energies

Article
Adaptive Sliding Mode Control for PMSG Wind
Turbine Systems
Sung-Won Lee and Kwan-Ho Chun *
Department of Electrical Engineering, Chungnam National University, Daejeon 34134, South Korea;
[email protected]
* Correspondence: [email protected]




Received: 30 November 2018; Accepted: 5 February 2019; Published: 13 February 2019


Abstract: In this paper, variable speed PMSG wind turbine systems with unknown system parameters,
such as vicious friction coefficient and total inertia, are considered. The errors and variations of
wind speed are modeled as a disturbance in mechanical torque. In general, the optimum rotating
speed is given based on the MPPT (Maximum Power Point Tracking) algorithm and the designed
controller tracks the reference (optimum) rotating speed in spite of these parametric uncertainties and
disturbances. In order to have a desired rotor speed, a sliding mode current controller is proposed
to have robustly stabilizing torque input. From the robustly stabilizing q-axis current iq , q-axis
voltage input uq is obtained. Additionally, the d-axis control input ud is designed to regulate the
d-axis current id . The adaptive estimator, for the total inertia J and the viscous friction coefficient
F, is designed by a backstepping control technique. The robust stability of the closed-loop system
is shown using a Lyapunov function. The proposed controller is verified via a simulation using
MATLAB/Simulink.

Keywords: adaptive sliding mode control; backstepping control; permanent magnet synchronous
generator (PMSG) wind turbine

1. Introduction
Wind turbines in wind power generation systems can be classified as inductive or synchronous
depending on the rotor of the generator [1]. Initially, the wind power generation industry used
doubly-fed induction generators, for which the operating speed of the generator can be controlled
using a slip. However, permanent magnet synchronous generators are increasingly being adopted for
scenarios requiring high generation capacity, easy maintenance management, and cost efficiency, such
as offshore wind power generation complexes. Permanent magnet synchronous generators (PMSG)
have the advantages of reduced production costs due to developments in processing technology,
reduced operating noise due to the use of permanent magnets in the rotors, reduced generator weight,
and reduced maintenance costs. They also have the advantage of being able to create high torque output
compared to doubly-fed induction generators due to high magnetic flux density and efficiency [2–7].
Wind turbines can be designed to operate at fixed speed or at variable speed [2,8]. Fixed-speed
operating systems transform wind energy into electrical energy using induction generators directly
connected to a three-phase power grid. The rotor of the wind turbine is coupled to the generator
shaft with a fixed ratio gearbox and operates at constant speed. By contrast, variable speed operating
systems are systems which vary the rotation speed of the generator’s rotors, and they are mainly
used in synchronous generators. Variable operating wind power generators do not connect directly
with the grid, instead connecting with it through a converter, and the generator is operated through
aerodynamic control to adjust the torque, speed, and power. This reduces the mechanical stress and

Energies 2019, 12, 595; doi:10.3390/en12040595 www.mdpi.com/journal/energies

Energies 2019, 12, 595 2 of 17

aerodynamic noise, and the generator can be controlled such that the wind power turbine operates at
the maximum output coefficient over a wider range of wind speeds [2,9].
Here, the control method for energy production at maximum efficiency is called maximum power
point tracking (MPPT) [10]. An MPPT controller calculates the optimal rotor rotation speed for varying
wind speeds. Therefore, the location and accuracy of the sensors which measure wind speed can be
considered as important factors. To extract maximum power, linear controllers are designed based
on an approximated linear model, such as conventional vector control with proportional-integral
loops [3,9]. However, these control strategies may not provide satisfactory performances due to the
system nonlinearity of the PMSG and wide-range operation points. To improve the performance,
a feedback linearizing control based MPPT is proposed [11], where the mechanical rotation speed
controller and current controllers are designed via linear control methods. However, this results
in a complex control law and has weak robustness against parameter uncertainties and external
disturbances. Generally, wind power turbine rotors have very large diameters, and wind speeds
vary according to the location on the turbine from which wind speeds are measured, so wind speed
measurements may not be a suitable value for maximum energy production, depending on the
measurement location. Therefore, wind speed measurement is more suitable for systems with severe
noise due to the environmental uncertainty caused by measurement location and by noise inherent in
their sensors. Because wind power turbine systems have significant uncertainty in their mechanical
elements, control techniques are required to overcome this problem.
Sliding mode control (SMC) has been frequently used as a robust control for disturbance and
uncertainty among the various robust control methods. Some papers on the application of SMC to
wind energy conversion system have been presented in recent decades [11–22]. In the early studies
of [11,12], SMC is applied for MPPT in the wind energy conversion system with uncertainties. In [12],
an optimal torque SMC strategy for a variable speed wind turbine system is proposed to implement
MPPT tasks. In other words, SMC strategy was applied for controlling electromagnetic torque in
MPPT for PMSG. In reference [11], an input-output linearization technique and SMC are applied to a
wind energy conversion system with unstructured uncertainties. In [13], an SMC based on Enhanced
Exponential Reaching Law was proposed and investigated on a grid-connected PMSG wind turbine
system. In [14,15], SMC was used to improve robustness under various operating conditions such
as parameter changes or load variation, and to reduce errors between desired command values and
actual current for d- and q-axis currents. In [16], an induction generator (IG) speed drive was designed
with the application of a sliding mode controller and a proposed artificial neural network controller.
In [17], optimum torque and rotor speed were studied to be obtained by defining the appropriate
sliding surface. In [18], an adaptive SMC was studied, first by designing an SMC for reducing the rotor
speed error in PMSG, followed by a design for controller gain and mechanical torque estimation.
For uncertain nonlinear systems with known bounds, classical sliding mode control provides
robustness. Recently, the adaptive SMC designs adapt the switching gain online without any
predefined knowledge of the bound of uncertainty [23,24]. To avoid over-estimation of gain,
super twisting controllers [25,26] and Time-Delayed controllers [27–30] are proposed. [31] tracked the
control of a class of uncertain nonlinear systems where the upper bound of the system uncertainty has
explicit dependency on the system states.
In this paper, an adaptive robust sliding mode controller is proposed for a wind turbine system
with parametric uncertainties and external disturbance. In the real system, some parameters, such as,
mechanical inertia, viscous friction, stator resistance, inductance, are affected by operating conditions
and manufacturing tolerance managed within predefined bounds. To improve the robustness and
remove the effect of the noise and uncertainty which exist in wind speed measurements, an adaptive
sliding mode controller is proposed. The wind speed sensor error is translated to an external torque
disturbance with amplitude bound. To reject the torque disturbance, an SMC is designed for the
nominal plant using the q-axis current as a virtual input. Next, the real physical control inputs, q-axis
voltage and d-axis voltage, are designed to track the designed q-axis current and to regulate the
Energies 2019, 12, 595 3 of 17

d-axisEnergies
current,2019,respectively.
12, x FOR PEER An adaptation law is designed based on back-stepping technique3while
REVIEW of 18
accounting for unknown parameters of the wind turbine system. The proposed controller is a dynamic
accounting
controller whichfor unknown parameters
is composed of an SMC of the wind
controller turbine
and system.
parameter The proposed
adaptation. controllerspeed
The rotation is a
dynamic controller which is composed of an SMC controller and parameter adaptation. The
error of rotor exponentially converges to the reference speed in spite of the external disturbance and rotation
speedparameters.
unknown error of rotor exponentially converges to the reference speed in spite of the external disturbance
and unknown parameters.
The remainder of this paper is organized as follows. Chapter 2 briefly describes the wind energy
The remainder of this paper is organized as follows. Chapter 2 briefly describes the wind energy
conversion system in wind power generation systems. Chapter 3 defines the problem of noise and
conversion system in wind power generation systems. Chapter 3 defines the problem of noise and
uncertainty in controller design, and describes the design of a controller to resolve this problem.
uncertainty in controller design, and describes the design of a controller to resolve this problem.
Chapter 4 verifies the controller through simulations using MATLAB/Simulink, and Chapter 5
Chapter 4 verifies the controller through simulations using MATLAB/Simulink, and Chapter 5
concludes the paper.
concludes the paper.

2. Modeling of a Wind Energy Conversion System

2. Modeling of a Wind Energy Conversion System
The wind power
The wind generation
power system
generation model
system is shown
model inin
is shown Figure
Figure1.1.ItItisisaasystem
system that convertswind
that converts wind
energy into mechanical energy, and creates electrical energy through a generator. This system
energy into mechanical energy, and creates electrical energy through a generator. This system can be can be
broadly
broadly divided
divided into into
two two parts:
parts: the the generator
generator side
side and
and the
the electricalgrid
electrical gridside.
side. This
This study
study focuses
focusesonon
controlling the generator side.
controlling the generator side.

Figure 1. Wind Energy Conversion System.

Figure 1. Wind Energy Conversion System.
2.1. Wind Turbine Modeling
2.1. Wind Turbine Modeling
The kinetic energy of wind and the power of the wind can be expressed by Equations (1) and (2),
The kinetic energy of wind and the power of the wind can be expressed by Equations (1) and (2),
1 2 1 1
E= mv 1= 2 (ρAvt )v2 =2 ρAv 3
t, (1)
2 E  mv2  (  Avt )v 2 1  Av 3t ,
1
(1)
2 2 2
E 1 1 1
Pw = = ρAv3 = ρ(πR2b )v3 = ρπR2b v3 , (2)
t 2E 1 32 1 3 21
Pw    Av   ( Rb )v   Rb2 v3 ,
2
(2)
where E is the wind’s kinetic energy, ρ is tthe 2air density, 2 A is the area 2 that the wind passes through, v is
where Eofisthe
the velocity wind’st is
thewind, kinetic energy,
the time, Pw is is thepower
the air density, is thewhich
of the Awind area that
canthe wind passes
potentially through,
be used, and
Rb is the radius
v is the of the
velocity ofwind power
the wind, t is turbine
the time,through which of
Pw is the power thethe
wind
windpasses.
which The power of the
can potentially wind
be used,
expressed in Equation (2) only shows the maximum potential power that the wind
and Rb is the radius of the wind power turbine through which the wind passes. The power of the wind has. In actuality,
only part of this potential power can be used as electricity via the wind power turbine. The ratio
expressed in Equation (2) only shows the maximum potential power that the wind has. In actuality,
of the wind’s power and the mechanical power which can be generated by the turbine is the power
only part of this potential power can be used as electricity via the wind power turbine. The ratio of
coefficient C p . The
the wind’s powermaximum
and thevalue of the power
mechanical outputwhich coefficient
can beis generated
59.26%, which
by theisturbine
referredis to
theaspower
Betz’s
limit;coefficient
however, Cit actually exists in the range of approximately 25 to 45%, which can be expressed as
p . The maximum value of the output coefficient is 59.26%, which is referred to as Betz's
follows [32,33]:
limit; however, it actually exists in the range of approximately Pm 25 to 45%, which can be expressed as
follows [32,33]: Cp = , (3)
Pw
Pm c
, c )e
C p (λ, β) = c1 (c2 λ1C p−c3Pβ −
− λ5
i , (3)
i
4
w (4)
1 1 0.035
λi = λ+0.08β − β3 +1
.
 Energies 2019, 12, x FOR PEER REVIEW 4 of 18

c5
1 
i
C p ( ,  )  c1 (c2  c3   c4 )e ,
Energies 2019, 12, 595 i 4 of 17
(4)
1 1 0.035
  .
i   0.08  3  1
where Pm is the output mechanical power, λ is the tip speed ratio, β is the blade pitch angle, c1 = 0.5,
where
c2 = 116, Pcm3 is=the output
0.4, c4 =mechanical
5 and c5 = power,  is the
21. The windtip power
speed ratio,  is mechanical
turbine the blade pitch angle,that
energy 0.5 , be
c1  can
, c3  the
c2  116from
extracted c4  5 ,and
0.4 wind 21 . The wind
c5 expressed
can be power turbine
via Equations mechanical
(2) and energybelow:
(3) as shown that can be extracted
from the wind can be expressed via Equations (2) and (3) as shown below:
1 2 3
Pm = 1ρπRb C p ( λ, β ) v , (5)
2
Pm   Rb C p ( ,  )v3 ,
2
(5)
2
Output power according to the rotor rotation speed for each wind speed is shown in Figure 2.
Output power according to the rotor rotation speed for each wind speed is shown in Figure 2.

Figure2.2.Maximum
Figure MaximumPower
Power Point Tracking with
Point Tracking withvariable
variablewind
windspeeds.
speeds.

It can
It can bebeseen seenthat
that the maximumoutput
the maximum output energy
energy in Figure
in Figure 2 varies
2 varies according
according to wind tospeed.
wind The
speed.
Thegenerator’s
generator’s rotation speed at the maximum output energy is referred to as
rotation speed at the maximum output energy is referred to as the optimal rotor speed the optimal rotor speed
( opt
(ωopt =ωreref f ),), and
and the
the technique
techniqueforformaintaining
maintainingthis thisspeed
speedatateach eachwind
windspeed,
speed,obtaining
obtaining maximum
maximum
output
output energy, is MPPT (Maximum Power Point Tracking) [10,34]. According to the variation of of
energy, is MPPT (Maximum Power Point Tracking) [10,34]. According to the variation , β,
typical characteristics of aerodynamic power coefficient correspond to tip speed ratio are illustrated
typical characteristics of aerodynamic power coefficient correspond to tip speed ratio are illustrated
in Figure
in Figure 3. 3. When Whenβ isismaintained
maintainedasas aaconstant,
constant, we we can
can see see the
the power
powercoefficient
coefficientC pChas
p has only
only oneone
maximum value C pmax that corresponds to the optimal value of C p [34]. Therefore, C pmax can be
maximum value C p max that corresponds to the optimal value of C p [34]. Therefore, C p max can be
expressed as
expressed as
C p−opt (λopt , β) = C pmax ,
C p  opt (opt ,  )  C p maxω, mopt Rb (6)
λopt = v ,
 R
opt (6)
opt opt  m b ,
where λopt and ωm are the optimal values of tip speed ratio v and rotor speed, respectively. By choosing
the optimal values ofopt tip speed ratio for the maximum power coefficient, maximum power can be
where opt and m are the optimal values of tip speed ratio and rotor speed, respectively. By
extracted from the optimal rotor speed.
choosing the optimal values of tip speed ratio for the maximum power coefficient, maximum power
The wind power turbine’s mechanical energy can be expressed as a product of the torque and
can be extracted from the optimal rotor speed.
rotation speed:
Pm = Tm ωm , (7)

where Tm is the wind power turbine’s mechanical torque, and ωm is the wind power turbine rotor’s
rotation speed. From Equations (5) and (7), the turbine’s torque is

ρπR2 C p (λ, β)v3

Tm = . (8)
2ωm

The wind power generator’s basic dynamic equation is

dωm
Tm = Te + Fωm + J , (9)
dt
Energies 2019, 12, 595 5 of 17

where F is2019,
Energies the12,
viscous friction
x FOR PEER coefficient, J is the total inertia, and Te is the electromagnetic torque.
REVIEW 5 of 18

Figure
Figure 3. 3. CharacteristicsofofPower
Characteristics Powercoefficient
coefficient versus
versus Tip
Tip speed
speedratio
ratiofor
fordifferent
differentblade pitch
blade angles.
pitch angles.

2.2. PMSG Modeling

The wind power turbine’s mechanical energy can be expressed as a product of the torque and
rotation speed:
Normally, the currents and the voltages in 3-phase systems are represented in dq frame using dq
transformation. In dq transformation, the d axis Tm
Pm  is set,as the generator rotor’s N pole, and the q axis
(7) is
m
perpendicular to the electrical angle, and according to the stator’s 3-phase current and the electrical
where inTmaissynchronously
elements the wind power turbine’s
rotating mechanical
coordinate torque,
axis, and m is the
are transformed wind power
as shown turbine
in Figure rotor’s
4. The PMSG
rotation speed. From Equations (5) and (7), the turbine’s torque
kinetic equation, having been dq transformed, is shown below [1,4,7,9]: is

 R 2C ( ,  )v3
didTm  R p . 1 (8)
= − id 2+mPωm iq + ud , (10)
dt L L
The wind power generator’s di
basic dynamic equation is
q R 1 1
= − iq − Pωm id + Pψm ωm + uq , (11)
dt L dLm L
Tm  Te  F m  J , (9)
where id is the d-axis current flowing to the stator,iq is dt the q-axis current flowing to the stator, ud
is the voltage for the stator’s d axis, u
where F is the viscous friction coefficient, q J is the totalvoltage
input is the input inertia, for
andthe
Te stator’s q axis, ωm is the
is the electromagnetic
torque. rotor speed, R is the resistance, L is the inductance, P is the number of pole pairs, and ψm
generator’s
Energies
is the 2019, 12,
magnetic x FOR
flux of PEER REVIEW
the PMSG. 6 of 18

2.2. PMSG Modeling

Normally, the currents and the voltages in 3-phase systems are represented in dq frame using dq
transformation. In dq transformation, the d axis is set as the generator rotor’s N pole, and the q axis is
perpendicular to the electrical angle, and according to the stator’s 3-phase current and the electrical
elements in a synchronously rotating coordinate axis, are transformed as shown in Figure 4. The
PMSG kinetic equation, having been dq transformed, is shown below [1,4,7,9]:
did R 1
  id  Pm iq  ud , (10)
dt L L

diq R 1 1
 iq  Pmid  P mm  uq , (11)
dt L L L

where id is the d-axis current flowing to the stator, iq is the q-axis current flowing to the stator, ud
is the input Figure
voltage4. for the stator’s
Schematic d axis,
diagram is the input
of theu Permanent voltage
Magnet for the stator’s
Synchronous q axis, m is the
Generator.
Figure 4. Schematic diagram of theq Permanent Magnet Synchronous Generator.
generator’s rotor speed, R is the resistance, L is the inductance, P is the number of pole pairs,
Then, the electromagnetic torque is
and Then,
m is the
the magnetic flux of torque
electromagnetic the PMSG.
is
3
TTe e= 3 P((ψmmiqiq+
( Ld( LdL−
q )iq id ) , d
L q ) i q i ), (12)
(12)
2

where Te is the electromagnetic torque of the generator. If the inductance of each axis has the same
value ( L  Ld  Lq ), then Equation (12) can be simplified as

Te  1.5 P m iq . (13)

Substitute Equation (13) into Equation (9), then the dynamics of the wind turbine are represented as
Energies 2019, 12, 595 6 of 17

where Te is the electromagnetic torque of the generator. If the inductance of each axis has the same
value (L = Ld = Lq ), then Equation (12) can be simplified as

Te = 1.5Pψm iq . (13)

Substitute Equation (13) into Equation (9), then the dynamics of the wind turbine are represented as

dωm 1

= Tm − 1.5Pψm iq − Fωm . (14)
dt J

From Equations (10), (11), and (14), the whole model is

dωm
dt = − J 1.5Pψm iq − J ωm + J ( Tm + ∆Tm ),
1 F 1
diq R 1 1
dt = − L iq − Pωm id + L Pψm ωm + L uq ,
(15)
did R 1
dt = − L id + Pωm iq + L ud .

In this model, the viscous friction coefficient F and the total inertia J are known a priori. However,
in practical cases, the true values are not the same as the designed ones, because of production variation
and changes according to time. Furthermore, the mechanical torque Tm includes disturbances such as
wind speed changes. In the steady state, the mechanical torque Tm is obtained using wind speed v
and rotating speed ωm from Equation (8). Thus, ∆Tm represents the time varying signal with bounded
amplitude, which is induced by the measurement error of v and the rotation speed error. For the
stable and robust operation, we should consider the unknown parameters F and J in the controller
design stage.

3. Controller Design
For the nominal turbine’s mechanical torque Tm , we can obtain from the measured wind speed
v and rotor rotation speed ωm given by MPPT algorithm. However, the measured wind speed has
some errors due to sensor noise and may not represent the effective wind speed according to the
measurement sensor’s location. When the torque input error is ∆Tm , the actual mechanical torque
input is expressed as the sum of the known nominal value Tm and the disturbance ∆Tm . Under these
∗ to generate the maximum power.
conditions, the controller should track the reference rotation speed ωm
To have a robustly stabilizing controller, we find a reference current input to stabilize the speed
error based on the sliding mode control technique. The obtained control is the reference signal of
q-axis current iq to control the rotation speed. Since the q-axis current iq is not the control input,
the q-axis voltage uq is designed to make iq track the reference signal. The other control input uq ,
the d-axis voltage, regulates the d-axis current id . For the unknown parameters, a parameter estimator
is designed based on backstepping.

3.1. Sliding Mode Controller Design for the Rotation Speed Error Regulation
A robust torque controller is designed based on the sliding mode control technique to overcome
torque disturbances. In general, sliding mode controllers achieve excellent tracking performances
and robustness against modeling uncertainty and disturbances. If we define the sliding surface S as
a rotating speed error and the system satisfies stable sliding mode condition, then the rotating error
slides on the sliding surface, i.e., rotating speed error remains 0. To make a system satisfy the condition
. .
of achieving a stable sliding mode, S · S < 0, where S is the sliding mode plane, and S is the time
derivative of S [35–38].
If the uncertainties in the mechanical torque are taken into account, the wind turbine PMSG
system is represented as

dωm 1 F 1
= − 1.5Pψm iq − ωm + ( Tm + ∆Tm ). (16)
dt J J J
Energies 2019, 12, 595 7 of 17

Define the rotor’s rotation speed error (speed tracking error)

∗
z1 = ω m − ω m , (17)

where ωm ∗ is the reference rotation speed which is obtained by MPPT algorithm based on the measured

wind speed, which may include sensor noises and wind speed variation. To have a torque controller,
take the time derivative of Equation (17)
. . . ∗
z1 = ωm − ωm ,
. ∗
= − 1J 1.5Pψm iq − FJ ωm + 1J ( Tm + ∆Tm ) − ω m , (18)
. ∗
= 1J u(t) − FJ ωm + 1J ( Tm + ∆Tm ) − ω m .

For the simple notations, the quantity −1.5Pψm iq is defined as uτ (t), which can be interpreted as
torque control input to track the reference rotation speed ωm∗ . To have a robustly stabilizing sliding

mode controller, we define the sliding surface as the speed tracking error

∗
S ( t ) = z1 = ω − ω m . (19)

Assumption 1. For the mechanical torque disturbance ∆Tm (t), there exists a positive real number δ that
satisfies Equation (20).
|∆Tm (t)| ≤ δ. (20)

The Assumption 1 implies that the influence on the torque by wind speed variation, sensor noises,
and model uncertainties is limited by a bounded magnitude. Given the practical environment, this
assumption is reasonable and acceptable. Now, we propose a torque control input u∗τ (t) for a given δ,
. ∗
u∗τ (t) = Fωm + J ω m − Tm − γsgn(z1 ) − c1 Jz1 , (21)

where c1 > 0, η > 0, and γ = δ + √η .

2

Proposition 1. Suppose that a system (18) satisfies assumption 1 for a positive real number δ. Then, the error
system (18) is robustly exponentially stable by the feedback input u∗τ (t) given as (21).

Proof of Proposition 1. Define a Lyapunov function candidate as

1 2
V1 = S . (22)
2
Take the time derivative and by Assumption 1, we have
. . .
V1 = S ·S = z1 · z1 ,
. ∗

= z1 1J u∗τ (t) − FJ ωm + 1J ( Tm + ∆Tm ) − ω m ,
. ∗ . ∗
   
= z1 1J Fωm + J ω m − Tm − γsgn(z1 ) − c1 Jz1 − FJ ωm + 1J ( Tm + ∆Tm ) − ω m , (23)
 
= z1 −c1 z1 + 1J (∆Tm − γsgn(z1 )) ,
≤ −c1 z21 − √η | z1 |.
2

.
Since V ≤ 0, the rotation speed error z1 converges to 0 exponentially as time goes to infinity. 
. ∗
 
As in the conventional SMC, if we choose u∗τ (t) = Fωm + J ω m − Tm − δ + √η sgn(z1 ) − c1 Jz1
2
. 1
for some positive constant α, then we can show that V 1 ≤ −ηV 12 and the control law u∗ (t) drives the
speed error z1 to zero in finite time tr ≤ α2 |z1 (0)|.
Energies 2019, 12, 595 8 of 17

Remark 1. From Equation (21), we have the reference q-axis current iq∗ to control the torque of PMSG as in the
following Equation.
1  . ∗

iq∗ = − Fωm + J ω m − Tm − γsgn(z1 ) − c1 Jz1 . (24)
1.5Pψm

3.2. Voltage Controller Design and Adaptative Estimation for Unknown Parameters
In practical cases, many parameters are different from the designed values or changes over
time. In this subsection, we assume that the viscous friction coefficient F and the total inertia J in
Equation (16) have unknown values due to manufacturing tolerance. If the coefficients F and J are
substituted with estimated values, the torque control input Equation (20) is
. ∗
ûτ ∗ (t) = F̂ωm + Ĵ ω m − Tm − γsgn(z1 ) − c1 Ĵz1 , (25)

where F̂ is the estimated viscous friction F, and Ĵ is the estimated total inertia J.
Define the input error z2 as
z2 (t) = uτ (t) − û∗τ (t). (26)

Substitute Equation (26) into Equation (18); we have

. . . ∗
z1 = ωm − ωm ,
. ∗
= 1J uτ (t) − FJ ωm + 1J ( Tm + ∆Tm ) − ω m ,
. ∗
= 1J (z2 + û∗τ (t)) − FJ ωm + 1J ( Tm + ∆Tm ) − ω m , (27)
. ∗ . ∗
 
= 1J z2 + 1J F̂ωm + Ĵ ω m − Tm − γsgn(z1 ) − c1 Ĵz1 − FJ ωm + 1J ( Tm + ∆Tm ) − ω m ,
. ∗
   
= −c1 z1 + 1J z2 + (∆Tm − γsgn(z1 )) + e J c1 z1 − ω m − Fω
e m ,

J = J − Ĵ, and Fe = F − F̂. Take derivative Equation (25),

where e
. . .
d ∗ . . ∗ .. ∗ .
dt ( ûτ ( t )) = F̂ωm + F̂ ω m + Ĵ ω m + Ĵ ω m − 2γδ(z1 ) − c1 Ĵz1 − c1 Ĵ z1 ,
. .  . 
. ∗ .. ∗ . ∗
 
= F̂ωm + Ĵ ω m + Ĵ ω m + F̂ ω m + − F̂ − Ĵ + c1 Ĵ c1 z1 + FJ − c1 z2 + 1J F̂ − Ĵc1 (∆Tm − γsgn(z1 ))


(28)
. ∗ . ∗
    
−2γδ(z1 ) + JJ c1 z2 − c1 Ĵ c1 z1 − ω m + F̂ c1 z1 − ω m + FJ c1 Ĵωm − z2 + F̂ωm .
e e

From Equations (27) and (28), the time derivative of z2 is

. . .∗
z2 = uτ (t) − ûτ (t),
  . .
. ∗ .. ∗ . ∗

= −1.5Pψm − Pωm id − RL iq − PψLm ωm + L1 uq − F̂ωm + Ĵ ω m + Ĵ ω m + F̂ ω m − 2γδ(z1 )
 .     (29)
− − F̂ − Ĵ + c1 Ĵ c1 z1 + J − c1 z2 + J F̂ − Ĵc1 (∆Tm − γsgn(z1 ))
F 1

. ∗ . ∗
    
− JJ c1 z2 − c1 Ĵ c1 z1 − ω m + F̂ c1 z1 − ω m − FJ c1 Ĵωm − z2 + F̂ωm .
e e

Remark 2. In practical control systems, it is impossible to implement the signum function sgn(·) and a
generalized function δ(·). By replacing the signum function with sigmoid function tanh(·), we obtain similar
.∗
results. If we use a sigmoid function tanh(·), the Dirac delta function δ(·) will not appear in ûτ (t). However,
because of this approximation, the states are absolutely bounded instead of displaying exponential convergence.
Please refer to Appendix A.
Assumption 2. The variation ranges for total inertia J and viscous friction coefficient F are known, i.e.
J ∈ [ Jmin , Jmax ] and F ∈ [ Fmin , Fmax ].
Energies 2019, 12, 595 9 of 17

As observed in Equation (27) and (29), the error system contains control input voltage uq and time
. .
derivatives Ĵ and F̂. For some Lyapunov function, if we can make the time derivative of Lyapunov
. .
function be negative by choosing appropriate functions for uq , Ĵ, and F̂, then the error system is stable
in the sense of Lyapunov. Additionally, since PMSG uses a permanent magnet, the d-axis current may
be zero. Therefore, the d-axis current is regulated by control input voltage ud . In other words, if d-axis
current tracking error is z3 such that z3 = id − id∗ then z3 goes to zero as time goes to infinity.
Now we propose the d,q-axis control input voltage ud , uq , and the estimator of Ĵ, F̂ as follows:
 . . . 
L . ∗ .. ∗
uq = 1.5Pψm − F̂ωm − −
Ĵ ω m Ĵ ω m + c1 Ĵz1 − c21 Ĵz1 + 2γδ(z1 )
(30)
.
∗
  
L 2γ  
+ 1.5Pψm
c1 z2 − F̂ ω m − c1 z1 + Jmin
 F̂ − Ĵc1 sgn(z2 ) + c2 z2

+ PLωm id + Riq + Pψm ωm ,

ud = Rid − PLωm iq − c3 z3 , (31)

.
. ∗
 
Ĵ = −c1 z22 + (c1 Ĵz2 − F̂z2 + z1 ) c1 z1 − ω m , (32)
.

F̂ = −z2 c1 Ĵωm − z2 − F̂ωm − z1 ωm . (33)

Proposition 2. Suppose that Assumption 1 and Assumption 2 are satisfied. By the control inputs (30) and (31)
and the parameter estimator (32) and (33), the tracking error states, z1 , z2 , z3 exponentially decay to the origin
and the estimated parameters Ĵ, F̂ converge to some bounded values.

Proof of Proposition 2. Define a Lyapunov function candidate as

1 2 1 2 1 2 1 2 1
V2 = z + z + z + e J + Fe2 . (34)
2 1 2 2 2 3 2J 2J

Equation (34) is differentiated as follows:

. . .
. . .
V2 z1 z1 + z2 z2 + z3 z3 − JJ Ĵ − FJ F̂
e e
.∗
    
= z1 −c1 z1 + 1J z2 + (∆Tm − γsgn(z1 )) + e J c1 z1 − ω m − Fω
e m
   . .
.∗ .. ∗ .∗

Pψ
+z2 −1.5Pψm − Pωm id − RL iq − L m ωm + L1 uq − F̂ωm + Ĵ ω m + Ĵ ω m + F̂ ω m − 2γδ(z1 )
  .    
+z2 − − F̂ − Ĵ + c1 Ĵ c1 z1 − FJ − c1 z2 − 1J F̂ − Ĵc1 (∆Tm − γsgn(z1 ))

(35)
.∗ .∗
    
− JJ z2 c1 z2 − c1 Ĵ c1 z1 − ω m + F̂ c1 z1 − ω m − FJ z2 c1 Ĵωm − z2 + F̂ωm
e e


  e. .
+z3 − RL id + Pωm iq + L1 ud − JJ Ĵ − FJ F̂.
e

By Assumption 1, we have
|∆Tm − γsgn(z1 )| ≤ 2γ. (36)
.
Hence,V 2 can be written as:
.    . ∗
 
V2 ≤ z1 −c1 z1 + 1J z2 + (∆Tm − γsgn(z1 )) + e J c1 z1 − ω m − Fω e m
   . .
. ∗ .. ∗ . ∗

+z2 −1.5Pψm − Pωm id − RL iq − PψLm ωm + L1 uq − F̂ωm + Ĵ ω m + Ĵ ω m + F̂ ω m − 2γδ(z1 )
  .       (37)
+z2 − − F̂ − Ĵ + c1 Ĵ c1 z1 − FJ − c1 z2 + 2γ  F̂ − Ĵc1 sgn(z2 ) + z3 − R id + Pωm iq + 1 ud
 
J L L
 .  .
. ∗ . ∗
   
J Fe


− J c1 z2 − c1 Ĵ c1 z1 − ω m z2 + F̂ c1 z1 − ω m z2 + Ĵ − J c1 Ĵωm z2 − z2 + F̂ωm z2 + F̂
e
Energies 2019, 12, 595 10 of 17

From Equations (30)–(33), Equation (37) is rewritten as

.  
≤ −c1 z21 + 1J z1 z2 + 1J z1 (∆Tm − γsgn(z1 )) − FJ z22 + 2γ F̂ − Ĵc1  1J − Jmin1
sgn(z2 )z2 − c3 z23 ,

V2
  (38)
≤ − 1J (γ − |∆Tm |)|z1 | − 2γ F̂ − Ĵc1  Jmin
1
− 1J |z2 | + 1J z1 z2 − c1 z21 − FJ z22 − c2 z22 − c3 z23 .
 

By the Cauchy-Schwarz inequality,

.  1     
F −1
≤ − 1J (γ − |∆Tm |)|z1 | − 2γ F̂ − Ĵc1  Jmin − 1J |z2 | − c1 − 1J z21 − c2 + z22 − c3 z23 ,

V2 J
    (39)
≤ − c1 − 1J z21 − c2 + F−J 1 z22 − c3 z23 .
n o .
Fmax 1− Fmin
For c1 , c2 and c3 such that c1 > Jmin1
, c2 > max 1−Jmin , Jmin , c3 > 0, V 2 ≤ 0.
Thus, the error states z1 , z2 and z3 decay to zero exponentially. Furthermore, if the error state
 0 . .
z = z1 z2 z3 is zero, then Ĵ and F̂ are also zero. By LaSalle’s Invariance Principle, we can say
the estimator output Ĵ, F̂ converges to some bounded point. 
Hence, it can be concluded that the proposed controller (30)–(33) of the model of the wind energy
conversion system with uncertainties (10), (11) and (16) guarantees that the rotating speed converges
to the reference rotating speed ωm ∗ exponentially. However, the parameter estimation converges to

some values which may not the true ones.

4. Simulation Results
In order to verify the proposed adaptive sliding mode controller, a PMSG wind turbine system is
simulated using MATLAB/Simulink, where the system parameters and the designed controller gains
are presented in Tables 1 and 2, respectively. The torque input used in the simulation is depicted in
Figure 5, which starts at 1000[ N · m] and drops to 900[ N · m] at time t = 1[sec]. The reference rotor
speeds are 75[rad/s] and 70[rad/s] for each torque value. The sinusoidal input 5 sin(44t) + 5 sin(20t) +
5 sin(52t) is also applied as an external disturbance ∆Tm which is bounded by δ = 17. The initial
values of the controller states F̂ and Ĵ are assumed to be zero.

Table 1. The parameters of the wind turbine and the PMSG.

Parameter Value Unit

The resistance of the PMSG ( R) 0.15 Ω
The inductance of the PMSG ( L) 5.3 mH
The magnetic flux of the PMSG (ψm ) 1.314 wb
The number of pole pairs ( P) 4 -
The total inertia ( J ) 100 kg · m2
The minimum of inertia ( Jmin ) 80 kg · m2
The viscous friction coefficient ( F ) 10 N · m · s/rad
The upper bound of disturbance (δ) 17 -

Table 2. The control parameters.

Parameter Value
γ 20
c1 20
c2 20
c3 10
 Energies 2019, 12, x FOR PEER REVIEW c1 20 12 of 18
c2 20
cc1 20
3 10
c2 20
Energies 2019, 12, 595
c3 10 11 of 17

Figure 5. The mechanical torque with external disturbance.

For the performanceFigureanalysis,

Figure
5.5.The we use an
Themechanical
mechanical SMCwith
torque
torque controller
with externaldesigned
external based on nominal values
disturbance.
disturbance.
J nom  90 and Fnom  9 without parameter adaption. As the simulation results, each of the system
ForFor
states thethe
and theperformance
performance analysis,
estimated analysis,
values we use
arewe
shownuseinan
an SMC 6.controller
SMC
Figure In Figuredesigned
controller designed
6a, based
basedon
the controller onnominal
wind nominal
turbinevalues
values
rotor
JnomJspeeds
=nom 
9090
andand
 nom
F F= 
9 9 without
without parameter
parameter adaption.
adaption. As As
thethe simulation
simulation results,
results, each
each of
andnomthe reference rotor speed  are shown as a solid line and a dotted line,
* of
the the system
system states
m m
andstates and the Theestimated
values arevalues are
inshown
Figurein 6. Figure 6. In6a,Figure
and6a, at the
timecontroller
instant twind 1secturbine
 , itspeeds rotorωm
the estimated shown In Figure the controller wind turbine rotor
respectively. reference rotor speed is initially 75[rad/s], changes
andspeeds m and
the reference rotorthespeed ωm∗ are shown as a solid
reference rotor speed  m* are lineshown
and a dotted as a solid line and a dotted
line, respectively. The reference line,
to 70[rad/s]. In the presence of a torque disturbance and unknown parameters, the rotor rotating
rotor speed
respectively. is initially 75[rad/s],
The referenceconverges
speed m exponentially
and
rotor speedat time
to is
the
instant
initially
reference
t = 1
75[rad/s], [ sec
speed and
] , it
 m* at
changes
in time
to
instant
a short t  1sec
70[rad/s].
time using  , it changes
In the
adaptive
presence
SMC.
of a torque disturbance and unknown parameters, the rotor rotating speed ωm exponentially converges
to 70[rad/s]. In the presence of a torque disturbance and unknown
However, in the case ∗of generic SMC, the rotor speed converged to 74.8[rad/s] and 69.8[rad/s] with parameters, the rotor rotating
to the reference speed ωm in a short time using adaptive SMC. However, in the case of generic SMC,
speed
remaining m exponentially converges
The to q- the
andreference speed  m*i ,in aand
shortreference
time using adaptive i* , idrefSMC.
the rotor speedsteady
convergedstate error.
to 74.8[rad/s] and d-axis currents
69.8[rad/s] q id
with remaining steady state values
error. qThe q-are
and
However,
plotted in in the case
Figure 6b,c.ofThe
generic
actualSMC, the
currents∗
rotor
reare speedlines,
f solid converged
and the toreference
74.8[rad/s] and 69.8[rad/s]
values are dotted with
lines.
d-axis currents iq , id and reference values iq , id are plotted in Figure 6b,c. The actual currents are solid
remaining
The q-, d-axis steady state converge
currents error. The to q-
theand d-axis values
reference currents iq , id and reference
in exponential way. Thevaluesconversion iq* , idrefrate areis
lines, and the reference values are dotted lines. The q-, d-axis currents converge to the reference values
plotted in Figure 6b,c. The actual currents are solid lines, and the reference
determined by choosing proper control parameters c1 , c2 , c3 , as shown in the proof of Proposition 2. values are dotted lines.
in exponential way. The conversion rate is determined by choosing proper control parameters c1 , c2 , c3 ,
The
The q-, d-axis currents
control converge
voltagetou2. theandreference values inuexponential way. The conversion Duerate is
as shown in theinputs
proof q-axis
of Proposition d-axis voltage
q The control inputs q-axis d arevoltage depicted uqinand
Figure 6d,e.
d-axis voltage touthe
d are
determined by choosing proper control parameters c , c , c , as shown in the proof of Proposition 2.
z 2 , chattering
depicted in Figureis6d,e.
observed
Due during
to the za2 ,reaching
chattering phase. 1The
is observed2 result
3 of parameter
during a reaching estimator
phase. isThe shown result in of
The control inputs q-axis voltage uq and d-axis voltage ud are depicted in Figure 6d,e. Due to the
Figure 6d,e.
parameter As mentioned
estimator is shownbefore,
in Figurethe 6d,e.
estimated parametersbefore,
As mentioned converge theto some bounded
estimated valuesconverge
parameters which
2 , chattering
zmay
to some different is
bounded observed
from
values trueduring
thewhich may adifferent
ones. reachingfrom phase.theThetrueresult
ones.of parameter estimator is shown in
Figure 6d,e. As mentioned before, the estimated parameters converge to some bounded values which
may different from the true ones.

(a)

6. Cont.
Figure(a)
Energies 2019, 12, 595 12 of 17
Energies 2019, 12, x FOR PEER REVIEW 13 of 18

(b) (c)

(d) (e)

(f) (g)

Figure
Figure 6. The
6. The Simulationresults
Simulation resultsof ofthe
theproposed
proposed adaptive
adaptive SMC
SMCand andSMC:
SMC:(a)(a)The Therotor speed
rotor mωandand
speed m
*∗
thethe reference
reference rotor
rotor speedω
speed ∗ *
.; (b)
mm; (b) Theq-axis
The q-axiscurrent
currentiqiqand
andreference value iiqq .;; (c)
reference value (c) The
The d-axis
d-axiscurrent
current id
re f
andidreference
and reference
valuevalue idref The
id ; (d) .; (d)q-axis
The q-axis voltage
voltage inputinput d-axis voltage input ud and ureference
q .; (e) The d-axis voltage input
uq ; (e)uThe d and
re f
value id ; (f)
reference Thei ref
value real value
.; (f) The of
realthe total
value the totalJ inertia
of inertia and estimator Ĵ; (g) The
J and estimator real
Ĵ .; (g) value
The of theofviscous
real value the
d
friction coefficient F and estimator F̂.
viscous friction coefficient F and estimator F̂ .
5. Conclusions
5. Conclusions
An adaptive robust sliding mode controller is proposed for a PMSG wind turbine system which
An adaptive robust sliding mode controller is proposed for a PMSG wind turbine system which
has parametric uncertainties and external torque disturbance. The torque disturbance includes the
has parametric uncertainties and external torque disturbance. The torque disturbance includes the
wind speed measurement error and mechanical vibrations, which influence the mechanical torque.
wind speed measurement error and mechanical vibrations, which influence the mechanical torque.
The control purpose is to track the reference rotor speed which is given by the MPPT algorithm and
The control purpose is to track the reference rotor speed which is given by the MPPT algorithm and
to regulate the d-axis current. To obtain a robust controller, a q-axis current is used as virtual control
to regulate the d-axis current. To obtain a robust controller, a q-axis current is used as virtual control
input and q-axis
input and q-axis voltage is is
voltage designed
designedfor forthetheq-axis
q-axiscurrent
currenttototrack
trackthe
thedesigned
designed virtual
virtual input using an
input using
adaptive SMC. Regarding
an adaptive SMC. Regardingthe d-axis
the current, d-axis voltage
d-axis current, is designed
d-axis voltage for d-axis
is designed forcurrent
d-axis regulation.
current
Forregulation.
the unknown For the unknown parameters J and F, an adaptation law is designed to make the
parameters J and F, an adaptation law is designed to make the closed loopclosed
system
be loop systemstable.
Lyapunov be Lyapunov stable. The performance
The performance of the controllerof theiscontroller is verifiedgenerator
verified through through generator
rotor speed
rotor speed
tracking tracking viausing
via a simulation a simulation using MATLAB/Simulink
MATLAB/Simulink (R2017b, MathWorks).
(R2017b, MathWorks). Theconverge
The error states error
states
to the converge
origin to the origin
exponentially, andexponentially,
estimated parametersand estimated parameters
converged to someconverged
boundedtovalues.
some bounded
In addition,
forvalues. In addition,
a practical for a practical
implementation implementation
sigmoid sigmoid boundedness
function, ultimate function, ultimate boundedness
of error system isofshown
error in
system isA.shown in Appendix A.
Appendix
In the
In the future,
future, it will
it will be necessary
be necessary to extendto extend to a complete
to a complete wind power
wind power generation
generation system system
including
including a grid side converter
a grid side converter connected to a grid.connected to a grid.
Author Contributions: Conceptualization, K-H.C.; Formal analysis, K-H.C.; Software, Sung-W.L.; Visualization,
Author Contributions:
Sung-Won Conceptualization,
Lee; Writing – review & editing,K-H.C.;
K-H.C. Formal analysis, K-H.C.; Software, Sung-W.L.; Visualization,
Sung-Won Lee; Writing – review & editing, K-H.C.
Funding: This research was funded by the Korea Electrical Power Corporation through the Korea Electrical
Funding: This research was funded by the Korea Electrical Power Corporation through the Korea Electrical
Engineering and Science Research Institute, grant number R18XA06-56.
Engineering and Science Research Institute, grant number R18XA06-56.
Conflicts
Conflicts of Interest:
of Interest: TheThe authorsdeclare
authors declareno
noconflict
conflict of
of interest.
interest.
Energies 2019, 12, 595 13 of 17

Appendix A
For the practical environments, all propositions are modified using sigmoid function tanh(·).
The torque control input in Equation (21) is
 
. ∗ z
u∗τ (t) = Fωm + J ωm − Tm − γtanh 1 − c1 Jz1 , (A1)
φ

where φ is the positive scalar to determine the boundary layer thickness.

Proposition A1. Suppose that a system (18) satisfies assumption 1 for a positive real number δ. Then,
the rotating speed error z1 is ultimately bounded by the feedback input u∗ (t) given as (A1).

Proof of Proposition A1. Define a Lyapunov function candidate as

1 2
V3 = S . (A2)
2

Take the time derivative; we have

. . .
V3 = S · S = z1 · z1 ,
. ∗
 
= z1 1J u∗ (t) − FJ ωm + 1J ( Tm + ∆Tm ) − ω m ,
    (A3)
= z1 −c1 z1 + 1J ∆Tm − γtanh zφ1 ,
2γ
≤ −c1 z21 + J | z1 |.

.
2γ
Since V 3 ≤ 0 for all z1 , , |z1 | ≤ Jc , the solution z1 is ultimately bounded. 
1
The control input Equation (25) is rewritten as the following equation:
. ∗
ûτ ∗ (t) = F̂ωm + Ĵ ω m − Tm − γtanh(z1 ) − c1 Ĵz1 (A4)

Substitute Equation (A4) into Equation (27), we have

. . . ∗
z1 = ωm − ωm ,
. ∗ . ∗
 
= 1J z2 + 1J F̂ωm + Ĵ ω m − Tm − γtanh( zφ1 ) − c1 Ĵz1 − FJ ωm + 1J ( Tm + ∆Tm ) − ω m , (A5)
. ∗
      
= −c1 z1 + 1J z2 + ∆Tm − γtanh zφ1 +e J c1 z1 − ω m − Fωe m .

Take derivative Equation (A4),

. . .
. . ∗ .. ∗
 . .
d ∗
dt ( ûτ ( t )) = F̂ωm + F̂ ω m + Ĵ ω m + Ĵ ω m − γφ 1 − tanh2 ( zφ1 ) z1 − c1 Ĵz1 − c1 Ĵ z1 ,
. .
. ∗ .. ∗ . ∗

= F̂ωm + Ĵ ω m − c1 z1 + Ĵ ω m + c21 Ĵz1 − c1 z2 + FJ z2 + F̂ ω m − F̂c1 z1
    
− γφ 1 − tanh2 ( zφ1 ) −c1 z1 + 1J z2 + 1J ∆Tm − γtanh zφ1 (A6)
      
− ∆Tm − γtanh zφ1 Ĵ F̂ Fe γ
J c1 − J + J c1 Ĵωm − z2 − F̂ωm + ωm φ 1 − tanh ( φ )
2 z1

. ∗ . ∗ . ∗
       
+ JJ c1 z2 − c1 Ĵ c1 z1 − ω m + F̂ c1 z1 − ω m − γφ 1 − tanh2 ( zφ1 ) c1 z1 − ω m
e
Energies 2019, 12, 595 14 of 17

From Equations (A5) and (A6), the time derivative of z2 is

. . .∗
z2 = u(t) − ûτ (t),
 
= −1.5Pψm − Pωm id − RL iq − PψLm ωm + L1 uq
. .
. ∗ .. ∗ . ∗

− F̂ωm − Ĵ ω m − c1 z1 − Ĵ ω m − F̂ ω m + F̂c1 z1 − c21 Ĵz1 + c1 z2 − FJ z2
     (A7)
+ γφ 1 − tanh2 ( zφ1 ) −c1 z1 + 1J z2 + 1J ∆Tm − γtanh zφ1
  
e  
+ 1J ∆Tm − γtanh zφ1 Ĵc1 − F̂ − FJ c1 Ĵωm − z2 − F̂ωm + ωm γφ 1 − tanh2 ( zφ1 )
. ∗ . ∗ . ∗
       
− JJ c1 z2 − c1 Ĵ c1 z1 − ω m + F̂ c1 z1 − ω m − γφ 1 − tanh2 ( zφ1 ) c1 z1 − ω m .
e

Now we propose the d,q-axis control input voltage ud , uq , and the estimator of Ĵ, F̂ as follows:
 . . 
. ∗ .. ∗ . ∗
  
uq = L
1.5Pψm − F̂ωm − Ĵ − c1 z1 −
ωm + c1 z2 −
Ĵ ω m − c21 Ĵz1
+ F̂c1 z1 − F̂ ω m γ
φ 1 − tanh2 ( zφ1 ) c1 z1
     (A8)
L 2γ  γ 2 z1 z2
+ 1.5Pψ Jmin  Ĵc1 − F̂ + φ 1 − tanh ( φ ) tanh( θ ) + c2 z2 + PLωm id + Riq + Pψm ωm ,

m

ud = Rid − PLωm iq − c3 z3 , (A9)

.  
 . ∗
  . ∗
 γ z . ∗
 . ∗
Ĵ = −c1 z22 + c1 Ĵz2 c1 z1 − ω m − F̂z2 c1 z1 − ω m + z2 1 − tanh2 ( 1 ) c1 z1 − ω m + c1 z21 − ω m z1 , (A10)
φ φ
.   
γ 2 z1
F̂ = z22 − c1 Ĵ − F̂ + 1 − tanh ( ) ω m z2 − z1 ω m , (A11)
φ φ
where φ and θ are positive scalars to determine each boundary layer thickness.

Proposition A2. Suppose that Assumption 1 and Assumption 2 are satisfied. By the control inputs (A8) and
(A9) and the parameter estimator (A10) and (A11), the tracking error states, z1 , z2 , z3 are ultimately bounded.

Proof of Proposition A2. Define a Lyapunov function candidate as

1 2 1 2 1 2 1 2 1
V4 = z + z + z + e J + Fe2 (A12)
2 1 2 2 2 3 2J 2J

Equation (A12) is differentiated as follows:

. . .
. . .
V4 = z1 z1 + z2 z2 + z3 z3 − JJ Ĵ − FJ F̂,
e e

. ∗
       
= z1 −c1 z1 + 1J z2 + ∆Tm − γtanh zφ1 +e J c1 z1 − ω m − Fω e m
   . .
. ∗

R Pψm 1
+z2 −1.5Pψm − Pωm id − L iq − L ωm + L uq − F̂ωm − Ĵ ω m − c1 z1
.. ∗ . ∗
    
+z2 − Ĵ ω m − F̂ ω m + F̂c1 z1 − c21 Ĵz1 + c1 z2 − FJ z2 + Jφ γ
1 − tanh2 zφ1 z2
      (A13)
+z2 γφ 1 − tanh2 ( zφ1 ) −c1 z1 + 1J ∆Tm − γtanh zφ1
  
e   
+z2 1J ∆Tm − γtanh zφ1 Ĵc1 − F̂ − FJ z2 c1 Ĵωm − z2 − F̂ωm + ωm γφ 1 − tanh2 ( zφ1 )
. ∗ . ∗ . ∗
       
− JJ z2 c1 z2 − c1 Ĵ c1 z1 − ω m + F̂ c1 z1 − ω m − γφ 1 − tanh2 ( zφ1 ) c1 z1 − ω m
e

  e. .
+z3 − RL id + Pωm iq + L1 ud − JJ Ĵ − FJ F̂.
e
Energies 2019, 12, 595 15 of 17

.
By using the Equation (36), V 4 can be written as:
.       
V4 ≤ −c1 z21 + 1J z1 z2 + 1J ∆Tm − γtanh zφ1 z1 + − FJ + Jφ γ
1 − tanh2 zφ1 z22
  
+z2 −1.5Pψm − Pωm id − RL iq − PψLm ωm + L1 uq
 . . 
. ∗ .. ∗ . ∗
  
+z2 − F̂ωm − Ĵ ω m − c1 z1 − Ĵ ω m − c21 Ĵz1 + c1 z2 − F̂ ω m + F̂c1 z1 − γφ 1 − tanh2 ( zφ1 ) c1 z1
      (A14)
+ |zJ2 | 2γ γφ 1 − tanh2 ( zφ1 ) + Ĵc1 − F̂  + z3 − RL id + Pωm iq + L1 ud
 
   .
− FJ c1 Ĵωm z2 − z22 − F̂ωm z2 + ωm z2 γφ 1 − tanh2 ( zφ1 ) + z1 ωm + F̂
e

 .
. ∗ . ∗ . ∗ . ∗
      
− JJ c1 z22 − c1 Ĵz2 c1 z1 − ω m + F̂z2 c1 z1 − ω m − z2 γφ 1 − tanh2 ( zφ1 ) c1 z1 − ω m − c1 z21 + ω m z1 + Ĵ
e

From Equations (A8)–(A11), Equation (A14) is rewritten as:

.        
V4 ≤ 1
J ∆Tm − γtanh zφ1 z1 − c1 z21 + 1J z1 z2 + Jφ γ
1 − tanh2 zφ1 − FJ − c2 z22 − c3 z23
    

+2γ γφ 1 − tanh2 zφ1 + Ĵc1 − F̂  1J sgn(z2 ) − Jmin
1
tanh zθ2 z2 ,
 
    
≤ 1J ∆Tm − γtanh zφ1 z1 − c1 z21 + 1J z1 z2 + Jφ γ
− FJ − c2 z22 − c3 z23 (A15)
  
+2γ γφ + Ĵc1 − F̂  1J + Jmin1
| z2 |,
 
    
≤ 2γ F
γ  1 1 2 1 γ 2 2
J | z1 | + 2γ  φ + Ĵc1 − F̂  J + Jmin | z2 | − c1 z1 + J z1 z2 + Jφ − J − c2 z2 − c3 z3 .

By the Cauchy-Schwarz inequality,

       
. 1 2 2γ 1 γ γ  1 1
V 4 ≤ − c1 − z1 + | z1 | − c2 − + F − 1 z22 + 2γ + Ĵc1 − F̂  + |z2 | − c3 z23 . (A16)
J J J φ φ J Jmin
    
Let α1 = c1 − 1J , β 1 = 2γ
J , α2 = c2 −
1 γ
+ F − 1 , β 2 = 2γ γφ + Ĵc1 − F̂  1J + 1
. Then we
 
J φ Jmin
can rewrite Equation (A16) as
.
V 4 ≤ −α1 z21 + β 1 |z1 | − α2 z22 + β 2 |z2 | − c3 z23 . (A17)

Therefore, the states are ultimately bounded. 

References
1. Kundur, P. Power System Stability and Control; McGraw-Hill: New York, NY, USA, 1994.
2. Chen, Z.; Guerrero, J.M.; Blaabjerg, F. A review of the state of the art of power electronics for wind turbines.
IEEE Trans. Power Electron. 2009, 24, 1859–1875. [CrossRef]
3. Li, S.; Haskew, T.A.; Swatloski, R.P.; Gathings, W. Optimal and direct-surrent vector control of direct-driven
PMSG wind turbines. IEEE Trans. Power Electron. 2012, 27, 2325–2337. [CrossRef]
4. Chinchilla, M.; Arnaltes, S.; Burgos, J.C. Control of permanent-magnet generators applied to variable-speed
wind energy systems connected to the grid. IEEE Trans. Energy Convers. 2006, 21, 130–135. [CrossRef]
5. Grauers, A. Efficiency of three wind energy generator systems. IEEE Trans. Energy Convers. 1996, 11, 650–657.
[CrossRef]
6. Chen, Y.; Pillay, P.; Khan, A. PM wind generator topologies. IEEE Trans. Ind. Appl. 2005, 41, 1619–1626.
[CrossRef]
7. Michalke, G.; Hansen, A.D.; Harkopf, T. Control strategy of a variable speed wind turbine with multipole
permanent magnet synchronous generator. In Proceedings of the 2007 European Wind Energy Conference
and Exhibition, Milan, Italy, 7–10 May 2007.
8. Zavadil, R.; Miller, N.; Ellis, A.; Muljadi, E. Making connections: Wind generation challenges and progress.
IEEE Power Energy. Mag. 2005, 3, 26–37. [CrossRef]
Energies 2019, 12, 595 16 of 17

9. Li, S.; Haskew, T.A.; Xu, L. Conventional and novel control designs for direct driven PMSG wind turbines.
Electr. Power Syst. Res. 2010, 80, 328–338. [CrossRef]
10. Abdullah, M.A.; Yatim, A.H.M.; Tan, C.W. A study of maximum power point tracking algorithms for wind
energy system. In Proceedings of the IEEE First Conference on Clean Energy and Techonology (CET), Kuala
Lumpur, Malaysia, 27–29 June 2011; pp. 321–326.
11. Delfino, F.; Pampararo, F.; Procopio, R.; Rossi, M. A Feedback Linearization Control Scheme for the
Integration of Wind Energy Conversion Systems into Distribution Grids. IEEE Syst. J. 2012, 6, 85–93.
[CrossRef]
12. Mao, J.; Wu, A.; Wu, G.; Zhang, X. Maximum power point tracking in variable speed wind turbine system
via optimal torque sliding mode control strategy. In Proceedings of the 34th Chinese Control Conference
(CCC), Hangzhou, China, 28–30 July 2015; pp. 7967–7971.
13. Mozayan, S.M.; Saad, M.; Vahedi, H.; Fortin-Blanchette, H.; Soltani, M. Sliding mode control of PMSG
wind turbine based on enhanced exponential reaching law. IEEE Trans. Ind. Electron. 2016, 63, 6148–6159.
[CrossRef]
14. Merzoug, M.S.; Banalla, H.; Louze, L. Sliding mode control (SMC) of permanent magnet synchronous
generators (PMSG). Energy Procedia 2012, 18, 43–52. [CrossRef]
15. Ciampichetti, S.; Corradini, M.L.; lppoliti, G. Sliding Mode Control of Permanent Magnet Synchronous
Generator for Wind Turbines. In Proceedings of the 37th Annual Conference on IEEE Industrial Electronics
Society, Melbourne, Australia, 7–10 November 2011; pp. 740–745.
16. Hong, C.-M.; Huang, C.-H.; Cheng, F.-S.; Chih-Ming, H.; Cong-Hui, H.; Fu-Sheng, C. Sliding mode control
for variable-speed wind turbine generation system using artificial neural network. Energy Procedia 2014, 61,
1626–1629. [CrossRef]
17. Beltran, B.; Ahmed-Ali, T.; El Hachemi Benbouzid, M. Sliding mode power control of variable-speed wind
energy conversion systems. IEEE Trans. Energy Convers. 2008, 23, 551–558. [CrossRef]
18. Merabet, A.; Beguenane, R.; Thongam, J.S.; Hussein, I. Adaptive Sliding Mode Speed Control for Wind
Turbine Systems. In Proceedings of the on 37th annual conference on IEEE Industrial Electronics Society,
Melbourne, Australia, 7–10 November 2011; pp. 2461–2466.
19. Sungwon, L.; Kwanho, C. Adaptive sliding mode controller design for MPPT control in wind turbine PMS
generator with torque uncertainties and unknown parameter. J. Inst. Control Robot. Syst. 2018, 24, 532–539.
20. Barambones, O. Sliding mode control strategy for wind turbine power maximization. Energies 2012, 5,
2310–2330. [CrossRef]
21. El Magri, A.; Giri, F.; El Fadili, A.; Dugard, L. Adaptive nonlinear control of wind energy conversion system
with PMS generator. In Proceedings of the 11th IFAC International Workshop on Adaptation and Learning
in Control and Signal Processing, Caen, France, 3–5 July 2013; pp. 318–325.
22. Tohidi, A.; Hajieghrary, H.; Hsieh, M.A. Adaptive disturbance rejection control scheme for DFIG-based wind
turbine: Theory and experiments. IEEE Trans. Ind. Appl. 2016, 52, 2006–2015. [CrossRef]
23. Plestan, F.; Shtessel, Y.; Bregeault, V.; Poznyak, A. New methodologies for adaptive sliding mode control.
Int. J. Control 2010, 83, 1907–1919. [CrossRef]
24. Plestan, F.; Shtessel, Y.; Bregeault, V.; Poznyak, A. Sliding mode control with gain adaptation—Application
to an electropneumatic actuator. Control Eng. Pract. 2013, 21, 679–688. [CrossRef]
25. Utkin, V.I.; Poznyak, A.S. Adaptive sliding mode control with application to super-twist algorithm:
Equivalent control method. Automatica 2013, 49, 39–47. [CrossRef]
26. Moreno, J.A.; Negrete, D.Y.; Torres-González, V.; Fridman, L. Adaptive continuous twisting algorithm. Int. J.
Control 2016, 89, 1798–1806. [CrossRef]
27. Roy, S.; Kar, I.N. Adaptive robust control of uncertain Euler-Lagrange systems with past data: A time-delayed
approach. In Proceedings of the 2016 IEEE International Conference on Robotics and Automation (ICRA),
Stockholm, Sweden, 16–21 May 2016; pp. 5715–5720.
28. Roy, S.; Kar, I.N. Adaptive sliding mode control of a class of nonlinear systems with artificial delay. J. Frankl.
Inst. 2017, 345, 8156–8179. [CrossRef]
29. Roy, S.; Kar, I.N.; Jinoh, L.; Tsagarakis, N.G.; Caldwell, D.G. Adaptive-robust control of a class of EL systems
with parametric variations using artificially delayed input and position feedback. IEEE Trans. Control
Syst. Technol. 2017, 1–13. [CrossRef]
Energies 2019, 12, 595 17 of 17

30. Roy, S.; Kar, I.N.; Jinoh, L.; Maolin, J. Adaptive-Robust Time-Delay Control for a Class of Uncertain
Eular-Lagrange Systems. IEEE Trnas. Ind. Electron. 2017, 64, 7109–7119. [CrossRef]
31. Roy, S.; Roy, S.B.; Kar, I.N. A new design methodology of adaptive sliding mode control for a class of
nonlinear systems with state dependent uncertainty bound. In Proceedings of the 15th International
Workshop on Variable Structure Systems (VSS), Graz, Austria, 9–11 July 2018; pp. 414–419.
32. Hau, E. Wind Turbines—Fundamentals, Technologies, Application and Economics, 2nd ed.; Springer:
Berlin/Heidelberg, Germany, 2006.
33. Heier, S. Grid Integration of Wind Energy: Onshore and Offshore Conversion Systems; John Wiley & Sons:
Hoboken, NJ, USA, 2014.
34. Abdullah, M.A.; Yatim, A.H.M.; Tan, C.W.; Saidur, R. A review of maximum power point tracking algorithms
for wind energy systems. Renew. Sustain. Energy Rev. 2012, 16, 3220–3227. [CrossRef]
35. Khalil, H.K. Nonlinear Systems, 3rd ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2002.
36. Perruquetti, W.; Barbot, J.P. Sliding Mode Control in Engineering; CRC Press: Boca Raton, FL, USA, 2002.
37. Utkin, V.; Guldner, J.; Jingxin, S. Sliding Mode Control in Electro-Mechanical Systems; CRC Press: Boca Raton,
FL, USA, 2009; pp. 325–355.
38. Krstic, M.; Kanelakopoulos, I.; Kokotovic, P. Nonlinear and Adaptive Control Design; Wiley: New York, NY,
USA, 1995.

© 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/).