Energy and Power Engineering, 2013, 5, 418-422
doi:10.4236/epe.2013.54B081 Published Online July 2013 (http://www.scirp.org/journal/epe)
Small-Signal Stability Analysis of Wind Power System
Based on DFIG
Bin Sun, Zhengyou He, Yong Jia, Kai Liao
School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China
Received April, 2013
ABSTRACT
This paper focuses on the small-signal stability of power syste m integrated with DFIG-based wind farm. The model of
DFIG for small-signal stability analysis has built; the 3-generator 9-bus WECC test system is modified to investigate
the impacts of large scale integration of wind power on power system small-signal stab ility. Different oscillatory modes
are obtained with their eigenvalue, frequency and damping ratio, the results from eigenvalue analysis are presented to
demonstrate the small-signal stability of power system is enhanced with the increasing output of the wind farm.
Keywords: Small-Signal Stability; DFIG; Wind Power; Power System; Wind Farm
1. Introduction
As of the end of 2011, the installed capacity of wind
power in China has reached 62.36 GW [1], the planning
and construction of large scale wind farm will be the
inevitable trend of the development of wind power in
China. Wind power has obvious characteristics of ran-
domness, volatility and intermittent, this development
affects the small-signal stability of the traditional power
system, so it is significant to study the impacts of large
scale integration of wind power on power system
small-signal stability.
In [2], an aggregated wind farm model with DFIG
based wind turbine for small-signal stability study is
proposed to deal with the situation that the wind turbin es
operate receiving different incoming wind speeds. In [3],
the 10-generator 39-bus New England test system is ap-
plied to assess the effect of the large scale wind farm on
power system small-signal stability. In [4], the effect of
wind power on the oscillations is investigated by gradu-
ally replacing the power generated by the synchronous
generators in the system by power from either constant or
variable speed wind turbines, while observing the
movement of the eigenvalues through the complex plane.
The aim of this paper is to study the small-signal sta-
bility of wind power system based on DFIG. This paper
is structured as follows, with DFIG model in section II,
and section III describe the modified 3-generator 9-bus
WECC system applied in the simulation and analysis.
The small-signal stability analysis of DFIG based wind
farm is described in section IV, with discussions and
conclusions presented in section V.
2. Modeling of DFIG
There are mainly three kinds of wind turbine widely used
nowadays: the constant speed wind turbine with squirrel
cage induction generator, the variable speed wind turbine
with doubly fed induction generator (DFIG) and the di-
rect drive synchronous generator. The main differences
between the three schemes are the generating system and
the way in which the aerodyna mic efficiency of the rotor
is limited during wind speed variations [5, 6]. As a mat-
ter of fact, each of the wind generating systems has its
own advantages and disadvantages. Almost all newer
larger wind generating system being produced are vari-
able speed systems based on DFIGs, so DFIG based
wind farm is the main object in this study.
As is well known, the rotor terminals of a DFIG are
fed with a symmetrical three-phase voltage of variable
frequency and amplitude fed through a voltage source
converter usually equipped with IGBT based power elec-
tronic circuitry [7]. The basic topolog y of DFIG is shown
in Figure 1.
2.1. Turbine and Drive Train
The mechanical power input to the wind turbine is con-
sidered as constant, so the pitch angle do not change
during the period of study. In this paper, the two-mass
drive train model [8] is considered and the dynamics can
be expressed by the following d ifferential equations [9]:

1
2
r

s
he
g
dTT
dt H


(1)

t
bt r
d
dt



  (2)

B. SUN ET AL. 419

1
2
t
msh
t
dTT
dt H



(3)
where ωr is the generator angular speed, ωt is the wind
turbine angular speed, ωb is the electrical base speed, θt is
the shaft twist angle, Hg is the inertia constant of the ge-
nerator, Ht is the inertia constant of turbine, Tsh is the
shaft torque, Te is the electrical torque, Tm is the me-
chanical torque.
2.2. Generator
The most common way of representing DFIG for the
purpose of simulation and control is in terms of direct
and quadrature axes (dq axes) quantities, which form a
reference frame that rotate synchronously with the stator
flux vector [10]. The various variables are defined as:
dsmrr s qr
eK

 , qsmrr s dr
eK

, 2
()
s
ssm rr
LL LL
 ,
rrrr
TLR, mrrm rr
K
LL, e sb

 
e, and 12s. Here, ds and qs are d-axis and
q-axis voltage behind transient reactance, respectively,
dr
2
2mrr r
RKR
RRRe 

and qr

are d-axis and q-axis rotor fluxes, respec-
tively, Lss is the stator self-inductance, Lrr is the rotor
self-inductance, Lm is the mutual inductance between
rotor and stator, Rr is the rotor resistance, and Rs is the
stator resistance.
For balanced and unsaturated conditions, the corre-
sponding DFIG model can be expressed as [10]:
1
1
ss dsr
dsssqsds
e
qsdsmrr dr
rs
Ldi Ri Lie
dt
evKv
T
s






 
 

(4)
1
1
qs
ss r
qsss dsqs
e
dsqsmrr qr
rs
di
LRi Lie
dt
evKv
T





s



 


(5)
2
11
1
ds r
qsdsqsmrr qr
erss
de RieeK v
dt T



 

 (6)
2
11
1
qs r
dsqsdsmrr dr
erss
de RieeK v
dt T



  

 (7)
where ids and iqs are d-axis and q-axis stator currents,
respectively.
2.3. Converter
The converter model in DFIG comprises of two pulse-
width modulation invertors connected back to back via a
dc link. The rotor side converter (RSC) and the grid side
converter (GSC) act as a controlled voltage sources. The
RSC injects an ac voltage at slip frequency to the rotor,
whereas the GSC injects an ac voltage at power fre-
quency to the gr id and maintains the dc link vo ltage con-
stant. The power balance equation for the converter
model can be written as follows:
rgd
PPP
c

 (8)
where Pr, Pg and Pdc are the active power at RSC, GSC
and dc link, respectively, which can be expressed as fol-
lows:
rdrdrqrq
Pvivi
r

 (9)
g
dg dgqg qg
Pvi vi

 (10)
dc
dcdc dcdc
dv
Pvi Cv
dt
 (11)
where vdr and vqr are d-axis and q-axis rotor voltages,
respectively, idr and iqr are d-axis and q-axis rotor cur-
rents, respectively, vdg and vqg are d-axis and q-axis volt-
age of the GSC, respectively, idg and iqg are d-aixs and
q-axis of the GSC, respectively, vdc and idc are the voltage
and current of the dc link capacitor, respectively, and C
is the capacitance of the dc capacitor. The direction of
the currents and power flow are demonstrated in Figure
1.
2.4. Controllers for DFIG
As mentioned above, there are two back to back convert-
ers named RSC and GSC in DFIG system, some meas-
ures should be taken to control the tw o converters so that
the DFIG can works in the appropriate condition. There
are many kinds of control strategy we can choose to con-
trol the RSC and GSC, in this paper, we adopt the model
of controllers built in [1 1], in which a decou pling control
for the active power and reactive power of DFIG was
developed.
3. Generic Test System
In this study, the 3-generator 9-bus WECC test system
has been modified to assess the impact of a wind power
plant based on DFIG on small-signal stab ility. The single
line diagram is shown in Figure 2. The system contains 3
synchronous generators and 3 constant impedance loads.
The large scale wind farm selected for simulation and
Figure 1. The basic topology of DFIG.

B. SUN ET AL.
420
Figure 2. Single line diagram of the analyzed case network.
analysis consists of 50 DFIG-based wind turbines and the
capacity of each turbine is 1.5 MW, so the total capacity
of this wind farm is 75 MW. Here, a single equivalent
model was used to represent all individual units within
the wind power plant in order to avoid increasing com-
putation time, so the modified WECC test system con-
tains 3 synchronous generators and a wind turbine repre-
sented by a DFIG.
The base capacity and frequency of the system are 100
MVA and 50 Hz. The active power and reactive power of
3 loads are adjusted, the active power and reactive power
of bus 5 are 1.200 and 0.075 pu respectively, the active
power and reactive power of bus 6 are 0.700 and 0.070
pu respectively, the active power and reactive power of
bus 8 are 2.250 and 0.350 pu respectively. In order to
ensure the power flow of the system unchanged, the out-
put of synchronous generator G3 should be adjusted at
the same time when increasing the output of wind farm.
The single wind turbine parameters are as follows:
UN=0.69 kV, SN=1.5 MW, f=50 Hz, Rs=0.01 pu, Xs=0.10
pu, Rr=0.01 pu, Xr=0.08 pu, Xm=4 pu, Hm=3 kWs/kVA,
Kp=10, Tp=3 s, Kv=10, TE=0.01s.
4. Small-Signal Stability Analysis
Small-signal stability is the ability of the power system to
maintain synchronism when subjected to small distur-
bances [12]. The disturbance is considered small enough
to permit the equations that describe the resulting re-
sponse of the system to be linearized and expressed in
state-space form. Then, by calculating the eigenvalues of
the linearized model, the small-signal stability character-
istics of the system can be evaluated. In today’s practical
power systems, the small-signal stability problem is usu-
ally on of insufficient damping of system oscillations.
Small-signal stability analysis using linear techniques
provides valuable information about the inherent dy-
namic charateristic of the power system and assists in its
design.
The most direct way to assess small-signal stability of
the model of power system is via eigenvalue analysis.
When the wind farm output is 60% of its total capacity,
i.e., the output of wind farm is 45 MW, at the same time,
adjust the output of G3 to 40 MW in order to keep the
power flow constant. The eigenvalues associated with
this case are given in Table 1. We can get 8 oscillatory
modes, most of their oscillatory frequencies are between
0.1 Hz and 2.5 Hz, so most of them are belong to low-
frequency modes. As a matter of fact, the low-frequency
oscillation is of great harm to the safe and stable opera-
tion of the power system, so these low-frequency oscilla-
tion modes must be suppressed or eliminate by taking
some efficient measures.
The eigenvalues will change with the increase of the
output of wind farm. Since the damping ratios of mode I
and mode II are relatively small, take them as an example,
the eigenvalue, and frequency and damping ratio of mode
I and mode II shifting with the output of wind farm va-
rying are given in Table 2 and Table 3, the eigenvalue of
mode I and mode II shifting with the output of wind farm
increasing are shown in Figure 3 and Figure 4.
We can learn from two tables and two figures above,
with the increase of the output of wind farm, the eigen-
values move to the left on the complex plane, the damp-
ing ratios become bigger and bigger, i.e., the small-signal
stability of the power system is constantly enhanced.
When the wind far m output at rated power, the system is
the most stable. When the wind farm output below rated
power, the larger the difference between output power
and rated power, the more unstable the system is.
5. Conclusion
In this paper, we build the model of DFIG for small-
signal stability an alysis and modify the 3-generator 9-bus
WECC test system to study the impact of large scale in-
tegration of wind power on power system small-signal
stability first. Then, different oscillatory modes are got
by us with their eigenvalue, frequency and damping ratio.
Table 1. Main eigenvalues, frequenc ies and damping Ratios
of the system.
Oscillatory
Mode No.Eigenvalue Frequency ( Hz ) Damping ratio ( % )
1 －1.168±j10.764 1.713 10.788
2 －0.268±j7.628 1.214 3.511
3 －5.472±j7.935 1.263 56.770
4 －5.171±j7.794 1.240 55.285
5 －5.240±j7.891 1.256 55.319
6 －3.647±j0.633 0.101 98.527
7 －0.492±j1.082 0.174 41.393
8 －0.463±j0.717 0.114 54.247

B. SUN ET AL. 421
Table 2. The eigenvalue, frequency and damping ratio of
mode I shifting with the output of wind farm varying.
Wind Farm
Output ( % ) Eigenvalue Frequency of oscilla-
tion ( Hz ) Damping Ratio ( % )
0 －0.667±j11.159 1.776 5.967
10 －0.726±j11.105 1.767 6.524
20 －0.795±j11.045 1.758 7.179
30 －0.875±j10.979 1.747 7.945
40 －0.965±j10.908 1.736 8.812
50 －1.064±j10.835 1.724 9.773
60 －1.168±j10.764 1.713 10.788
70 －1.271±j10.700 1.703 11.796
80 －1.364±j10.647 1.695 12.707
90 －1.440±j10.609 1.688 13.450
100 －1.528±j10.565 1.681 14.314
Table 3. The eigenvalue, frequency and damping ratio of
Mode II shifting with the output of wind farm varyi ng.
Wind Farm
Output ( % ) Eigenvalue Frequency of oscillation ( Hz ) Damping Ratio (%)
0 －0.223±j7.648 1.217 2.915
10 －0.226±j7.643 1.216 2.956
20 －0.231±j7.639 1.216 3.023
30 －0.238±j7.635 1.215 3.116
40 －0.247±j7.631 1.215 3.235
50 －0.257±j7.629 1.214 3.367
60 －0.268±j7.628 1.214 3.511
70 －0.279±j7.629 1.214 3.655
80 －0.288±j7.632 1.215 3.771
90 －0.296±j7.636 1.215 3.873
100 －0.306±j7.640 1.216 4.002
Figure 3. The eigenvalue of mode I shifting with the output
of wind farm incre asing.
Figure 4. The eigenvalue of mode II shifting with the output
of wind Farm varying.
At last, we study the relationship between the output of
wind farm and the small-signal stability of power system,
and get some preliminary conclusions. Since the 3-gen-
erator 9-bus WECC test system is just a small system,
further research should be done in some larger systems.
6. Acknowledgements
The authors would like to thank the National High-tech
Research and Development Program of China. (Grant No.
2012AA050208) for financial support.
REFERENCES
[1] China Wind Energy Association, “Statistics reports on
China wind power construction achievements in 2011,”.
[2] H. Wang, Y. Zhang and Q. Zhou, “Wind Farm Model
with DFIG for Small Signal Stability Study,” Electric
Utility Deregulation and Restructuring and Power Tech-
nologies (DRPT), 2011 4th International Conference on,
6-9 July 2011, pp. 303-307.
[3] C. Wang, L. Shi, L. Yao, L Wang and Y. Ni, “Sma ll Sig-
nal Stability Analysis of the Large-scale Wind Farm with
DFIGs,” Proceedings of the CSEE, Vol. 30, No. 4, 2010,
pp. 63-70.
[4] J. G. Slootweg and W. L. Kling, “The Imapct of Large
Scale Wind Power Generation on Power System Oscilla-
tions,” Electric Power System Research, Vol. 67, No. 1,
2003, pp. 9-20. doi:10.1016/S0378-7796(03)00089-0
[5] L. Erlich and F. Shewarega, “Introduction of Wind Power
Generation into the First Course in Power Systems,”
IEEE Power Engineering Society General Meeting, June
2007, pp. 1-8
[6] J. G. Slootweg, “Wind Power Modelling and Impact on
Power System Dynamics,” Ph,D. Thesis, Technische Un-
iversiteit Delft, Delft, 2003.
[7] L. Erlich, J. Kretschmann, J. Fortmann, S. Muel-
ler-Engelhardt and H. Wrede, “Modeling of Wind Tur-
bines Based on Doubly-fed Induction Generators for
Power System Stability Studies,” IEEE Transactions on
Power Systems. Vol. 22, No. 3, 2007, pp. 909-919.
doi:10.1109/TPWRS.2007.901607

B. SUN ET AL.
422
[8] S. K. Salman and A. L. J. Teo, “Windmill Modeling Con-
sideration and Factors Infulencing the Stability of a
Grid-connected Wind Power Based Embedded Genera-
tor,” IEEE Transactions on Power Systems. Vol. 18, No.
2, 2003, pp. 793-802. doi:10.1109/TPWRS.2003.811180
[9] F. Mei and B. C. Pal, “Modal Analysis of Grid Connected
Doubly Fed Induction Generator,” IEEE Transactions on
Energy Conversion. Vol. 22, No. 3, 2007, pp. 728-736.
doi:10.1109/TEC.2006.881080
[10] F. Mei and B. C. Pal, “Modeling and Small Signal Analy-
sis of a Grid Connected Doubly Fed Induction Genera-
tor,” IEEE Power Engineering Society General Meeting,
San Francisco, 12-16 June 2005,pp. 2101-2108.
[11] F. Wu, X. Zhang, K. Godfrey and P. Ju, “Modeling and
Control of Wind Turbine with Doubly Fed Induction Ge-
nerator” 2006 IEEE PES Power System Conference and
Exposition, 29 October-1 November 2006, pp. 1404-1409.
doi:10.1109/PSCE.2006.296507
[12] P. Kundur, “Power System Stability and Control”
McGraw-Hill, Inc., New York, 1994.