blades oscillate in response to the wind
shear. Additionally, the amplitude of the wind speed increases with the height for different rotating
angles from 0◦ to 360◦ . The wind speed in Equation (2) can be rewritten with respect to the speed at
the turbine hub as follows [21]:  α
z
V ( z ) = Vh . (3)
H
Considering the blade’s rotation and the unique wind shear at each angle, instead of expressing
the wind shear in only the z direction, the wind shear in Equation (6) can be expressed as a function of
the blade rotational angle ψb and radial distance r from the rotor axis. The wind shear torque can be
modeled as [22]:
r cos ψb + H α
!
V (r, ψb ) = Vh (4)
H
which can be written as
!α
r cos ψb + H r cos ψb
!
V (r, ψb ) = Vh = Vh +1 . (5)
H H

Equation (5) can be represented as a function of the wind shear disturbance F(r, ψb ):

V (r, ψb ) = Vh (F(r, ψb ) + 1). (6)

Due to wind shear, the wind speed can be expanded using Fourier transform as shown in the
following Equation [22]:

 α r cos ψ + α(α−1) r 2 cos2 ψ

     

H b 2 H b
ϑws (r, ψb ) = Vh  α(α−1)(α−2)  3 . (7)
 
H cos ψb
 + r 3 
6

In low altitude, wind shear is influenced by the large amount of change in wind speed due to
the friction of the surface. In addition, the longer the blade length of the wind turbine, the wider the
turning radius of the rotor, and the larger the wind shear effect is. Figure 3 shows the wind shear effect
of the turbine blade length, r hub height, and H empirical wind shear index as in Equation (7).
 Energies 2017, 10, x FOR PEER REVIEW 5 of 26

In low altitude, wind shear is influenced by the large amount of change in wind speed due to
the friction of the surface. In addition, the longer the blade length of the wind turbine, the wider the
Energies
turning 2019, 12, 1907of the rotor, and the larger the wind shear effect is. Figure 3 shows the wind5 shear
radius of 25
effect of the turbine blade length, r hub height, and H empirical wind shear index as in Equation (7).

(a)

(b)

(c)
Figure3.3.Wind
Figure Wind shear
shear comparison
comparison with
with parameter
parameter adjustment:
adjustment: (a) Wind
(a) Wind shearshear by radius;
by rotor rotor radius; (b)
(b) wind
windby
shear shear
hub by hub height;
height; (c) wind(c)shear
windwith
shear withshear
wind windindex.
shear index.

InInthe
thecase
caseofofaalarge-diameter
large-diameterwind
windturbine,
turbine,the
theblades
bladesexperience
experienceaawide
widerange
rangeofofwind
windspeeds
speeds
inineach revolution. With three-blade
each revolution. With three-blade wind turbines, the frequent rotation through this wide range
turbines, the frequent rotation through this wide range of
ofspeeds
speedsresults
resultsinintorque
torqueand
andpower
poweroscillations
oscillationsatatthree
threetimes
timesthe
therotor
rotorrotational
rotational speed,
speed, called
called 3p
3p frequency.
frequency.
The tower radius (a), the distance of the blade origin from the tower midline (x), and the lateral
distance of the blade from the tower midline (y) affect the tower shadow amplitude, as shown in
Figure 4.
 Energies 2017, 10, x FOR PEER REVIEW 6 of 26
Energies 2017, 10, x FOR PEER REVIEW 6 of 26

The tower radius (a), the distance of the blade origin from the tower midline (𝑥), and the lateral
The tower radius (a), the distance of the blade origin from the tower midline (𝑥), and the lateral
Energies 12, the
2019, of
distance 1907blade from the tower midline (𝑦) affect the tower shadow amplitude, as shown6in
of 25
distance of the blade from the tower midline (𝑦) affect the tower shadow amplitude, as shown in
Figure 4.
Figure 4.

Figure 4. Dimensions used in the tower shadow formula [12].

Figure 4. Dimensions used in the tower shadow formula [12].
Figure 4. Dimensions used in the tower shadow formula [12].
In front of the tower, the disturbance is nonnegligible and affects the wind speed and turbine
output In front of the
power.
In frontThe
tower,
wind
of the
the reduces
speed
tower,
disturbance
by aisis
the disturbance
nonnegligible
value
nonnegligible
andof
of ∆ in front affects the wind
the tower
and affects and speed
the wind withinand
the turbine
speed and turbineW
region
output power. The wind speed reduces by a value of Δ in front of the tower and within the region
of 3–6 times
output of theThe
power. tower
winddiameter, as shown
speed reduces by ain Figure
value Δ in front of the tower and within the region
of 5.
W of 3–6 times of the tower diameter, as shown in Figure 5.
W of 3–6 times of the tower diameter, as shown in Figure 5.
VV0
0
∆
∆
yy

W
W
xx

Figure 5. Wind speed pattern in front of the tower.

Figure5.5.Wind
Figure Windspeed
speedpattern
pattern in
in front of the
the tower.
tower.

The tower shadowasasa afunction

function of x and yy is is given by [23]
The tower
The tower shadow as a functionofofxxand
shadow and y isgiven
given byby [23]
[23]

(( ))
 aa222 xx222++ yy222 
  

VV (( ))
= Vh1 + a x + y  
V (xx,, yy) =
x, y =VVhh11++ (x22 + y22)222  (8) (8)
 ((xx ++ yy )) 
2 2
(8)

V (x, y) = Vh + Vtower (x, y). (9)

Expressing the tower shadowVas x (( ))

, yy ==VVhof
V ax,function +r and
Vtower (( ))
ψb , x , y .
h + Vtower x, y .
(9)
(9)
 
2 2 sin2 ψ − x2
Expressing the tower shadow as a function of ar rand ψ b ,b
Vtoweras(r,a ψ
Expressing the tower shadow function of  r and ψ b , 
b , x) = m (10)
2
aa 2 ( rr 2 sin
2 ψb 2+ x
sin 2ψψbb −− xx ))
2 2 2
2r sin 2

VVtower ( rr,,ψψb ,,xx) == m ( 2

tower ( ) m r 22sin 22ψ + x 22 22 (10)

where m = 1 + 8H2 .
α(α−1)R2 b
(( r sin ψbb + x )) (10)
Figure 6 shows wind shear and tower shadow effects at the tip of each blade at an average wind
speed of 8 m/s. The tower shadow shape was calculated for different radial circulating points, from
8 m to 20 m in diameter, measured from the hub center, with an incremental distance of 8 m, as shown
in Figure 6. When the diameter increases, the points on this circle perimeter experience short tower
shadow intervals. The inner curve in Figure 6 indicates the circle with a diameter of 20 m, which has the
lowest tower shadow interval. In contrast, the outer and widest curve in Figure 7 illustrates the shadow
 Figure
where 6 shows wind 8H 2shear
. and tower shadow effects at the tip of each blade at an average wind
speedFigure
of 8 m/s. The tower shadow
6 shows wind shear and shape was
tower calculated
shadow forat
effects different
the tip ofradial
eachcirculating
blade at anpoints,
averagefrom
wind8
m to 20 m in diameter, measured from the hub center, with an incremental distance
speed of 8 m/s. The tower shadow shape was calculated for different radial circulating points, from 8 of 8 m, as shown
in
mFigure
to 20 m6.inWhen the diameter
diameter, measured increases,
from thethe
hubpoints
center,onwith
thisancircle perimeter
incremental experience
distance short
of 8 m, tower
as shown
shadow
in Figure intervals.
WhenThe the inner curve in Figurethe
6 indicates
points onthe circle with a diameter of 20 m,short
which has
Energies 2019, 12,6.
1907 diameter increases, this circle perimeter experience tower
7 of 25
the
shadow intervals. The inner curve in Figure 6 indicates the circle with a diameter of 20 m, which the
lowest tower shadow interval. In contrast, the outer and widest curve in Figure 7 illustrates has
shadow
the lowest effect at shadow
tower a radiusinterval.
of 4 m. In The parameters
contrast, usedand
the outer in widest
this test 𝑅 Figure
are in
curve 10 m,7𝐻illustrates
25 m, 𝛼the
0.15,
effect at a𝑎 radius
shadow 0.75 m,at
of
effect 4and
m. x =parameters
The
a radius 3m.
of 4 m. The usedparameters are Rin
in this testused = this
10 m, H=
test are25𝑅m, α10=m,0.15,
𝐻 a25 = m, 𝛼 m,
0.75
x = 3𝑎 m.0.75 m, and x = 3m.
and 0.15,

Figure 6. Tower shadow effect on each blade.

Figure
Figure 6. 6. Tower
Tower shadoweffect
shadow effecton
oneach
eachblade.
blade.

Figure 7. Tower shadow effect for different r.

Figure 7. Tower shadow effect for different r.

If the radial distance r is kept Figure 7. Tower

constant andshadow effect for different
the distance betweenr. the blade origin to the tower
If the radial distance r is kept constant and the distance between the blade origin to the tower
midline x varies from 3 m to 5 m, as Figure 8 illustrates, the tower shadow effect increases when the
midline x varies
If the radialfrom 3 m to
distance 5 m,
r is keptasconstant
Figure 8 and
illustrates, the tower
the distance shadow
between the effect
blade increases
origin to when the
the tower
distance x
distance
is small. Therefore, the blades should be installed far from the tower to reduce the tower
midline xx varies
is small. Therefore,
from 3 m to 5the
m, blades
as Figureshould be installed
8 illustrates, far from
the tower the tower
shadow effecttoincreases
reduce the tower
when the
shadow
shadow effect. TheThe
effect. tower
towershadow
shadow is is
given
given byby
distance x is small. Therefore, the blades should be installed far from the tower to reduce the tower
shadow effect. The tower shadow 3 " is given by 2
3  ψb
2 2

a22 RR 2sin  2a22Ra22 R 2
! #
mVmVh X a sin ψ
V
3mV
=
Vts ts= 3R h
2 
2 3 
 2 b
sin 22
ψ
ln
ln  2x2 2 + 1 −R2 sin2 ψ 2+ x22 . .
R h bb==1 1 sin a ψ b  R sin
+
b 1 −
x ψ b  R sin2a ψRb + x 
2 2 2b 2 
(11)
(11)
Vts =
3R 2 bthe
=1 
 ln

sin 2 ψ b torque 2  R sin ψ + x 2 
+1 − 2 2
 Ttxcan be expressed

. (11)
For any wind speed value, turbine as: b 
Cp (λ)
Tt = 0.5ρAVh3 . (12)
ωr

The wind shear module implements the disturbance in torque due to wind shear given as

Cp (λ)
Tws = ρAVh Rvws (13)
λ
and the tower shadow module implements the disturbance in torque due to the tower shadow
expressed as
Cp (λ)
Tts = ρAVh Rvts . (14)
λ
Due to turbine torque, wind shear, and tower shadow, the resultant aerodynamic torque is defined
as Equation (15a).
Energies 2017,12,
Energies2019, 10,1907
x FOR PEER REVIEW 8 of
of 25
26

Figure 8. Effect
Figure8. Effect of
of distance
distancefrom
fromtower midlinex𝑥on
towermidline ontower
towershadow.
shadow.

However, the wind

For any wind speedspeed
value,atthe
the hub level
turbine is not
torque accurate
Tt can when modeling
be expressed as: the turbine torques.
Hence, a wind speed estimation method can be used to estimate the average wind speed to resemble
p (λ )
C can
real characteristics. The resultant torque in Equation (15a) be expressed as Equation (15b):
Tt = 0.5 ρAVh 3
. (12)
C (λ)
3 p
Cpω(λr)
Tres = 0.5ρAVh + ρAVh R(vws + vts ). (15a)
ωr λ
The wind shear module implements the disturbance in torque due to wind shear given as
Cp (λ) Cp (λ)
Tres = 0.5ρAVo3 + ρAVo R(vws + vts ). (15b)
ωr C p (λλ)
T = ρAV h Rv ws (13)
2.3. Dynamic of the Rotating Masses in wsthe Wind Turbine λ
The wind turbine rotor can be coupled directly to the generator shaft, or through a gearbox.
and the tower shadow module implements the disturbance in torque due to the tower shadow
The model traditionally used to represent the dynamic behavior of the wind group is the one that uses
Energies 2017,
expressed as 10, x FOR PEER REVIEW 9 of 26
a system composed of two masses and their respective moments of inertia, with one mass representing
therepresenting
complete wind the turbine
complete and the other
wind theand
turbine generator rotor.
the other theThese masses
generator are connected
rotor. by anare
These masses axis
C pin(λ ) 9. Aasgearbox
that has its stiffness
connected andthat
by an axis damping
T = ρAVh
has itsmodeled, as shown
stiffness and damping Figure
modeled,
Rvts . shown with an n 9.
in Figure ratio is used to
A gearbox (14)
couple
with the
an low-speed side to
n ratio is used of couple thetslow-speed
the turbine sideλof the
to the high-speed sideturbine
of the generator.
to the high-speed side of the
generator.
Due to turbine torque, wind shear, and tower shadow, the resultant aerodynamic torque is
Jt
defined as Equation (15a).
However, the wind speed at the hub level is not accurate when modeling the turbine torques.
ωt T t method can beGear
Hence, a wind speed estimation ratio
used Jg
Tgtheωaverage
to estimate wind speed to
k in 1:n
resemble real characteristics. The resultant torque Equation (15a) can beg expressed as Equation
(15b):

C p (λ D
) tg C p (λ )
Tres = 0.5ρAVh3 + ρAVh R(vws + vts ). (15a)
ωr λ Dg
Dt
C p (λ ) model
Figure9.9.Dynamic
Figure of aaC p (λturbine.
wind )
turbine.
Tres = 0.5 ρAVo3
Dynamic model of
+ ρAVo
wind
(
R vws + vts ). (15b)
2.4. The Wind Speed Simulator ωr λ
2.4. The Wind Speed Simulator
To model the torque fluctuation of a running wind turbine, a stochastic model for wind speed is
To model the torque fluctuation of a running wind turbine, a stochastic model for wind speed is
required. This model is a sum of harmonics and is expressed as [24]
required.
2.3. DynamicThis model
of the is a sum
Rotating of harmonics
Masses in the Wind and is expressed as [24]
Turbine
N
 
The wind turbine rotor can be coupleddirectly
Ai sin(ωi t). 
XN to the generator shaft, or through a gearbox. The

 
υ(t) = Vo 1 + (16)
υ ( t ) =theV odynamic
model traditionally used to represent 1 + behavior ω ithe
A i sin( of t ) wind group is the one that(16)
uses a
 i=1 .
system composed of two masses and their irespective =1 moments  of inertia, with one mass

For each selected harmonic component, the amplitude can be written as a function of Dryden
spectra φu as [25]
Energies 2019, 12, 1907 9 of 25

For each selected harmonic component, the amplitude can be written as a function of Dryden
spectra φu as [25]
r
2 1
Ai = [ϕu (ωi ) + ϕu (ωi+1 )]·[ωi+1 − ωi ] (17)
π 2
The longitudinal Dryden power spectrum is defined in terms of turbulence intensity σ, turbulence
length, and harmonic frequency as follows [26]:
Energies 2017, 10, x FOR PEER REVIEW 10 of 26
Lu 2σ2 1
φu (ω) = . (18)
Vo π 1 + (Lu ω 2
0 = idr Rr + (ω e − ω r )λqrVo+) pλdr (21)

3. Induction Motor Control

0 = i qr R r + (ω e − ω r ) λ dr + p λ qr . (22)
The induction machine can be modeled by writing the dynamic equations in the d-q synchronous
reference frameThe
form as [27]
electromagnetic torque in the induction machine is fundamentally defined as
vds = ids Rs − ωe λqs + pλds (19)
3 P
Te = λ dr i qs
vqs = iqs2Rs 2+ ωe λds + pλqs (23) (20)
= K i ds i qs = K t i qs .
0 = idr Rr + (ωe − ωr )λqr + pλdr (21)
The motor torque and0 flux Rr + (ωeto− ω
= iqrconnected r )λdr + pλqr .
the power source are controlled through the (22)
back-to-back converter. Whereas the motor flux is kept constant, the torque is controlled to provide
The electromagnetic torque in theToinduction
the wind turbine characteristics. machine is
generate a continuous fundamentally
variable defined
torque, the torque as iqs is
current
continuously controlled as defined in Equation (23). In experiment, the turbine reference torque and
2 2 λdr iqs
3P
wind speed pattern are implemented Te by=Equations (15b) and (16), respectively. The torque current
*
(23)
reference iqs is then calculated from the= qs .dividing it by the torque constant Kt ,
qs = Kt iby
Kids iequation
torque
as shown in Figure 10a.
The motorIn torque and
fact, the flux reference
motor connected to the
d-axis power
current source
is kept are controlled
constant at the value through the back-to-back
that maintains the
converter. Whereas
minimum the ironmotor
loss. In flux is kept
contrast, in theconstant, the torque
grid-side converter, theisDC-link
controlled toisprovide
voltage the wind
kept constant, and turbine
the output
characteristics. power factor
To generate is adjustedvariable
a continuous to unity by controlling
torque, the the grid-side
torque converter
current iqs is q-axis current andcontrolled
continuously
the d-axis current. The dc-link voltage is controlled to the desired value by using a
as defined proportional-integral
in Equation (23). (PI) In controller
experiment, the turbine reference torque and wind speed pattern
and the change in the dc-link voltage represents a change in the
∗
are implemented by Equations
q-axis current. (15b) and (16),
A current feed-forward respectively.
control loop is also usedThe torque
here current
to improve voltageiqs is then
reference
the dc-link
calculated from thetotorque
response equation by
load disturbance. For dividing
unity power it by thethe
factor, torque
demand for the K
constant t , as current
d-axis shownisinzero.
Figure 10a.
Figure 10b shows a control block diagram of the grid-side converter.
DFIG

IM
Vdc

abc/dq ωr
var* vbr* vcr* θe

1
iqr idr s
SVPWM ωe
ωr
e∗ e∗ +
v
qs v ds ω sl +
1
Tr Slip
e Calculation
i ds
-
Kp + Ki
s + e*
ids ∗
.. Τturbine
Kp + K i +
s iqse *
iqse -
Κt

(a)

Figure 10. Cont.

Energies 2019, 12, 1907 10 of 25
Energies 2017, 10, x FOR PEER REVIEW 11 of 26

(b)
Figure
Figure 10. 10. Block
Block diagram
diagram forfor
(a)(a)induction
induction motor
motorcontrol and (b)
control and grid-side converter.converter.
(b) grid-side
4. Control Scheme of DFIG
In fact, the motor reference d-axis current is kept constant at the value that maintains the minimum
To control the active and reactive power in the stator side, a d-q reference frame
iron loss. In contrast, in the grid-side converter, the DC-link voltage is kept constant, and the output
synchronization with the stator flux is selected. The flux vector in the stator is aligned with the
power factor isThe
d-axis. adjusted to unity
flux linkage by controlling
of the rotor as [28] converter q-axis current and the d-axis
the grid-side
and stator is defined
current. The dc-link voltage is controlled to the desired value by using a proportional-integral (PI)
controller and the change in theλdc-link s = λdsvoltage
= Lm ims = Ls ids a+change
represents Lm idr in the q-axis current.
(24) A current
feed-forward control loop is also used here to improve the dc-link voltage response to load disturbance.
For unity power factor, the demand for the d-axis L2m current is zero. Figure 10b shows a control block
diagram of the grid-side converter. λ dr = ims + σLr idr (25)
Ls
4. Control Scheme of DFIG
λqr = σLr idr (26)
To control the active and reactive power in the stator side, a d-q reference frame synchronization
with the stator flux is selected. The flux vector in the stator is aligned with the d-axis. The flux linkage
L2
of the rotor and stator is defined as [28] σ =1− m . (27)
Lr Ls
λs = λds = Lm ims = Ls ids + Lm idr (24)
Rotor voltages in the d-q reference frame can be written in terms of rotor and stator
magnetizing currents. 2 Lm
λdr = ims + σLr idr (25)
Ls di
vdr = Rridr + σLr dr − ωslσLriqr (28)
λqr = σLdtr idr (26)

L2m
σ =di1 − . 2 (27)
vqr = Rr iqr +σLr
qr LLs i + Lm i )
+ ωLslr(σ (29)
r dr ms
dt Ls
Rotor voltages in the d-q reference frame can be written in terms of rotor and stator
magnetizing currents.
The stator flux angle is expressed as

didr
vdr = Rr idr + σLr − ωsl σLr iqr (28)
dt
diqr L2m
vqr = Rr iqr + σLr + ωsl (σLr idr + ims ) (29)
dt Ls
The stator flux angle is expressed as
Z
λsds = (vsds − Rs isds )dt (30)

Z
λsqs = (vsqs − Rs isqs )dt (31)
Energies 2019, 12, 1907 11 of 25

λsqs
−1
θe = tan (32)
λsds
where the superscript s represents the parameters in the stationary reference frame.
Regulating the q-axis component of the rotor current controls either the active power of the DFIG
stator side or the developed torque in the generator.

3 3 Lm
Ps = (vqs iqs + vds ids ) = − · · vqs iqr (33)
2 2 Ls
To directly control the stator-side reactive power, the rotor d-axis current component should be
regulated. This reactive power can be expressed as

3 3 Lm
Qs = (vqs ids − vds iqs ) = · · vqs (ims − idr ). (34)
2 2 Ls

The stator active and reactive powers are proportional to the current components iqr , and
idr , respectively.
The DFIG wind turbine system configuration consists of a DFIG connected to the utility grid
through its stator. The back-to-back Pulse-Width Modulation (PWM) power converter enables a
bidirectional power-flow control. As a result, the DFIG can operate either in sub-synchronous (ωr < ωe )
or in super-synchronous modes (ωr > ωe ). By calculating the stator active power, the control block
diagram of the stator power is presented in the figure. The PI controller gains for the active and reactive
power controllers are Kpp , Kip , Kpq , and Kiq , respectively. The rotor reference q-axis current i∗qr is then
calculated as the output of the active power controller and input for the inner current control loop.
The rotor instantaneous q-axis current is then calculated from the sensed three-phase rotor currents
and controlled to produce a reference q-axis rotor voltage. Normally, the output reactive power of the
wind power conversion system is controlled as zero to keep unity power factor of the stator voltage
and current. The stator reactive power is controlled to the desired value to produce the reference d-axis
rotor voltage as shown in Figure 11b. One can see how the outer stator power feedback loop produces
the rotor reference d-axis current i∗dr for the inner current feedback control loop. The reference q-axis
rotor voltage is then produced by controlling the rotor d-axis component. The schematic diagram of
the simplified control scheme is shown in Figure 11c.
The DFIG optimum power P∗ is the reference power value determined by the wind speed and
rotor angular speed of the power controller loop. A simple proportional-integral (PI) controller can be
utilized to regulate the d-q component of the rotor current only if the reference frames for both the
measured and reference current vectors match [29].
To achieve the full control of the grid-side current, the dc-link voltage must be boosted to a
level higher than the amplitude of the line-line voltage. The power flow of the grid-side converter is
controlled so as to keep the dc-link voltage constant. To maintain the dc-link voltage constant and to
ensure the reactive power flowing into the grid at null, the grid-side converter currents are controlled
using the d-q vector control approach. The dc-link voltage is controlled to the desired value by using an
IP controller and the change in the dc-link voltage represents a change in the q-axis current. A simple
voltage control scheme is shown in Figure 12a. A current feed-forward control loop is also used here to
improve the dc-link voltage response to load disturbance. For unity power factor, the demand for the
d-axis current is zero. Figure 12b shows a control block diagram of the grid-side converter.
Energies 2017, 12,
Energies 2019, 10, 1907
x FOR PEER REVIEW 12 of
13 of 26
25

Ps* iqr* iqr - 3 ⋅ Lm ⋅ v Ps

Kpp+ Kip State feedback
+ s + Current controller 2 Ls
qs

-
P
s
iqr -

(a)

ims
Qs* i*dr idr- +
3 Lm
Qs
Kpq+ Kiq State feedback ⋅ ⋅ vqs
+ s Current controller 2 Ls
-
Q
s

(b)

DFIG 3-φ
DFIG
Vdc Grid

var* vbr* vcr* ωr

abc/dq θsl 1
s eab ebc
θ
r abc/dq
θe
iqr idr
SVPWM

v
e∗
dr vqre ∗ eqe ede
Synchronization mode
Decoupling
iqre iqre *٠ =
+ - 1
+ State feedback
Current controller +ie * 2 e
qr Voltage - Vqs
controller
1
State feedback + idr
e* + eqsrc
+ Current controller
- -i e 2
dr
P*
Decoupling
+
Power
controller -
P
meas
Running mode Q
Reactive power - meas
controller
+ Q*

(c)

11. DFIG
Figure 11. DFIGrotor-side
rotor-side(a)(a)
active power
active control,
power (b) reactive
control, power
(b) reactive control,
power and (c)and
control, the (c)
total control
the total
block diagram
control of the DFIG.
block diagram of the DFIG.

To achieve the full control of the grid-side current, the dc-link voltage must be boosted to a level
higher than the amplitude of the line-line voltage. The power flow of the grid-side converter is
controlled so as to keep the dc-link voltage constant. To maintain the dc-link voltage constant and to
ensure the reactive power flowing into the grid at null, the grid-side converter currents are
controlled using the d-q vector control approach. The dc-link voltage is controlled to the desired
value by using an IP controller and the change in the dc-link voltage represents a change in the
q-axis current. A simple voltage control scheme is shown in Figure 12a. A current feed-forward
control loop is also used here to improve the dc-link voltage response to load disturbance. For unity
Energies 2017, 10, x FOR PEER REVIEW 14 of 26

Energies 2019, 12, 1907 13 of 25

power factor, the demand for the d-axis current is zero. Figure 12b shows a control block diagram of
the grid-side converter.

θe

ωL ω L

(a)

(b)
Figure 12.
Figure 12. (a)
(a) DFIG
DFIG grid-side
grid-side control,
control, (b)
(b) voltage
voltage controller
controllerblock
blockdiagram.
diagram.

5.
5. Wind
Wind Speed
Speed Estimation
Estimation
5.1. Support Vector Regression
5.1. Support Vector Regression
The SVR is an algorithm that estimates a function that predicts unknown mapping between
The SVR is an algorithm that estimates a function that predicts unknown mapping between a
a system’s l-dimensional inputs and real output using a series of training data. When the map is
system’s l-dimensional inputs and real output using a series of training data. When the map is
accurately constructed, the relation between the inputs and output is then used to predict the target
accurately constructed, the relation between the inputs and output is then used to predict the target
output. The regression algorithm approximates the unknown desired function in the following
output. The regression algorithm approximates the unknown desired function in the following form
form [30]:
[30]:
f (x) = (w.Φ(x)) + b. (35)
f ( x) = (w.Φ( x)) + b . to determine the target function.
These values are extracted from the available training data
(35)
The first step
These is to obtain
values both wfrom
are extracted and bthe to available
minimizetraining
the real data
data to
differences
determineoutside ε-insensitive
the function.
the target The
region, which are determined by slack variables ξ and ξ∗ . To minimize the empirical risk, the slack
first step is to obtain both w and b to minimize the real data differences outside the ε -insensitive
variables are applied as [31] ∗
region, which are determined by slack variables ξ and ξ . To minimize the empirical risk, the
n
slack variables are applied as [31] 1 2
X
Rreg ( f ) = kwk + C Γ( f (xi ) − yi ) (36)
2
i=n
1
1 2
subject to
2 i =1

Rreg ( f ) = w + C Γ( f ( xi ) − yi ) (36)
yi − w.Φ(xi ) − b ≤ ε + ξi
∗
(37)
w.Φ(xi ) + b − yi ≤ ε + ξi
Energies 2019, 12, 1907 14 of 25

i = 1, 2, . . . . . . . . . . . . n ξi , ξ∗i ≥ 0.

The penalty factor C determines the compromise value between reduction to minimize both
empirical and global errors and the model complexity term kwk2 . SVR reduces the error to zero if the
penalty factor is very high. This option leads to a complex model. However, a low C leads to high
estimation errors.
The regression problem in Equation (24) is then solved using dual multipliers:
n
X
f (x) = (αi − α∗i ).K(xi , x) + b (38)
i=1

where K(xi , x j ) = Φ(xi )T Φ(x j ) is the kernel function that enables a dot product in high-dimensional
feature space using low-dimensional space data input without calculating the function. The radial
base function that is used in the framework is expressed as shown in the following equation:

|xi − x|2
K(xi , x) = exp{− }. (39)
σ2

5.2. Particle Swarm Optimization

The PSO method is a population-based search technique that starts with a population of random
solutions called particles [32]. Every solution within the swarm is called a particle [33]. The swarm
is arbitrarily initialized and updated in each reiteration t to fully adapt the fitness function. The ith
d-dimensional particle is characterized by its location vector Pti = Pti1 , Pti2 , . . . , Ptid and its speed vector
vti = vti1 , vti2 , . . . , vtid . Each particle knows its best personal location Pti,Best and the entire population’s
best global solution GtBest . The position of the population at iteration t is Pt = Pt1 , Pt2 , , . . . , PtN , where N
is the population size. Each particle updates its location according to Equations (40) and (42):

vti +1 = wt vti + c1 r1 (pti,Best − pti ) + c2 r2 (pti,Best − pti ) (40)

pti +1 = pti + vti +1 , i = 1, 2, 3, . . . , N. (41)

The constriction coefficients introduced in [33,34] are used to set c1 and c2 :

2
x= (42)
φ − 1 + φ2 − 4φ
p

c1 = xφ1 , c2 = xφ2 (43)

where φ = φ1 + φ2 > 4. In this paper, φ1 = φ2 = 2.05; thus,c1 = c2 = 1.4962. Location and speed
boundaries are adjusted such that if a particle tries to pass the allowable boundaries, a limiting
procedure brings it back within the permissible limit. If the best personal solution Pi,t
Best
has a higher
fitness than the current best global solution Gt−1
Best
, then GtBest is equal to Pi,t
Best
[35]. The end points are
determined when the best global solution results in a permissible fitness or when a predetermined
number of iterations is achieved.
The wind speed estimation using PSO-SVR method can be described in steps as follows:

• Enter original data for estimation, known as the data preparation step;
• Initialize the particles with random velocities and positions, known as particles optimization;
• Perform an offline training process for SVR with training samples assessing each particle fitness
value of the PSO for the SVR;
• Update each particle velocity and position until the termination condition is satisfied;
• Construct and retrain the SVR estimation model based on the optimal parameters.
 • Initialize the particles with random velocities and positions, known as particles
optimization;
• Perform an offline training process for SVR with training samples assessing each
particle fitness value of the PSO for the SVR;
• 12,
Energies 2019, Update
1907 each particle velocity and position until the termination condition is satisfied;
15 of 25
• Construct and retrain the SVR estimation model based on the optimal parameters.
Training
Trainingdata
dataare
areobtained
obtainedfrom
fromthe
theturbine
turbinepower
powerequation
equationwith
withpre-specified
pre-specifiedrotor
rotorspeed
speedand
and
wind
windvelocity
velocitysamples.
samples.During
Duringon-line
on-lineoperation,
operation,the
theturbine
turbinepower
powerisiscalculated
calculatedas
as

dω
JωJωm m dωmm +
Pt P=t = +PPgg . . (44)
(44)
dt
dt

For For each each sample,sample, thethe rotor rotorspeed

speedandandthe thecorresponding
corresponding turbine
turbine power
power (
ω( iω, ip, pi , i,j
j | i =
= 1 . .50
,....
1, . 50,
, j =
j=1 1, .50
,..... . . 50
) ) are
are combined
combined as a as a for
pair pair
thefor the of
inputs inputs
SVR of SVR as
estimator estimator
shown as
in
shown in Figure 13a. For any operating condition, if the turbine power and the rotor speed are
Figure 13a. For any operating condition, if the turbine power and the rotor speed are known, the
known, the wind speed can be calculated as described in the framework of a PSO-SVR method in
wind speed can be calculated as described in the framework of a PSO-SVR method in Figure 13b
Figure 13b [36].
[36].

Energies 2017, 10, x FOR PEER REVIEW 17 of 26

(a)

(b)
Figure
Figure 13. Wind
Windspeed
speedestimation
estimation process:
process: (a) Training
(a) Training samples,
samples, (b) implementation
(b) implementation process
process diagram.
diagram.

The detailed implementation process of the proposed method is shown in Figure 14.

Online Implementation

Initialization and parameter PSO fitness calculation

 (b)
Figure
Energies 13.1907
2019, 12, Wind speed estimation process: (a) Training samples, (b) implementation process
16 of 25
diagram.

The
The detailed implementation process of the proposed method is shown in Figure 14.

Online Implementation

Initialization and parameter PSO fitness calculation

Training Data
setting of PSO using SVR

Is termination criterion Updating the inertial

reached? weight

Going back to PSO Best SVR parameters

Test data
fitness calculation are set

Figure 14.
Figure Implementation process
14. Implementation process diagram.
diagram.

6. Experimental Results
6. Experimental Results
It is desirable that the wind turbine simulator is implemented in hardware (M-G set) to test the
It is desirable that the wind turbine simulator is implemented in hardware (M-G set) to test the
WECS operation in the laboratory. The hardware setup consists of the DFIG-driven by the SCIM, PWM
WECS operation in the laboratory. The hardware setup consists of the DFIG-driven by the SCIM,
converters, and digital signal processor (DSP) control boards as shown in Figure 15. The specifications
PWM converters, and digital signal processor (DSP) control boards as shown in Figure 15. The
of the wind turbine blade modeled for the simulator, the SCIM motor ratings and parameters, and the
specifications of the wind turbine blade modeled for the simulator, the SCIM motor ratings and
experimental setup are listed in Tables A1 and A2 in the Appendix A. Figure 16 shows the turbine
parameters, and the experimental setup are listed in Tables A1 and A2 in the Appendix. Figure 16
performance for turbulent wind speed, of which the spectrum is similar to the real wind speed pattern.
shows the turbine performance for turbulent wind speed, of which the spectrum is similar to the real
High estimation accuracy and very high PSO-SVR dynamic performance is illustrated in Figure 16a.
wind speed pattern. High estimation accuracy and very high PSO-SVR dynamic performance is
It is noticeable that the estimated wind speed has a slight difference from the real value due to the
illustrated in Figure 16a. It is noticeable that the estimated wind speed has a slight difference from
trade-off
Energies between
2017, 10, x FORthe minimizing
PEER REVIEW error and the model complexity. The generator rotational18speed
of 26
the real value due to the trade-off between the minimizing error and the model complexity. The
follows the wind speed pattern in all wind speed values, as shown in Figure 16b. Figure 16c plots the
generator rotational
generator output speed
power follows
versus the wind
variable wind speed
speed. pattern in all wind speed values, as shown in
Figure 16b. Figure 16c plots the generator output power versus variable wind speed.

Turbine
simulator DFIG
Power
AC-DC-AC
Supply
Converter

DSP
Oscilloscope Board

Layout of
Figure 15. Layout
Figure of the
the experimental
experimental equipment.
equipment.

By adding the effect of wind shear and tower shadow, the total turbine torque oscillates in the same
frequency. However, the inertia of the turbineReal
leads to speed
wind the reduction of the effect of both disturbances
as presented in Figure 17a,b.

7 [m/s]

Estimated wind speed

(a) 2 [m/s]/div
Energies 2019, 12, 1907 Figure 15. Layout of the experimental equipment. 17 of 25

Real wind speed

7 [m/s]

Estimated wind speed

(a) 2 [m/s]/div

800 [rpm]

(b) 100 [rpm]/div

1000 [W]

(c) 500 [W]/div

Figure 16. Experimental results for continuously variable wind speed: (a) Real and estimated wind
Figure
Energies 2017, 16.
speed, (b)xExperimental
10, results
FOR PEERspeed,
generator REVIEWand for continuously
(c) generator variable wind speed: (a) Real and estimated wind
power. 19 of 26
speed, (b) generator speed, and (c) generator power.

By adding the effect of wind shear and tower shadow, the total turbine torque oscillates in the
same frequency. However, the inertia of the turbine leads to the reduction of the effect of both
[10Nm]
disturbances as presented in Figure 17a,b.

(a) 4 [Nm]/div

[1600W]

100 [W]/div
(b) 1 [sec]/div
Figure17.
Figure Towershadow
17.Tower shadowand
andshear
sheareffects:
effects: (a)
(a)Generator
Generatortorque
torque(Nm),
(Nm),(b)
(b)generator
generator power
power (kW).
(kW).

ItIt can
can be
be observed
observedthat
thatperiodic
periodiccomponents
components areare
concentrated on frequency
concentrated multiples
on frequency of theofblade
multiples the
passing frequency, mainly 3p, 6p . . . etc., as it is shown in Figure 18. It should be
blade passing frequency, mainly 3p, 6p … etc., as it is shown in Figure 18. It should be expected expected thatthat
the
amplitude
the amplitude decreases when
decreases theythey
when increase in order.
increase in order.
Equation (16) is then used to generate a stochastic model of the wind speed to model the real
nature of instantaneous wind speed, as shown in Figure 19a. The turbulent wind speed spectrum is
generated by the wind speed simulator. The parameters are chosen as follows: Vo = 9 [m/s], N = 15
in (11), L = 90 [m] and σ = 0.2 in Equation (13). It is obvious that the tower shadow and the wind
shear are quite significant for the variations of the turbine torque and power. The range of harmonic
content in the wind speed profile is from 0.1 Hz to 10 Hz. From Figure 19, it can be observed that the
turbine performance for turbulent wind speed continuously follows the wind speed pattern, as shown
in Figure 19b–e. During this low wind speed, the pitch angle controller keepsivthe pitch angle position
5 [Hz]/d

0 [Hz]

Figure 18. Turbine torque spectrum.

Equation (16) is then used to generate a stochastic model of the wind speed to model the real
 (a) 4 [Nm]/div

Energies 2019, 12, 1907 18 of 25

[1600W]
in the minimum value, which is zero. However, when the average wind speed increases to 12 m/s, the
generator rotational speed and generator output power increase to the maximum limits, which activate
100 [W]/div
the pitch angle controller to increase the pitch angle to(b) the predetermined position based on the wind
1 [sec]/div
speed value. In this case, the generator output power and the generator rotating speed are constant
Figure
and equal to17.
theTower
ratedshadow
values.and shear20
Figure effects: (a) Generator
depicts torquebetween
the transition (Nm), (b)the
generator
maximumpower (kW). point
power
tracking (MPPT) mode and constant power mode. The generator output power is constant when the
speedItgoes
can higher
be observed
than 13that
m/speriodic
in Figurecomponents
20a, as shown arein concentrated
Figure 20c. Theongenerator
frequencyspeed,
multiples
rotor of the
q-axis
blade passing
current, torquefrequency, mainly
and have the same3p, 6p …
trend asetc., as it is20b,
in Figure shown in e,
d and Figure 18. It should be expected that
respectively.
the amplitude decreases when they increase in order.

5 [Hz]/d iv

0 [Hz]
Energies 2017, 10, x FOR PEER REVIEW 20 of 26
Figure
Figure 18. Turbine torque
18. Turbine torque spectrum.
spectrum.

Equation (16) is then used to generate a stochastic model of the wind speed to model the real
nature of instantaneous9 [m/s]
wind speed, as shown in Figure 19a. The turbulent wind speed spectrum is
generated by the wind speed simulator. The parameters are chosen2 as[m/s] /div V = 9 [ m / s ], N = 15
follows: o
(a)
in (11), L = 90 [m] and σ = 0.2 in Equation (13). It is obvious that the tower shadow and the wind
shear are quite significant for the variations of the turbine torque and power. The range of harmonic
1800[rpm]
content in the wind speed profile is from 0.1 Hz to 10 Hz. From Figure 19, it can be observed that the
100 [rpm]/div
turbine performance for turbulent wind speed continuously follows the wind speed pattern, as
(b)
shown in Figure 19b–e. During this low wind speed, the pitch angle controller keeps the pitch angle
position in the minimum value, which is zero. However, when the average wind speed increases to
0[W]
12 m/s, the generator rotational speed and generator output power increase to the maximum limits,
750 [W]/div
which activate the pitch angle controller to increase (c)the pitch angle to the predetermined position
based on the wind speed value. In this case, the generator output power and the generator rotating
speed are constant and equal to the rated values. Figure 20 depicts the transition between the
0[A]
maximum power point tracking (MPPT) mode and constant power mode. The generator output
6 [A]/div
power is constant when the speed goes higher than 13(d) m/s in Figure 20a, as shown in Figure 20c. The
generator speed, rotor q-axis current, torque and have the same trend as in Figure 20b, d and e,
respectively.
0 [Nm]
8 [Nm]/div
(e) 2 [s]/div
Figure 19. Turbulent wind speed: (a) Wind speed, (b) generator speed, (c) generator power, (d) q-axis
Figure 19. Turbulent wind speed: (a) Wind speed, (b) generator speed, (c) generator power, (d) q-axis
current, and (e) and generator torque.
current, and (e) and generator torque.
The wind variation is modeled as a sum of wide range harmonics with frequencies 0.1–10 Hz.
The spectrum of the turbulent wind speed in Figure 21 shows the low frequency components of the
wind speed. The measured
12 [m/s]stator and rotor currents are shown in Figure 22a,b. From the stator current,
it is obvious that the currents are almost sinusoidal. 2 [m/s]/div

(a)

1800 [rpm]
100 [rpm]/div
(b)
 0 [Nm]
8 [Nm]/div
(e) 2 [s]/div

Figure
Energies 2019,19.
12,Turbulent
1907 wind speed: (a) Wind speed, (b) generator speed, (c) generator power, (d) q-axis19 of 25
current, and (e) and generator torque.

12 [m/s]
2 [m/s]/div

(a)

1800 [rpm]
100 [rpm]/div
(b)

0[W]
750 [W]/div
(c)

0[A]
6 [A]/div
(d)

0 [Nm]
Energies 2017, 10, x FOR PEER REVIEW 21 of 26
8 [N/m]/div

The wind variation is modeled as a sum of wide(e)range harmonics with 2 [s]/div

frequencies 0.1–10 Hz.
The spectrum of the turbulent wind speed in Figure 21 shows the low frequency components of the
Figure20.
Figure Turbulenthigh
20.Turbulent highwind
windspeed
speedand
andpitch
pitchangle
anglecontrol:
control:(a)
(a)Wind
Windspeed,
speed,(b)
(b)generator
generatorspeed,
speed,
wind speed. The measured stator and rotor currents are shown in Figure 22a,b. From the stator
(c) generator power, (d) q-axis current, and (e) and generator torque.
(c) generator power, (d) q-axis current, and (e) and generator torque.
current, it is obvious that the currents are almost sinusoidal.

5[Hz]/ div

0 [Hz]

Figure 21. Spectrum of the turbulent wind speed.

The operating modes of a wind turbine are obtained, as in Figure 23. This curve is divided into
four regions, as follows:

- Region AB, when the wind speed is less than the cut-in speed when the rotational speed is less
0 [A]
than the minimum rotational speed for optimum operation.
- Region BC, when the wind speed is higher than the cut-in speed and less than the rated value.
The output power is given by Popt = Kopt υ3 . 5 [A/div]
- Region CD, when the rotational speed approaches (a) to its rated value.
5 [ms/div]
- Region DE, when the wind speed is beyond the limits and the generator output power is controlled
to its rated value. The blade pitch controller is activated in this region.

0 [A]

5 [A/div]
(b) 0.2 [sec/div]

Figure 22. DFIG currents after synchronization: (a) Stator currents, (b) rotor currents.

The operating modes of a wind turbine are obtained, as in Figure 23. This curve is divided into
four regions, as follows:
 5[Hz]/ div

0 [Hz]
Energies 2019, 12, 1907 20 of 25
Figure 21. Spectrum of the turbulent wind speed.

0 [A]

5 [A/div]
(a) 5 [ms/div]

0 [A]

5 [A/div]
(b) 0.2 [sec/div]
Energies 2017, 10, x FOR PEER REVIEW 22 of 26
Figure22.
Figure DFIGcurrents
22.DFIG currentsafter
aftersynchronization:
synchronization:(a)(a)Stator
Statorcurrents,
currents,(b)
(b)rotor
rotorcurrents.
currents.

The operating modes of a wind turbine are obtained, as in Figure 23. This curve is divided into
four regions, as follows: 12 [m/s]

- Region AB, when the wind speed is less than the cut-in speed when the rotational
speed is less than the minimum rotational(a)
speed for3 optimum
[m/s]/div operation.
- Region BC, when the wind speed is higher than the cut-in speed and less than the
rated value. The output power is given by Popt = K optυ .
3

1800 [rpm]
- Region CD, when the rotational speed approaches to its rated value.
- Region DE, when the wind speed is beyond the limits and the generator output power
(b)
is controlled to its rated value. The blade pitch 100 [rpm]/div
controller is activated in this region.

0 [W]

(c) 750 [W]/div

0 [A]

1 [s]/div (d) 6 [A]/div

Figure 23. The simulator performance in power operation modes: (a) Wind speed, (b) generator speed,
Figure 23. The
(c) generator simulator
power, performance
and (d) in power operation modes: (a) Wind speed, (b) generator
q-axis current.
speed, (c) generator power, and (d) q-axis current.
Figure 24a,b show the accuracy of the proposed wind estimation algorithm to estimate the wind
Figure 24a,b showerror
speed. The estimation the in
accuracy of the
continuous proposed
change wind
of wind estimation
speed is in thealgorithm to m/s,
range of 0.2 estimate
whichthe
is
wind speed.a The
considered smallestimation
value witherror in continuous
respect change of wind
to using an anemometer speed is in
or measuring the
the range wind
average of 0.2speed
m/s,
which is considered
in different locations.a small value with respect to using an anemometer or measuring the average
wind speed in different locations.

6 [m/s]

(a) 0.5 [m/s]/div

0 [m/s]
 Figure 23. The simulator performance in power operation modes: (a) Wind speed, (b) generator
speed, (c) generator power, and (d) q-axis current.

Figure 24a,b show the accuracy of the proposed wind estimation algorithm to estimate the
wind speed. The estimation error in continuous change of wind speed is in the range of 0.2 m/s,
Energies 2019, 12, 1907 21 of 25
which is considered a small value with respect to using an anemometer or measuring the average
wind speed in different locations.

6 [m/s]

(a) 0.5 [m/s]/div

0 [m/s]

5 [s]/div (b) 0.125 [m/s]/div

Figure 24. Wind speed estimation accuracy: (a) Real and estimated wind speed and (b) estimation error.
Figure 24. Wind speed estimation accuracy: (a) Real and estimated wind speed and (b) estimation
error.
7. Conclusions
In this paper, a novel sensorless-based modeling for wind energy system is presented. It is
7. Conclusion
developed using a torque-controlled SCIM as the wind turbine simulator. Stochastic wind speed
In thiswind
profiles, paper, a novel
shear, sensorless-based
and torque modeling
shadow effects for wind energy system is presented. It is
are simulated:
developed using a torque-controlled SCIM as the wind turbine simulator. Stochastic wind speed
1. The
profiles, turbine
wind steady-state
shear, and torquecharacteristics
shadow effects areare
modeled using the turbine torque equation, which is a
simulated:
function of the wind speed, tip-speed ratio, and blade pitch angle.
1. The turbine steady-state characteristics are modeled using the turbine torque equation, which is
2. A derivation for the wind shear and tower shadow torque components are obtained as a function
a function of the wind speed, tip-speed ratio, and blade pitch angle.
of the turbine dimensions and the estimated effective wind speed.

The main cause of system oscillations is tower shadow, wind shear shows significant oscillations
on the total output torque, and the spectrum analysis shows the oscillation frequencies, which are from
the third-order components. The electromagnetic transient models of DFIG and control strategies were
presented and analyzed.
All the characteristics are simulated based on the estimated wind speed using the PSO-SVR
method, which gives an accurate simulation of the dynamic characteristics of wind turbine.

Author Contributions: A.G.A.-K. conceived the idea and conducted the experimental part. S.A. performed the
analysis, K.S. and A.A. wrote and review the original manuscript.
Funding: This research received no external funding.
Acknowledgments: The author would like to thank Deanship of Scientific Research at Majmaah University for
supporting this work under Project Number No. 38/120.
Conflicts of Interest: The authors declare no conflict of interest.

Nomenclature
Pt Turbine power
ρ standard air specific density [kg/m3 ]
Cp power conversion coefficient
β pitch angle of turbine blades [degree]
υ wind speed [m/s]
λ tip-speed ratio (TSR)
R blade radius [m]
h corresponding height of point under consideration
α empirical parameter, which increases with the terrain roughness
z the elevation above ground
H hub height
Vh wind speed at hub height
Energies 2019, 12, 1907 22 of 25

ψb blade rotational angle

r radial distance from the rotor axis
F(r, ψb ) the wind shear disturbance
a the tower radius
x the distance from the blade origin to the tower midline
y the lateral distance of the blade from the tower midline
Vts the tower shadow
Tt turbine torque
Tts disturbance in torque due to tower shadow
Tws disturbance in torque due to wind shear
JT moment of inertia of the wind turbine
JG moment of inertia of the generator
k stiffness of coupling shaft
D coupling shaft damping
n transmission ratio of the gearbox
Pg the generator power
ωm the turbine speed
J the system moment of inertia
υ(t) instantaneous wind speed
Vo effective average value of wind speed
N harmonic samples order
ωi harmonic frequency
Ai harmonic amplitude
σ turbulence intensity
Lu turbulence length
vds , vqs the d-q voltages in the stator
ids , iqs the d-q currents in the stator
ωe the frequency in the rotor
λds , λqs the d-q flux linkages in the stator side
Lm equivalent magnetizing inductance
Ls equivalent stator self-inductance
Lr equivalent rotor self-inductance
λds , λqs equivalent d-q flux linkage in the stator
λdr , λqr equivalent rotor d -q flux linkage
ims , ids , idr magnetizing d-axis currents in the stator and rotor
Kt the torque constant
Ps , Qs Stator active and reactive power
vdr , vqr the d-q voltages in the rotor
Rr rotor equivalent resistance
ωsl slip angular frequency
Rs the stator resistance
θe the synchronous frame angle
w the SVR weight vector
b the bias
ε the allowable error
C the penalty factor
αi , α∗i Lagrange multipliers.
c1 and wt is the velocity weight at iteration t.
c2 social learning factor
r1 , r2 cognition factor are random numbers in the range [0, 1]
Energies 2019, 12, 1907 23 of 25

Appendix A
The specification of the induction machine used for the test is three-phase, four poles, 230 V, 50 Hz, 3 kW.

Table A1. Parameters of turbine blade model.

Parameters Value
Blade radius 0.95 m
Max. power conv. coeff. 0.45
Optimal tip-speed ratio 7
Cut-in speed 4 m/s
Rated wind speed 13 m/s

Table A2. Parameters of 3kW SCIG.

Parameters Value
Stator resistance 0.93 Ω
Rotor resistance 0.533 Ω
Iron loss resistance 190 Ω
Stator leakage inductance 0.003 H
Rotor leakage inductance 0.003 H
Mutual inductance 0.076 H

References
1. Gan, L.K.; Shek, J.K.H.; Mueller, M.A. Modeling and characterization of downwind tower shadow effects
using a wind turbine emulator. IEEE Trans. Ind. Electron. 2017, 64, 7087–7097. [CrossRef]
2. de Kooning, J.D.M.; Vandoorn, T.L.; van de Vyver, J.; Meersman, B.; Vandevelde, L. Shaft speed ripples in
wind turbines caused by tower shadow and wind shear. IET Renew. Power Gener. 2014, 8, 195–202. [CrossRef]
3. Weihao, H.; Chi, S.; Zhe, C. Impact of wind shear and tower shadow effects on power system with large-scale
wind power penetration. In Proceedings of the IECON 2011 37th Annual Conference of the IEEE Industrial
Electronics Society, Melbourne, Australia, 7–10 November 2011; pp. 878–883.
4. Ahshan, R.; Iqbal, M.T.; Mann, G.K.I. Controller for a small induction generator-based wind turbine.
Appl. Engery 2008, 85, 218–227. [CrossRef]
5. Martinez, F.; Herrero, L.C.; Pablo, S.D. Open-loop wind turbine emulator. Renew. Energy 2014, 63, 212–221.
[CrossRef]
6. Kojabadi, H.M.; Chang, L.; Boutot, T. Development of a novel wind turbine simulator for wind energy
conversion systems using an inverter-controlled induction motor. IEEE Trans. Energy Convers. 2004, 19,
547–552. [CrossRef]
7. Mihet-Popa, L.; Blaabjerg, F.; Boldea, I. Wind turbine generator modeling and simulation where rotational
speed is the controlled variable. IEEE Trans. Ind. Appl. 2004, 40, 3–10. [CrossRef]
8. Monfared, M.; Kojabadi, H.M.; Rastegar, H. Static and dynamic wind turbine simulator using a converter
controlled DC motor. Renew. Energy 2008, 33, 906–913. [CrossRef]
9. Sajadi, A.; Rosłaniec, Ł.; Kłos, M. An emulator for fixed pitch wind turbine studies. Renew. Energy 2016, 87,
391–402. [CrossRef]
10. Hughes, F.M.; Anaya-Lara, O.; Ramtharan, G. Influence of tower shadow and wind turbulence on the
performance of power system stabilizers for DFIG-based wind farms. IEEE Trans. Energy Convers. 2008, 23,
519–528. [CrossRef]
11. Wan, S.; Cheng, L.; Sheng, X. Effects of yaw error on wind turbine running characteristics based on the
equivalent wind speed model. Energies 2015, 8, 6286–6301. [CrossRef]
12. Tan, J.; Hu, W.; Wang, X. Effect of tower shadow and wind shear in a wind farm on AC tie-line power
oscillations of interconnected power systems. Energies 2013, 6, 6352–6372. [CrossRef]
13. Shamshirband, S.; Petkovic, D.; Anuar, N.B.; Kiah, M.L.M.; Akib, S.; Gani, A.; Cojbasic, Z.; Nikolic, V.
Sensorless estimation of wind speed by adaptive neuro-fuzzy methodology. Int. J. Electr. Power Energy Syst.
2014, 62, 490–495. [CrossRef]
Energies 2019, 12, 1907 24 of 25

14. Mohandes, M.; Rehman, S.; Rahman, S. Estimation of wind speed profile using adaptive neuro-fuzzy
inference system (ANS). Appl. Energy 2011, 88, 4024–4032. [CrossRef]
15. Li, H.; Shi, K.; McLaren, P. Neural-network-based sensorless maximum wind energy capture with
compensated power coefficient. IEEE Trans. Ind. Appl. 2015, 41, 1548–1556. [CrossRef]
16. Abo-Khalil, A.G.; Abo-Zied, H. Sensorless control for d-fig wind turbine based on support vector regression.
In Proceedings of the IECON 2012 38th Annual Conference on IEEE Industrial Electronics Society, Montreal,
QC, Canada, 25–28 October 2012; pp. 3475–3480.
17. Abo-Khalil, A.G.; Lee, D.C. MPPT control of wind generation systems based on estimated wind speed using
SVR. IEEE Trans. Ind. Electron. 2008, 55, 1489–1490. [CrossRef]
18. Abo-Khalil, A.G. Model-based optimal efficiency control of induction generators for wind power systems.
In Proceedings of the 2011 IEEE International Conference on Industrial Technology (ICIT), Auburn, AL, USA,
14–16 March 2011; pp. 191–197.
19. Abo-Khalil, A.G. Impacts of wind farms on power system stability. In Modeling and Control Aspects of Wind
Power Systems; IntechOpen: London, UK, 2013; ISBN 980-953-307-562-9.
20. Abo-Khalil, A.G. Synchronization of DFIG output voltage to utility grid in wind power system. Renew. Energy
2012, 44, 193–198. [CrossRef]
21. Dolan, D.; Lehn, P. Simulation model of wind turbine 3p torque oscillations due to wind shear and tower
shadow. IEEE Trans. Energy Convers. 2006, 21, 717–724. [CrossRef]
22. Manwell, J.F.; McGowan, J.G.; Rogers, A.L. Wind Energy Explained: Theory, Design and Application;
Wiley: Hoboken, NJ, USA, 2010.
23. Tummala, A.; Velamati, R.K.; Sinha, D.K.; Indraja, V.; Krishna, V.H. A review on small scale wind turbines,
Renew. Sustain. Energy Rev. 2016, 56, 1351–1371. [CrossRef]
24. Lei, Y.; Mullane, A.; Lightbody, G.; Yacamini, R. Modeling of the wind turbine with a doubly fed induction
generator for grid integration studies. IEEE Trans. Energy Convers. 2006, 21, 257–264. [CrossRef]
25. Gregory, I.M.; Chowdhry, R.S.; McMinn, J.D.; Shaughnessy, J.D. Hypersonic Vehicle Model and Control Law
Development Using H∞ and µ Synthesis; Langley Research Center: Hampton, VA, USA, 1994.
26. Abo-Khalil, A.G.; Kim, H.G.; Lee, D.C.; Seok, J.K. Maximum output power control of wind generation system
considering loss minimization of machines. In Proceedings of the 30th Annual Conference of IEEE Industrial
Electronics Society, Busan, Korea, 2–6 November 2004; pp. 1676–1681.
27. Abo-Khalil, A.G.; Lee, D.C. Grid connection of doubly-fed induction generators in wind Energy conversion
system. In Proceedings of the 2006 CES/IEEE 5th International Power Electronics and Motion Control
Conference, Shanghai, China, 14–16 August 2006; Volume 2, pp. 1–5.
28. Park, H.G.; Abo-Khalil, A.G.; Lee, D.C.; Son, K.M. Torque ripple elimination for doubly fed induction
motors under unbalanced source voltage. In Proceedings of the 2007 7th International Conference on Power
Electronics and Drive Systems, Bangkok, Thailand, 27–30 November 2007; Volume 7, pp. 1301–1306.
29. Abo-Khalil, A.G.; Park, H.G.; Lee, D.C. Loss minimization control for doubly fed induction generators in
variable speed wind turbines. In Proceedings of the IECON 2007 33rd Annual Conference of the IEEE
Industrial Electronics Society, Taipei, Taiwan, 5–8 November 2007; pp. 1109–1114.
30. Smola, A.J.; Schölkopf, B. A tutorial on support vector regression. Stat. Comput. 2004, 14, 199–222. [CrossRef]
31. Nuller, K.R.; Smola, A.J.; Ratsch, G.; Scholkopf, B.; Kohlmorgen, J.; Vapnik, V. Predicting time series with
support vector machine. In International Conference on Artificial Neural Networks; Springer: Berlin/Heidelberg,
Germany, 1997; pp. 999–1004.
32. Senthil, M.A.; Rao, M.V.C.; Chandramohan, A. Competitive approaches to PSO algorithms via
new-acceleration coefficient variant with mutation operators. In Proceedings of the Sixth International
Conference on Computational Intelligence and Multimedia Applications (ICCIMA’05), Las Vegas, NV, USA,
16–18 August 2005.
33. Vukosavic, S.N.; Levi, E. Robust DSP-based efficiency optimization of a variable speed induction motor
drive. IEEE Trans. Ind. Electron. 2003, 50, 560–570. [CrossRef]
34. Lin, S.W.; Ying, K.C.; Chen, S.C.; Lee, Z.J. Particle swarm optimization for parameter determination and
feature selection of support vector machines. Expert Syst. Appl. 2008, 35, 1817–1824. [CrossRef]
Energies 2019, 12, 1907 25 of 25

35. Cho, K.R.; Seok, J.K.; Lee, D.C. Mechanical parameter identification of servo systems using robust support
vector regression. In Proceedings of the 2004 IEEE 35th Annual Power Electronics Specialists Conference
(IEEE Cat. No.04CH37551), Aachen, Germany, 20–25 June 2004; Volume 5, pp. 3425–3430.
36. Liu, H.-H.; Chang, L.-C.; Li, C.-W.; Yang, C.-H. Particle swarm optimization-based support vector regression
for tourist arrivals forecasting. Comput. Intell. Neurosci. 2018, 2018, 1–13. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).