MODELING AND STUDY OF WIND TURBINE DRIVEN

DFIG USING THREE LEVEL INVERTER

ZERZOURI NORA*, TOURAB WAFA*, BENALIA NADIA*, BENSIALI NADIA*.

*Department of Electrical Engineering, Annaba University, Email: nzerzouri@yahoo.fr

Abstract

In recent years, wind energy has become one of the most important and promising sources of renewable energy, which
demands additional transmission capacity and better means of maintaining system reliability. The doubly fed induction
generator is an important type of wind generator (WG) due to its robustness and versatility. Its accurate and efficient
modeling is very important in distribution system planning and analysis studies. This paper deals with the analysis, modeling
and control of a grid connected doubly fed induction generator (DFIG) wind turbine. A mathematical model is derived for
modeling of doubly fed induction generator wind turbine. And a control structure using standard proportional integral PI
controller and a field-oriented control strategy based on a reference frame rotating synchronously with the rotor flux for
variable speed wind turbines using doubly fed induction generator and for attaining injected rotor voltages is described and
simulated. The modeling of the machine considers operating conditions below and above synchronous speed, which are
actually accomplished by means of double side PWM converters joining the machine rotor to the grid. The control is
especially designed to reduce the harmonic distortion of the generated currents and active power ripples. The simulation
results verify the control strategy that decouples successfully active and reactive power generated by the machine and clearly
show the improvement of the quality of the energy produced when using three-level inverter rated rotor.

Keywords:

Renewable Generation, Wind Turbine, Doubly Fed Induction Generator, Rotor Side Converter, Grid Side Converter, Distortion
Harmonics.

I. INTRODUCTION:
In the recent years, renewable energy systems have attracted the great interest because conventional sources of energy are
limited and a number of problems associated with their use, like environment pollution, large grid requirements etc.
Governments of the whole world are forced for the alternative energy sources such as wind power, solar energy and small
hydro-electric power [1]. Among the above given choices, wind energy is a realistic way of harnessing the natural energy.
Wind energy has been intensively investigated in recent years in many different countries, which resulted in several different
configurations like fixed speed system with a SCIG, the variable speed system with permanent magnet synchronous
generator (PMSG) and the variable speed system with a DFIG to improve the efficiency, power rating, cost benefit
effectiveness etc. [2-3]. Wind is highly variable in nature, so variable speed DFIG based WECS offers many advantages
compared to the fixed speed squirrel cage induction generators, such as reduced converter rating, cost and losses in result of
that an improved efficiency, easy implementation of power factor correction, variable speed operation and four quadrants
active and reactive power control capabilities [ 4-5]. Due to variable speed operation, total energy output is much more in
case of DFIG-based WECS, so capacity utilization factor is improved and cost of per unit energy is reduced. A wind energy
conversion system using DFIG is shown in Fig.1.

Fig.1. DFIG connected to grid with back-to-back converters in rotor circuit

In the above system, the variable speed DFIG stator is directly connected to the grid and the rotor connected to a back-to-
back partially rated (20-30% of machine rating) voltage source power converter [6]. Vector oriented control enables
decoupled control of active and reactive power flowing between the DFIG and the grid [7]. Traditionally the converter
connected to the grid is called grid side converter (GSC), and the converter connected to the DFIG rotor circuit is called rotor
side converter (RSC). Basically, the main role of the GSC is to control the DC bus voltage and the reactive power at the grid,
and the RSC to control active and reactive power production at the stator. This is done by implementing an MPPT algorithm
in the RSC, e.g. the optimal tip speed method, which ensures harvesting of maximum available mechanical power at a unit
power factor. Both stator and rotor are capable of power generation; however the direction of active power flow through the
rotor circuit is dependent on the wind speed and accordingly the generator speed, which can operate within ±33% around
the synchronous speed [8].
This paper explicitly models the DFIG and examines the current and voltage relationships associated with the stator and
rotor. The dq0 transformation is given in the paper for aided analysis. Finally, a scheme is presented to find the desired active
and reactive power. The two powers are decoupled to find their individual effects using stator side flux orientated technique
where the rotor is fed by a three level inverter. Simulations and results from Simulink/MATLAB validate the efficacy of the
presented methodology.

II. WIND TURBINE MODEL :

The aerodynamic model of a wind turbine is determined by its power speed characteristics [9]. For a horizontal axis wind
turbine, the mechanical power output Pt (w) that a turbine can produce is given by the following equation:

1
Pt = ρA C p ( λ , β ) v 3 (1)
2

Where, ρis the air density (kg/m3), v is the wind speed (m/s), A is the area covered by the rotor (m2), and Cp is the power
coefficient which is a function of both tip speed ratio, λ, and blade pitch angle β (deg). In this work, the CP equation is
approximated using a non-linear function according to [10].

−21
116
C p ( λ , β )=0,517− (
λi
−0,4 β−5 e ) λi
+ 0,0068 λ (2)

Which

1 1 0,035
= − 3
λi λ+0,08 β β +1

The tip speed ratio λ for

ΩR
λ=
v
(3)

Where Ω is the angular velocity of wind turbine and R is blade radius. The turbine power characteristics are illustrated as
shown in Figure 2. These characteristics are plotted at pitch angle β=0 (deg).
0.5

0.45

1beta=0
0.4 2beta=2
3beta=4
0.35 4beta=6
5beta=8
6beta=10
0.3 7beta=12
8beta=14
coefficient Cp

9beta=16
0.25

0.2

0.15

0.1

0.05

0
0 5 10 15 20 25
Vitesse de Vent(m/s)

Figure 2. Power coefficient C p versus tip speed ratio 𝝀

III. MODELING OF DFIG :

The DFIG model is developed based on the following assumptions. The equations are derived in the synchronous reference
frame using direct (d) and quadrature (q) axis representation. The stator current is assumed positive when flowing toward
the machine. The m-axis was assumed to be 90 lag of t-axis in the direction of rotation .

Stator voltage
dϕ ds

{V ds =Rs I ds+
dt
dϕ
−ϕqs ωs

V qs=R s I qs + qs +ϕ ds ω s
dt
(4)

Rotor voltage
dϕdr

{V dr =R r I dr +
dt
dϕ
−ϕqr ( ω s−ω r )

V qr=R r I qr + qr +ϕ dr ( ω s−ω r )
dt
(5)

Where the flux linkages are given by the expressions

ϕds =Ls I ds + M I dr

{ϕqs =Ls I qs + M I qr
ϕdr =Lr I dr + M I ds
ϕqr =Lr I qr + M I qs
(6)

The developed electromagnetic torque is given by

3
C e = PM (I qs I dr−I ds I qr ) (7)
2

To easily control the production of electricity from wind, we will achieve an independent control of active and reactive power
by the stator flux orientation. The idea is to align along the axis of the rotating frame [11] stator flux. We therefore: ϕqs =0
and consequently ϕds = ϕs.
This choice is not random but is justified by the fact that the machine is often coupled with a powerful network voltage and
constant frequency, which leads to a finding stator flux of the machine. Neglecting the resistance of the stator windings,
often accepted hypothesis for high power machines: The systems of equations (4) and (6) can be simplified as follows:
V ds =0
{V qs =V s =ϕ s ω s
(8)

ϕ s=L s I ds+ M I dr

{ 0=Ls I qs + M I qr
ϕdr =Lr I dr + M I ds
ϕqr =Lr I qr + M I qs
(9)

The stator active and reactive power in the orthogonal coordinate system can be written:
3

{P s= ( V ds I ds+ V qs I qs)
2
3
Q s= (V qs I ds −V ds I qs)
2
(10)

Under the assumption of a stator flux oriented, this system of equations can be simplified as:
3

{Ps = V s I qs
2
3
Qs= V s I ds
2
(11)
The rotor currents can be derived in the following form:
M I dr −ϕds

{I ds=

M
Ls
I qs= I qr
Ls
(12)

By introducing these expressions in the equations of the rotor voltages are found:

d I dr

{ V dr=Rr I dr +δ Lr

V qr =Rr I qr +δ Lr
d I qr
dt
dt
−ω g δ Lr I qr

M
+ω g δ Lr I dr + ϕ ds
Ls
(13)

It is clear from the stator power expressions (11), that stator active power is controlled by Iqr, while Idr controls stator reactive
power. The RSC control strategy is illustrated in Figure 3. For the stator to operate at unity power factor, the flow of reactive
power from the stator side to grid is set to zero, thus Iqr*can be determined from (10), so Idr* can be calculated from (12).
Now the actual rotor currents Idr and Iqr are compared with the reference rotor currents Idr * and Iqr * before being processed
using inner PI current controllers, in a similar manner to the ones used for GSC, then added with cross-terms, to finally
generate the control signals for the RSC[12].

Figure 3.Vector control of DFIG with grid and rotor side converters.

IV. MODELING AND CONTROL OF THE THREE LEVELS STRUCTURE NPC VSI
The three phases three-level NPC VSI is constituted by three arms and two DC voltage sources. Every arm has four
bi-directional switches in series and two diodes.
 Figure 4. Equivalent circuit of NPC inverter back to back.

For the mathematical model, the switching states of the A phase, Sa is written:

S a=2 ⟹ S 1 a=1 , S 2 a=0 , S 3 a=0

{S a=1 ⟹ S 1 a=0 , S2 a=0 , S3 a =1

S a=0 ⟹ S1 a=0 , S2 a=1, S3 a=0
(14)

Sa1, Sa2, Sa3 are the connection functions which describe the state of the first arm (1a= upper half of the arm, 2a= lower half
of the arm and 3a= neutral point of the arm). Similarly, the phases of the switching states B and C, Sb and Sc are also
decomposed. Assuming that the three-phase voltages (VA, VB and VC) of the NPC three level converters is balanced, using the
DC voltage (Vdc1 and Vdc2), relations between the voltages (VA, VB and VC) with connection functions (S1a, S2a, S3a) are
defined as follows:

VA S1 a S2 a
1 2−1−1
[ ] [ ]([ ] [ ] )
V B 3 −1 2−1
VC
=
−1−1 2
S1 b
S1 c
.V dc 1 − S2 b . V dc 2
S2c
(15)

The inverter is controlled by with Pulse modulation strategy which uses two bipolar carriers. This strategy is characterized by
two parameters:
Modulation index m defined as ratio between the carrier frequency fp and the reference voltage frequency:
fp
m= (16)
f
Modulation rate r is the ratio between the magnitudes Vm of the reference voltage and three times of the carrier’s
magnitude Upm:
Vr
r= ≤1 (17)
Vp
V. SIMULATION RESULTS

Simulation results were performed using software Matlab / Simulink. To validate the controls proposed in this paper, we
present two operating points in hypo-or hyper-synchronous fashion. For these simulations, we consider that the wind system
is steady and working in the zone of optimal functioning (zone 2), that is to say that it produces maximum power versus
speed of the wind. The reference voltage of the DC link is set at 1200V. We have vary the stator reactive power Qs playing on
its reference value. The parameters of the DFIG are given in Table 1.

1) Wind power system operation mode hypo synchronous.

For this mode, a wind velocity equal to 8ms -1 is applied on wind turbine blades, which corresponds to a speed of DFIG in
MPPT control approximately 1380 tr/min, a shift 30% hypo synchronous mode, as shown in Figure 5.
Figure 6 shows the temporal evolution of the various power supply Qs, Ps and Pr, Qr. At time t = 2s, the wind system
operates as a unity power factor and reactive power Qs_ref imposed zero. From time t = 2s to t = 3,5s, we set the stator
reactive power reference equal to -1MVar corresponding to changes in the reference direct current rotor and also on the
rotor reactive power Qr. From time t = 3.5s we set the stator reference reactive power equal to 1 MVar, again, different
reactive powers are properly regulated.
1400

1200

1000

G en erator s peed N r(tr/m in)

800

600

400

200

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time(s)

Figure 5.Generator speed

3 1.2

Pr(MW)
1

2
0.8
Qs(MVAR)

0.6
1

0.4

Ps(MW)
0 0.2

-1
-0.2

-0.4
-2

-0.6 Qr(MVar)

-3 -0.8
1 1.5 2 2.5 3 3.5 4 4.5 5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time(s) Time(s)

3000

2500
2500 Idr(A)

2000
2000 Iqr(A)
1500
1500

1000
Rotor phase c urent Iar(A )

1000
500

500
0

0
-500

-500
-1000

-1000
-1500

-1500
-2000

-2000 -2500
1 1.5 2 2.5 3 3.5 4 4.5 5 1 1. 5 2 2.5 3 3.5 4 4.5 5
Time(s) Time(s)

Figure 6. Ps, Qs, Pr, Qr, ird, irq and iar (hyposynchronous)

2) Wind power system operation mode hyper synchronous

For this mode, a wind speed equal to 11m/s corresponds to the nominal operation of the turbine is applied on wind turbine
blades, which corresponds to a speed of DFIG in MPPT control approximately 1950 tr /min, a shift of -30% hyper synchronous
mode, as shown in Figure 7 and Figure 8. The same stator reactive power variations have been applied to this mode as the
previous. We also note that the d q axis rotor currents are perfectly decoupled through the decoupling strategy by ensuring
natural PWM. We also observe the rotor phase current follows perfectly the reference obtained by the orientation of the
stator flux. Both modes are validated and independent control of different powers of the wind system.
2000

1800

1600
G e n e ra t o r s p e e d N r(t r/m in )

1400

1200

1000

800

600

400

200

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time(s)

Figure 7.Generator speed

 15
1

10 Qr(MVAR)
0.5

5
Qs(MVAR) 0

Pr(MW)
0
-0.5

-5
-1
Ps(MW)

-10
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time(s) Time(s)

4000
4000

Iqr(A)

3000
3000

2000

2000

1000

Rotor phase curent Iar(A)

1000
0

Idr(A)
-1000
0

-2000

-1000

-3000

-2000 -4000
1 1.5 2 2.5 3 3.5 4 4.5 5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time(s)

Figure 8.Ps, Qs, Pr, Qr, ird, irq and iar (hypersynchronous)

Figure 9 and Figure 10 show the harmonic spectrum of output phase stator current obtained using Fast Fourier Transform
(FFT) technique for PWM for 2-level and 3-level inverter respectively. It can be clear observed that all the lower order
harmonics are reduced for 3-level inverter (THD = 3.88%) when compared to 2-level inverter (THD = 4.09%).

(a) (b)
4
x 10
1.5
800

600
1

400

0.5
Stator phase current Ias(A)

200
Rotor phase voltage Var(V)

0
0

-200
-0.5

-400

-1

-600
Time(s)

-800 -1.5
1.6 1.65 1.7 1.75 1.8 1.85 1.9 1.95 2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time(s) Time(s)

Fundamental (50Hz) = 2348 , THD= 4.09%

2000

1500
Mag

1000

500

0
0 10 20 30 40 50 60
Harmonic order

(c)
Figure 9. (a) Rotor phase voltage, (b) Stator phase current, (c) Spectrum harmonics using 2-level

(a) (b)
 4
x 10
800 1.5

600
1

400

0.5
200

Rotor phase voltage Var(V)

Stator phase current Ias(A)

0 0

-200
-0.5

-400

-1
-600

-800 -1.5
1.6 1.65 1.7 1.75 1.8 1.85 1.9 1.95 2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time(s) Time(s)

(c)

Fundamental (50Hz) = 2348 , THD= 3.88%

2000
Mag (% of Fundamental)

1500

1000

500

0
0 10 20 30 40 50 60
Harmonic order

Figure 10. (a) Rotor phase voltage, (b) Stator phase current, (c) Spectrum harmonics using 3-level

Table 1. 3MW WTG Induction Machine Parameters

parametre value

Rotor resistor per phase 2,97mΩ

Rotor resistor per phase 3,82mΩ

Inductance of the stator winding 121 mH

Inductan ce of the retor winding 57,3 mH

Mutual Inductance 12,12 mH

Number of pole pairs 2

Inertia 114 kg.m2

Rated power 3MW

Rated voltage 690V

Table 2.The parameters of wind turbine

paramètre value

Nombre of blade 3

Rated speed 11m/s

 Inertia

Gearbox ratio 114 kg.m2

Rotor Diameter 100

45m

VI. CONCLUSION

This paper presents simulation results of Field orientation method control for active and reactive power control of a
DFIG, using the modulation strategy of the PWM inverter for two and three level. With results obtained from simulation,
it is clear that for the same operation condition, the DFIG active and reactive power control with field oriented control
using in the three level had better performance than the PWM technique two level and that is clear in the spectrum of
phase stator current harmonics which the use of the three level, it is reduced of harmonics more than two level inverter.

REFERENCES

[1] Y. Lei, A. Mullane, G.Lightbody, and R. Yacamini, “Modeling of the wind turbine with a doubly-fed induction generator for
grid integration studies,” IEEE Trans. Energy Conversion, vol. 21, no. 1, pp. 257-264, Mar. 2006.
[2] Yifan Tang and Longya Xu, “A flexible active and reactive power control strategy for a variable speed constant frequency
generating system,” IEEE Trans. on Power Electronics, vol. 10, no. 4, pp. 472-478, July 1995.
[3] A. Ostadi, A. Yazdani, and R. K. Varma, “Modeling and Stability Analysis of a DFIG-Based Wind-Power Generator
Interfaced With a Series-Compensated Line,” IEEE Trans. on Power Delivery, vol. 24, no.3, pp. 1504-1514, July 2009.
[4] S. Muller, M. Deicke, and R. W. De Doncker, “Doubly fed induction generator systems for wind turbines,” IEEE Ind. Appl.
Mag., vol. 8, no.3, pp. 26–33, May/Jun. 2002.
[5] G. Ramtharan, J. B. Ekanayake, and N. Jenkins, “Frequency support from doubly fed induction generator wind turbines,”
IET Renewable Power Generation, vol.1, no. 1, pp. 3–9, 2007.
[6] A. Tapia, G. Tapia, J. X. Ostolaza, and J. R. Saenz, "Modeling and control of a wind turbine driven doubly fed induction
generator," IEEE Trans. On Energy Conversion, vol.18, no. 2,pp. 194-204, 2003.
[7] S. Li and T. A. Haskew, "Analysis of Decoupled d-q Vector Control in DFIG Back-to-Back PWM Converter," in Power
Engineering Society General Meeting, 2007. IEEE, 2007, pp. 1-7.
[8] R. Datta and V. T. Ranganathan, "Variable-speed wind power generation using doubly fed wound rotor induction machine-
a comparison with alternative schemes," IEEE Trans. on Energy Conversion, vol. 17, no. 3, pp. 414-421, 2002.
[9] LIU Qihui,HE Yikang,ZHANG Jianhua, “Investigation of Control for AC-exited VSCF Wind Power Generation System
Connected to Grid,” Proceedings of the CSEE, vol. 26, no. 23, 2006, pp. 109-115.
[10] Ma Wen,Yu Junjie,Wu Guoxiang,Chen Guocheng. “Doubly-fed Wind Power Generation System Maximum Power Point
Tracking Control Strategy,” Transactions of china electrotechnical society, vol. 24, no. 4, 2008, pp. 202-208.
[11] B. Belabbas, ”Hybrid Fuzzy sliding mode performance control applied to a DFIG system for the production and integrated
wind energy into a power grid based three-level converters” BALKAN JOURNAL OF ELECTRICAL & COMPUTER ENGINEERING
2013, Vol.1, No.2
[12] Mohammad Sleiman, Bachir Kedjar, Abdelhamid Hamadi, Kamal Al-Haddad, Hadi Y. Kanaan. “ Modeling, Control and
Simulation of DFIG for Maximum Power Point Tracking”. Control  Conference (ASCC), 2013 9th Asian, Publication Year: 2013,
Page(s): 1 – 6,  IEEE CONFERENCE PUBLICATIONS.