Proceedings of the IEEE

International Conference on Automation and Logistics

August 18 - 21, 2007, Jinan, China

The Dynamic Model of Doubly-fed Induction Generator

Based on Wind Turbine
Xingjia Yao, Hongxia Sui and Zuoxia Xing Dayong Liu
Wind Power Technique Institute Electric Power Department of Changchun
Shenyang University of Technology
Shenyang City ,Liaoning Province, China Changchun city of Jilin Province
[email protected] [email protected]
Abstract- According to grid codes issued by utilities, tripping When the voltage is in the shadow area, the turbine should
of wind turbines following grid faults is not allowed, and it is supply reactive power.
necessary to provide reactive power support to grid sometime. So
this paper sets up a complete DFIG dynamic model, including the
generator model, converter model and converter protection, grid
model etc. to study the transient characters. The method of
sensorless direct torque control (DTC) and model referencing
adaptive system (MRAS) method were used in the converter
control. At last, the results of simulation on DFIG transient
power control performance based on 1.5MW SUT-1500 wind
turbine prototype indicated that the DTC control and the MRAC
principle improve the system phase-ground fault ride-through
capacity ,and fulfil the dynamic regulating of the DFIG external
fault .

Index Terms- DFIG, Dynamic model, DTC

I. INTRODUCTION
Fig.1 E.On Netz requirements for wind farm behavior during faults
The worldwide are all attach important to the
development of wind power, so the wind turbines technology
development is at very fast speed, and the scale of the wind Then we must get to know the transient character of
farms are also increasing. The nation “tenth-five year” DFIG when its connected network is fault. So it is important
programming and 2015 year future layout have indicated the to set up the dynamic model of a DFIG. For a dynamic model
distance market in China, and its wind power installed of DFIG ,the control of its converter is the key.[3,4,5] only
capacity will be 1260 MW by the end of 2010 year and give the rotor side converter model of DFIG or the certain part
40000MW by the end of 2020 year>@. control, these can not depict completely the whole dynamic
With the development of the wind farm, its proportion in course of DFIG. In this paper, the dynamic model of the
the power system is increasing, its running performance will DFIG, the control model of rotor side converter and stator side
be important to the system .For example, the active power and converter for DFIG will be set up respectively, as well as the
reactive power of the fixed-speed wind turbines with the protection of the converter. Then a complete model of DFIG
induction generator can not be controlled when the network will be founded in PSCAD platform. The DTC control
faults, and its disconnect will impact the network’s resume, strategy and model referencing adaptive system ˄MRAS˅
even aggravate the network breakdown. Even if the wind
are used in the control of rotor side converter., and the
turbines with DFIG can improve the performance of fault,
voltage-current doubly loop control is used in the stator side
whereas its drawback make it difficult to connect the
converter. At last, the simulation based on the SUT-1500 will
transmission networks considering the stability and security of
prove the model’s validity.
the power system ,especially for the large MW capacity wind
farms with the DFIG. So the transmission system operators II. WIND TURBINE MODEL
have already been revising their grid codes for the connection
The wind turbines with DFIG are one of the mainstream
of large wind farms in outside countries. In Northern German,
type of the wind turbine. One of the advantages is the
where the concentration of wind turbines is high, the grid
converter capacity, it only be 1/3 generator power, reduce the
operator (E. On Netz) already has set requirements for the
cost of the system, but improve the electric power quality>6@.
behaviour of wind turbines. Instead of disconnecting them
from the grid, the turbines should be able to follow the The DFIG block configuration is shown in Fig.2>7@.
characteristic shown in Fig.1>[email protected] when the griG voltage
goes below the curve, the turbine is allowed to disconnected.

1-4244-1531-4/07/$25.00 © 2007 IEEE. 1023

 The function of AMC unit is to calculate measured torque
according to
3
T p n (<dr i qr  <qr idr ) (4)
2
The function of UCT unit is the transformation from -
coordinates reference frame to a-b-c there phase coordinates
reference frame.

Fig.2 The DFIG basic configuration

A. DFIG Model
DFIG model has been depicted in many papers. So here
the doubly fed induction generator based on variable-speed
wind turbines use dynamic fifth-order model of induction
machine in a two-axis d-q reference frame rotating at
synchronous speed. This five order model need assumed
following>8@:
1) The equations are derived on the synchronous reference
frame using direct(d) and quadratere(q) axis representation .
2) The q-axis is assumed to be 90 degrees ahead of the d-
axis in the direction of rotation. Fig.3 Doubly-fed VSCF wind turbine sensorless DTC control scheme

3) The q component of the stator voltage ( s ) used within

v Flux controller DMC can generate IGBT bridge switch
the model is equal to the real part of the generator busbar signal logic state through 3 flux trigger comparator,
voltage. controlling rotor side bridge circuit and output the
The five order function are : corresponding space voltage vector time sequence diagram.
The function of AZS unit is selector of zero vector state,
­ 1 d
°°vds Rs ids \ qs  \ ds takes the pole of insert zero vector to control flux motion
Z s dt average speed, thus control torque output.
® 1 d
(1) Torque regulator ATR is one tolerance Schmitt trigger.
°vqs Rs iqs  \ ds  \ qs According to the compare of torque reference and feedback, it
°¯ Z s dt can control the torque switch on DMC or AZS zero state.
regulating the switch on the work state and zero state by
­ 1 d
°°vdr Rr idr  s\ qr  \ turns ,and the rotor flux motion can be run a little then stop a
Z s dt dr little, keep the torque within dynamic fluctuating tolerance,
® 1 d
(2)
thus the torque is controlled.
°vqr Rr iqr  s\ dr  \ qr The torque reference is selected from the look-up table
°¯ Z s dt based on wind turbine optimum power output curve,
B. Rotor Side Converter Control Model Maximum Power Point Tracking (MPPT) principle and
In this paper, the converter control strategy used the direct current generator speed.
torque control (DTC) and the speed calculated used the model The measured generator speed is sent to DMC unit and to
referencing adaptive system (MARS) . judge whether in sub-synchronous or in super-synchronous
The configuration diagram of converter DTC no speed state.
sensor control system is showed in Figure 3, the function of The speed can be calculated by the model referencing
adaptive system (MRAS), where the output of a reference
each unit is described as following>9,10,11@:
model is compared with the output of an adjustable or
The function of flux model unit AMM is to calculate rotor
flux - reference coordinates components according to the adaptive model until the errors between the two models vanish
to zero. A block diagram for speed estimation by the MRAS
following formula, and integrated step by step.
­<dr
technique is shown in Fig 4. Consider the voltage model’s
° ³ edr dt ³ (u dr  i dr Rr )dt (3) stator-side equations, the reference model can be defined
®
°̄<qr ³e qr dt ³ (u qr  i qr Rr )dt as>12@:

1024
ª\ dr º Lr ­°ªv dr º ª( Rs  VLs s ) 0 º ªidr º ½° (5)
« » ®« »  « « »¾
«¬\ qr »¼ Lm °̄«¬v qr »¼ ¬ 0 ( Rs  VLs s )»¼ «¬iqr »¼ °¿
This model receives stator voltage and current signals and
calculates the rotor flux vector signals.
The adaptive model of the current model flux equation can
be defined as:
ª 1 º
« T  Zr »
ª\ dr º ª\ dr º L ªidr º
« » « r »« »  m « » (6)
¬«\ qr ¼»  » ¬«\ qr ¼» Tr ¬«iqr »¼
«Z 1
«¬ r Tr »¼
This model can calculate fluxes from the input stator
currents only if the speed signal Z r is known. According to Fig 5 The variable speed regulated pitch wind turbine speed control curve.
the correct speed signal , ideally, the fluxes calculated from
the reference model and those calculated from the adaptive D Grid Side Converter Control Model
model should be match each other, that is
\
, dr
\ˆ dr and As the rotor side VSC, it requires a dc power supply. The dc
\ qr \ˆ qr \ˆ
dr and qr \ˆ voltage is usually generated using another voltage sourced
,where are the adaptive model converter connected to the ac grid at the generator stator
outputs. An adaptation algorithm with P-I control, as indicated terminals. The grid PWM converter is operated so as to keep
below, can be used to tune the speed Ẑ r so that the error
the dc voltage on the capacitor shown in Fig.2 at a constant
value. In effect ,this means that the grid side converter is
[ 0. supplying the real power demands of the rotor side converter.
And a feedback controller is used in which the error between
the desired and ordered currents is passed through a
proportional-integral controller which controls the output
voltage of conventional Sinusoidal PWM converter. Its
configuration is shown in Fig.7[13].

Vdc

* Voltage
V dc outloop Active
controller current Current Voltage Pulse
command Inloop command signal
SPWM
Reactive controller
current
command

Fig 4 Speed estimation by model referencing adaptive control (MRAC) Current fed
principle

Fig.7 Grid side PWM converter control system configuration

C. Rotor Speed Controller Model
About the rotor speed control principle, the Fig 5 can E. DFIG Protection System Model
show the detail. In low winds it is maximize energy capture
following an optimum tip speed ratio load line and get a The goal of the protection system is to protect the wind
maximum power coefficient. The load line is a quadratic turbine form damage caused by the high current that can occur
curve on the torque speed plane which shown by BC. when the terminal voltage drops as a result of a short circuit in
Alternatively a look-up table may be specified. In high wind the grid. It also has a task of preventing islanding. The anti-
speed, above rated speed, the blade pitch is adjust to maintain islanding protection of the wind turbine acts in response to
the chosen operating point, once rated torque is reached at voltage and/or frequency deviations or phase angle frequency.
point H, the torque demand is kept constant for all high wind The grid side of the converter measures the grid voltage with a
speed, and pitch control regulated the rotor speed. A small high sampling frequency. There are criteria implemented in
margin should be allowed between points H and D to prevent the protection system for determining whether an island
excessive mode switching between below and above rated exists. If these criteria met, the wind turbine is disconnected
control modes. [14].
D. Power System Model

1025
 Based on the assumption of single machine connected to is the dc voltage between rotor side converter and grid side
the infinite power system, the model can be presented as converter side converter. It can be seen form the result of the
following Fig.6>15@. and here the infinite system is assumed simulation that dynamic characters of DFIG. When the extern
10KV voltage level, its neutral point is grounded . fault occurred from the grid, the grid side converter can ensure
the DC voltage invariable , and the DFIG is not only
disconnected from the grid ,but also to supply a certain
reactive power to sustain the voltage. That is to say this
dynamic model meet the requirement of the Fig.1. it can ride-
through the voltage dip caused by external phased-ground in
0.5 seconds.

Fig 6 Power system model

III. MODEL SIMULATION

After building the above wind turbine model, first do

simulation based on the SUT-1500 prototype doubly-fed
generator wind power system design parameters, funded by
MOST “863” program “Mega-watt VSCF wind turbine”,
The parameters used in process of simulation are presented
in Table I.
TABLE I
PARAMETERS USED IN DOUBLY-FED GENERATOR WIND POWER
SYSTEM SIMULATION
Doubly-fed Rated power 1500kW
generator Rated stator voltage 690V AC
Frequency 50Hz
Rated stator current 1284A
Fig.7 The DFIG terminal voltage and current when the connected network
Rated speed 1810rpm
fault in A phase
Synchronous speed 1500rpm
Speed range 900-2000rpm
Equivalent circuit R1 0.0018Ohm
parameters X1 0.0227Ohm
Rm 0.05Ohm
Xm 1.37Ohm
X2' 0.0368Ohm
R2' 0.00206Ohm
Transformer Transformer nominal 2MW
power
Transformer winding 1 Ph-ph Vrms :690V
R1:0.0027 p.u.
L1:0.08 p.u.
Transformer winding 2 Ph-ph Vrms :10.5kV
R1:0.0027 p.u.
L1:0.08 p.u. Fig.8 The DC voltage of converters

The simulation analysis was carried out in PSCAD IV. CONCLUSION

system simulator. The system simulator was running with the In this paper, the model of DFIG and converters system
time step 't 20ns , and the fault is in the A phase at the are set up ,and the control of rotor side converter uses
fifth second , and the duration of fault is 0.5 second. The sensorless DTC and MRAS method. The simulation of the
simulation result is shown in Fig.7 Fig.8 and Fig.9. transient analysis when the power system takeV the phase-
In this simulation, theP3 and Q3 are the active power and ground fault is completed. The result of the simulation
reactive power of the DFIG respectively, and Speed is the indicate that the DTC control and MRAC principle used in the
DFIG’s revolving angle velocity.Tm3 and Ttur3 are the DFIG can improve the power system fault ride-through
electromagnetic torque and machine torque respectively, Vt3 capacity ,that is to say when the system fault is phase ground
is the DFIG transformer network side terminal line-to-line in a little time ,the DFIG can implement the running of
voltage. The Va , Vb and Vc are the phase voltage of DFIG, disconnected network. And this give the theory base for future
study the ride through the network fault to wind farm.
I a , I b and I c are the phase current of the DFIG .and Ecap

1026
 Turbines. IEEE Transactions on Power System, Vol.18, No.2, MAY
2003.
[4] HU Jia-bing, HE Yi-kang. Dynamic Modeling and Robust Current
Control of Wind-turbine Driven DFIG During External AC Voltage Dip. J
Zhejiang University SCIENCE A 2006 7(10).PP: 1757-1764.
[5] A. Perdana, O. Carlson, and J. Persson. Dynamic Response of Grid-
Connected Wind Turbine With Doubly Fed Induction Generator during
Disturbances. Nordic Workshop on Power and Industrial Electronics.
TRONDHEIM-2004.
[6] Yao Xingjia, Liu Guangde, Xing Zuoxia, WANG Chao. “Discussions on
General Design of Large Scale Variable Speed Wind Turbine”[J]. Journal
of Shenyang University of Technology,2006, Vol.28, No.1, pp:196-201.
[7] Johan Morren, Student Member, IEEE, and Sjoerd W. H. de Haan,
Member, IEEE. Ride through of Wind Turbines with Doubly-Fed
Induction Generator During a Voltage Dip. IEEE Transaction on Energy
Conversion, Vol. 20, No. 2, JUNE 2005 , pp:435-441.
[8] O. Anaya-Lara X. Wu P. Cartwright J. B. Ekanayake and N. Jenkins
Performance of Doubly Fed Induction Generator (DFIG) During Network
Faults. Wind Engineering. Vol. 29,No.1,2005, pp:49-66.
[9] Lei Y. ; Mullane A. ; Lightbody G. ; Yacamini R. ;Modeling of the Wind
Turbine With a Doubly Fed Induction Generator for Grid Integration
Studies, IEEE Transactions on Energy Conversion: Accepted for future
publication,Volume PP, Issue 99, 2005 ,pp:1 – 8.
[10]Lascu, Cristian Trzynadlwski, Andrzej M. ,Combining the Principles of
Sliding Mode, Direct Torque Control, and Space-Vector Modulation in a
High-Performance Sensorless AC Drive[J]. IEEE Transactions on
Industry Applications vol.40 (1 ) .
[11]Xing Zuo-Xia, Zheng Qiong-Lin , Liu Guang-De,Yao Xing-Jia. DTC
Control of Doubly-fed Variable Speed Constant Frequency Wind Turbine.
The Press of Liaoning Technical Engineering University.Vol.25,No.4,
Aug.2006. pp:556-559.
[12]Bimal K. Bose. Modern Power Electonics and AC Drives. Pearson
Education North Asia Limited and China Machine Press, 2002, pp:386-
393.
[13]Zhao Rende. Investigation of the AC-Excited Power Supply for a
Variable-Speed Constant-Frequency Wind Energy Generation System
with DFIG . Zhejiang University PH.D. Dissertation. May,2005.
[14]Thomas Ackermann. Wind Power in Power Systems. John Wiley & Sons,
Ltd.2005.pp: 567-574.
[15]Ekanayake, J.B.; Holdsworth, L.; XueGuang Wu; Jenkins, N.; Dynamic
modeling of doubly fed induction generator wind turbines,Power Systems,
IEEE Transactions on Volume 18, Issue 2, May 2003, pp:803 – 809.

Fig.9 The simulation result of the DFIG connected network fault

REFERENCES

[1] http://wap.cfi.net.cn/newspage.aspx?id=20070309001595&AspxAutoDet
ectCookieSupport=1.
[2] Holdsworth, L., Charalambous, I., Ekanayake, J.B. and Jenkins, N. Power
system fault ride through capabilities of induction generator based wind
turbines Wind Engineering.Vol. 28, No. 4, 2004 , PP: 399-412.
[3] Janaka B. Ekanayake, Lee Holdsworth, XueGuang Wu, and Nicholas
Jedkins. Dynamic Modeling of Doubly Fed Induction Generator Wind

1027