4th International Conference on Electrical Engineering (ICEE 2015)

IGEE, Boumerdes, December 13th-15th 2015

New Control strategies of a Small Size Wind

Generation System for Battery Charging
Mayouf Messaoud

Dahmani Aissa

Department of Electrical Engineering Faculty of Applied

Sciences, University of Tiaret
Algeria
mayouf71@yahoo.fr

Department of Electrical Engineering, Faculty of

Engineer's Sciences, University of Batna
Algeria
aissa.dahmani@gmail.com

AbstractThis paper presents new control strategies of a

small size wind generation system for battery charging. The
output voltage of the direct drive PMSG is connected to the
battery through a switch mode rectifier. A DC-DC boost
converter is used to regulate the battery bank current in order to
achieve maximum power from the wind. A maximum powertracking algorithm calculates the current command that
corresponds to maximum power output of the turbine. The DCDC converter uses this current command to calculate the duty
cycle witch is necessary to control the pulse width modulated
(PWM) active switching device (IGPT). The system overview and
modeling are presented including characteristics of wind turbine,
generator, batteries, power converter, control system, and
supervisory system. A simulation of the system is performed
using MATLAB/SIMULINK.
Keywords Battery charging; Permanent magnet synchronous
generator (PMSG) ; Maximum power point tracking (MPPT) ;
PWM; DC-DC boost converter

I. INTRODUCTION
Variable speed wind turbines have many advantages over
fixed speed generation such as increased energy capture,
operation at maximum power point, improved efficiency and
power quality [1]. However, the presence of gearbox matching
the wind turbine to the generator still causes problems. The
gearbox suffers from faults and requires regular maintenance.
The reliability of the variable speed wind turbine can be
improved significantly by using a direct drive permanent
magnet synchronous generator. PMSG has received much
attention in wind energy application because of its property of
self-excitation, which allows an operation at a high power
factor and high efficiency [2].
A wind energy conversion system with PMSG can be used
basically in three distinct applications: standalone systems,
hybrid systems, and grid connected systems [3]. For a
standalone system, the output voltage of the load side converter
has to be controlled in terms of amplitude and frequency.
Previous publications related to PMSG based variable speed
wind turbine are mostly concentrated on grid connected system
[1], [2]. For electric generation in isolated systems, many
countries are affluent in renewable energy resources; however
they are located in remote areas where power grid is not
available.

To resolve this problem, small vertical axis wind turbines

for battery charging are very useful. The advantages of this
kind of turbines is that, they can rotate with reduced wind
speed. However, cost effectiveness is very important for such
applications and topologies used for higher power generation
are not viable [3], [4]. The direct connection of a three-phase
diode controlled rectifier to the batteries is a common practice
adopted by some manufactories; Although there is simplicity
and robustness, several problems associated with this solution
result, such as the reduction of batteries useful life and increase
of power losses [5], and also, the wind turbine does not operate
in its maximum electrical power in all operating conditions.
Therefore, it is necessary to provide variable wind
generation configuration; which allows the use of wind turbine
for its maximum power coefficient in a large range of wind
speed, optimizing the use of the available energy. This is
possible by inserting a DC-DC converter between the rectifier
output and the battery to adapt the generator voltage to the
battery bank voltage [5], [6]. In reference [5], two regulation
loops are used in the control system, an external controller
regulates the output voltage to avoid overvoltage, and internal
controller regulates average inductor current to achieve the
maximum power point. However, the state of charge and
discharge will be briefly discussed in this paper. We will focus
on the energetic behavior of the wind conversion topology
studied, using proposed MPPT algorithms for power
maximizing, with and without speed sensor. Thus, we will
reduce the control system to one controller.
II. SYSTEM CONFIGURATION
The proposed system is shown in Fig. 1. It is composed of a
vertical axis wind turbine directly connected to a permanent
magnet synchronous generator (PMSG). The generator is
cascaded with a battery charger composed by a three-phase
rectifier and a DC-DC boost converter. In order to know the
global behavior of the production system, we suppose that the
batteries are composed of an ideal electric supply source Ebat in
series with a resistance Rbat. The battery charger operates to
obtain maximum power transferred to the batteries and can also
limit voltage levels of the battery bank. The converter duty
cycle is changed in accordance with control system, which
receives
a
reference
signal
from
the
supervisory system.

2015 IEEE

Idc1
+

GSAP

D2

L
UL
K

Vdc1

U0

Idc2

0.16

Ibat
Q0/k
Vdc2
D1

Cp-max = 0.15

RL

0.12
0.1

Ebat
Cp

Rbat

0.14

MPPT

0.08
0.06

PWM

0.04

Idc1

TSR-opt = 0.78
0.02

Current control

Fig. 1. General scheme of the proposed system

III.

(1)

With:
(2)

Where is the air density (typically 1.225Kg/m3 at sea

level with standard conditions, i.e. temperature of 15C and
atmospheric pressure of 101.325Kpa), S is the area swept by
the rotor blades (in m2), VV is the wind speed (in m/s) and Cp
is the so-called ''power coefficient'' of the wind turbine
(dimensionless), with R being the radius of the turbine blades
(in m) and being the angular speed of the turbine rotor (in
rad/s). As can be derived from equation (1), the power
coefficient Cp is a nonlinear function of the tip-speed ratio
TSR or (dimensionless). Therefore, if the air density, swept
area, and wind speed are constant, the output power of the
turbine will be a function of power coefficient of the turbine. A
generic equation is used to model the power coefficient Cp( ),
based on the modeling vertical axis turbine characteristics used
in [6]:

Cp

0.4

0.6

0.8

1.2

1.4

Fig. 2. Power coefficient versus tip-speed ratio

SYSTEM MODELING

1
. .S .Cp ( ).Vv3
2
R.
Vv

0.2

TSR

A. Wind Turbine model

The proposed model is based on the steady state
characteristics of the vertical axis wind turbine. The output
mechanical power available from a wind turbine is given by
the following expression:

pm

Thus, the control of the wind energy conversion system

load results in a variable-speed operation of the turbine rotor,
such that the maximum power is extracted continuously from
the wind. The target optimum power from a wind turbine can
be written as:

Pm

opt

1
. .S .C p
2

opt

.(

R.

)3

K opt .

3
opt

(4)

opt

Where,

K opt

1
. .S .C p
2
opt

opt

opt

.(

)3

(5)

opt

.Vv

(6)

B. PMSG Model
In order to get a dynamical model for the electrical
generator that easily allows us to define the generator control
system, the equations of the generator are projected on a
reference coordinate system rotating synchronously with the
magnet flux as shown in Fig. 4.
q

d=r
iSd

iSq
VSq

(3)
0.2121 3 0.0856 2 0.2539
The characteristic function Cp vs. , is illustrated in Fig. 2.

As can be seen from Fig. 2, at TSR-opt, Cp has its

maximum value which results in the optimum efficiency;
therefore, maximum power is captured from wind by the wind
turbine. For each wind speed, there exists a specific point in the
wind turbine characteristic, this point is known as the
maximum power point MPPT), where the output power is
maximized.

opt

VSd
n
s

S1

Fig. 3. PMSG model in d-q references

In the synchronous machines with sinusoidal distribution of

conductors, flux d and q are linear functions of currents id
and iq situated on the rotor. And they are given by the
equations:

L d id
Lq i q
q

voltage and current as following:

(7)

v dc1

Where:

3 6

max
Eab

f eff

.P.

(13)

Ld: Stator inductance in direct axis (H).

I dc

Lq : Stator inductance in quadrature axis (H).

Rs : Stator phase winding resistance ( ).
f:

The wind turbine driven PMSG can be represented in the

rotor reference frame as:

Vq

Rs I d
Rs I q

Lq

Ld

d
Id
dt

d
Iq
dt

Lq I q
(8)

Ld I d

3
P(
2

i )

d q

V dc 2
(9)

q d

I bat dt
Q0

The electromagnetic torque in the rotor reference frame

may be written as:

3
P [( Lq
2

Ld )i d i q

iq

(10)

The dynamics of the machine is given by the following

mechanical equation:

Cm

C em

d
dt

(11)

Where:
Cm: Motor torque developed by the turbine shaft (Nm).
Cem: Electromagnetic torque developed by the generator (Nm).
: Friction torque.
J: Moment of inertia referred to the shaft of the generator
(Kgm2)
f: Friction coefficient.
The relation between the rotor angular velocity of the
generator and the mechanical angular velocity of the wind
turbine rotor is expressed as:

P
.
2

(12)

The diode rectifier is the most commonly used topology in

power electronic applications. For a three phase system it is
consisting of six diodes. The diode rectifier can only be used in
one quadrant, it is simple and it is not possible to control it.
The DC voltage and current output depend on the generator

(15)

Is a term that indicates the state of discharge.

D. DC-DC Boost Converter Model

The standard unidirectional topology of the DC-DC boost

converter of Fig. 1, consist of a switching-mode power device
containing basically two semiconductor switches (a rectifier
diode D2 and a power transistor K with its corresponding antiparallel diode D1) and an inductor L. The output DC voltage is
produced at a level higher than its input DC voltage [6], [7].
This converter acts as an interface between the full-wave
rectifier bridge and the battery bank, by employing pulse-width
modulation (PWM) control techniques. Operation of the DCDC converter in the continuous (current) conduction mode
(CCM), i.e. the current flowing continuously in the inductor
during the entire switching cycle, facilitates the development of
the state-space model because only two switch states are
possible during a switching cycle. Electric equations that
describe the dynamics of the DC-DC boost converter over a
commutation
period
are
given
by:

Vdc 2

: mechanical angular velocity of the turbine (rd/s).

E bat

I bat dt

R bat .I bat
Q0
K is a constant depends on the battery.

P: Number of poles pairs of PMSG.

C em

(14)

C. Battery Model and State of Charge

The battery model is inspired from Ford batteries model[6].
It is formed by an ideal voltage supply Ebat in series with
resistance Rbat and capacity Q(Ah) as illustrate in Fig. 1. The
electrical model is given by the following expression[6]:

The electromagnetic torque is expressed by:

Cem

Ig

Where: Vdc1 and Idc1: Medium output voltage and current of

rectifier. Ea and Ig: Output generator voltage and current.

Amplitude of the flux linkages (v.s/rad).

Vd

I dc 2

1
1
(1

Vdc1

) I dc1

vdc1 U 0 U L

(16)
(17)
(18)

Where UL and Ud0 are respectively, inductor (L) and power

switch (K) voltages. We define Rg, and Rdc1 as variable
resistances of whole circuit located respectively upstream and
downstream of the rectifier, they vary depending on the duty
cycle, and load resistance R. Current Idc1 and Idc2 are given by:

I dc1

Vdc1
Rdc1

(19)

Vdc 2
RL

I dc 2

U bat
RL

(20)

It is possible to simplify the control block diagram in Fig. 4

by eliminating clearing and disturbance terms. The reduced
scheme is shown in Fig. 5.

Based on conservation energy principal, we can write:

Rg I g2

Rdc1 I dc2 1

Rg I g2

Rdc1 (

Idc1-ref

RL I dc2 2

) I g2

) 2 .(

RL (1

PI

(21)

UL

Umax

Umin

Idc1

) I g2

(22)

We deduce the expression of the resistance Rg:

Fig. 5. Simplified diagram of control loops

Rg

18

) 2 RL

(1

(23)

In the aim to get the maximum power extracting by the

generator, we will express the power supplied by the generator
as a function of duty cycle. The electric power supplied by the
generator is given by:

Pg

Eg

3R g

( Rs

Rg )

Xs

) RL

)2

eff

)2 RL

[ (1
18

RS ]2

XS

(25)

The regulating current Idc1 depends on the power Pdc1 which

is selected as variable reference, and the measurable voltage
Vdc1. From equations 4 and 13 giving optimal wind power Popt
and voltage Vdc1, according to the rotor speed, it is possible to
express the reference current Idc1-ref using the following
expression:
C max
p

The search of the maximum power Pg.-max is equivalent to

the search of duty cycle canceling its derivative, which gives:

2 RS

opt

Pg

XS

I dc1

( RS

eff
2

)2 RS
XS

XS

RS )2

I dc1

XS

(27)

Vdc1

(29)

3. 6

eff

.P.

.C pmax . .R 4 .H

ref ( )

3. 6 .

opt

eff

.P

(30)

Where is a constant defined by:

.C pmax . .R 4 .H

CONTROL OF DC-DC BOOST CONVERTER WITH

MAXIMUM POWER EXTRACTION
The DC-DC boost converter input current can be controlled
by controlling the duty cycle of the switch (K) at any wind
speed to extract maximum power from the wind turbine. Fig. 4
shows the control block diagram of DC-DC boost converter.

3. 6 .

opt

eff

.P

(31)

The control algorithm includes then the following steps:

Measure rotor speed.
Determine input reference current of DC-DC boost
converter
B. Sensorless Control strategy

Idc1-ref
-

( )

. .R 4 .H .

opt

(26)

IV.

ref

Popt

RL

3(k
max

(28)

A. Control strategy with speed sensor

(24)

(k

(1

Pdcref
Vdc1

I dcref1

By replacing Rg and Eg by their expressions, we can

express the power Pg by:

Pg

Since the boost converter input current Idc1 depends on the

power Pdc1, we can consider this power as reference stat
variable
according
to
following
relation:

1-

max

PI

Ebat

min

Idc1
Ebat

Vdc1

Fig. 4. Control block diagram off dc-dc boost converter

The senseless control strategy proposed in this work is based

on the indirect piloting of the battery current charging. This can
simplify wind power conversion system and minimize its cost,
without reducing the energizing efficiency. We know that the
PMSG electromotive force value is given according to the rotor
speed; both parameters are then bound to output rectifier
voltage. The mechanical speed sensor can be suppressed and
replaced by a simple voltage measure which can be assimilated

to the rotation speed according to following equation:

3. 6

max
.Eab

f eff

. P.

(32)

Where is a constant given by:

3. 6

f eff

.P

(33)

Using equations (30) and (32), it is possible to express the

target optimum boost converter input current Idc1-ref according
to the voltage Vdc1 by:

I dc1

( )

ref

Vdc1

)2

.Vdc2 1

(34)

Where is a constant given by:

.C pmax . .R 4 .H .(

3. 6.

opt

eff

.p

)3 (35)

The control algorithm includes then the following steps:

Measure of output diode rectifier voltage.
Determine the boost converter input reference current.
For an optimal working rgime, the output rectifier voltage
value can be determined according to the wind speed by the
following expression:

3. 3

Vdc1

. p.

f max

opt

.Vv

(36)

While neglecting losses, the output rectifier voltage can be

written according to the battery voltage Ubat and duty cycle
by:

min
dc1

(1

).U bat

(37)
According to the wind system application conditions, the
simple DC-DC boost converter imposes some limitations. The
converter input voltage possess minimal and maximal stops,
determined by the battery voltage Ubat and duty cycle stops min
and max, The minimum input voltage of DC-DC converter is
defined by:
max

0.32

V.

RESULTS AND DISCUSSIONS

The model of the PMSG based variable speed wind turbine

system is built using Matlab/Simpower dynamic system
simulation software. The parameters of the Turbine and PMSG
used are given in Table 1. The power converter and the control
algorithm are also implemented and included in the model. The
wind profile used during this study is shown in Fig. 6.
Extractable maximum power of the generator according to the
rotor speed illustrated in Fig. 7 is located above the optimal
power of the turbine and the electric power of the battery,
witch expresses the possibility of associating a wide range of
settings of PMSG with the whole farm system studied. As
shown in Fig. 8, powers supplied follow the wind speed
profile, and they are close to the optimal power for each
variation of wind speed. As the wind speed decreased or
increased, the speed sensor controller tracks maximum power
better than senseless controller. Fig. 9 shows optimum powers
versus wind speed, where extractable power of PMSG is very
important than those provided by the control algorithms. The
integrated value of the maximum power at the end of the
simulation is the maximum energy available from the wind.
Simulation results show that the overall energy captured is
optimized by the control strategy throughout the range of wind
speed as can be seen in Fig. 10.
16
14

wind speed (m/s)

vdc1 ( )

m
(40)
s
For the duty cycle lower limit that is supposed equal to 0.1,
there is no problem posed, because the wind values gotten are
very big.
Vvmin

12
10
8
6
4
2
0

20

40

60

80

100

120

Time (s)

Fig 6. Wind profile used for system simulation

Vdcmin
1

(1

max

).U bat

3. 3

. p.

max

opt

.Vvmin

(38)
800

Therefore, the minimal wind speed can be calculated

according to the battery voltage:

optima l pow er of

ma ximum
ex tractiing powe r
of ge ne ra tor

700

turbine

600
rated powe r

Vvmin

(1 max ).U bat . .R

3. 6. p. eff . opt

(39)

powers(w)

500
ba tte ry
powe r

400
300
200
100

Admitting that duty cycle maximal value is equal to 0.99; it

is possible to find the minimal speed of wind assuring the good
functioning of the wind power conversion system for different
battery voltages. For a battery voltage equal 300v, the
minimum wind speed is:

10

15

20
25
rotor spe ed (rd/s

30

35

Fig. 7. Powers versus rotor speed

40

45

boost converter is designed without measuring wind and

generator speed, which is of great importance for small size
and low cost wind turbine. These advantages favor the use of
this configuration and control strategy in isolated sites.

700
optimal pow e r
pow e r supplie d w ith spee d se ns or

600

Powers (w)

500

Future perspectives for additional studies include

development of an MPPT algorithm, tests with prototype using
big size wind generation systems.

400

300

200

TABLE I.

100

PARAMETERS OF THE TURBINE-GENERATOR SYSTEM

pow er supplie d w ithout spee d se ns or

10

20

30
Time (s)

Wind turbine
40

50

60

Fig. 8. Powers tracking performance of the system

Density of air
Area swept by blades, S
Optimum coefficient, Kopt

1.225 Kg/m3
2 m2
0.0474 Nm/(rd/s)2
PMSG

1800
wind power
maximised power by MPPT
maximum extracting power of GSAP

1600
1400

Powers (w)

1200
1000
800
600
400
200
0

10

15

wind speed (m/s)

Fig. 9. Powers versus wind speed

No, of poles
Rated current
Rated voltage
Armature resistance, RS
Magnet flux linkage
Stator inductance, LS
Rated power
Inertia

34
4.8 A
90 V
1.137
0.15 Wb
2.7 mH
600 W
0.1 Nm

DC-DC converter
filtering voltage capacity
Smoothing inductance
Cutting frequency
Diode threshold voltage
Transistor threshold voltage
Diode conduction resistance
Transistor conduction resistance

3.3 mF
2.5 mH
5 kHz
0.65 V
0V
20.7e-3
85e-3

2.5

x 10

Optimal Energy
Battery Energy maximised by MPPT

REFERENCES
E=20835 (J)

Energy (J)

[1]
1.5

E=20719 (J)

[2]

E=4475 (J)

0.5

[3]

E=4367 (J)

10

20

30
Time (s)

40

50

60

Fig. 10. Maximum energy available from the system

VI. CONCLUSION
Simple control strategies applied in small size wind
generator systems for charging batteries have been proposed. A
complete modeling and simulation of turbine, generator,
converter, and battery was developed. The MPPT system was
designed and studied via simulated results. The main target of
the proposed system is the exploitation of the available wind
energy at low speeds in an optimum operating point without
compromising the efficiency at higher wind speeds. Although,
the energetic behavior is slightly better in the control strategy
with speed sensor; the sensorless control algorithm of DC-DC

[4]

[5]

[6]

[7]

[8]

S.Vijayalakshmi, Saikumar.S, Saravanan.S,R.V.Sandip, V. Sridhar:

Modeling and control of a Wind Turbine using Permanent Magnet
Synchronous Generator, International Journal of Engineering Science
and Technology (IJEST), Vol. 3 No. 3 March 2011
M.Yin, G. Li, M. Zhou, C. Zhao: Modeling of the Wind Turbine with
a Permanent Magnet Synchronous Generator for Integration, 1-42441298-6/07/2007 IEEE.
A.M De Broe, S. Drouilhet, V. Gevorgian, A peak power tracker for
small wind turbines in battery charging applications, IEEE Trans. On
Energy Conversion, Vol. 14, No. 4,pp. 1630-1635 , Dec. 1999.
[4] Y. Chen, P. Pillay, A. Khan: PM Wind Generator Topologies,
IEEE Transaction on. Industry application, vol. 41, No. 6, Nov/Dec.
2005.
Herminio M. Oliveira Filho, Demercil S. Oliveira Jr, Ren P. T.
Bascop, Carlos E. A. Silva, ivanildo J. Almeida, On the study of wind
energy conversion system applied to battery charging using multiblade
turbine, 2009 Brazilian Power Electronics Conference, Bonito (MS) Brazil - ISSN 2175-8603
D.Rekioua,Z, Roumila, T.Rekioua, Etude d'une centrale hybride
photovoltaique,olien,diesel, Revue des energies renouvelables.Vol.11
N4 (2004), pp. 623-633
N. T. Hai, S.Ho. Jang, H.G. Park, D.C. Lee: Sensorless Control of PM
Synchronous Generators for Micro Wind Turbines, 2nd IEEE
International Conference on Power and Energy (PECon 08), Dec. 1-3,
2008, Johor Baharu, Malaysia.
M.Mayouf, R.Abdessemed, Comparative Study of a Small Size Wind
Generation System Efficiency for battery charging, Serbian Journal of
Electrical Engineering, SJEE , Vol. 10, No.2,pp 261-274,