TAP CHi KHOA HQC I& CONG NGHE CAC TRU''''ONG hAHIQC KY TEIUAT * St̂ 86 2012 SYNCHRONIZATION AND POWER FLOW CONTROL OF DFIG FOR WIND ENERGY SYSTEM DIEU KHIEN HOA DONG BQ VA DONG CONG SUAT CUA DFIG TRONG H[.]
Trang 1SYNCHRONIZATION AND POWER FLOW CONTROL
OF DFIG FOR WIND ENERGY SYSTEM
DIEU KHIEN HOA DONG BQ VA DONG CONG SUAT CUA DFIG
TRONG HE THONG N A N G H 'ONCi GIO
Phan Quoc Dzung, i\guyen Bao Anh, Le Minit Phuong, Le Dinh Khoa
Ha Chi Minh University ofTevlinology Received October 31 "•' lU 11
TOM TAT
Bai bao trinh bay cac thuat toan d4 diSu khi4n bo bi§n doi phia rotor (RSC) cua he thong may
phat cam t/ng ngudn k^p (DFIG) bao gdm giai thu$t dieu khiSn tru-c ti4p cdng suat (DPC) de di&u
twoi Trong ci hai giai thuat tren, cdc khdi diSu chinh ti le • tich phan d6u khong dux?c siy dung Ban
chat cua hai thuat toan la chon Iwa vector dien ap thich hap clio bd bi4n doi phia rotor K4t qua thiK
nghiem tren mdy 1,5 kW duxyc trinh bay, su dijng phan mdm MATLAB/SiMUUNK va vi di6u khiSn
DSpace 1103
ABSTRACT
This paper presents algorithms for controlling rotor-side converter (RSC) of a double-fed
induction generator (DFIG) system including direct power control (DPC) strategy for controlling power
flow and direct virtual torque control (DVTC) strategy for synchronizing DFIG with grid There is no
proportional-integral (PI) regulator in both of strategies The essence of two strategies is selection of
appropnate voltage vectors on the rotor-side converter The experimental results on a 1 5'kW machine
are explained using MA TLAB/SIMULINK together DSpace 1103
Keyword Direct Power Control (DPC), Direct Virtual Torque Control (DVTC), Direct Torque
Control (DTC), Artificial Neural Network (ANN), Doubly-Fed Induction Generator (DFIG), Rotor-side
converter (RSC), Gnd-side converter (GSC)
Nomenclature vg (grid voltage), is (stator current), ir (rotor current), 9r (rotor angle) H'g 4's
(grid and stator flux), 4^9 Tv (rotor flux and virtual torque), Ps Qs (active and reactive power), Lm
(mutual inductance) Ls Lr (stator and rotor inductance), Rs Rr (stator and rotor resistance), p (pole
pairs), w (frame speed), wr (rotor speed), tos (synchronous speed)
I INTRODUCTION turbines have many advantages when compared
, _ , , , to wind turbines using fixed speed induct''^"
In recent \cars fossil fuel is exploited
and used more and more, this leads to the
generatoib, which are variable ;.peed operation
, four-quadrant active and reactive power exhaustion of fossd fuel and the a.r pol ution by ^apabiliues, lower converter costs and louer
greenhouse gas So, human are researching and |^^^^^ [,j ^ ,<,|,„^^,i, ji3„,,„, „f ,
developing new technologies to use the types of DpiQ.based wind enerav a^ncrauon" system h
renewable energy Among the types of shown in Figure 1 "' "
renewable energy, wind energy is applied in
large power plans such as large wind farms
Wind turbine technology can be divided (icarx>\ ( (,.|^ into fixed speed category (generally with
squirrel-cage induction generator) and variable WiiuliuibiiK
speed one (with doubly-fed induction (n.'iicraiiir MICIIIIK-MVIC d ,^\-.\\.k
generator) For many wind farms, wind turbines annciicr i.iiii\jrii.'r
based on doubly-fed induction generator
(DFIG) with converters rated at 25% - 30% of l^ig /- Schematic diagram of a DFlG-hase
Trang 2A DFIG has a wounded rotor which is
connected to grid through a back-to-back
converter, and the stator is connected to grid
directly Controlling DFIG is implemented
through controlling back-to-back converter, thai
is rotor-side converter (RSC) and grid-side
converter (GSC) Objectives are good responses
of current, voltage and power
n MATHEMATICAL MODEL OF DFIG
Equivalent circuit of DFIG is shown in
Figure 2
i / y)\\lJ l.n Lr.r i(V)-i<),)\\l,' •.•
Fig 2 Equivalent circuit of DFIG
In a frame which rotates at a speed of to,
voltage equation of stator and rotor are
' dt
dy/^
~dt~
Flux equation of stator and rotor are
V J = ^asK + ^n,i'' + 'r ) = k's + AnC
And mechanical equations of DFIG are
L=-P¥sX'
(1) (2)
(3) (4)
(5) (6)
p dt
III DVTC ALGORITHM
When the stator is connected to grid
which has a constant magnitude and frequency
of voltage, voltage vectors are applied to the
RSC, the algorithm is called DTC algorithm
According to (5), the expression depicts the
electromagnetic torque is shown In equation
(7), torque is a function of stator and rotor flux
and the angle between them So the torque can
be controlled through the control of magnitude
and angle y of rotor flux and stator flux [4]
When the stator is about lo connect to grid, there is no current on stator According (5), electromagnetic torque is equal zero and the stator flux vector is collinear to the rotor flux
So, a virtual torque is defined as given in equation (8) [3] and this algorithm is called DVTC algorithm
The variation of Ihc rotor flux vector is achieved through the application of the appropriate rotor voltage vector as (9)
The rotor flux vector rotates in the same direction as the rotor voltage and with a rotation speed proportional to the rotor voltage magnitude as shown in Figure 3
Fig 3 Rotor flux vector and rotor \oltage vector
DVTC algorithm is implemented as a look-up table as shown in Table 1 that generates the switching states of the power semiconductors corresponding to the appropriate rotor voltage vector The choice of rotor voltage vector depends on fogic outputs of rotor flux and torque hysteresis controllers and
on the position of rotor flux vector in rotor plane divided into six sectors
Tabic I Optimal switching table for DVTC algorithm
N
S,,= l S,-0
S i - I
S i - 0
S i - I
1 V,
v»
V , i
v,<
2 3
V,] V,j Vrl V,,
V r t v„
v„
4
V r t v„
V,
V,
5
V,6 V,4
V r l
6
V , i
V„ v,>
Trang 3Connection of DFIG to grid requires the
adjustment of stator voltage to be synchronized
with the grid voltage The objecdve of RSC
control is meeting the synchronization
conditions, these conditions are equality in
phase, frequency and magnitude between tlie
grid and stator voltage vector [3]
The first two synchronization conditions
can be met through bringing to collinearity grid
flux and rotor flux vectors b> eliminating the
angle 5 between them Other-word, virtual
torque will be controlled so that it is equal to
zero
The third synchronization condition is
equality in magnitude between grid voltage and
stator voltage The amplitude of stator voltage
is indirectly controlled through adjusting the
rotor flux An appropriate rotor flux reference
as (10) is calculated and used to control rotor
flux in order to guarantee the required stator
voltage amplitude
The overall control strategy scheme is
shown in Figure 4
Fig 4 DVTC scheme for DFIG grid connection
IV DPC ALGORITHM
In stator grid-connected mode, DFIG
exchanges power with grid If stator resistance
can be neglected, the stator active and reactive
power can be expressed as (11) and (12) [ 1 ]
3 co,L„
laL^L,
3 w, I ,\,L,
,(^ It// isin6
^ k | ( f | v : i c o s ^ - | < | )
(11)
(12)
According to (10) and (II), active power and reactive power can be controlled by adjusting v',' sin^ and L H C O S ^ They are components of rotor flux vector at the perpendicular and ihe same direction of the stator flux respectively as shown in Figure 5
Fig 5 Relation of tlator and rotor flux in slalionary and rotor reference frame
Similarly DVTC algorithm, two above components of rotor flux can be controlled through selecting the appropriate rotor vector voltage
States of active power and reactive power which are calculated by two 3-level hysteresis comparator and the position of stator flux in rotor reference fi^me are used to determine the appropnate rotor voltage vector An optimal switching table is given in Table 2
Tahie 2 Optirral switching table for DPC algorithm
N
Sy=l
Sy=0
Sy=-I
S,.= l Sp=0 S,.=-l Sp=l Sp=0
Sp=l Sp=0 Sp=-I
1 Vre
V,-
V,s Vrf
Vrl
Vri V,3
2 Vrl
v,^
Vr3
Vr7
V , j
v^
v^
V,4
3
V , ; Vr-,
V , j
Vrt,
V n
Vrf,
4 V,5 Vr4
V r i
V,7
Vrt
v„,
5 Vr4 V,s
Vr-,
Vrl
V,t
6
V , V.fr V,i Vr4 V,7 " V,-
V , , V,3
The overall control strategy scheme is shown in Figure 6
:- P5
Trang 4V EXPERIMENTAL RESULTS
An experimental model will verify two
above algorithms That model consists power
components, processor and support circuits
Algorithms will be implemented on
DSpace 1103 which connect to computer
through MATLAB/SIMULINK software The
emulation of wind turbine is done by means of
a 2 2 -kW/1500-rpm induction motor The
DFIG is a I 5-kW;i500-rpm wound rotor
induction machine The RSC are IGBT-bascd
converter Hall-effect sensors are used lo
measure current (HX 20-P) and voltage
([,V25-
P)-DFIG and IM parameters are sliouii in
Table 3
Table 3 Parameters of DFIG and IM
Rated power
Rated stator \ ollasc
Rated rotor \ ollaue
Rated stator frequency
Nominal speed
Rated stator current
Poles pair
DFIU l.SkW
415 V
145 V
50 Hz
3.7 A
2
IM 2.2 kW
380 V 50llz
1435 rpm
5 A
2 The overall experimental model is shown
in Figure 7
Fig 7 Experimental model of DFIG ywlcm
The rotor shaft operates at speed of 1350
rpm (10% slip), four cases are examined in ihis
experimental model They include:
1) Case study I: synchronizing DFIG
with grid
2) Case study 2 distributing aclixe
power P = 1OOOW
3) Case study 3: distributing active
power P = 500W and reactive power Q ^ 500
The responses of three cases are shown in following figures
i ' J :
! Ji
(c)
Fig S Responses of system in case I
(a) Grid (red) and Slalor (green) line-to-line voltage
(b) Reference (red) and real rotor flux (green) (c) Virtual torque
\ CONCLUSION
These algorithms used to operate the DFIG system have been proposed in this paper Both of algorithms have simple structure because there is no need of PI regulator, thus problems related lo parameter tuning and machine parameter dependence are eliminated For DPC algorithm, the only machine parameter required to control system is stator resistance which has small effect on responses
of algorithm and can be neglected The method selects appropnate \oltage vectors based on the stator flux position and active and reactive power errors, thus the difficulties associated with rotor flux estimation are removed
Trang 5The experimental model gives good
responses as shown in above figures The
results are similar to-tlie reference signal, and it
K,i
proves that both of algorithms are suitable for controlling DFIG system
0^
(b) Stator power
Fig 9 Responses of system in case 2
(b) Stator power Fig 10 Responses of system in case 3
REFERENCES
1 Lie Xu, Phillip Cartwright, "Direct Active and Reactive Power Control of DFIG for Wind Energy
Generation", IEEE Transactions on Energy' Conversion, Vol 21, No.3 September 2006
2 Rajib Datta, V.T Ranganathan, "Direct Power Control of Grid-connected Wound Rotor Induction Machine Without Rotor Position Sensors", IEEE Transactions on Power Electronics, Vol, 16, No.3, May 200!
3 Jihen Arbi, Manel Jebali-Ben Ghorbal, llhem Slama-Belkhodja, Lotfi Charaabi, "Direct Virtual
Torque Control for Doubly Fed Induction Generator Grid Connection", IEEE Transactions on Industrial Electronics, Vol 56, No 10, October 2009
4 Z Mahi, C Serban, H.Siguerdidjane, "Direct Torque Control of a Doubly Fed Induction Generator
of a Variable Speed Wind Turbine Power Regulation", European Wind Energy Conference, Milan, May 2007
5 S Muller, M Deicke, Rik W De Doncker, "Doubly Fed Induction Generator System for Wind
Turbines", IEEE Industiy Application Magazine, May/June 2002
6 C Belfedal, S.Moreau, G.Champenois, T AIlaoui, M.Denai, "Comparison of PI and Direct Power
Control with SVM of Doubly Fed Induction Generator", Journal of Electrical & Electronics Engineering Istanbul University, 2008
7 Luna A, Lima F.K A, Rodriguez P, Watanabe E.H, Teodorescu R, 'Comparison of Power Control
Strategies for DFIG Wind Turbines", Proceedings of The 34th Annual Conference of the IEEE Industrial Electronics Society lECON 200S IEEE 2008
8 F Poitiers, M Machmoum, R Le Doeuff, M.E Zaim, "Control of a Doubly Fed Induction
Generator for Wind Energy Conversion Systems", Australian Universities Power Engineering Conference, 2001
9 D Aouzellag, K Ghedamsi, E.M Berkouk, "Power Control of a Variable Speed Wind Turbine Driving a DFIG", International Conference on Renewable Energies and Power Quality 2006 Author's address.'Phan Quoc Dzung - Tel: 0903657486 - Email: phan_quoc_dung@yahoo com
Faculty of Electrical- Electronics Engineering