1

Voltage Profile Improvement for Distributed

Wind Generation using D-STATCOM
Naruttam Kumar Roy, M. J. Hossain, Student Member, IEEE, and H. R. Pota

 definite conclusion was not given in this paper. A capacity

Abstract-- This paper presents the application of FACTS credit of wind power is calculated in [4]. An overview of
devices for the enhancement of dynamic voltage stability in dynamic models for transient stability analysis of transmission
distribution networks with distributed wind generation. The and distribution systems with DG is given in [5]. The authors
analysis is carried over a test distribution system representative
state that the inclusion of squirrel cage induction generator
of the Kumamoto area in Japan. The detailed mathematical
modelling of the system is also presented. Firstly, this paper (SCIG) does not have a large influence on transient stability.
provides simulation results showing the effects of higher and But it significantly affects the voltage stability of the system
lower penetration of distributed wind generation on the voltage [6], [7]. Problems related to voltage instability in power
dynamics in a faulted system. Then, a distribution static systems are one of the major concerns in power system
synchronous compensator (D-STATCOM) is used to improve the planning and operation [8]. If voltage limits and voltage
voltage profile of the system. This analysis shows that D-
stability are the determining factors for the system operation,
STATCOM has significant performance to improve the voltage
dynamics of distribution system compared to shunt capacitor. additional sources of reactive power can be installed at critical
location in order to smooth the voltage profile and to increase
Index Terms--Distributed generation (DG), distribution the reserves against the loss of voltage stability. Among
system, eigenvalues, wind turbine, shunt capacitor, D- different FACTS devices, static synchronous compensator
STATCOM. STATCOM is being increasingly used for enhancing dynamic
voltage stability. STATCOMs with a suitable control strategy
I. INTRODUCTION have the potential to significantly increase the transient

T HE integration of renewable power offers a promising

solution, and is consequently the subject of increasing
research efforts around the world. During the next ten years,
stability margin as well as voltage stability of the system [9].
Application of D-STATCOM compensator to improve power
quality (PQ) in low voltage (LV) grids with distributed energy
renewable energy will emerge as a major enabler of the smart resources (DERs) is given in [10]. However, the dynamics of
grid for the integration of small and medium sized renewable wind farms are not investigated in this paper.
energy into the electricity grids. Multiple mini-grids, Recent work tends to focus on assessing how distributed
connected into the distribution level network, will form new generation can be integrated into existing grids without
systems exhibiting potentially new collective (emergent) affecting the system stability. However, the effect of a
behaviour. Distributed supply based on renewable energy potentially rising penetration of DG upon the distribution
sources is one main element in all scenarios for the worldwide systems has attracted much less attention. Very little work has
energy supply in the future. Solar energy, wind energy, bio examined its impact on distribution network. The aim of this
fuel, and hydro power will be the basic components of a long- proposed work is to discuss the potential impact of the DG on
term sustainable energy supply. The integration of distributed distribution systems, i.e., how the integration of wind
generators into supply networks has already become a relevant generation affects the behaviour of the distribution system.
issue in many countries. The costs of energy production of Moreover, the effects of voltage rise under fault condition
renewable sources will become more and more competitive as with different penetration of DG are also analysed and to
compared to the increasing cost for fossil fuels in the coming enhance the voltage profile a comparative study between
years. Important developments have occurred in the last few shunt capacitor and D-STATCOM is provided through non-
years in the fields of electrical system technology for linear simulations.
distributed generation [1], [2].
Some analytic and utility case studies on the impacts of II. MATHEMATICAL MODEL
wind power on existing power systems are described in [3]. A The basic features of a distribution system can be
represented through the widely used Kumamoto 15 bus
Naruttam Kumar Roy, M. J. Hossain, and H. R. Pota are with the School distribution test system as shown in Fig. 1 which is a
of Engineering and Information Technology (SEIT), The University of New benchmark system. For the analysis in this paper, it is
South Wales at Australian Defence Force Academy (UNSW@ADFA),
Canberra, ACT 2600, Australia. E-mail: [email protected], M.
[email protected], H. [email protected] .

978-1-4577-1002-5/11/$26.00 ©2011 IEEE

 2

1 2 3 4 5 6

15 14 13 12

11 10 9 8 7

Fig. 1. Single line diagram of Modified Kumamoto 15-bus distribution system [16]

automatic voltage regulator, Xdg is the direct-axis synchronous

modified by connecting a small synchronous generator at bus reactance, X dgc is the direct axis transient reactance,
1 and considering bus 15 connected to the main grid (infinite
Vt c  X dg
( E qg c I qg ) 2 is the terminal voltage of
c I dg ) 2  ( X dg
bus) and a wind generator is also connected at bus 10. Here,
the base power is 10 MVA and the base voltage is 6.6 kV, the generator, Idg and Iqg are direct and quadrature axis
total load on the system is 6.3 MW, 0.0446 Mvar. The line currents of the generator, respectively.
and load data of the test system and machine parameters are
B. Wind Generator
given in Appendix.
The most important components of a constant speed wind
A. Synchronous Generator are rotor, drive train, and the generator. The rotor of the wind
With some typical assumptions, the synchronous generator turbine, with radius Ri , converts the energy from the wind to
can be modeled by the following set of differential equations
[11]: Z m . The power from the
the rotor shaft, rotating at the speed
G Z sZ  Z s (1) wind depends on the wind speed, Vw , the air density, U , and
D Z the swept area Awt . From the available power in the swept
Z  Z  s ( Pm  Eqg c I qg ) (2)
2H 2H area, the power on the rotor is given based on the power
1 coefficient c p (O , T ) , which depends on the pitch angle of
c
E qg c ) I dg ]
[ K A (Vref  V0 )  ( X dg  X dg (3)
Tdc0 the blade, T , and the ratio between the speed of the blade tip
1 Zm R
V0 Vt  V0  (4) and the wind speed ratio, O r
Vw
. The aerodynamic
Tr
where į is the power angle of the generator, Ȧ is the rotor torque applied to the rotor for the turbine by the effective
speed with respect to synchronous reference, H is the inertia wind passing through the rotor is given in [12], [13]:
constant of the generator, Pm is the mechanical input power to U
Tae Awt c p (Or , T )Vw3 (5)
the generator which is assumed to be constant, D is the 2Z m
damping constant of the generator, E qgc is the quadrature-axis
where c p is approximated by the following relation [14]:
transient voltage, KA is the gain of the exciter amplifier, Vref is
the reference voltage, V0 is the output of terminal voltage cp 0.44  0.0167T sin ª« S (O  3) º»  0.00184(O  3)T
(6)
transducer, Tdc0 is the direct-axis open-circuit transient time ¬15  0.3T ¼
constant of the generator, Tr is the time constant of the A two-mass drive train model of a wind turbine generator
system (WTGS) is used in this paper as the drive train
 3

modelling can satisfactorily reproduce the dynamic the direct and quadrature axis transient voltages respectively,
characteristics of WTGS. The drive train is attached to the ids and iqs are the direct and quadrature axis currents
wind turbine converts the aerodynamic torque Tae on the
respectively, and Z s is the synchronous speed.
rotor into the torque on the low speed-shaft, which is scaled
down through the gearbox to the torque on the high-speed C. D-STATCOM
shaft. The first mass term stands for the blades, hub and low- The equivalent circuit of D-STATCOM is shown in Fig.
speed shaft, while the second mass term stands for the high The dynamics of the voltage source is governed by the
speed shaft having the inertia constants H m and H g . The charging and discharging of a large (non-ideal) capacitor.
Firing angle Įl controls active power exchange in the
shafts are interconnected by the gear ratio, N g , combined
converter as well as charge and discharge of dc capacitors and
with torsion stiffness, K s , and torsion damping, Dm and consequently dc-bus voltage ȣdcl.
The dynamics for lth STATCOM can be described by the
D g resulting in torsion angle O . The normal grid frequency following equation [9]:
is f . The dynamics of the shaft can be represented as follows P Q
Qdcl (t )  sl  dcl (15)
[12], [13]: ClQ dcl Rcl Cl
1
Z m >Tae  K s O  DmZm @ (7) for l=1,2, . . . , p, where p is total number of D-
2H m STATCOMs, vdcl is the capacitor voltage, Cl is the dc
capacitor, RCl is the internal resistance of the capacitor, and Psl
Z g
1
2H g
>
K s O  Tae  D g Z g @ (8) is the power supplied by the system to the D-STATCOM to
charge the capacitor. The ac side D-STATCOM
§ · voltage El kl vdcl ‘D l , where Įl the bus angle of the D-
1
O 2Sf ¨ Z m  Z ¸ (9) STATCOM in the reduced network and kl is a constant
¨ Ng
g ¸
© ¹ associated with the inverter of the D-STATCOM.
The induction generator gets the power from the gear box The terminal voltage of D-STATCOM is measured using a
through the stiff shaft. The relationship between the transducer with first order dynamic,
mechanical torque and torsional angle is given by: Q tm Ps
Tm K s O (10) Qtm  l l
 K mlQ tl (16)
l
Tml
The mechanical torque depends on torsion stiffness (Ks),
torsion angle (Ȝ), rotor speed (Ȧg) etc. In this work, time
domain simulation is carried out with full non-linear model of
the wind turbine generation system considering variable
mechanical torque.

A simplified transient model of a SCIG can be described by

the following algebraic-differential equations [15]:

E qrc 
1
Toc
>
E qrc  ( X  X c)ids  sZ s TocE drc @ (11)

E drc 
1
Toc
>
E drc  ( X  X c)iqs  sZ s TocE qrc @ (12)

1
s Tm  Te  (13)
2H g
v ds  jv qs  Rs  jX cids  jiqs   j E qrc  jE drc  (14)
Here, X c X m X r is the transient reactance, R is the
Xs  s
Xm  Xr
stator resistance which is assumed to be zero, X r is the rotor
reactance, X m is the magnetizing reactance, X X s  X m is Fig. 2. D-STATCOM equivalent circuit [9]
the rotor open circuit reactance, Toc is transient open circuit
time constant, Tm is the mechanical torque, s is the slip,
Te E drc ids  E qrc iqs is the electrical torque, Edrc and Eqrc are
 4

where, Q tm l
is the sensor output, Qt l
is the voltage at the and participation factors, it is clear that it is related to reactive
power mismatch.
connection point of D-STATCOM, K ml is a constant, and
TABLEI
Tml is the time constant of the voltage transducer. EIGENVALUES OF DOMINATING MODES

Modes Eigenvalues
1.000 Without Wind Generator 18 -0.16229
With Wind Generator 19 -0.0000113
0.995
TABLE II
Voltage (pu)

PARTICIPATION FACTORS
0.990
Modes Participation factors

0.985 18 ǻEqǯ= 1.0 ǻEdǯ=0.38805

19 ǻEqǯ=0.30

0.980 1.1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1.0
Node
0.9
Fig. 3. Nodal voltages without and with distributed wind
Voltage (pu)

20% Wind penetration

generation
0.8
40

30 0.7

20
0.6
10

0
0.5
0 1 2 3 4 5 6 7 8
-10 Time (sec)
-20 Fig. 5. Voltage profile of bus 10 with 20% wind penetration
-30
1.1
-40

-45 -40 -35 -30 -25 -20 -15 -10 -5 0 1.0

Fig. 4: Eigenvalues of modified Kumamoto 15 bus

distribution test system with wind generation 0.9
Voltage (pu)

50% Wind penetration

III. EIGENVALUES AND PARTICIPATION FACTORS OF THE 0.8

SYSTEM
The modified Kumamoto 15-bus distribution network 0.7

shown in Figure 1 is used to exhibit the effectiveness of the

proposed analysis. Firstly, the wind farm is installed at bus 10 0.6
to provide 20 percent of the total power. The nodal voltages
without and with installing DGs are given in Figure 3. The 0.5
0 1 2 3 4 5 6 7 8
integration of wind generator decreases the voltage at the
Time (sec)
adjacent nodes. The eigenvalues of the system with wind
generation are shown in Fig. 4. From the eigenvalues it is Fig. 6. Voltage profile of bus 10 with 50% wind penetration
found that modes 18 and 19 make the system marginally
IV. DYNAMIC CHARACTERISTICS OF THE DISTRIBUTION
stable. The eigenvalues and participation factors of these two
SYSTEM
modes are given in Table I and Table II, respectively.
From Table-II it is clear that direct and quadrature axis To analyse the voltage profile, the test system is tested with
transient voltages are responsible for this situation. Here, the different penetration levels of wind generation under fault
marginally stable modes are monotonic. From the eigenvalues condition with outage of the line connecting buses 3 and 12
for 2 sec. A fault is applied at t = 1 second and removed at t =
 5

3 second. The terminal voltage of wind generator for 20%, V. CONCLUSION

50%, and 80% wind generation penetration under pre-fault, The application of D-STATCOM to improve the voltage
faulted, and post-fault conditions are shown in Fig.5, Fig. 6, profile of a distribution network with distributed wind
and Fig. 7, respectively. These show that higher penetration of generation is investigated in this paper. The impacts of wind
distributed wind generation causes significant voltage rise and farm dynamics on a distribution network are also analysed,
oscillations in distribution system. However, in this analysis, which shows that the integration of wind generation
the system returns to its pre-fault condition in case of high significantly affects the voltage stability of the system. The
penetration also. In other cases, it may become unstable with analysis also shows that the high penetration of DG in the
high penetration in case of shortage of reactive power. It also distribution network has a significant impact on the voltage
depends on the operating conditions, system configuration etc. rise. Finally, it can be said that D-STATCOM can be used as
an effective device to enhance the voltage profile of the
1.1 distribution networks with distributed wind generation.

1.0
VI. APPENDIX
APPENDIX A
0.9 TABLE III
Voltage (pu)

LINE AND LOAD DATA OF KUMAMOTO DISTRIBUTION SYSTEM [17]

0.8 80% wind penetration
Sending Ending R X B Pload Qload
Node Node (p.u) (p.u) (p.u) (p.u) (p.u)
0.7
1 2 0.00315 0.075207 0.00000 0.02080 0.0021

0.6 2 3 0.00033 0.001849 0.00150 0.04950 0.0051

3 4 0.00667 0.030808 0.03525 0.09580 0.0098
4 5 0.00579 0.014949 0.00250 0.04420 0.0045
0.5
5 6 0.01414 0.036547 0.00000 0.01130 0.0012
0 1 2 3 4 5 6 7 8 4 7 0.00800 0.036961 0.03120 0.06380 0.0066
Time (sec) 7 8 0.00900 0.041575 0.00000 0.03230 0.0033
8 9 0.00700 0.032346 0.00150 0.02130 0.0022

Fig. 7. Voltage profile of bus 10 with 80% wind penetration 9 10 0.00367 0.01694 0.00350 0.02800 0.0029
10 11 0.00900 0.041575 0.00200 0.21700 0.0022
3 12 0.02750 0.127043 0.00000 0.01320 0.0014
1.1
12 13 0.03150 0.081405 0.00000 0.00290 0.0003
13 14 0.03965 0.102984 0.00000 0.01610 0.0016
1.0
14 15 0.01061 0.004153 0.00000 0.01390 0.0014

0.9
Bus voltage=6.6 kV, Base MVA= 10 MVA
Voltage (pu)

0.8 APPENDIX B
GENERATOR AND INDUCTION MOTOR PARAMETERS
Without Capacitor or D-STATCOM
0.7 With Shunt Capacitor
The machine parameters used for the system are given below:
With D-STATCOM
Synchronous generator parameters:
0.6
X d =0.1, X q =0.69 X dc =0.031, Tdc0 =5s, H=0.04s, D=2
Automatic Voltage Regulator (AVR) Parameters:
0.5
KA = 5, Tr = 0.1
Induction generator parameters:
0 1 2 3 4 5 6 7 8 Power: 2 MW, Rs=0.0121 pu
Time (sec) Voltage: 690 V, Xs=0.0742 pu
Fig. 8. Voltage profile of bus 10 for 80% wind penetration Frequency, f =50 Hz , Xm=2.7626 pu
Self-damping: 0.008 pu, Rr=0.008 pu
without and with compensating devices Rated slip: 0.02, Xr=0.1761 pu
Two mass model Parameters:
To improve the voltage profile a shunt capacitor is Hm=2.6 s, Hg=0.22 s, Dm=3 pu, Ks=141 pu, Gearbox ratio: 23.75
connected at bus 10 with 80% wind penetration. From Fig. 8, D-STATCOM Parameters:
Capacity: 5 MVA
it is seen that still there exists a large voltage dip in the voltage C=300ȝF
profile. A D-STATCOM is connected at bus 10 to make a RC=0.01 pu
comparison with the performance of shunt capacitor. Fig. 8
shows the superiority of D-STATCOM to enhance the voltage VII. REFERENCES
compared to shunt capacitor. Similar results are true for [1] W. Freitas, J.C.M. Vieira, A. Morelato, L.C.P. da Silva, V.F. da Costa,
different level of wind penetration also. and F.A.B. Lemos, “Comparative analysis between synchronous and
induction machines for distributed generation applications,” IEEE
Transactions on Power Systems, vol. 21, no. 1, pp. 301–311, 2006.
 6

[2] Stratis Tapanlis and Michael Wollny, “Advanced active and reactive [10] Irena Wasiak, Rozmyslaw Mienski, Ryszard Pawelek, Piotr Gburczyk,
power control for mini grids,” World Climate & Energy Event, pp.111- “Application of DSTATCOM compensators for mitigation of power
117, Rio de Janeiro, Brazil, 2009. quality disturbances in low voltage grid with distributed generation,” in
[3] B. Parsons and M. Milligan, “Grids impacts of wind power: summary of Proc. 9th International Conference on Electrical Power Quality and
recent studies in the United States,” NREL/CP-500-34318, National Utilisation, 9-11 October, 2007, Barcelona.
Renewable Energy Laboratory, 2003. [11] A. R. Bergen, Voltage Stability of Electric Power Systems, New Jersey:
[4] Yi Zhang, Sadrul Ula and Yibin Zhang, “Wind power availability and Prentice-Hall, 1986.
increased capacity credit for multiple wind farms,” IEEE PES General [12] T. Ackermann, Wind Power in Power Systems, John Wiley & Sons Ltd,
Meeting, Minneapolis, Minnesota, USA, 2010. England, 2005.
[5] J. C. Boemer, M. Gibescu, W.L Kling, “Dynamic models for transient [13] M. J. Hossain, H. R. Pota, V. Ugrinovskii, and R. A. Ramos, “A novel
stability analysis of transmission and distribution systems with STATCOM control to augment LVRT capability of fixed speed wind
distributed generation: an overview,” IEEE Bucharest Power Tech generators,” in Proc. 48th IEEE Conference on Decision and Control,
Conference, pp. 1-8, Bucharest, Romania, 2009. Shanghai, P. R. China, pp. 7843-7848, 2009.
[6] Genevieve Coath and Majid Al-Dabbagh, “Effect of Steady-State Wind [14] E. S. Abdin and W. Xu, “Control design and dynamic performance
Turbine Generator Models on Power Flow Convergence and Voltage analysis of wind turbine-induction generator unit,” IEEE Trans. on
Stability Limit,” Australasian Universities Power Engineering Energy Conversion, vol. 15, no. 1, pp. 91-96, 2000.
Conference (AUPEC 2005), Tasmania, Australia, 2005. [15] C. W. Taylor, Power System Voltage Stability, New York: McGraw-Hill,
[7] Ching-Yin Lee, Li-Chieh Chen, Shao-Hong Tsai, Wen-Tsan Liu, and 1994.
Yuan-Kang Wu, “The Impact of SCIG Wind Farm Connecting into a [16] Fracisco M. González-Longatt, “Impact of distributed generation over
Distribution System,” Asia-Pacific Power and Energy Engineering power losses on distribution system,” in Proc. 9th International
Conference (APPEEC 2009), pp. 1-7, China, 2009. Conference on Electrical Power Quality and Utilisation, 9-11 October,
[8] Per-Anders LÖf, David J Hill, Stefan Arnborg, and GÖran Andersson, 2007, Barcelona.
“On the analysis of long-term voltage stability,” International Journal of [17] S. Li, K. Tomsovic, and T. Hiyama, “Load following functions using
Electrical Power & Energy Systems, vol. 15, no. 4, pp. 229-237, 1993. distributed energy resources,” in Proc. of the IEEE PES Summer
[9] M. J. Hossain, H. R. Pota, and C. Kumble, “Decentralized robust static Meeting, vol. 3, pp. 1756 – 1761, 16-20 July 2000, Seattle, WA , USA.
synchronous compensator control for wind farms to augment dynamic
transfer capability,” Journal of renewable and sustainable energy, vol.
2, , pp. 022701-20, 2010.