Transient Stability Evaluation of Power Systems

with Large Amounts of Distributed Generation

A.Khosravi M.Jazaeri S.A.Mousavi
University of Semnan, IRAN University of Semnan, IRAN Shiraz Electric Distribution Company, Shiraz, IRAN
Khosravi1983@yahoo.com m.jazaeri@semnan.ac.ir Mousavi.ahmad@live.com

Abstract-It is expected that increasing amount of distributed high control performance, such as fast dynamic response with
generation (DG) will be connected to electrical power systems in low harmonic distortion. Furthermore with this control
the near future. When these generations are in small amounts,
their impacts on power system dynamics will be negligible. But if strategy, the induction generator can operate with a
the penetration level becomes higher, distributed generations comparatively wide slip range, which allows making better
may start to influence the dynamic behavior of a power system use of available wind energy [3].
as a whole and its impacts is no longer restricted to distribution In a traditional power system, synchronous generators have
network. the most important task in the secure operation and the
In this paper transient stability of a power system is stability of power system. During a disturbance, the speed
evaluated when large amounts of wind farm with the DFIG governor and AVR/exciter regulate the frequency (active
technology is connected to the system. We use Digsilent software power) and terminal voltage (reactive power) of the SG
as our simulating program. Simulation studies are carried out to
demonstrate and compare the transient performance of IEEE 10 respectively back to steady state operating point. The stability
machine-39 bus test system with and without wind power of the SG and the power network is therefore maintained.
integration during a severe grid fault. However the controllability of the SG controllers is restricted
Index terms -- transient stability, distributed generation, by the operating limits of the SG [4].
power system But in the networks consists of large amounts of wind
generation, the stability is an important problem. Compared
I. INTRODUCTION to conventional SGs it is commonly known that the Wind
Turbine Generators (WTGs) have a relatively weak capability
The power of wind has been used to serve people from
to ride through grid faults because of the use of induction
ancient time. In the modern world, it has again emerged as
generators and a power electronic interface.
one of the best sources of energy. Without using any kind of
This paper investigates transient stability of a power system
fossil fuels, the wind can produce electricity without any air
based on IEEE 10 machine-39 bus system, in which the SGs
pollution and noise.
are replaced by large wind farms with the same generation
Many countries are now using this kind of electricity
capacity. The penetration level of wind generation is
production and there has been a worldwide growth in this
increased and the stability at any step is compared by the
exploiting of wind energy. The total operating wind power
previous case. Simulation studies are carried out in Digsilent
capacity in the world has increased from approximately
software during severe faults.
2000MW in 1990 to well over 16000 by the end of 2000. By
the end of December 2009 the global installed wind energy
capacity has grown to more than 157GW which is about 1.5% II. WIND TURBINE AND GENERATOR MODELS
of worldwide electricity usage having doubled in three years At first the system components such as wind turbine and
between 2005 and 2008. DFIG, are modelled with the appropriate degree models for
Large scale wind farms are connected to electric power our simulation process.
transmission network. Smaller facilities are used to provide
electricity to isolated locations. At this point of time, the A. Wind Turbine Model
majority of wind turbines are equipped with Doubly Fed The wind turbine aerodynamics can be characterized by the
Induction Generators (DFIG). DFIG is a kind of induction well-known nondimensional curves of the power coefficient
generator with wounded rotor. The stator terminals are C p as a function of both tip speed ratio λ and the blade pitch
directly connected to grid while the rotor winding is angle β . The tip speed ratio is the ratio between the blade tip
connected via slip rings to a converter [1]. speed Vt and the wind speed U w [2, 5].
The converter consists of a Rotor-Side Converter (RSC)
and a Grid-Side Converter (GSC) connected back to back by Vt
λ= (1)
a DC link capacitor and only needs to handle a fraction (25- Uw
30%) of the total power to achieve full control of the
generator [2]. The rotor converter uses a high frequency
switching pulsewide-modulated PWM converter to achieve
Vt = ωt × R (2)
 Where R is the blade length (m) and ωt is the wind turbine
rotor speed.
Using the coefficient C p , the aerodynamic power Pae of the
turbine is determined by:
1 (3)
ρπR 2Vt 3C p (λ , β )
Pae =
2
Where ρ is the air density kg m3 .
Combining (1) and (3), the torque available from the wind Fig. 2. Equivalent circuit of induction generator
turbine Tt is determined directly from the aerodynamic power
according to:
3 p Rr Vr2 (6)
Paeπ Te =
Tt = = ρR 3Vt 2C p (λ , β ) (4) 2ω s s ( R + Rr ) + ( X + X ) 2
ωt 2λ e
s
e r

The power extracted from the wind is maximized if the jX m ( Rs + jX s ) (7)

Re + jX e =
rotor speed is such that C p is maximum. Rs + j ( X s + X m )

Dynamic of the rotor can be expressed by:

B. Induction Generator Model
The configuration of a DFIG driven by a wind turbine is
shown in Fig. 1. The wind turbine is connected to the DFIG d p
ωr = (Tm − Te − Fωr ) (8)
through a mechanical shaft system, which consists of a low- dt 2j
speed shaft and a high-speed shaft with a gearbox in between.
The wound rotor induction machine in this configuration is d
θ r = ωr (9)
fed from both stator and rotor sides. The stator is directly dt
connected to the grid while the rotor is connected to the grid
through a converter.
Where, Rs , Rr , X s , X m , X r are stator and rotor resistances,
In order to produce electrical power at constant voltage and
stator reactance, magnetizing reactance and rotor reactance
frequency to the utility grid for a wide operating range from
respectively. Also, p is the number of pole pairs, s is slip, Te
subsynchronous to supersynchronous speeds, the power flow
is electromagnetic torque, Tm is shaft mechanical torque, J is
between the rotor circuit and the grid must be controlled both
moment of inertia and F stand for the combined rotor and
in magnitude and direction. Therefore the converter consists
load viscous friction coefficient.
of two four-quadrant IGBT PWM converters connected back
to back by a DC link capacitor. The crow-bar is used to short
circuit the rotor-side converter (RSC) to protect it from over- C. Aggregated Wind Farm Model
current during transient grid disturbances [2].
An aggregated wind farm model should be used for system
The equivalent circuit of the induction generator is shown dynamic behavior analysis in order to reduce problem of dimension
in Fig. 2. According to this model, voltage and torque and complexity.
equations are as follow [4]:
The instantaneous power generated by the DFIG wind turbine
dependents on the actual value of rotor speed, rather than on the
jX mVt (5) wind speed. Therefore the sum of the mechanical power of the
Vr =
Rs + j ( X s + X m ) individual wind turbines would introduce an error. It would be
assumed that there exist an instantaneous relation between wind
speed and generated power, which is, however, not true. Therefore
an aggregated model of a wind farm with available speed wind
turbine, the rotor speed of the individual turbines is kept track of and
the electrical power of the individual turbines is added, rather than
the mechanical power [6].
The electrical power system of individual turbines is represented
by a simplified model. The aggregated electrical power is the sum of
the outputs of the electrical power of the individual turbines
according to the incoming wind speed. Thus the different incoming
wind speed of aggregated wind turbines is considered. To obtain the
Fig. 1. Configuration of a DFIG driven by a wind turbine equivalent wind, the output power of each wind turbine is the total
output power. Assuming the per-unit power curve of the equivalent
wind turbine is equal to the one of the individual wind turbine, the
equivalent wind can be calculated from the total output power, using
the equivalent power curve. The wind is applied to an aggregated
generation system with the same per-unit model as the aggregated
wind turbine, mechanical parameters and electrical parameters can
be obtained from identified system.

III. DYNAMIC ANALYSIS

In a synchronous machine the armature winding is
distributed around the armature environment. The excitation
winding is excited with a DC current. So the field flux
altitude has no change by the time variations [7].
If we have N ph as the number of round per phase of the
armature winding, the induced emf in each phase is given by: As it can be seen in Fig. 4 the rotor angle δ has been
increased. In a power system when a fault is occurred, all of
E = 2πk w f r N phφ (10) the synchronous generators contribute in compensation of the
existing surplus current causes by the fault in the network. If
E = 2 Kωφ there is a confident penetration level of wind generation with
DFIG technology in the network, this kind of generation has
Where φ is flux in each phase of armature winding, f r is unchangeable output current at its connection point to the
rotating frequency of the rotor, kω is constant and ω is the grid. So the extended current has to be provided by the
rotor rotating speed in radians. synchronous generators. In this manner the synchronous
The rotating frequency is as below: machine overloads more than the case without any
penetration level of DGs in the system, the rotor angle has
pnr (11) greater oscillations and the rotor speed grows up faster than
fr =
2 the previous case.
Where p is the number of stator poles and nr is the rotating The DG penetration level in the system is defined as [6, 8,
speed in RPM. 9]:
The phasor diagram of a synchronous generator is shown in
Fig. 3. In this diagram Vt is terminal voltage of synchronous
machine, I a is the armature current, E is the no-load PL (%) = ∑P DG
× 100 (12)
generated electromotive force and X d and X q are the direct ∑P Load

and quadrature axis reactances of synchronous machine,

respectively. Where PL is the penetration level of the DG in the power
When a fault occurs in the power system, the system inertia system, PDG is the total DG power generation and PLoad is
starts to increase. So the rotor speed of the synchronous the total load demand.
generators becomes higher. As we have in (10) if the rotor Compared to the conventional SGs, it is commonly known
speed ω grows up the generated electromotive force of the that the wind turbine generators (WTGs) have a relatively
synchronous machine increases. So the output current of the weak capability to ride through grid faults because of the use
armature has an incremental value. As it can be seen in Fig. 3 of induction generators and a power electronic interface. In
if we assume θ and I f are constant, the phasor diagram could the contingencies, for protection purposes, the DFIG
be changed as shown in Fig. 4. converters have to be tripped from the system and
consequently the WTG may have to be disconnected from the
system.
Fig. 5 represents the reactive power output of a DFIG when
a fault is occurred in the system. As it can be seen in Fig. 5,
as soon as a fault occurs in the system, the DFIG starts to
consume a large amount of reactive power and this causes
extracting more current from synchronous generators and
corrupts the transient stability.
The active power loss due to disconnecting the DFIG from
the grid and also the more reactive power consumed by the
induction generator during the fault occurrence, cause a more
loading condition on synchronous generators.
 TABLE I
CHARACTERISTICS OF THE NEW ENGLAND TEST SYSTEM

System characteristic Value

Number of buses 39
Number of generators 10
Number of loads 19
Number of transmission lines 46
Total generation 6140.7MW/1264.3MVAr
Total load 6097.1MW/1408.7MVAr

The synchronous generator at bus 33 has the most rotor

angle and rotor speed variations.
Fig. 5. Reactive power of DFIG after fault occurrence Fig. 7 and 8 represent the rotor angle and rotor speed
variations with various penetration levels.
As it can be seen in Fig. 7 the rotor angle variations of the
IV. CASE STUDY
synchronous generator at bus 33 increase by adding
To investigate the impact of distributed generation on distributed generation to the system. The system is still stable
power system transient stability, a model of a power system for 18% of DG penetration level and none of the synchronous
of a certain scale is necessary in order to arrive at results that generators has lost its synchronism. The rotor speed
can be considered representative. The widely used New deviations also grow up by increasing the penetration level.
England test system is depicted in Fig. 6 and some of its Fig. 9 and 10 show the rotor angle and rotor speed variation
characteristics are given in table 1 [8]. of generator at bus 33 with the DG penetration level of 31%.
Dig Silent power factory is used as our simulating diagram. As it can be seen in the figures this generator has been lost
The results are extracted and diagrams are drawn by
MATLAB.
During the simulation, a 150 ms three-phase short circuit is
applied at buses 16. The penetration level of DG has been
increased step by step by replacing the synchronous
generators at buses 36, 37 and 38 consequently with DFIG.
The synchronous generator at bus 36 has 560MW and 100
MVar, the synchronous generator at bus 37 has 540MW and
0.65MVar and the synchronous generator at bus 38 has
830MW and 22.6Mvar of active and reactive power
generation respectively. By replacing the DFIG at each bus
the penetration level would become to 9%, 18% and 31%
respectively. The active and reactive power generation
capacity of the DFIG is the same as each synchronous
generator at these buses.
Fig. 7. rotor angle variations of the synchronous generator at bus 33

Fig. 6. One-line diagram of the 39-bus New England test system [9] Fig. 8. rotor speed variation of the synchronous generator at bus 33
its synchronism and the power system becomes unstable in V. CONCLUSION
this case. In this paper the power system transient stability is
When the network is going unstable the voltage at bus 16 investigated when a fault is applied in some buses. The
has serious oscillations after the fault cleared. Fig. 11 shows penetration level of DG implementation is raised by reduction
these oscillations at bus 16. of centralized generation with replacement of distributed
As it can be seen in these figures if the penetration level of generation.
DGs grow up, the network weaken in contrast with a fault In this regard, the detailed wind energy conversion system
and the synchronous generators would lose the synchronism model, induction generator model are used and the impact of
faster than the case without any DG in the system. aggregated doubly fed induction generators on the transient
stability is illustrated by simulation and confirmed by
analytical solution as well.
Simulation results show that ever the more penetration
level of wind farm we have in the system, the worse transient
stability and more rotor angle and rotor speed oscillations we
have. Also we could not have an unlimited DG generation in
the power system and at a certain penetration level, the power
network would become unstable and the synchronous
generators will lose their synchronism because of the weak
capability of the wind turbines in fault detection and their
cooperation in supplying additional current due to fault
occurrence.

Fig. 9. rotor angle variation of the synchronous generator at bus 33 REFERENCES

[1] L.H. Hansen, et al., “Conceptual survey of generators and power
electronics for wind turbines,” Risø National Laboratory, Roskilde,
Denmark, Dec 2001.
[2] W. Qiao, and R. G. Harley, “Effect of grid connected DFIG wind
turbines on power system transient stability,” in proc. Power and
Energy Society General Meeting – Conversion and Delivery of
Electrical Energy in the 21st century, IEEE, Pittsburgh, PA, July 2008,
pp. 1-7.
[3] M. V. A. Nunes, and J.A. Pecas, “Influence of the variable speed wind
generators in transient stability margins of the conventional generators
integrated in electrical grids,” IEEE trans. Energy Conversion, vol. 19,
no. 4, pp 692-701, Dec 2004.
[4] P. Kundur, “Power System Stability and Control,” McGraw-Hill, New
York, 1994.
[5] T. Ackerman, “Wind Power in Power Systems,” Wiley, New York,
May 2005.
[6] Y. Yang, and X. Zha, “Aggregating wind farm with DFIG in power
system online analysis,” in proc. IEEE IPEMC 2009 Power Electronics
and Motion Control Conference, Wuhan, China, May 17-20, 2009, pp.
2233-2237.
Fig. 10. rotor speed variation of the synchronous generator at bus 33 [7] P. S. BimBahara, “Electrical Machinery Theory, Performance and
Application,” Ghaem press, Tehran, Iran, 2006.
[8] V. V. Thong, et al., “Influence of large penetration of distributed
generation on N-1 safety operation,” in proc. IEEE PES 2004 Power
Engineering Society General Meeting, IEEE, vol. 2, pp. 2177-2181,
2004.
[9] M. Reza, et al., “Impact of distributed generation penetration levels on
power system transient stability,” in proc. IEEE PES 2004 Power
Engineering Society General Meeting, IEEE, vol. 2, pp. 2150-2155,
2004.
[10] V. Akhmatov, and H. Knudsen, “An aggregate model of a grid
connected, large scale, offshore wind farm for power stability
investigations-importance of windmill mechanical system,” Electrical
Power and Energy Systems, no. 4, pp 709-717, 2002.
[11] G. N. Kariniotakis, and G. S. Stavrakakis, “A general simulation for
the accurate assessment of isolated diesel-wind turbines interaction,
part I: A general multi machine power system model,” IEEE trans.
Energy Conversion, vol. 10, pp 577-583, Sept. 1995.

Fig. 11. voltage oscillations of bus 16 at unstable case