Eval uat ion of t he DF I G Wi nd Tur bi ne

Bui l t-i n Model i n PSS/ E






Master of Science Thesis in the Programme of Electric Power Engineering


MOHAMMAD SE YEDI


Division of Electric Power Engineering
Department of Environment and Energy
CHALMERS UNIVERSITY OF TECHNOLOGY
Göteborg, Sweden, June 2009
ii
























iii



Evaluation of the DFIG Wind Turbine Built -in Model in PSS/E






MOHAMMAD SEYEDI





In partial fulfillment for the award of Master of Science degree in Electric Power Engineering,
in the Department of Environment and Energy, Division of Electric Power Engineering,
Chalmers
University of Technology, Göteborg, Sweden









© Mohammad Seyedi, June 2009.











Examiner: Anh Tuan Le
Division of Electric Power Engineering
Chalmers University of Technology
SE-412 96 Göteborg
Sweden
Telephone + 46 (0)772 1000
iv



v



Abstract

Due to growth of environmental concern, more electricity must be generated from renewable
energy sources. One of the most cost efficient alternatives is wind energy. Large wind farms
with several hundred megawatts of rated power have been connected to grid. Nowadays,
many transmission system operators require wind farms to maintain zero reactive power
exchange with the grid during normal operation. Furthermore, like other conventional power
plants wind turbines must be able to remain connected to the grid during grid faults. In order
to study impact of wind power generation on the power system valid models of wind turbines
are needed.
The purpose of this study is to evaluate DFIG wind turbine built-in model in PSS/E. In this
report, the responses of the model subjected to grid disturbances are investigated. At first, the
typical dynamic model of a DFIG wind turbine is introduced. Secondly the built-in dynamic
model for DFIG wind turbine in PSS/E is illustrated and it is used to study the dynamic
behavior of DFIG subjected to a symmetrical short circuit at the point of common coupling of
a large wind farm. The load flow study for a network where a wind farm consisting 67
numbers of DFIG wind turbines and each one with 1.5MW rated power connected to the grid
are presented and the output results from load flow study is used to perform dynamic
simulation. The voltage, current, output power, speed and pitch angle profiles are presented in
the dynamic study.
The dynamic simulation results shows that the DFIG dynamic model presented in PSS/E is
not able to fully represent the limitations of the wind turbine. There are some deficiencies in
the model, for example the protection systems for converter and rotor are not included in the
model, under/over voltage protection and rotor current limits are not considered. The upper
and lower limitation in the voltage control loop will affect the voltage stability and there is no
direct relation between the maximum over load on converters and the limitation in voltage
control loop.



Keywords: wind turbine, modeling, doubly fed Induction generator, power system stability,
voltage stability, aggregated model.


vi



vii

Acknowledgements

This work has been done at the Division of Electric Power Engineering, Department of
Energy and Environment at Chalmers University of Technology, Göteborg, Sweden.
The financial support provided by Svenska Vindbolaget is gratefully acknowledged. I would
like to thank my supervisor Abram Perdana for his patience, good guidance, excellent
supervision, critical comments and revising the thesis report to give it a better shape. I
acknowledge the motivation words and good support of my examiner Anh Tuan Le. Special
thanks goes to Ola Carlson. the electric power engineering master`s programme coordinator
for best advices, good supporting and constant guidance during studying.
I would like to express my gratitude to Richard Larsson from Svenska Vindbolaget for his
good negotiation and technical discussions during the project.
I owe a great debt of appreciation to my family, my wife Asrin and our lovely daughter,
Dilan. They are really a treasure for me; they are my success and my inspiration. They made
me the one I am today and I am very thankful for that. I couldn´t imagine a life without them
and without their support.















viii


ix


Contents
Abstract ...................................................................................................................................... v
Acknowledgements .................................................................................................................. vii
1 Introduction ......................................................................................................................... 1
2 Wind Turbine ...................................................................................................................... 3
2.1 Fixed-Speed Wind Turbine ......................................................................................... 3
2.2 Wind Turbine with Variable Rotor Resistance............................................................ 4
2.3 Variable-Speed Wind Turbine with Doubly-Fed Induction Generator ....................... 4
2.4 Variable-Speed Wind Turbine with Full Converter .................................................... 5
3 Wind Turbine with Doubly Fed Induction Generator (DFIG) ............................................ 6
3.1 DFIG and Converter Dynamic Model ......................................................................... 6
3.1.1 DFIG Model .................................................................................................................... 7
3.1.2 Frequency Converter Model .......................................................................................... 11
3.1.2.1 Rotor-side Converter ................................................................................................. 12
3.1.2.2 Grid-side Converter ................................................................................................... 14
3.2 Mechanical Model and Turbine Control ................................................................... 16
3.2.1 Turbine Speed Control Model ....................................................................................... 16
3.2.2 Drive Train Model ......................................................................................................... 16
3.2.3 Aerodynamic and Pitch control Model .......................................................................... 18
3.3 PSS/E Simulation Software ....................................................................................... 20
3.4 Built-in Wind Turbine Model in PSS/E .................................................................... 21
3.4.1 Generator/Converter Model........................................................................................... 22
3.4.2 Electrical (Converter) Control Model ........................................................................... 24
4 Model Simulation .............................................................................................................. 26
4.1 Grid Codes and fault-ride through capability ............................................................ 26
4.2 Test System................................................................................................................ 27
4.3 Load Flow .................................................................................................................. 27
4.4 Dynamic Simulation .................................................................................................. 29
4.4.1 Constant reactive power without closed-loop terminal voltage control ...................... 29
4.4.2 Constant reactive power with closed-loop terminal voltage control ........................... 34
4.5 Simulation result summary ........................................................................................ 36
5 Conclusion and Future Works .......................................................................................... 39
x

5.1 Conclusions ............................................................................................................... 39
5.2 Future works .............................................................................................................. 39
References ................................................................................................................................ 41
Appendices ............................................................................................................................... 42
A .Wind Turbine and Network Parameters .............................................................................. 42





















xi


1


1 I ntroduction

The utilization of wind turbine to produce electricity is increasing rapidly in different parts of
the world. It has become one of the main alternatives for non pollutant and environmentally
friendly type for power generation in Sweden and a large amount of wind power are
connected to the Swedish national grid and some others will be connected in next coming
years.
Not until recently, the contribution of wind power generation on the system stability was
considered to be small. However with increasing in the wind farm capacity it is clear that
disconnecting a large wind farm will result in losing of a big part of power generation in grid,
which can aggravate instability problems. Due to increasing portion of wind power, wind
turbines have to contribute in reactive power support during steady state as well as during
transient conditions. Transmission system operators (TSO) in countries where the amount of
wind power generation is significant have established a special regulation and added to the
grid codes that specify a number of requirements for wind farms to stay connected to the grid
during disturbances.

In order to establish stability and power quality, the Swedish TSO, Svenska Kraftnät (Svk
), prescribes performance for production units larger than 1.5 MW. Practically this regulation
comprehends all new modern wind power turbines. The grid connection procedure is
changing and adaption to a large scale wind power expansion is continually made. This
adaption wills most likely lead to requirements of information regarding the wind power
units` electrical behavior and then to determine the impact oI adding wind generation. and
establish how the system can be upgraded, SvK needs adequate wind-turbine generator
dynamic models and simulation results. The Svk uses PSS/E as a simulator tool to study the
dynamic behavior of the power system. The new version of PSS/E has added a dynamic
model for doubly fed induction generator wind turbine. This wind turbine is referred as type 3
per classification of the western electricity coordinating council wind generator modeling
group (WECC WGMG) and it is the only model that is publicly available and it is approved
by WECC and GE [8]. The model is used for short term voltage stability according to the
requirements by SvK which is the basic part of this study.
The mathematical equations for a generic model are presented and some simplifications
have been done to get final equations and then the block diagrams of the DFIG which are
used in the simulator tool have explained. The fault scenarios according the Svenska
KraItnät`s grid code have simulated to evaluate the dynamic behavior of a wind farm model.
According to the dynamic simulations results the ride-through capability of the wind farm has
evaluated.
2

After explanation the generic model, the dynamic model of DFIG in PSS/E has introduced
and the procedure for controlling active and reactive power in the model has illustrated. The
advantages and drawbacks with the model are also discussed.

3


2 Wind Turbine

The wind turbines are divided into four main types: fixed speed wind turbine with induction
generator, variable-speed wind turbine with variable rotor resistance, variable speed wind
turbine with doubly fed induction generator (DFIG) and variable speed wind turbine with full
converter (FCWT) [3].
2.1 Fixed-Speed Wind Turbine

In fixed speed wind turbine, the generator is a squirrel-cage induction generator which is
directly connected to the grid as in Figure 3.1. The rotor of a fixed-speed wind turbine rotates
at a fixed speed determined by the frequency of the grid, the gear ratio and the pole pairs of
generator. A fixed-speed wind turbine is connected to the grid through a soft-starter. The
induction generator absorbs reactive power from the grid, so capacitor bank is necessary to
provide reactive power compensation. A gear box is used to transform power from the turbine
with lower-rotational speed to the generator rotor with high-rotational speed. The generator
terminal voltage is increased with a step-up unit transformer to a medium voltage level.
This type of wind turbines has the advantage of being simple, robust and more cost-efficient
compared to the other wind turbine types. However, the reactive power consumption cannot
be controlled. Another drawback with the fixed speed wind turbine is that wind speed
fluctuation is transmitted into the mechanical torque and it is finally transferred to the
electrical power on the grid. The fluctuation in the delivered power to the grid can lead to
large voltage fluctuation where the wind farm connected to a weak grid [4].
Risk of loss of synchronism due to over-speed in case of voltage dips and increasing of
reactive power consumption, especially after fault clearance, is other disadvantages of this
type of wind turbine. The Induction generator dynamics cannot be improved. Preventive or
corrective measures which can be done are to limit acceleration during voltage dip through
improvement of pitch control and to provide reactive power support during and after clearing
of fault, via FACTS devices. Compatibility with grid code requirements not satisfied.
Installation of FACTS devices, such as SVCs and STATCOMs, is a potential but it can be
costly.
4

Gear
Box
Gen
Capacitpor
Bank
Unit
Transformer

Figure 3.1: Fixed speed wind turbine
2.2 Wind Turbine with Variable Rotor Resistance

In wind turbine with variable rotor resistance the generator is an induction machine with
wound rotor and it is directly connected to the grid. Like a fixed-speed wind turbine, a
capacitor bank is necessary to provide reactive power for induction machine. By changing the
rotor resistance magnitude, the rotor speed can be regulated in a short range up to 10% higher
than synchronous speed [5]. The structure of this type of wind turbine is shown in Figure 3.2.

Gear
Box
Gen
Capacitpor
Bank
Unit
Transformer
Variable Resistor

Figure 3.2: Variable-speed wind turbine with variable rotor resistance
2.3 Variable-Speed Wind Turbine with Doubly-Fed I nduction Generator

A power electronic converter is used in the variable-speed wind turbine with doubly fed
induction generator as shown in Figure 3.3. The converter is connected to the generator rotor
and the stator is directly connected to the grid.
Compared to the one with full converter, this type is more suitable for high power wind
turbines since only a fraction (typically 20-30%) of the total power goes through the
5

converter. This gives possibility to design the converter with smaller size, and as a result,
lower cost and lower power electronic losses.
Gear
Box
Gen
Unit
Transformer
AC
AC DC
DC
Power Electronic
converters

Figure 3.3: Variable-speed with doubly fed induction generator

2.4 Variable-Speed Wind Turbine with Full Converter

In variable-speed wind turbine with full converter, the generator can be a squirrel-cage induction
or a synchronous generator which is connected to the grid via a power electronic converter as
shown in Figure 3.4. The whole power output from generator goes through the converter and
therefore the converter is rated at full power. The voltage level and the reactive power can be
regulated by using power electronic converters.
Gear
Box
Gen
Unit
Transformer
AC
AC DC
DC
Power Electronic
converters


Figure 3.4: Full converter wind turbine

The converter is connected to the stator. This converter has to be able to convert full power from
the generator. The gear box can be removed or simplified if a multi-pole synchronous generator is
employed. The main advantages of this type of wind turbine are the most favorable dynamic
behavior during disturbances, with minimum transients at fault occurrence and clearing. It has
also the capability to enhance active and reactive power control and therefore the compatibility
with grid code requirements can be satisfied [5].

6

3 Wind Turbine with Doubly Fed I nduction Generator (DF I G)

The objective of this chapter is to provide an overview on the dynamic model of different
parts in a DFIG wind turbine. At first, the main mathematic equations, that describe the
relationship between voltage and fluxes in an induction machine and they are basic equations
for establish dynamic model, are presented with a short description of the other electrical
parts. Model of aerodynamic and mechanical parts are also presented. Later, the simplified
model which is used in PSS/E software is introduced and the mechanism for controlling
active and reactive power in this model is illustrated.
3.1 DF I G and Converter Dynamic Model

An induction generator with power supply on the rotor is a main part of the DFIG wind
turbine. The DFIG is modeled by means of an equivalent circuit model. The main components
of a DFIG wind turbine are shown in Figure 3.1. In DFIG wind turbine, the stator directly
connected to the grid, while the rotor circuit is connected to a power converter by means of
slip rings and brushes. The rotor current is regulated by the power converter to control the
electromagnetic torque and field current and thus the stator output voltage. The DFIG can
operate in either sub-synchronous or in super-synchronous operation modes due to capability
of converter to operate in bi-directional operation power mode.
=
=
3~
3~
Rotor Side
Converter
Grid Side
Converter
G
r
i
d
DGIG
r
e
m
T
s
u

Figure 3.1: Doubly fed induction generator
The general block diagram for controlling a DFIG is shown in Figure 3.2 [3]. The basic parts
of model consist of the generator and drive train, the turbine rotor, the grid side converter with
dc-link capacitor, the pitch controller and the rotor side controller. The active and reactive
power is controlled by the rotor side converter, while the grid-side converter is used to control
the dc voltage of dc-link capacitor.

7


Turbine
rortor
Wind
Model
Drive-Train
Model
Pitch
Controller
Generator
Model
Interface Network
Model
Rotor-Side
Converter
Power/Speed
Controller
Reactive
Power
Controller
DC link&
Grid-Side
Converter
v
wind
T
m
e
t
|
P
g e
t
T
e
e
g
i i
sq sd
,
u u
sq sd
,
Q
g
P
g
i
ref
rd
i
ref
rq
u
dc
u
u
r
q
r
d
,
i i
rq rd
,
i
i
c
q
c
d
,
i
r
q
i
r
d
g
,
e
u
u
s
q
s
d
,
u u
rq rd
,
I
U

Figure 3.2: Control block diagram of DFIG wind turbine
3.1.1 DF I G Model

The conventional T-equivalent circuit is used to represent the dynamic model of induction
generator (IG). This model consists of stator winding and rotor winding with slip rings. The
three phase stator windings directly connected to the grid through a step-up transformer. The
rotor also has a three phase winding but it is connected to the rotor power supply through slip
rings and brushes. The power supply of the rotor comes from the rotor-side converter.

The rotor current is controlled by the rotor-side converter and consequently the active and
reactive power can be controlled. The equivalent diagram of the stator and rotor and the grid-
side and rotor-side converters is shown in Figure 3.3.



8

§
=
§
=
f
R
GSC RSC
dc
C
+
-
dc
v
+
-
+
-
f
L
Grid Filter
DC Link
s
f
u
G
r
i
d

g
E
s
f
i
+ ÷
r
r
j ) (
1
e e
r
R
ì s
L
s
R
m
L
+
-
g
i
+
-
+
s
i r
i
r
u
s
u
m
i
+
-
dt
d
r
+
+
dt
d
s
+
-
~ ~
-
ì r
L
r
i
+
s
j
1
e + -
Induction Generator

Figure 3.3: Dynamic equivalent circuit of the DFIG and converters
Based on the equivalent circuit presented in Figure 3.3, the electrical dynamics equations
using the space vector approach can be derived. By applying KirchhoII`s voltage law in the
stator and rotor and in per unit system it can be written as
s
s
s
s s
j
dt
d
R u i v e
v
1
+ + =

(3.1)


v
v
e e
r
r
r
r
r r
j
dt
d
R u i ) (
1
÷ ÷ + =
(3.2)

where:
=
s
R stator resistance
ì s
L

stator leakage inductance


=
r
R rotor resistance =
ì r
L

rotor leakage inductance
=
m
L

mutual inductance = ÷ v e e
r
r
) (
1
rotor back emf
=
s
u stator voltage =
r
u rotor voltage
=
s
v stator Flux =
r
v rotor Flux
=
1
e synchronous frequency =
r
e rotor frequency
= ÷ =
r
e e e
1
2
Slip frequency
9

=
s
i Stator current =
r
i Rotor current
The subscripts s and r refer to quantities of the stator and rotor, respectively. The dynamic
model of an induction machine is usually presented by means of a so called fifth-order model
that represents IG by a system of five general differential equations of an idealized induction
machine.
i i
r
m
s
s
s
L L + = v (3.3)
i i
s
m
r
r
r
L L + = v
(3.4)
where
m s s
L L L + =
ì

m r r
L L L + =
ì


The electromechanical torque is

) . . ( ] Im[
*
i i
i
sd
sq
sq
sd s
s
s
e
T v v
v ÷ = =

(3.5)
The electrical power will be
} Im{
*
r
r
e
i P v =

(3.6)
In power system studies, it is desirable to reduce the complexity of the system by using
reduced-order models that can be obtained by assuming some of the derivatives as being
equal to zero. For example, a third-order model of IG is obtained when we neglect the stator
flux transients.

Under balanced network conditions, the amplitude and rotating speed of the stator and rotor
fluxes are assumed to be constant. The flux dynamics are canceled to get a fast response to the
commands from electric controls of converter. Consequently, in the dq reference frame, the
stator and rotor fluxes remain constant, i.e.
0 =
dt
d
s
v

0 =
dt
d
r
v
(3.7)
Thus the voltage equations can be written as
sq sd s sd
i R u v e
1
÷ =

(3.8)
sd sq s sq
i R u v e
1
+ =

(3.9)


10

rq rd s rd
s i R u v e
1
÷ =

(3.10)

rd rq s rq
s i R u v e
1
÷ =

(3.11)
If d-axis of the synchronous frame is fixed to the stator voltage, then

0 =
sd
u

(3.12)

The flux linkage equations are as follows:

rd m sd s sd
i L i L ÷ = v

(3.13)
rq m sq s sq
i L i L ÷ = v
(3.14)
sd m rd r rd
i L i L ÷ = v

(3.15)
sq m rq r rq
i L i L ÷ = v

(3.16)
We can also neglect the stator resistance because it is considerably small compared to
the stator reactance, hence

0 =
sd
v

(3.17)
Subsequently

rd
s
m
sd
i
L
L
i =

(3.18)



sq sd
u v e
1
÷ =

(3.19)
The electromechanical torque is calculated as:
rq rd rd rq e
i i T v v ÷ =

(3.20)


) (
rd sq rq sd m e
i i i i L T ÷ =

(3.21)

The active power of stator, rotor, grid-side converter and the injected active power to the grid
can be determined as:


(3.22)


(3.23)

(3.24)

11


c s g
P P P + = (3.25)

Where
s
P
,
r
P ,
c
P , and
g
P

are the stator, rotor, dc link grid-side converter active power and the
total generator active power, respectively .
If the losses in the converter ignored the converter power will be equal to the power injected
to the rotor (P
c
=P
r
),
thus it can be written as
rq rq rd rd sd sd g
i u i u i u P . . . + + = (3.26)
The reactive power can be calculated as:
sq sq s
i u Q . ÷ = (3.27)
rq rd rd rq c
i u i u P . . ÷ =

(3.28)
By assuming the grid-side converter reactive power is zero, the generated reactive power will
be
r s g
Q Q Q + =

(3.29)

rq rq rd rd sq sd g
i u i u i u Q . . . ÷ + ÷ =

(3.30)

3.1.2 F requency Converter Model

The induction machine of a DFIG wind turbine operates in a generating mode at sub-
synchronous and super-synchronous speeds where the speed range is limited by the maximum
voltage of the rotor-side frequency converter [6]. The rotor winding is supplied from a
converter consisting of back-to-back connected converter bridges with a dc link, as shown in
Figure 3. 4.

The grid-side converter operates at a network frequency and controls the voltage level in
the dc link circuit. If required, it is possible assign the grid-side converter to deliver reactive
power to the grid. The rotor-side converter operates at different frequencies (slip frequency),
depending on the rotor speed, and controls the flux of the DFIG and thus the active and
reactive power [6].

The rotor-side converter has an over-current protection called crowbar. The crowbar
protects the rotor-side converter, as well as the rotor circuit of the DFIG, against high currents
during grid disturbances. The location of crowbar resistor is shown in Figure 3.4, where RSC
is the rotor-side converter, GSC is the grid-side converter, CH is the chopper circuit and CR is
the crow bar circuit.
12



P Q, pf ,Ut
Q U
DC
GSC control
RSC control
P
i
t
c
h







a
n
d




P
o
w
e
r
DFIG
CR
DC Link
GSC RSC
Ut
CH
P
ref

Figure 3.4: Typical layout of a DFIG


3.1.2.1 Rotor-side Converter

The power rating of the rotor-side converter is determined according the maximum slip power
and reactive power control capability. The rotor-side converter can be simplified as a current-
controlled voltage source converter.

Rotor Control Mode
The reference rotor speed is obtained from power-speed characteristic curve as shown in
Figure 3.5. The rotor speed is then used as an input for a PI controller to get the stator active
power reference
P
ref
s
. The rotor reference current in dq reference frame is then calculated by
the following expression [3]
(4.31)
13

P
rated
e
min
e
max
e
sync
Maximum Speed
operation
Optimum speed
curve
Minimum speed
operation
Pitch controlled
Power Limitation
Power
Speed

Figure 3.5: Power-speed curve
It can be found from above equation that the stator active power is controlled by controlling
the rotor current in d direction. The magnitude of reference rotor current in q direction is
calculated as:

m
sd sq s
ref
rq
L
L
u i
i
e
e
1
1
+
÷ =

(3.32)
The control diagram for active and reactive power controls is shown in Figure 3.6.
a) Active power control
P
gen
+
-
e
ref
r
e
r
¯
P
ref
s
L
L L
m
) m ls
( + ÷
÷
v
s
i
ref
rd

b) Reactive power control
+
- Q
ref
s
Q
s
¯
i
ref
rq

Figure 3.6: Rotor current controller: (a) active and (b) reactive power control

14

Over-Cur rent Protection (Crowbar)
The new Swedish grid code requires grid connected wind turbines to remain connected to the
network even when there is a short-term power network disturbance. In other words, the wind
farm should have fault-ride-through capability. However, riding through grid faults with
DFIG will lead to high peak currents in the rotor and high power into the DC circuit. Thus the
rotor and rotor-side converter need to be protected against over-current and over voltage. In
order to protect against over current and over voltage, a DC-chopper parallel with DC-link
and an AC-crowbar installed in the rotor circuit are used.

When there is a short circuit in the grid, the voltage at the terminal of DFIG goes down
almost immediately and it will cause high transient rotor current. One possible solution to
avoid over-current is to use a rotor over-current protection, a so-called crowbar. The crowbar
is connected in parallel with the rotor-side converter as shown in Figure 3.7. The crowbar
deactivates the rotor-side converter and short-circuits the rotor winding, whenever the rotor
current exceeds the maximum limit value. When the grid fault is cleared, the rotor-side
converter is restarted.

Rotor Side
Converter
R
crow
i
crow


Figure 3.7: The crowbar connected between the rotor of the DFIG and the rotor-side
converter.
3.1.2.2 Grid-side Converter

The grid-side converter (GSC) is usually represented by a simplified model, a so-called
generic control scheme based on a set of PI controllers. The voltages in d and q direction are
calculated according the active and reactive power reference values. In a simplified model, the
switching dynamics of converter is neglected and an ideal control is assumed, meaning that
the converter is able to follow its demanded value at any time [6].

The converter can also be utilized to support grid reactive power during a fault and it can
also be used to enhance grid power quality. However, these abilities are seldom utilized since
they require a larger converter rating [6].

15

In some studies, such as in [9], the detailed model of GSC for dynamic simulation is
recommended. The author argued that if the detailed model is not used, the results for rotor
current calculation and dc link voltage are not accurate enough. Nevertheless, the majority of
DFIG studies use the simplified dc-link voltage controller. The dc-link capacitor voltage can
be calculated as a function of the input power to the dc link [6]. Accordingly, the main
objective of the GSC is to control the dc-link voltage assuming that the power factor is unity.
The equivalent circuit of the dc-link model is shown in Figure 3.8 where
i
r
and
i
g
are the dc
currents flowing from dc-link to the rotor-side converter and from dc-link to the grid-side
converter, respectively.
GSC
RSC
C
dc v
dc
+
-
=
=
§
§
i
r
i
g
i
c


Figure 3.8: dc-link equivalent circuit


The energy stored to the dc-link capacitor can be written as

(3.33)

Where P is the net power flow into the capacitor
C
dc
is the dc-link capacitor value and v
dc
is
the capacitor voltage. P equals to P
r
-P
g,
where P
r
is the power to the rotor and P
g
is the power
from grid. The aim of the grid-side converter control is to maintain the level of dc link voltage
v
dc
at a pre-set value
v
ref dc÷
by means of a PI controller. . In Figure 3.9 the simplified
model of dc-link and grid-side converter is represented.

(3.34)

PI
+
-
v
ref
dc
i
g
+ C
1
÷
v
dc
s
1
+
GSC
model
i
r

Figure 3.9: Simplified model of the dc-link and the grid-side converter.

16


3.2 Mechanical Model and Turbine Cont rol

The mechanical model of a DFIG wind turbine and its control in dynamic studies of power
systems usually consists of wind speed model, aerodynamic model; drive train model and
rotor speed and pitch angle controller subsystems. The mechanical model and control of the
wind turbine is simplified in this study. The input and out powers to the different parts of a
wind turbine is shown in Figure 3.10.

Wind
Energy
Turbine
Induction
Generetor
Pwind Paero
Drive
Train
Pmech Pelec
Grid


Figure 3.10: The input and out powers in WT

3.2.1 Turbine Speed Control Model

The basic strategy for the wind turbine control system is to get the maximum power from the
wind. The limiting factor that affects the amount of energy from wind is the equipment power
ratings. Accordingly, the mechanical input power has to be limited when the input power
from wind is higher than the equipment power ratings. Otherwise, it may destroy the
mechanical parts. To reduce the amount of input mechanical power the turbine blades must be
turned. This ability of controlling the mechanical input power is called pitch control of turbine
blades. The dynamics of the pitch control are moderately fast, and can have significant impact
on dynamic simulation results.
On the hand if the available wind power is less than the equipment ratings, the blades is set
at a certain pitch angle to get maximum power from wind. The turbine control model sends a
power order to the electrical control according the available wind speed, requesting the
converter to deliver this amount of power to the grid
3.2.2 Drive Train Model

The two-mass model is commonly used to model the rotor dynamic simulation. This model,
the drive-train consists of two inertias, i.e. turbine and generator inertias. The two inertias are
connected through a spring as shown in Figure 3.11. The mechanical power from the wind
power model and the electrical power from the generator/converter model are used as inputs
to calculate the rotor speed. The block diagram of the system is shown in Figure 3.12.
17


Turbine
Generator
J
t
J
g
e
t e
r
D
m
D
t
D
g
T
mech T
elec
K
s

Figure 3.11: Drive Train of Wind Turbine

E
E
E
E
÷
÷
+
-
-
+
+
+
+
+
+
-
-
+
Ht
2
1
s
1
s
1
s
1
Hg
2
1
K
e
t
P
mech
P
gen
e
g
T
mech
e
t
A
e
tg
A
e
tg
A
e
t
e
o
D
shaft
e
g
A
e
o
e
g
T
elec
E
o
tg

Figure 3.12: Two-mass torsional model

H
t
Turbine inertia constant [MW.sec/MVA]
H
g
Generator inertia constant [MW.sec/MVA]
K Gain factor
H = H
t
+H
g
Total inertia constant [MW.sec/MVA]
D
shaft
Shaft damping factor [pu]
Freq1 First shaft torsional resonant frequency [Hz]
K
s
Shaft stiffness



18

3.2.3 Aerodynamic and Pitch control Model

For a power system simulation involving grid disturbances and fault-ride through capability,
it is important to include the aerodynamic model. In this model, the mechanical power to the
turbine is a function of wind speed, blade pitch angle and shaft speed.

The function of the wind power module is to compute the wind turbine mechanical power
(shaft power) from the energy contained in the wind using the following formula:

(3.35)

is the mechanical power extracted from the wind, ȡҏ is the air density |kg/m
3
], A
r
is the
area swept by the rotor blades [m
2
] or , is the wind speed [m/sec], and Cp is the is
the power coeIIicient. which is a Iunction oI Ȝҏ and ș. Ȝҏ is the ratio oI the rotor blade tip speed
and the wind speed (v tip/v
w
ș is the blade pitch angle in degrees.

The relationship between blade tip speed and generator rotor speed is a constant ( ) and it is
defined as

(3.36)

Where r is the rotor blade radius; is turbine angular speed; and is wind speed.

C
p
is a characteristic of the wind turbine. There are different methods to calculate C
p
, for
instance, like blade element method, look-up table and analytical approximation.
In this project, the second method is used. C
p
is provided as a set of curves relating Cp to Ȝ,
with u ҏҏ as a parameter. The Cp curves for the GE wind turbine are shown in figure 3.13.


Figure 3.13: Wind power C
p
curves
For power levels below rated, the turbine speed is controlled primarily by the electric
power order to specify speed reference. For power levels above rated, the rotor speed is
19

controlled primarily by the pitch control, with the speed being allowed to rise above the
reference transiently.
Typically high wind speed threshold is 25 [m/sec]. The measured rotating turbine speed is
compared with the reference speed and the difference is integrated and if this value is higher
than a specified value, the unit is tripped. It has also the ability to trip the machine if turbine
speed falls below a specified value.

The pitch angle controller is shown in figure 3.14 including the pitch control and pitch
compensation. The block diagram to control of active power is illustrated in figure 3.15. The
pitch controller, pitch compensator and drive train models are included in the block diagram.
In the diagram is the voltage measured at the terminal of wind turbine, is the
current injected from the wind turbine to the grid, is the wind turbine rotating shaft
speed, is the wind speed, is the output electrical power measured at the terminal of
wind turbine, is mechanical input power to the shaft ,
P
cmd
is the command
power and is the pitch angle.
PI
+
-
u
From Turbine
Model
e
e
err
e
ref
From Converter
Control Model
Blade Pitch
Angle
u
cmd
Pitch Control
PI
+
-
The measured
electric power
P
e
P
err
P
ref
Set point
+
+
Pitch Compensation

Figure 3.14: Pitch control and pitch compensation block diagram

20

P
ref
Power
filter
e
ref
-
¯
Drive Train
Model
+
÷
Current
filter
¯
PI
e
err
e
wt
×
×
÷
V
term
V
term
I
inj P
e
P
e
P
e -
+
T
cmd
P
cmd
e
wt
I
cmd
e
wt
P
a
PI ¯
PI
¯
e
wt
w
s
e
err
Pitch Controller
P
e
P
mech
P
max
-
+
Pitch
Compensator
|
) , ( | ì
C
p
+
+
T
a

Figure 3.15: The block diagram for active power control in wind turbine

3.3 PSS/ E Simulation Software

PSS/E stands for Power System Simulator/Engineering and is mainly used for power system
studies. It can perform both phase vector and electromechanical simulations and has become
one of the most commonly used software within the power system industry. Many wind
turbine manufacturers provide PSS/E models that are e.g. used for verification of grid codes
however most of them are bound with non-disclosure agreements. Therefore, only one wind
turbine model is so far available publicly. PSS/E is developed by Siemens Power
Technologies International (PTI).

Power System Simulation for Engineering (PSS/E) is composed of a comprehensive set of
programs for studies of power system transmission network and generation performance in
both steady-state and dynamic conditions. PSS/E can be utilized to facilitate calculations for a
variety of analyses, including: Power flow and related network functions, optimal power flow,
balanced and unbalanced faults, network equivalent construction, dynamic simulation. In
addition to the steady-state and dynamic analyses, PSS/E also provides the user with a wide
range of auxiliary programs for installation, data input, output, manipulation and preparation.


21

Power Flow

A power flow study (also known as load-flow study) is an important tool involving numerical
analysis applied to a power system. Unlike traditional circuit analysis, a power flow study
usually uses simplified notation such as a one-line diagram and per-unit system, and focuses
on various forms of AC power (i.e.: reactive, real, and apparent).

Power flow studies are important because they allow for planning and future expansion of
existing as well as non-existing power systems. A power flow study also can be used to
determine the best and most effective design of power systems.

Dynamics
The dynamic simulation program includes all the functionality for transient, dynamic and
long term stability analysis. The purpose of the dynamics is to facilitate operation of all
dynamic stability analytical functions.

The dynamic modeling simulation is used to ensure the reliability of electricity supply and
to predict the performance of the system under a wide range of conditions and to identify any
problems and scope measures needed for reliability. The testing and validation of models will
enable the performance of wind turbine on the power system on their own and interacting
with other turbines, to be predicted [1]. The PSS/E version number 31 has a dynamic model
for doubly fed induction generator wind turbine in the library. The wind turbine is referred to
as Type 3 per classification of the Western Electricity Coordinating Council Wind Generator
Modeling Group (WECC WGMG). [8]

3.4 Built-in Wind Turbine Model in PSS/ E

The dynamic model for DFIG in PSS/E model is according to the recommendation by GE
Energy. In the following section, the main strategy of the model will be explained. There are
three considerations when the model is used for dynamic simulations
1. The model is mainly appropriate for bulk power system studies. The model is for
positive sequence phasor time-domain simulations.
2. The model assumes that the analysis is mainly focused on how the wind turbine-
generator (WTG) reacts to grid disturbances, such as faults on transmission system.

To construct a complete WTG model, four device models are used: generator/converter
model (WT3G), electrical control model (WT3E), mechanical control (wind turbine) module
(WT3T) and pitch control module (WT3P). These models are represented as four boxes in
Figure 3.16.
22

Converter
Control
Model
Generator/
Converter
Model
Pich Control
Model
Wind
Turbine
Model
) (P
I
p
) (Q
Eq
Q
P
gen
gen
,
Shaft Speed
Blade Pitch
S
p
e
e
d

O
r
d
e
r
P
o
w
e
r

O
r
d
e
r
V
regbus
V
term
P
g
e
n
V
term
X
eq
P,Q
I
inj


Figure 3.16: GE WTG dynamic model

3.4.1 Generator/ Converter Model

The generator/converter model (WT3G1) is the equivalent of the generator and the field
converter and provides the interface between the WTG and the network. The difference
between this model and conventional models is that this model has no mechanical variables
for the machine rotor. These variables are included in the turbine model, instead. The flux
dynamics have been neglected in order to get a faster response to the higher level commands
from the electrical controls through the converter.
The generator is modeled as a controlled-current source as shown in Figure 3.17. This
current controller delivers the amount of required current that must be injected to the grid in
response to the flux and active current commands which are generated by the converter
control model [3]. This part represents DFIG and rotor converter model.
In case of over voltage, the amount of injected current to the grid is reduced to mitigate
over voltage. The two first-order low-pass filters represent the electronic control systems.
They have a time constant of 20 msec. There is a phase locked loop (PLL) in the actual
converter controls to synchronize the generator rotor currents with stator, however due to
extremely fast response of PLL dynamics relative to the generator/field converter time frame;
it is not shown in the model.

The input in the upper part is the excitation current, which adjusts the output voltage. The
input in the lower part is the active current, which specifies the amount of active power which
should be delivered to the grid. X
´
in the model represents the effective equivalent reactance
of generator and T represents the transfer function of generator-network. The vector diagram
of the voltages and currents is shown in Figure 3.18
23




Figure 3.17: Generator and rotor converter model
Y
Im
"
E
term
V
E
Y
"
Y
V
Y
I
X
I
X
V
E
X
"
X
term
I
Re
u
o

Figure 3.18: Vector diagram of voltages and currents
24


3.4.2 Elect rical (Converter) Control Model

The electrical control model determines the amount of active and reactive power that must be
delivered by the generator to the grid. It compares the measured active power ( ( with the
ordered active power from turbine model ( and the measured reactive power at the
wind turbine terminal (with the ordered reactive power (from the reactive
power control model. The electrical control model provides the voltage and current
commands and sends them to the generator model by comparing the measured active and
reactive power with the required amount of the active and reactive power.
The control procedure for active power is specified in the turbine model. The model for
reactive power control is shown in Figure 3.19. Three different options are provided in the
model for reactive power control. These options can be selected by changing varflg switch
position. When the switch is in position 1, the voltage magnitude at a particular bus, often the
point of common coupling (PCC) with the transmission system is compared with the voltage
reference magnitude, and this voltage is regulated by sending a reactive command to all of the
WTGs.
The reference voltage magnitude is specified during data entering in load flow study. In
position -1 the reactive power command comes from power factor regulation and in position 0
the reactive power command is set to follow a certain amount of reference magnitude for
reactive power [7]. The reactive power command is compared with reactive power measured
at the terminal of wind turbine in another model that is shown in Figure 3.20. The output of PI
controller is the reference voltage. The reference voltage compared with voltage magnitude at
the terminal point of WTG to make the voltage command ( for the generator model.
The overall electrical model is presented in simplified model of Figure 3.21
V
ref
V
reg
PF ref
P
gen
-1
1
0
Q
max
Q
min
Q
max
Q
min
Q
cmd
Q
ref
lg varf
Power
Factor
Regulator
Reactive
Power
Control

Figure 3.19: Converter control model
25

Q
cmd
Q
gen
V
term
PI
V
max
V
min
Vref
_
I
0
1
E
"
qcmd
lg vltf
+
To the Generator
control model
I V
max Q term
X +
I V
min Q term
X +

Figure 3.20: Voltage control model


V
ref
V
reg
PF
ref
P
gen
Q
ref
Q
gen
Electrical Control
Model
E
"
qcmd
To the Generator
control model
I
p
P
cmd
Active power
command from wind
turbine model
Voltage measued at
regulated bus
V
term
Voltage measued at
WTG terminal bus
R
e
a
c
t
i
v
e

P
o
w
e
r

C
o
n
t
r
o
l
A
c
t
i
v
e

P
o
w
e
r

C
o
n
t
r
o
l
V
term
Voltage measued at
WTG terminal bus
P
mech
Mechanical Power
From Pitch control
Voltage Command
Current Command

Figure 3.21 Overall DFIG electrical control model



26

4 Model Simulation

In this section, the load flow and dynamic simulation of a DFIG wind turbine model in PSS/E
subjected to a grid fault is presented.
4.1 Grid Codes and fault-ride through capability

In the past the wind farms were disconnected from the grid following a grid fault and then
there were no requirements for wind farms to contribute in voltage stability [1]. Such wind
farms usually had small power ratings. Recently, wind turbines grow in size. Big wind farm
consisting several units of wind turbines are connected together to make wind farms with high
power rating. As the wind farm power rating is higher disconnection of large wind farms from
the grid during disturbance will lead to loss of large amount of power from the grid [1].

Transmission system operators in the countries where the wind power is a part of main
electricity source added some regulations to grid codes, which require the wind farms to
continue their operation during a grid fault for a certain time period.

Accordingly during a grid fault, wind turbines should be able to stay connected to the
power network and. The Swedish technical document which specifies the requirements for
connecting wind farms to grid is SvKFS 2005:2. The fault-ride-through on the low voltage is
a basic part the specification.

This grid code states the wind turbine must not be disconnected from the grid when there is
a three-phase fault in the closest bus in the 400 KV grids and for fault duration of 250
milliseconds with voltage recovery to 90% of rated voltage. Therefore the wind farm must
stay connected to the grid as long as the voltage of the grid where the farm is connected is
above the criteria limit. There are different requirements for wind parks with a rated active
power more than 100 MW and medium size ones with rated power between 1.50 and 100
MW. This study is focused on the medium sized wind parks in which a three phase short
circuit fault at point of common coupling causes a 75% voltage dip at the terminal of wind
turbine. The voltage profile is depicted if Figure 4.1.

20
40
60
80
100
0 0.20 0.40 0.60 0.80 1.00
U(%)
t(s)
1.20 1.40 1.60
Prated>100MW
20
40
60
80
100
0 0.20 0.40 0.60 0.80 1.00
U(%)
t(s)
1.20 1.40 1.60
1.5MW<Prated<100MW
25%
25%
0
.
7
5
90%
90%


Figure 4.1: Voltage profile for wind farm to perform fault-ride through
27


4.2 Test System

A test power system configuration for load flow and dynamic simulation studies consist of a
wind farm with several wind turbines. In this study, however, the wind farm is presented by a
single equivalent machine connected to a single equivalent unit of transformer. This
representation is called aggregated model of wind farm. The aggregated model is considered
to be sufficient for analyzing the response of the wind park to grid disturbances and for
performance evaluation of wind turbine fault ride-through capability. The high voltage side
of unit transformer connected to a wind farm transmission line to transfer power to a wind
farm substation. The voltage level is stepped-up in the substation to a level that is proper for
connecting to the grid.
The rated active and reactive power of the single machine which represents the wind farm
is the sum of rated active and reactive power of wind turbines inside the wind farm. In the
same way, the rated active power and reactive power for single unit transformer obtained by
multiplying the number of individual unit transformers and the rated active and reactive
power of each unit transformer. A single bus represents the bus collector. The wind farm has
an active power rating of 100 MW, which consists of 67 wind turbines; each of them has rated
power of 1.5 MW. The configuration of the power system is shown in Figure 4.2. The
connection between wind farm and main substation is through more than one line in case of
disconnecting of one line the wind farm dose not disconnected from the grid. However in this
test system it is assumed the disturbance is on the PCC and not on the transmission line
between wind farm and the main substation and therefore it is shown as a single line.
Unit
Transformer
Substation
Transformer
Step-up Voltage
Substation
LV Bus HV Bus
PCC
Wind Farm
DFIG
GRID
0.600kv 34.5kv
34.5kv
230.0kv 230.0kv
Load
200MW
Infinte Bus
Prated
100MW


Figure 4.2: Power system configuration
4.3 Load Flow

The load flow provides the initial conditions for dynamic simulation. The load flow model of
a wind farm serves two purposes: First, it serves as the basis for load flow studies including
thermal, voltage, and other analyses. Second, it serves as initial conditions for stability
analysis. The quantities that can be obtained from load flow study are the bus voltage
28

magnitudes, bus voltage angle, active and reactive powers that flow through transmission
lines.

The maximum and minimum active and reactive power limits according to the capability
curve of generators are among the most important input data that must be specified during
data preparation. The capability curve for a 1.5 MW GE wind turbine generator is shown in
Figure 4.3[7]
1.0
0.5
0.12
+0.296pu -0.436pu
Under excited
Over excited
PF=0.96 PF=0.90
P
Q

Figure 4.3: GE 1.5 MW DFIG capability curve

A wind turbine generator in load flow is treated as a conventional generator and is specified
on the existing generator record of the power flow raw data file. In the end of this data record,
a wind control mode must be specified, as follows:
- 0 for a non wind turbine generator.
- 1 for a wind turbine generator that participates in voltage control provided with upper
and lower reactive power limits.
- 2 for a wind turbine generator with a constant power factor.
- 3 for a wind turbine generator that operates at a fixed active and reactive power.
The load flow diagrams of the test system are shown in Figure 4.4

Figure 4.4: The active and reactive power flow (MW,MVAr)

29

4.4 Dynamic Simulation

The first step in dynamic simulation is to enter the dynamic model data, which is saved in a
file. This file contains a group of records, each of which defines the location of a dynamic
equipment model in the network along with the constant parameters of the model. As it was
mentioned in previous chapter, the PSS/E version 31 provides a dynamic model for a DFIG
wind turbine. The model includes generator, electrical control, and wind turbine and pitch
control. In this simulation, the infinite grid is represented by a constant internal voltage
generator model (GENCLS). GENCLS is the classical constant voltage behind transient
reactance generator model in PSS/E. Appendix 1 contains model parameters that are used in
this simulation.
Dynamic simulation is performed based on the load flow data that provide the transmission
network, load, and generator data. In this study, a number of simulations are performed to
investigate the model response subjected to grid disturbances. The disturbance in the
simulation study is a three-phase symmetrical short-circuit fault on one of the lines
connecting PCC bus and grid bus. The fault lasts for 250 milliseconds and is cleared by
disconnecting the faulted line. During the fault the voltage at PCC is dropped to zero and thus
no electrical power transfer from wind turbine to the grid.

4.4.1 Constant reactive power without closed-loop terminal voltage control

When the reactive control mode of the model is set to a constant reactive power (varflg=0),
the current in q-direction is also constant during the fault. Figure 4.5 shows that during the
fault on PCC voltage drops to zero. After fault clearing, the voltage gradually recovers to
around 0.95 pu. The wind turbine terminal voltage has a similar pattern, except that during the
fault the retaining terminal voltage is 0.25 pu, which is considerably higher that the PCC
voltage. When faulted line is disconnected from the network, the total impedance between the
turbine terminal and PCC increases and therefore the voltage drop will increase and the
recovered voltage will be less than its pre-fault magnitude.
30


Figure 4.5: Voltage profile

Figure 4.6: Electrical, mechanical and aeromechanical power

As seen in figure 4.6, when the fault occurred the electrical power suddenly decreases to a
very low value (0.06433pu or 6.433 MW) and the difference between mechanical input power
and electrical output power make an increasing in the rotor speed and therefore the rotor starts
to accelerate. The torsional oscillation in the drive-train model is reflected in the output
power of the wind turbine. The frequency of the oscillation is 1.8 Hz in this particular case.
0.5 1 1.5 2 2.5 3 3.5
0
0.2
0.4
0.6
0.8
1
1.2
time[sec]
V
o
l
t
a
g
e
[
p
u
]
Voltage Profile


PCC
WTG
0.5 1 1.5 2 2.5 3 3.5 4
-0.5
0
0.5
1
1.5
2
Time [sec]
Pelec, Pemach, Paero [pu]


Pelec
Paero
Pmech
31

The peak of the mechanical power is considerably high, which is up to 2 pu during the first
cycle of the oscillation. The scale of this magnitude indicates that the mechanical stress in the
drive train is quite high.


Figure 4.7: Active, reactive and mechanical power profile
During the fault, the mechanical input power is reduced by pitch control mechanism. The
P
mech
plot shows the damped mechanical power output of the wind turbine due to the pitch
controller action. It is assumed that the wind turbine works at its rated power at the fault
instant. As the fault duration and the transient of voltage recovery is short, the wind speed can
be assumed constant in the grid fault simulations. It can be seen in figure 4.7 prior to the fault,
the reactive power is zero and the wind farm operated at unity power factor. During the fault,
the wind turbine inject reactive power of 0.3671 pu (36.71 MVAr) to the grid.
The output current profile is shown in figure 4.8. It can be seen the current level during
fault is around 1.5 pu and as the magnetizing impedance of generator is much bigger than its
stator and rotor impedances it can be assumed whole current goes through the rotor-side
converter and rotor winding. The converters are normally designed for a 10% overload and
therefore flowing of the high current during fault will likely violate the rotor converter current
limit and can damage the IGBTs inside converters.
0.5 1 1.5 2 2.5 3 3.5 4
0
0.2
0.4
0.6
0.8
1
Time [sec]
Active and reactive power [pu]


P
Q
32


Figure 4.8: Current profile
The performance of the model for longer fault duration time has also been investigated.
The simulation results, which is not presented here, show that even when the fault occurs in
much longer time period, in this case up to 3 seconds, the turbine is able to ride through the
fault despite the fact that the current injected by the generator and hence the converter is
considerably high i.e. up to 1.5 pu. This may raise a question regarding the validity of the
model to realistically simulate the limitation of the rotor-side converter. In other words, the
model seems to produce too optimistic simulation results.


0 0.5 1 1.5 2 2.5 3 3.5 4
1
1.5
2
2.5
3
3.5
4
time[sec]
C
u
r
r
e
n
t
[
p
u
]
Output Current


I
33


Figure 4.10: Wind turbine and generator speed profile
It can be seen from figure 4.10 in the two mass model for drive train the speed of generator is
oscillating and it is not identical with the turbine speed.


Figure 4.11: Pitch angle profile

It can be seen in Figure 4.11, the pitch angle is increased during the fault in order to reduce
the power input from wind turbine. The pitch angle is oscillating; this is because the pitch
angle is controlled by the speed of the turbine. After fault clearing, the turbine speed goes
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
1.12
1.14
1.16
1.18
1.2
1.22
1.24
1.26
time[sec]
S
p
e
e
d
[
p
u
]
Generator and Turbine Power


Gen speed
Turbine speed
0.5 1 1.5 2 2.5 3 3.5 4
9
10
11
12
13
14
15
time [sec]
P
i
t
c
h


[
D
e
g
]

Pitch Angle


Pitch
34

down (see Figure 4.10), however the pitch angle continuously increases until few seconds.
This is due to the effect of the pitch compensation. The pitch compensation tries to reduce the
input power to the turbine by increasing a little more in pitch angle.


4.4.2 Constant reactive power with closed-loop terminal voltage control

In the second simulation study, the varflg is set to 0 (constant reactive power) and vltflg is set
to 1. This means that the terminal voltage control is enabled. The reactive power command is
equal to the reference value and the internal voltage magnitude is limited in the closed loop
controller. The upper and lower limits of the internal voltage magnitude are Vterm+XI
Qmax

and Vterm+XI
Qmin
, respectively. In the simulation, the voltage drop limits XI
Qmax
and XI
Qmin

must be set by user. It was found that using the typical voltage drop limit values given in
PSS/E, i.e. XI
Qmax
=0.4, the voltage is not able to recover following a grid fault (see Figure
4.12 and Figure 4.13). This is probably due to numerical instability since the upper and lower
limits of the voltage controller are not fixed; rather it is dependent on the terminal voltage
level (Vterm).
In order to avoid such a numerical instability, the maximum voltage drop limit value must
be increased. Figure 4.14 shows simulation result where the voltage is successfully recovered
with maximum voltage drop limit set at 0.55.


Figure 4.12: Voltage profile for upper limit value of 0.4pu
0.9 1 1.1 1.2 1.3 1.4 1.5
0
0.2
0.4
0.6
0.8
1
Time [sec]
Terminal voltage [pu]
35


Figure 4.13: Active and reactive power profile for upper limit value of 0.4pu


Figure 4.14: Voltage profile for upper limit value of 0.55pu

0.9 1 1.1 1.2 1.3 1.4
0
0.2
0.4
0.6
0.8
1
Time [sec]
Active and reactive power [pu]
0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
Time [sec]
Terminal voltage [pu]
36


Figure 4.15: Active and reactive power profile for upper limit value of 0.55pu


4.5 Simulation result summary

It was found from the simulations that the model has advantages and some drawbacks. The
advantages with the model are summarized as follows:
- The drive-train model is presented in two-mass model representation. The influences
of this model on the dynamic response of the wind turbine model are clearly shown in
the simulations. The magnitude of the mechanical power oscillation during the fault
indicates the scale of mechanical stress in the drive train. Thus, the impacts of frequent
expose of voltage dip on the lifetime of the drive train components need to be
investigated further.
- The pitch angle controller and its function for decreasing the aerodynamic power
during and following a grid fault are considered in the wind turbine model. The
simulation results emphasize the importance of the pitch controller model in fault
responses of a DFIG wind turbine. Without appropriate representation of the pitch
controller the aerodynamic input power would be constant and thereby would result in
less accurate predictions.
- The time step required to run the model is compatible with the standard time step in
PSS/E simulator. Consequently, this model does not require time step reduction in
order to avoid numerical instability during simulations.
0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2
Time [sec]
Active and reactive power [pu]
37

- The procedure for controlling voltage and reactive power is included in the model.
There are three different options provided in the model to control reactive
power/voltage i.e. constant voltage, constant power factor and constant reactive
power.
- There is possibility to define the power-speed curve in the model. This provides
flexibility for users to adjust the power-speed characteristic of the turbine according to
the wind turbine being investigated.

The main drawbacks with the model can be explained as follows:
- The generator is modeled as a simple voltage source behind transient reactance. As a
consequence, the generator losses are not included in the model.
- The rotor current and voltage are not represented in the model and therefore the
influence of rotor current limits on the wind turbine response during and following a
disturbance on the grid cannot be investigated.
- There is high risk for voltage instability following a grid fault when the closed loop
terminal voltage control is used. This is probably due to numerical instability in the
controller limiters which are not fixed; rather it is dependent on the terminal voltage
level (Vterm). To avoid such a problem, the upper limits must be increased with the
consequence of much higher reactive current during fault.
- The FRT is not included in the model. Therefore, regardless the fault duration and the
magnitude of voltage dip during grid fault, wind turbine remains intact. It has been
observed that during grid faults the generator/converter current can be very high,
while in reality such a high current magnitude are not permitted due to the limited
capability of the converter. That means the model provide too optimistic results and
therefore it does not depict realistic behavior of a typical wind turbine.
- Wind speed is assumed to be constant throughout simulations and thereby the effect of
wind speed changes on the turbine response cannot be simulated using this model.









38




39

5 Conclusion and Future Wor ks

5.1 Conclusions

The modeling of wind turbine generators for power system studies is an important aspect and
it is in a rapid evolution state. More accurate model can help the power system operators to
study the dynamic performance of the gird under disturbances and predict its behavior in a
more realistic way.
The PSS/E built-in model investigated in this study emphasizes a detailed representation of
reactive power and voltage control, while the generator model of the wind turbine is simply
modeled as a voltage source behind transient impedance. In this study the model has been
evaluated by simulating the model subjected to grid disturbances.
The simulation results show that the behavior of the DFIG wind turbine model in PSS/E is
dependent on the parameters used in the model. One of the most important parameters is the
closed loop terminal voltage control limits. This parameter has a significant effect on voltage
stability of the wind turbine after fault clearing. It was found that the under voltage and over
current protection is not implemented in the model. It is observed that there no limitation in
fault time duration and severity. The model does not differentiate between short and long-
term/permanent faults. While in reality, the wind farm must be disconnected from the grid
when disturbance is permanent otherwise it can be damaged by high amount of short circuit
current and also the under voltages.
On the other hand the model the model has number of advantages such as the possibility to
control the reactive power and voltage using different mode of operation. In general, the
model can be used to study dynamic simulation when it is connected to bulky power system,
however some considerations related with the model must be take into account. Accordingly
the model is applicable with care.

5.2 Future works

The results of this work can be basic point for future studies. Some basic parts of DFIG same
as rotor crowbar can be added to the dynamic model. The under/over voltage protection can
be considered. The dynamic model for the full converter and fixed-speed wind turbines also
can be introduced. Another subject that could be of interest is evaluation of the model for
islanding operation.


40






41

References

[1] EWEA (European Wind Energy Association) (2005). Large scale integration of wind energy in
the European power supply: analysis, issues and recommendations. Brussels: EWEA.
http://www.ewea.org/index.php?id=178

[2] Svenska kraItnät (2005). 'AIIärsverket svenska kraItnäts IöreskriIter och allmänna råd om
driftsäkerhetsteknisk utIormning av produktionsläggningar¨. Internet:
http://www.svk.se/Tekniska-krav/Foreskrifter/

[3] A.Perdana. 'Dynamic Models oI Wind Turbines.¨ Ph.D. dissertation. Department oI Energy
and Environment, Chalmers University of Technology, Göteborg, Sweden, 2009.

[4] Thomas Ackermann.¨ Wind Power in Power Systems.¨ Royal Institute oI Technology.
Stockholm, Sweden

[5] N.Rahmat Ullah. 'Win Power Added Value Ior Network Operation.¨ Ph.D. dissertation,
Department of Energy and Environment, Chalmers University of Technology, Göteborg,
Sweden, 2008.


[6] Slavomir Seman. ' Transient PerIormance Analysis oI WindPower Induction Generators.¨
Ph.D. dissertation, Department of Electrical and Communications Engineering, Laboratory of
Electro mechanics, Helsinki University of Technology, Helsinki 2006, Finland

[7] Nicholas W.Miller. 'Dynamic Modeling oI GE 1.5 and 3.6 Wind Turbine-Generators.¨ Version
4.1 Jan.2008.

[8] PSS/E Program Application Guide

[9] Akhmatov. V.2002. ' Modelling oI Variable-Speed Wind Turbines with Doubly-Fed Induction
Generator in Short Term Stability Investigations¨.Int.Workshop on Transmission Networks for
Offshore Wind Farms,2002,Stoclholm,Sweden





42

Appendices
A . Wind Turbine and Networ k Parameters

A.1 Doubly-fed induction generator (WT3G1)
Symbol Description Value
X
eq
Equivalent reactance for current injection(pu) 0.8
K
p
ll PLL first integrator gain 30
K
ip
ll PLL second integrator gain 0
PLLMX PLL maximum limit 0.1
Prated Turbine MW rating 1.5

A.2 Doubly-fed induction generator electrical control (WT3E1)
Symbol Description Value
Tfv Filter time constant in voltage regulator(sec) 0.15
K
pv
Proportional gain in voltage regulator(pu) 18
K
iv
Integrator gain in voltage regulator(pu) 5
X
c
Line drop compensation reactance(pu) 0.05
Tfp Filter time constant in torque regulator(sec) 0.05
K
pp
Proportional gain in torque regulator(pu) 3
K
ip
Integrator gain in torque regulator(pu) 0.6
PMX Max limit in torque regulator(pu) 1.12
PMN Min limit in torque regulator(pu) 0.10
QMX Max limit in voltage regulator (pu) 0.296
QMN Min limit in voltage regulator(pu) -0.436
IP
max
Max reactive current limit(pu) 1.10
T
rv
Voltage sensor time constant(sec) 0.05
RPMX Max power order derivative(pu) 0.45
43

RPMN Min power order derivative(pu) -0.45
T_Power Power filter time constant(sec) 5.0
K
qi
MVAR/Voltage gain 0.0
VMINCL Min voltage limit 0.90
Symbol Description Value

VMAXCL Max voltage limit 1.20
K
qv
Voltgae/MVAR gain 40
XIQmin Min limit of diff. between Vterm and E
q
¨
cmd -0.50
XIQmax Max limit of diff. between Vtrem and E
q
¨
cmd 0.40
T
v
Lag time constant in WindVar controller(sec) 0.05
T
p
Pelec filter in fast PF controller 0.05
Fn A portion of online wind turbines 1.0
WPMIN Shaft speed at Pmin(pu) 0.69
W
p
20 Shaft speed at 20% rated power(pu) 0.78
W
p
40 Shaft speed at 40% rated power(pu) 0.98
W
p
60 Shaft speed at 60% rated power(pu) 1.12
Pwp Minimum power for operation at W
p
100 speed (pu) 0.74
W
p
100 Shaft speed at 100% rated power(pu) 1.20

A.3 Turbine (WT3T1)
Symbol Description Value
VW Initial wind, pu of rated wind speed 1.25
H Total inertia constant, MW*sec/MVA 4.95
DAMP Machine damping factor , pu P/pu speed 0.00
Kaero Aerodynamic gain factor 0.0070
Theta2 Blade pitch at twice rated wind speed (deg) 21.98
44

Htfrac Turbine inertia fraction (Hturb/H) 0.875
Frec1 First shaft torsional resonant frequency(Hz) 1.80
DSHAFT Shaft damping factor (pu) 1.50

A.4 Pitch control (WT3P1)
Symbol Description Value
Tp Blade response time constant(sec) 0.30
Kpp Proportional gain of PI regulator(pu) 150
Kip Integrator gain of PI regulator(pu) 25
Kpc Proportional gain of the compensator(pu) 3.0
Kic Integrator gain of the compensator(pu) 30
TetaMin Lower pitch angle limit(deg) 0.0
TetaMax Upper pitch angle limit(deg) 27.0
RTeta Upper pitch angle rate limit(deg/sec) 10.0
PMX Power reference(pu on MBASE) 1.0



    ii    

 

Evaluation of the DFIG Wind Turbine Built-in Model in PSS/E

MOHAMMAD SEYEDI

In partial fulfillment for the award of Master of Science degree in Electric Power Engineering, in the Department of Environment and Energy, Division of Electric Power Engineering, Chalmers University of Technology, Göteborg, Sweden

© Mohammad Seyedi, June 2009.

Examiner: Anh Tuan Le Division of Electric Power Engineering Chalmers University of Technology SE-412 96 Göteborg Sweden Telephone + 46 (0)772 1000
iii    

 

 

iv    

A bstract
Due to growth of environmental concern, more electricity must be generated from renewable energy sources. One of the most cost efficient alternatives is wind energy. Large wind farms with several hundred megawatts of rated power have been connected to grid. Nowadays, many transmission system operators require wind farms to maintain zero reactive power exchange with the grid during normal operation. Furthermore, like other conventional power plants wind turbines must be able to remain connected to the grid during grid faults. In order to study impact of wind power generation on the power system valid models of wind turbines are needed. The purpose of this study is to evaluate DFIG wind turbine built-in model in PSS/E. In this report, the responses of the model subjected to grid disturbances are investigated. At first, the typical dynamic model of a DFIG wind turbine is introduced. Secondly the built-in dynamic model for DFIG wind turbine in PSS/E is illustrated and it is used to study the dynamic behavior of DFIG subjected to a symmetrical short circuit at the point of common coupling of a large wind farm. The load flow study for a network where a wind farm consisting 67 numbers of DFIG wind turbines and each one with 1.5MW rated power connected to the grid are presented and the output results from load flow study is used to perform dynamic simulation. The voltage, current, output power, speed and pitch angle profiles are presented in the dynamic study. The dynamic simulation results shows that the DFIG dynamic model presented in PSS/E is not able to fully represent the limitations of the wind turbine. There are some deficiencies in the model, for example the protection systems for converter and rotor are not included in the model, under/over voltage protection and rotor current limits are not considered. The upper and lower limitation in the voltage control loop will affect the voltage stability and there is no direct relation between the maximum over load on converters and the limitation in voltage control loop.

K eywords: wind turbine, modeling, doubly fed Induction generator, power system stability, voltage stability, aggregated model.

 

 

v    

      vi     .

I couldn´t imagine a life without them and without their support. good supporting and constant guidance during studying. I would like to express my gratitude to Richard Larsson from Svenska Vindbolaget for his good negotiation and technical discussions during the project. they are my success and my inspiration. They made me the one I am today and I am very thankful for that. Department of Energy and Environment at Chalmers University of Technology. I acknowledge the motivation words and good support of my examiner Anh Tuan Le. Göteborg. The financial support provided by Svenska Vindbolaget is gratefully acknowledged. I owe a great debt of appreciation to my family. excellent supervision. Dilan.                               vii     . my wife Asrin and our lovely daughter.Acknowledgements     This work has been done at the Division of Electric Power Engineering. good guidance. They are really a treasure for me. Sweden. critical comments and revising the thesis report to give it a better shape. I would like to thank my supervisor Abram Perdana for his patience. Special for best advices.

    viii     .

..  24   3.....................................................................4   3   3.........................  29   Constant  reactive  power  with  closed-­‐loop  terminal  voltage  control  .......................................................................................................................................1...................................................... 26   4................3   3..............................................................................1   3................3   4..................................................................3   2....................................................................................................................................4.......... 3   2.......................................... 27   Load Flow ...........  7   Frequency Converter Model  ....................................1.................1   2........... 29   Constant  reactive  power  without  closed-­‐loop  terminal  voltage  control  ..........................................  18   3...................................................4   Grid Codes and fault-ride through capability ...........................................................................................................................................................................................1   4.........................  16   Aerodynamic and Pitch control Model   .......... 26   Test System.........5   5   Simulation result summary ................................................4   PSS/E Simulation Software ...................................... 1   Wind Turbine ....................................................................................................... 3   Wind Turbine with Variable Rotor Resistance...................................1........  34   4..................2   4.. 20   Built-in Wind Turbine Model in PSS/E ...1   3.............. 4   Variable-Speed Wind Turbine with Doubly-Fed Induction Generator .   Contents Abstract ................... 21   Generator/Converter Model.........................................  12   Grid-side Converter  .............2   4............................................... v   Acknowledgements ........................1   3...........4.......... 6   DFIG Model  ...........................................................1........................................................................................... 5   DFIG and Converter Dynamic Model ...........................1   Fixed-Speed Wind Turbine .................. 6   3......... 36   Conclusion and Future Works ..................2..... vii   1   2   Introduction ....................................................................................................................................................................................................2   3........................................................................................................................2   3......1   3............................................... 16   Turbine Speed Control Model  ....................................2   4   Model Simulation..2   Mechanical Model and Turbine Control ..........4.....  14   Wind Turbine with Doubly Fed Induction Generator (DFIG)..............................2   3...............................................  22   Electrical (Converter) Control Model  ....... 27   Dynamic Simulation ...................4.....................................2...........2.......................................................................................................... 4   Variable-Speed Wind Turbine with Full Converter ......  16   Drive Train Model  .....................1   4.................... 39   ix     .............  11   Rotor-side Converter  ............................................2..............................................................................3   3.............................................................2..2   2.............

..................................5................1   5..............2   Conclusions ............................................ 41   Appendices ............................................................................................ 42                                             x     ...........................................................................Wind Turbine and Network Parameters ... 39   Future works ............... 42   A ........................................................................................ 39   References .................................................................................................................................................................................

  xi     .

  1 Introduction The utilization of wind turbine to produce electricity is increasing rapidly in different parts of the world. The model is used for short term voltage stability according to the requirements by SvK which is the basic part of this study. the contribution of wind power generation on the system stability was considered to be small. The grid connection procedure is changing and adaption to a large scale wind power expansion is continually made. Transmission system operators (TSO) in countries where the amount of wind power generation is significant have established a special regulation and added to the grid codes that specify a number of requirements for wind farms to stay connected to the grid during disturbances. This wind turbine is referred as type 3 per classification of the western electricity coordinating council wind generator modeling group (WECC WGMG) and it is the only model that is publicly available and it is approved by WECC and GE [8]. The fault scenarios according the Svenska have simulated to evaluate the dynamic behavior of a wind farm model. The mathematical equations for a generic model are presented and some simplifications have been done to get final equations and then the block diagrams of the DFIG which are used in the simulator tool have explained. wind turbines have to contribute in reactive power support during steady state as well as during transient conditions. SvK needs adequate wind-turbine generator dynamic models and simulation results. 1     . According to the dynamic simulations results the ride-through capability of the wind farm has evaluated. Svenska Kraftnät (Svk ).5 MW. However with increasing in the wind farm capacity it is clear that disconnecting a large wind farm will result in losing of a big part of power generation in grid. Practically this regulation comprehends all new modern wind power turbines. In order to establish stability and power quality. This adaption wills most likely lead to requirements of information regarding the wind power establish how the system can be upgraded. Not until recently. Due to increasing portion of wind power. which can aggravate instability problems. prescribes performance for production units larger than 1. The Svk uses PSS/E as a simulator tool to study the dynamic behavior of the power system. The new version of PSS/E has added a dynamic model for doubly fed induction generator wind turbine. the Swedish TSO. It has become one of the main alternatives for non pollutant and environmentally friendly type for power generation in Sweden and a large amount of wind power are connected to the Swedish national grid and some others will be connected in next coming years.

2     . The advantages and drawbacks with the model are also discussed. the dynamic model of DFIG in PSS/E has introduced and the procedure for controlling active and reactive power in the model has illustrated.After explanation the generic model.

This type of wind turbines has the advantage of being simple.1. 3     . is other disadvantages of this type of wind turbine. Another drawback with the fixed speed wind turbine is that wind speed fluctuation is transmitted into the mechanical torque and it is finally transferred to the electrical power on the grid. the gear ratio and the pole pairs of generator. Installation of FACTS devices.1 F ixed-Speed W ind T urbine In fixed speed wind turbine. so capacitor bank is necessary to provide reactive power compensation. Compatibility with grid code requirements not satisfied. Risk of loss of synchronism due to over-speed in case of voltage dips and increasing of reactive power consumption. The Induction generator dynamics cannot be improved. However. especially after fault clearance. The generator terminal voltage is increased with a step-up unit transformer to a medium voltage level. A gear box is used to transform power from the turbine with lower-rotational speed to the generator rotor with high-rotational speed. the reactive power consumption cannot be controlled. The induction generator absorbs reactive power from the grid.2   W ind T urbine The wind turbines are divided into four main types: fixed speed wind turbine with induction generator. A fixed-speed wind turbine is connected to the grid through a soft-starter. variable speed wind turbine with doubly fed induction generator (DFIG) and variable speed wind turbine with full converter (FCWT) [3]. The fluctuation in the delivered power to the grid can lead to large voltage fluctuation where the wind farm connected to a weak grid [4]. such as SVCs and STATCOMs. robust and more cost-efficient compared to the other wind turbine types. Preventive or corrective measures which can be done are to limit acceleration during voltage dip through improvement of pitch control and to provide reactive power support during and after clearing of fault. variable-speed wind turbine with variable rotor resistance. the generator is a squirrel-cage induction generator which is directly connected to the grid as in Figure 3. 2. The rotor of a fixed-speed wind turbine rotates at a fixed speed determined by the frequency of the grid. is a potential but it can be costly. via FACTS devices.

Gear Box Gen Unit Transformer Capacitpor Bank Figure 3. Gear Box Gen Unit Transformer Capacitpor Bank Variable Resistor Figure 3. the rotor speed can be regulated in a short range up to 10% higher than synchronous speed [5].3. By changing the rotor resistance magnitude.2 W ind T urbine with V ariable Rotor Resistance In wind turbine with variable rotor resistance the generator is an induction machine with wound rotor and it is directly connected to the grid.2: Variable-speed wind turbine with variable rotor resistance 2.3 V ariable-Speed W ind T urbine with Doubly-F ed Induction G enerator A power electronic converter is used in the variable-speed wind turbine with doubly fed induction generator as shown in Figure 3. The structure of this type of wind turbine is shown in Figure 3. Compared to the one with full converter. this type is more suitable for high power wind turbines since only a fraction (typically 20-30%) of the total power goes through the 4     .1: Fixed speed wind turbine 2. a capacitor bank is necessary to provide reactive power for induction machine. Like a fixed-speed wind turbine. The converter is connected to the generator rotor and the stator is directly connected to the grid.2.

The main advantages of this type of wind turbine are the most favorable dynamic behavior during disturbances.3: Variable-speed with doubly fed induction generator 2.4. with minimum transients at fault occurrence and clearing.4: Full converter wind turbine Unit Transformer The converter is connected to the stator. The gear box can be removed or simplified if a multi-pole synchronous generator is employed. This converter has to be able to convert full power from the generator. lower cost and lower power electronic losses. The whole power output from generator goes through the converter and therefore the converter is rated at full power. Gear Box Gen AC DC DC AC Power Electronic converters Figure 3. It has also the capability to enhance active and reactive power control and therefore the compatibility with grid code requirements can be satisfied [5].4 V ariable-Speed W ind T urbine with F ull Converter In variable-speed wind turbine with full converter. the generator can be a squirrel-cage induction or a synchronous generator which is connected to the grid via a power electronic converter as shown in Figure 3. The voltage level and the reactive power can be regulated by using power electronic converters. Gear Box Gen Unit Transformer AC DC DC AC Power Electronic converters Figure 3. This gives possibility to design the converter with smaller size. and as a result.converter.   5     .

the stator directly connected to the grid. the simplified model which is used in PSS/E software is introduced and the mechanism for controlling active and reactive power in this model is illustrated. r DGIG Tm 3~ = = 3~ Grid Side Converter Rotor Side Converter Figure 3. the main mathematic equations. The main components of a DFIG wind turbine are shown in Figure 3. The rotor current is regulated by the power converter to control the electromagnetic torque and field current and thus the stator output voltage. the pitch controller and the rotor side controller. the grid side converter with dc-link capacitor. Later.1.1: Doubly fed induction generator The general block diagram for controlling a DFIG is shown in Figure 3. The DFIG is modeled by means of an equivalent circuit model. The basic parts of model consist of the generator and drive train.1   D F I G and Converter Dynamic Model An induction generator with power supply on the rotor is a main part of the DFIG wind turbine. 6     Grid us . the turbine rotor. Model of aerodynamic and mechanical parts are also presented. At first. are presented with a short description of the other electrical parts.2 [3]. that describe the relationship between voltage and fluxes in an induction machine and they are basic equations for establish dynamic model. In DFIG wind turbine. while the grid-side converter is used to control the dc voltage of dc-link capacitor. The active and reactive power is controlled by the rotor side converter. 3. while the rotor circuit is connected to a power converter by means of slip rings and brushes.3   W ind T urbine with Doubly Fed Induction G enerator (D F I G) The objective of this chapter is to provide an overview on the dynamic model of different parts in a DFIG wind turbine. The DFIG can operate in either sub-synchronous or in super-synchronous operation modes due to capability of converter to operate in bi-directional operation power mode.

1 D F I G Model The conventional T-equivalent circuit is used to represent the dynamic model of induction generator (IG). u rq g . i sq u sd . i rq Network Model U .2: Control block diagram of DFIG wind turbine 3. u sq Interface I u sd .v wind Turbine rortor Wind Model t Tm Drive-Train Model t Pitch Controller Pg Te Generator Model g i sd .1. 7     i cd . u rq u rd . The power supply of the rotor comes from the rotor-side converter. The rotor current is controlled by the rotor-side converter and consequently the active and reactive power can be controlled.3. i cq i rd . The equivalent diagram of the stator and rotor and the gridside and rotor-side converters is shown in Figure 3. The rotor also has a three phase winding but it is connected to the rotor power supply through slip rings and brushes. u sq u rd . The three phase stator windings directly connected to the grid through a step-up transformer. This model consists of stator winding and rotor winding with slip rings. i rd Rotor-Side Converter u dc DC link& Grid-Side Converter i ref rd i ref rq Power/Speed Controller Pg Reactive Power Controller i rq Qg Figure 3.

1) ur where: Rr i r d dt r j( 1 r) r (3.2) Rs Rr Lm stator resistance rotor resistance mutual inductance stator voltage stator Flux synchronous frequency 1 Ls Lr ( r stator leakage inductance rotor leakage inductance 1) r rotor back emf us s ur r rotor voltage rotor Flux rotor frequency 1 2 r r Slip frequency 8     .3.3: Dynamic equivalent circuit of the DFIG and converters Based on the equivalent circuit presented in Figure 3. the electrical dynamics equations law in the stator and rotor and in per unit system it can be written as us Rs i s d dt s j 1 s (3.Induction Generator ig + Grid Rs + j 1 s- + ~ Ls d s + Eg - us - Lm i s i r Lr im Rr + d dt - + r ~ 1) ir r + j( r dt ur - Rf i sf Lf u sf Grid Filter + + = GSC Cdc vdc - = RSC DC Link Figure 3.

For example. i sq sq . i. in the dq reference frame.7) Thus the voltage equations can be written as usd usq Rs isd Rs isq 1 sq 1 sd (3. respectively.is Stator current ir Rotor current The subscripts s and r refer to quantities of the stator and rotor.3) r Lr i r Lm i s (3. the amplitude and rotating speed of the stator and rotor fluxes are assumed to be constant.6) In power system studies.4) where Ls Lr Ls Lr Lm Lm The electromechanical torque is Te Im[ s * ] s is ( sd . d dt s 0 d dt r 0 (3. the stator and rotor fluxes remain constant.e.5) The electrical power will be Pe Im{ r i * r } (3.9) 9     . it is desirable to reduce the complexity of the system by using reduced-order models that can be obtained by assuming some of the derivatives as being equal to zero. The flux dynamics are canceled to get a fast response to the commands from electric controls of converter. s L s i s Lm i r (3. Under balanced network conditions.8) (3. Consequently. The dynamic model of an induction machine is usually presented by means of a so called fifth-order model that represents IG by a system of five general differential equations of an idealized induction machine. i sd ) (3. a third-order model of IG is obtained when we neglect the stator flux transients.

22) (3.18) (3.20) (3.10) (3.15) (3.13) (3.16) Lr i rq Lm i sq We can also neglect the stator resistance because it is considerably small compared to the stator reactance.24) 10     .urd u rq Rs i rd Rs i rq s s 1 rq 1 rd (3.12) The flux linkage equations are as follows: sd sq Ls i sd Ls i sq rd Lmi rd Lm i rq (3.21) Te Lm (isd i rq isqi rd ) The active power of stator.14) rq Lr ird Lmisd (3. rotor.23) (3.11) If d-axis of the synchronous frame is fixed to the stator voltage. grid-side converter and the injected active power to the grid can be determined as: (3.19) u sd The electromechanical torque is calculated as: Te rq i rd rd i rq (3. then usd 0 (3. hence sd 0 (3.17) Subsequently i sd Lm i rd Ls 1 sq (3.

2 F requency Converter Model The induction machine of a DFIG wind turbine operates in a generating mode at subsynchronous and super-synchronous speeds where the speed range is limited by the maximum voltage of the rotor-side frequency converter [6]. CH is the chopper circuit and CR is the crow bar circuit.i rd u rq . rotor. and controls the flux of the DFIG and thus the active and reactive power [6].i rd u rd .28) By assuming the grid-side converter reactive power is zero. dc link grid-side converter active power and the total generator active power. Pr . depending on the rotor speed. If required. 11     .25) Where Ps .i rq (3. against high currents during grid disturbances. 4.i sq u rq .29) Qg usd . Pc .30) 3. respectively . where RSC is the rotor-side converter.i rd urq .isq u rd . The rotor-side converter operates at different frequencies (slip frequency). thus it can be written as Pg usd . it is possible assign the grid-side converter to deliver reactive power to the grid. The grid-side converter operates at a network frequency and controls the voltage level in the dc link circuit. and Pg are the stator.1. The location of crowbar resistor is shown in Figure 3.27) (3.i rq (3. The rotor winding is supplied from a converter consisting of back-to-back connected converter bridges with a dc link. The crowbar protects the rotor-side converter. GSC is the grid-side converter.Pg Ps Pc (3. as well as the rotor circuit of the DFIG.isd u rd . as shown in Figure 3.4.i rq (3.26) The reactive power can be calculated as: Qs Pc u sq . If the losses in the converter ignored the converter power will be equal to the power injected to the rotor (P c = P r). the generated reactive power will be Qg Qs Qr (3. The rotor-side converter has an over-current protection called crowbar.

5. Rotor Control Mode The reference rotor speed is obtained from power-speed characteristic curve as shown in Figure 3. pf . The rotor-side converter can be simplified as a currentcontrolled voltage source converter.1 Rotor-side Converter The power rating of the rotor-side converter is determined according the maximum slip power and reactive power control capability.2. The rotor reference current in dq reference frame is then calculated by the following expression [3] (4.RSC control Power GSC control Pref P Q.4: Typical layout of a DFIG 3.Ut Q UDC Pitch and DC Link CR RSC CH GSC Ut DFIG Figure 3.31) 12     .1. The rotor speed is then used as an input for a PI controller to get the stator active ref power reference P s .

The magnitude of reference rotor current in q direction is calculated as: i ref rq 1 Ls i sq 1 Lm u sd (3.5: Power-speed curve It can be found from above equation that the stator active power is controlled by controlling the rotor current in d direction.Power Prated Pitch controlled Power Limitation Maximum Speed operation Optimum speed curve Minimum speed operation Speed min sync max Figure 3.6.32) The control diagram for active and reactive power controls is shown in Figure 3.6: Rotor current controller: (a) active and (b) reactive power control 13     . a) Active power control P gen ref r + r P ref s ( L ls L m ) i ref rd Lm vs b) Reactive power control Q ref s + Q s i ref rq Figure 3.

The crowbar deactivates the rotor-side converter and short-circuits the rotor winding. i crow R crow Rotor Side Converter Figure 3. these abilities are seldom utilized since they require a larger converter rating [6]. One possible solution to avoid over-current is to use a rotor over-current protection. In a simplified model. a so-called crowbar.7: The crowbar connected between the rotor of the DFIG and the rotor-side converter. meaning that the converter is able to follow its demanded value at any time [6].2   Grid-side Converter The grid-side converter (GSC) is usually represented by a simplified model.7. The crowbar is connected in parallel with the rotor-side converter as shown in Figure 3.1. a so-called generic control scheme based on a set of PI controllers. In order to protect against over current and over voltage. whenever the rotor current exceeds the maximum limit value. In other words. Thus the rotor and rotor-side converter need to be protected against over-current and over voltage. However.O ver-C ur rent Protection (C rowbar) The new Swedish grid code requires grid connected wind turbines to remain connected to the network even when there is a short-term power network disturbance.2. However. The voltages in d and q direction are calculated according the active and reactive power reference values. the rotor-side converter is restarted. 3. riding through grid faults with DFIG will lead to high peak currents in the rotor and high power into the DC circuit. When the grid fault is cleared. the voltage at the terminal of DFIG goes down almost immediately and it will cause high transient rotor current. the switching dynamics of converter is neglected and an ideal control is assumed. 14     . a DC-chopper parallel with DC-link and an AC-crowbar installed in the rotor circuit are used. The converter can also be utilized to support grid reactive power during a fault and it can also be used to enhance grid power quality. the wind farm should have fault-ride-through capability. When there is a short circuit in the grid.

The equivalent circuit of the dc-link model is shown in Figure 3.9: Simplified model of the dc-link and the grid-side converter. Accordingly. (3. . such as in [9]. the detailed model of GSC for dynamic simulation is recommended.34) ref v dc + PI - GSC model ig + ir + 1 C 1 s v dc Figure 3. RSC ir Cdc ig + GSC =   ic v dc - =   Figure 3.9 the simplified model of dc-link and grid-side converter is represented. In Figure 3. The aim of the grid-side converter control is to maintain the level of dc link voltage v dc at a pre-set value v dc ref by means of a PI controller.In some studies.8: dc-link equivalent circuit The energy stored to the dc-link capacitor can be written as (3. the results for rotor current calculation and dc link voltage are not accurate enough. The author argued that if the detailed model is not used. The dc-link capacitor voltage can be calculated as a function of the input power to the dc link [6]. respectively. 15     . where P r is the power to the rotor and Pg is the power from grid. the main objective of the GSC is to control the dc-link voltage assuming that the power factor is unity. Nevertheless. the majority of DFIG studies use the simplified dc-link voltage controller.33) Where P is the net power flow into the capacitor C dc is the dc-link capacitor value and vdc is the capacitor voltage.8 where i r and i g are the dc currents flowing from dc-link to the rotor-side converter and from dc-link to the grid-side converter. P equals to P r-Pg.

aerodynamic model. requesting the converter to deliver this amount of power to the grid 3.2 M echanical Model and T urbine Control The mechanical model of a DFIG wind turbine and its control in dynamic studies of power systems usually consists of wind speed model. turbine and generator inertias. it may destroy the mechanical parts. and can have significant impact on dynamic simulation results.3.2 D rive T rain Model The two-mass model is commonly used to model the rotor dynamic simulation. the blades is set at a certain pitch angle to get maximum power from wind. i. The mechanical model and control of the wind turbine is simplified in this study.2.e. This model. On the hand if the available wind power is less than the equipment ratings.1   T urbine Speed Control Model The basic strategy for the wind turbine control system is to get the maximum power from the wind. Accordingly. The dynamics of the pitch control are moderately fast. the mechanical input power has to be limited when the input power from wind is higher than the equipment power ratings. The turbine control model sends a power order to the electrical control according the available wind speed.10. The input and out powers to the different parts of a wind turbine is shown in Figure 3. the drive-train consists of two inertias.2.11. The limiting factor that affects the amount of energy from wind is the equipment power ratings. This ability of controlling the mechanical input power is called pitch control of turbine blades. Pwind Paero Pmech Pelec Induction Generetor Wind Energy Turbine Drive Train   Grid   Figure 3.12. The mechanical power from the wind power model and the electrical power from the generator/converter model are used as inputs to calculate the rotor speed. drive train model and rotor speed and pitch angle controller subsystems. Otherwise. 16     . To reduce the amount of input mechanical power the turbine blades must be turned. The two inertias are connected through a spring as shown in Figure 3.10: The input and out powers in WT 3. The block diagram of the system is shown in Figure 3.

12: Two-mass torsional model Ht Hg K H = Ht+Hg Dshaft Freq1 Ks Turbine inertia constant [MW.11: Drive Train of Wind Turbine t o -­ -­ 1 2 Ht P mech T mech + 1 s tg + -­ t + + t tg D shaft P gen tg 1 s g + + K T elec g + -­ + 1 2 Hg 1 s g o Figure 3.Turbine Dm T mech t T elec Generator Dt Ks Dg r Jg   Jt Figure 3.sec/MVA] Generator inertia constant [MW.sec/MVA] Gain factor Total inertia constant [MW.sec/MVA] Shaft damping factor [pu] First shaft torsional resonant frequency [Hz] Shaft stiffness 17     .

with Cp curves for the GE wind turbine are shown in figure 3. like blade element method. There are different methods to calculate C p. The function of the wind power module is to compute the wind turbine mechanical power (shaft power) from the energy contained in the wind using the following formula: (3. the rotor speed is 18     . look-up table and analytical approximation. the turbine speed is controlled primarily by the electric power order to specify speed reference.13. is the wind speed [m/sec]. the mechanical power to the turbine is a function of wind speed.2. ]. C p is a characteristic of the wind turbine. blade pitch angle and shaft speed.13: Wind power Cp curves For power levels below rated.3. For power levels above rated. Figure 3. for instance. is turbine angular speed.35) 3 is the mechanical power extracted from the wind.36) Where r is the rotor blade radius. and is wind speed. it is important to include the aerodynamic model. In this model. C p is provided as a set of curves relating Cp to . In this project. A r is the 2 area swept by the rotor blades [m ] or . The relationship between blade tip speed and generator rotor speed is a constant ( ) and it is defined as (3.3 Aerodynamic and Pitch control Model For a power system simulation involving grid disturbances and fault-ride through capability. the second method is used. and Cp is the is and the wind speed (v tip/vw is the blade pitch angle in degrees.

The pitch controller. Typically high wind speed threshold is 25 [m/sec]. is mechanical input power to the shaft . The block diagram to control of active power is illustrated in figure 3.14 including the pitch control and pitch compensation.controlled primarily by the pitch control. The pitch angle controller is shown in figure 3. the unit is tripped.15. P cmd is the command power and is the pitch angle. is the output electrical power measured at the terminal of wind turbine.14: Pitch control and pitch compensation block diagram 19     . The measured rotating turbine speed is compared with the reference speed and the difference is integrated and if this value is higher than a specified value. pitch compensator and drive train models are included in the block diagram. is the current injected from the wind turbine to the grid. In the diagram is the voltage measured at the terminal of wind turbine. It has also the ability to trip the machine if turbine speed falls below a specified value. is the wind speed. is the wind turbine rotating shaft speed. From Turbine Model + - err PI + cmd + Blade Pitch Angle From Converter Control Model ref Pitch Control The measured electric power Pe + P err PI P ref Set point Pitch Compensation   Figure 3. with the speed being allowed to rise above the reference transiently.

output. including: Power flow and related network functions. dynamic simulation. Therefore. It can perform both phase vector and electromechanical simulations and has become one of the most commonly used software within the power system industry. ) + err Pitch Controller Figure 3. In addition to the steady-state and dynamic analyses.3   PSS/E Simulation Software PSS/E stands for Power System Simulator/Engineering and is mainly used for power system studies. 20     . manipulation and preparation.g. optimal power flow. Power System Simulation for Engineering (PSS/E) is composed of a comprehensive set of programs for studies of power system transmission network and generation performance in both steady-state and dynamic conditions. only one wind turbine model is so far available publicly. used for verification of grid codes however most of them are bound with non-disclosure agreements. Many wind turbine manufacturers provide PSS/E models that are e. network equivalent construction.15: The block diagram for active power control in wind turbine 3. PSS/E also provides the user with a wide range of auxiliary programs for installation. balanced and unbalanced faults.V term V term Pe I inj Current filter I cmd P e Power P ref filter ref P cmd wt T cmd Pe + Pa Ta Drive Train Model wt + - err PI P max Pmech wt ws wt Pitch Compensator - PI PI + Pe + C p( . PSS/E is developed by Siemens Power Technologies International (PTI). data input. PSS/E can be utilized to facilitate calculations for a variety of analyses.

4 Built-in W ind T urbine Model in PSS/E The dynamic model for DFIG in PSS/E model is according to the recommendation by GE Energy. four device models are used: generator/converter model (WT3G). such as faults on transmission system. These models are represented as four boxes in Figure 3.Power F low A power flow study (also known as load-flow study) is an important tool involving numerical analysis applied to a power system.: reactive. A power flow study also can be used to determine the best and most effective design of power systems. The dynamic modeling simulation is used to ensure the reliability of electricity supply and to predict the performance of the system under a wide range of conditions and to identify any problems and scope measures needed for reliability. real. The model is for positive sequence phasor time-domain simulations. The model is mainly appropriate for bulk power system studies. Dynamics The dynamic simulation program includes all the functionality for transient. to be predicted [1].e. The PSS/E version number 31 has a dynamic model for doubly fed induction generator wind turbine in the library.16. the main strategy of the model will be explained. 2.   To construct a complete WTG model. and apparent). The model assumes that the analysis is mainly focused on how the wind turbinegenerator (WTG) reacts to grid disturbances. The wind turbine is referred to as Type 3 per classification of the Western Electricity Coordinating Council Wind Generator Modeling Group (WECC WGMG). Power flow studies are important because they allow for planning and future expansion of existing as well as non-existing power systems. Unlike traditional circuit analysis. a power flow study usually uses simplified notation such as a one-line diagram and per-unit system. [8] 3. The purpose of the dynamics is to facilitate operation of all dynamic stability analytical functions. and focuses on various forms of AC power (i. 21     . There are three considerations when the model is used for dynamic simulations 1. electrical control model (WT3E). The testing and validation of models will enable the performance of wind turbine on the power system on their own and interacting with other turbines. In the following section. dynamic and long term stability analysis. mechanical control (wind turbine) module (WT3T) and pitch control module (WT3P).

instead.17.4. There is a phase locked loop (PLL) in the actual converter controls to synchronize the generator rotor currents with stator.18 22     P gen .Q P gen . The two first-order low-pass filters represent the electronic control systems. This part represents DFIG and rotor converter model. The flux dynamics have been neglected in order to get a faster response to the higher level commands from the electrical controls through the converter. The input in the upper part is the excitation current.V regbus V term V term I p ( P) Converter Control Model E q (Q ) Generator/ Converter Model P. the amount of injected current to the grid is reduced to mitigate over voltage. The difference between this model and conventional models is that this model has no mechanical variables for the machine rotor.1 Generator/Converter Model The generator/converter model (WT3G1) is the equivalent of the generator and the field converter and provides the interface between the WTG and the network. however due to extremely fast response of PLL dynamics relative to the generator/field converter time frame. These variables are included in the turbine model. it is not shown in the model. They have a time constant of 20 msec.16: GE WTG dynamic model 3. X in the model represents the effective equivalent reactance of generator and T represents the transfer function of generator-network. The input in the lower part is the active current. which adjusts the output voltage. Q gen Shaft Speed I inj X eq Power Order Pich Control Model Speed Order Blade Pitch Wind Turbine Model Figure 3. which specifies the amount of active power which should be delivered to the grid. This current controller delivers the amount of required current that must be injected to the grid in response to the flux and active current commands which are generated by the converter control model [3]. In case of over voltage. The generator is modeled as a controlled-current source as shown in Figure 3. The vector diagram of the voltages and currents is shown in Figure 3.

17: Generator and rotor converter model Y Im E" Y V Y E" V I V X X term E"X X I Y I ter m Re Figure 3.18: Vector diagram of voltages and currents 23     .Figure 3.

The reference voltage magnitude is specified during data entering in load flow study.20.4.19: Converter control model 24     . the voltage magnitude at a particular bus. The reference voltage compared with voltage magnitude at the terminal point of WTG to make the voltage command ( for the generator model. When the switch is in position 1. In position -1 the reactive power command comes from power factor regulation and in position 0 the reactive power command is set to follow a certain amount of reference magnitude for reactive power [7]. The output of PI controller is the reference voltage.21 V reg Qmax Reactive Power Control V ref Q min P F ref Power Factor Regulator 1 -1 0 Qmax varf lg Pgen Q cmd Q min Q ref Figure 3. often the point of common coupling (PCC) with the transmission system is compared with the voltage reference magnitude.2 E lectrical (Converter) Control Model The electrical control model determines the amount of active and reactive power that must be delivered by the generator to the grid. These options can be selected by changing varflg switch position. and this voltage is regulated by sending a reactive command to all of the WTGs. The overall electrical model is presented in simplified model of Figure 3. The model for reactive power control is shown in Figure 3. It compares the measured active power ( ( with the ordered active power from turbine model ( and the measured reactive power at the wind turbine terminal (with the ordered reactive power (from the reactive power control model. The electrical control model provides the voltage and current commands and sends them to the generator model by comparing the measured active and reactive power with the required amount of the active and reactive power. Three different options are provided in the model for reactive power control.3.19. The control procedure for active power is specified in the turbine model. The reactive power command is compared with reactive power measured at the terminal of wind turbine in another model that is shown in Figure 3.

V term V max V term X I Q max vltf lg 1 0 Q gen Q c md PI V min Vref _ + I V term X I Qmin E "qcmd To the Generator control model Figure 3.21 Overall DFIG electrical control model 25     .20: Voltage control model Electrical Control Model Voltage measued at regulated bus Voltage measued at WTG terminal bus V reg V term V ref P F ref Reactive Power Control E "qcmd Voltage Command Pgen Q gen Q ref Active power command from wind turbine model Mechanical Power From Pitch control To the Generator control model P cmd Active Power Control P mech V term Ip Current Command Voltage measued at WTG terminal bus   Figure 3.

The fault-ride-through on the low voltage is a basic part the specification.40 0. Big wind farm consisting several units of wind turbines are connected together to make wind farms with high power rating. wind turbines grow in size. The voltage profile is depicted if Figure 4. the load flow and dynamic simulation of a DFIG wind turbine model in PSS/E subjected to a grid fault is presented.20 0.4 Model Simulation In this section. which require the wind farms to continue their operation during a grid fault for a certain time period. Transmission system operators in the countries where the wind power is a part of main electricity source added some regulations to grid codes.60 t(s) Figure 4. Therefore the wind farm must stay connected to the grid as long as the voltage of the grid where the farm is connected is above the criteria limit.40 1. wind turbines should be able to stay connected to the power network and. This grid code states the wind turbine must not be disconnected from the grid when there is a three-phase fault in the closest bus in the 400 KV grids and for fault duration of 250 milliseconds with voltage recovery to 90% of rated voltage. 4. Recently. Accordingly during a grid fault. U(%) U(%) Prated>100MW 100 90% 100 1.40 1. This study is focused on the medium sized wind parks in which a three phase short circuit fault at point of common coupling causes a 75% voltage dip at the terminal of wind turbine.1.20 0.60 t(s) 0 0. Such wind farms usually had small power ratings.00 1.80 1.5MW<Prated<100MW 60 0.40 0. As the wind farm power rating is higher disconnection of large wind farms from the grid during disturbance will lead to loss of large amount of power from the grid [1]. There are different requirements for wind parks with a rated active power more than 100 MW and medium size ones with rated power between 1.60 0.75 80 80 90% 60 40 40 25% 20 25% 20   0 0.20 1.20 1.50 and 100 MW.1 Grid  Codes  and  fault-­ride  through  capability   In the past the wind farms were disconnected from the grid following a grid fault and then there were no requirements for wind farms to contribute in voltage stability [1].1: Voltage profile for wind farm to perform fault-ride through 26     .80 1. The Swedish technical document which specifies the requirements for connecting wind farms to grid is SvKFS 2005:2.60 0.00 1.

Step-up Voltage Substation GRID Infinte Bus Load 200MW 230. In the same way. each of them has rated power of 1. it serves as initial conditions for stability analysis.5kv Unit 34. This representation is called aggregated model of wind farm. which consists of 67 wind turbines.5 MW. however. Second. The load flow model of a wind farm serves two purposes: First. it serves as the basis for load flow studies including thermal.600kv Prated 100MW Figure 4. In this study.3 Load F low The load flow provides the initial conditions for dynamic simulation.0kv PCC HV Bus Wind Farm LV Bus DFIG Substation 230.2 T est System A test power system configuration for load flow and dynamic simulation studies consist of a wind farm with several wind turbines. the rated active power and reactive power for single unit transformer obtained by multiplying the number of individual unit transformers and the rated active and reactive power of each unit transformer.0kv Transformer 34.4. The aggregated model is considered to be sufficient for analyzing the response of the wind park to grid disturbances and for performance evaluation of wind turbine fault ride-through capability. the wind farm is presented by a single equivalent machine connected to a single equivalent unit of transformer. The rated active and reactive power of the single machine which represents the wind farm is the sum of rated active and reactive power of wind turbines inside the wind farm. However in this test system it is assumed the disturbance is on the PCC and not on the transmission line between wind farm and the main substation and therefore it is shown as a single line. voltage. The voltage level is stepped-up in the substation to a level that is proper for connecting to the grid. The wind farm has an active power rating of 100 MW. The configuration of the power system is shown in Figure 4.5kv Transformer0. The connection between wind farm and main substation is through more than one line in case of disconnecting of one line the wind farm dose not disconnected from the grid.2. A single bus represents the bus collector.2: Power system configuration 4. The quantities that can be obtained from load flow study are the bus voltage 27     . The high voltage side of unit transformer connected to a wind farm transmission line to transfer power to a wind farm substation. and other analyses.

The load flow diagrams of the test system are shown in Figure 4.4 Figure 4. a wind control mode must be specified. The capability curve for a 1.MVAr) 28     .3[7] P PF=0.magnitudes. The maximum and minimum active and reactive power limits according to the capability curve of generators are among the most important input data that must be specified during data preparation. In the end of this data record.5 Under excited -0. active and reactive powers that flow through transmission lines. 3 for a wind turbine generator that operates at a fixed active and reactive power.296pu Q Figure 4. 2 for a wind turbine generator with a constant power factor.5 MW GE wind turbine generator is shown in Figure 4.4: The active and reactive power flow (MW.5 MW DFIG capability curve A wind turbine generator in load flow is treated as a conventional generator and is specified on the existing generator record of the power flow raw data file.96 1.90 PF=0. bus voltage angle. as follows: 0 for a non wind turbine generator. 1 for a wind turbine generator that participates in voltage control provided with upper and lower reactive power limits.3: GE 1.436pu 0.12 Over excited +0.0 0.

the PSS/E version 31 provides a dynamic model for a DFIG wind turbine. In this simulation. 29     . During the fault the voltage at PCC is dropped to zero and thus no electrical power transfer from wind turbine to the grid.5 shows that during the fault on PCC voltage drops to zero. In this study. GENCLS is the classical constant voltage behind transient reactance generator model in PSS/E. and wind turbine and pitch control. each of which defines the location of a dynamic equipment model in the network along with the constant parameters of the model.4 Dynamic Simulation The first step in dynamic simulation is to enter the dynamic model data. the total impedance between the turbine terminal and PCC increases and therefore the voltage drop will increase and the recovered voltage will be less than its pre-fault magnitude. The fault lasts for 250 milliseconds and is cleared by disconnecting the faulted line. When faulted line is disconnected from the network. which is saved in a file.95 pu.1   Constant  reactive  power  without  closed-­loop  terminal  voltage  control   When the reactive control mode of the model is set to a constant reactive power (varflg=0). the infinite grid is represented by a constant internal voltage generator model (GENCLS). Appendix 1 contains model parameters that are used in this simulation.25 pu. load. The wind turbine terminal voltage has a similar pattern. Dynamic simulation is performed based on the load flow data that provide the transmission network. the current in q-direction is also constant during the fault. the voltage gradually recovers to around 0. The disturbance in the simulation study is a three-phase symmetrical short-circuit fault on one of the lines connecting PCC bus and grid bus. a number of simulations are performed to investigate the model response subjected to grid disturbances. which is considerably higher that the PCC voltage. After fault clearing. This file contains a group of records.4. Figure 4. As it was mentioned in previous chapter.4. electrical control. except that during the fault the retaining terminal voltage is 0. and generator data. The model includes generator. 4.

mechanical and aeromechanical power As seen in figure 4.5 0.Voltage Profile PCC WTG 1. The torsional oscillation in the drive-train model is reflected in the output power of the wind turbine.5 3 3.5 1 0.5 0 -0.5 2 2.5 Time [sec] 3 3.2 1 Voltage[pu] 0.06433pu or 6.8 Hz in this particular case.5 Figure 4.6: Electrical. when the fault occurred the electrical power suddenly decreases to a very low value (0. 30     .433 MW) and the difference between mechanical input power and electrical output power make an increasing in the rotor speed and therefore the rotor starts to accelerate. Paero [pu] 2 1.5 4 Pelec Paero Pmech Figure 4. The frequency of the oscillation is 1. Pemach.5 2 time[sec] 2.8 0.5 1 1.6.6 0.4 0.5 1 1.5: Voltage profile Pelec.2 0 0.

7 prior to the fault. During the fault. the wind speed can be assumed constant in the grid fault simulations.5 4 Figure 4. The converters are normally designed for a 10% overload and therefore flowing of the high current during fault will likely violate the rotor converter current limit and can damage the IGBTs inside converters.71 MVAr) to the grid.7: Active.The peak of the mechanical power is considerably high. the reactive power is zero and the wind farm operated at unity power factor. The output current profile is shown in figure 4. As the fault duration and the transient of voltage recovery is short.6 0. It is assumed that the wind turbine works at its rated power at the fault instant.4 0.2 0 0. The Pmech plot shows the damped mechanical power output of the wind turbine due to the pitch controller action. 31     . It can be seen the current level during fault is around 1. Active and reactive power [pu] P Q 1 0.3671 pu (36. which is up to 2 pu during the first cycle of the oscillation. the wind turbine inject reactive power of 0. The scale of this magnitude indicates that the mechanical stress in the drive train is quite high.8.8 0. It can be seen in figure 4.5 1 1.5 Time [sec] 3 3.5 pu and as the magnetizing impedance of generator is much bigger than its stator and rotor impedances it can be assumed whole current goes through the rotor-side converter and rotor winding. the mechanical input power is reduced by pitch control mechanism.5 2 2. reactive and mechanical power profile During the fault.

up to 1.5 4 Figure 4. in this case up to 3 seconds. In other words.5 3 Current[pu] 2. which is not presented here.5 2 1. the model seems to produce too optimistic simulation results.5 2 time[sec] 2.5 1 1.8: Current profile The performance of the model for longer fault duration time has also been investigated.e. This may raise a question regarding the validity of the model to realistically simulate the limitation of the rotor-side converter. show that even when the fault occurs in much longer time period.Output Current I 4 3. 32     . The simulation results.5 1 0 0.5 3 3.5 pu. the turbine is able to ride through the fault despite the fact that the current injected by the generator and hence the converter is considerably high i.

Pitch Angle Pitch 15 14 Pitch [Deg] 13 12 11 10 9 0.22 Speed[pu] 1.Generator and Turbine Power Gen speed Turbine speed 1. this is because the pitch angle is controlled by the speed of the turbine.26 1.5 2 2.11: Pitch angle profile It can be seen in Figure 4.16 1.11. the turbine speed goes 33     .24 1.5 3 3.5 5 Figure 4.5 2 2.5 4 4.2 1.5 1 1.14 1. After fault clearing.10: Wind turbine and generator speed profile It can be seen from figure 4. The pitch angle is oscillating.10 in the two mass model for drive train the speed of generator is oscillating and it is not identical with the turbine speed.5 1 1.5 4 time [sec] Figure 4.12 0 0.5 time[sec] 3 3. the pitch angle is increased during the fault in order to reduce the power input from wind turbine.18 1.

XIQmax=0.4 0.9 1 1.4pu 34     . 4. The reactive power command is equal to the reference value and the internal voltage magnitude is limited in the closed loop controller. The upper and lower limits of the internal voltage magnitude are Vterm+XIQmax and Vterm+XIQmin.5 Figure 4.2 0 0. the maximum voltage drop limit value must be increased.13). Terminal voltage [pu] 1 0.55.12 and Figure 4.2 Constant  reactive  power  with  closed-­loop  terminal  voltage  control   In the second simulation study. The pitch compensation tries to reduce the input power to the turbine by increasing a little more in pitch angle.12: Voltage profile for upper limit value of 0. It was found that using the typical voltage drop limit values given in PSS/E.e.2 Time [sec] 1. Figure 4. This is probably due to numerical instability since the upper and lower limits of the voltage controller are not fixed. i.1 1. the voltage is not able to recover following a grid fault (see Figure 4.6 0.3 1. respectively.4. however the pitch angle continuously increases until few seconds. rather it is dependent on the terminal voltage level (Vterm). This means that the terminal voltage control is enabled.8 0.10). In the simulation.14 shows simulation result where the voltage is successfully recovered with maximum voltage drop limit set at 0.down (see Figure 4.4.4 1. This is due to the effect of the pitch compensation. the voltage drop limits XIQmax and XIQmin must be set by user. the varflg is set to 0 (constant reactive power) and vltflg is set to 1. In order to avoid such a numerical instability.

6 0.9 1 1.55pu 35     .8 0.8 2 Figure 4.4 0.4 Figure 4.4 Time [sec] 1.3 1.Active and reactive power [pu] 1 0.1 1.2 0 0.8 1 1.6 0.4pu Terminal voltage [pu] 1 0.14: Voltage profile for upper limit value of 0.4 0.2 Time [sec] 1.6 1.13: Active and reactive power profile for upper limit value of 0.2 0 0.2 1.8 0.

the impacts of frequent expose of voltage dip on the lifetime of the drive train components need to be investigated further.Active and reactive power [pu] 1. Thus.4 0.4 Time [sec] 1.2 0 0. The simulation results emphasize the importance of the pitch controller model in fault responses of a DFIG wind turbine.6 1. The advantages with the model are summarized as follows: The drive-train model is presented in two-mass model representation. this model does not require time step reduction in order to avoid numerical instability during simulations.2 1 0. The time step required to run the model is compatible with the standard time step in PSS/E simulator. 36     .8 2 Figure 4.5 Simulation result summary It was found from the simulations that the model has advantages and some drawbacks. The pitch angle controller and its function for decreasing the aerodynamic power during and following a grid fault are considered in the wind turbine model. Consequently.55pu 4. The influences of this model on the dynamic response of the wind turbine model are clearly shown in the simulations.6 0.8 1 1. The magnitude of the mechanical power oscillation during the fault indicates the scale of mechanical stress in the drive train.2 1.8 0.15: Active and reactive power profile for upper limit value of 0. Without appropriate representation of the pitch controller the aerodynamic input power would be constant and thereby would result in less accurate predictions.

The procedure for controlling voltage and reactive power is included in the model. It has been observed that during grid faults the generator/converter current can be very high. This is probably due to numerical instability in the controller limiters which are not fixed. The FRT is not included in the model. To avoid such a problem. This provides flexibility for users to adjust the power-speed characteristic of the turbine according to the wind turbine being investigated. rather it is dependent on the terminal voltage level (Vterm). while in reality such a high current magnitude are not permitted due to the limited capability of the converter. constant power factor and constant reactive power. As a consequence. constant voltage. That means the model provide too optimistic results and therefore it does not depict realistic behavior of a typical wind turbine. Wind speed is assumed to be constant throughout simulations and thereby the effect of wind speed changes on the turbine response cannot be simulated using this model. The rotor current and voltage are not represented in the model and therefore the influence of rotor current limits on the wind turbine response during and following a disturbance on the grid cannot be investigated. the generator losses are not included in the model. There are three different options provided in the model to control reactive power/voltage i. the upper limits must be increased with the consequence of much higher reactive current during fault. wind turbine remains intact.e. There is possibility to define the power-speed curve in the model. 37     . regardless the fault duration and the magnitude of voltage dip during grid fault. The main drawbacks with the model can be explained as follows: The generator is modeled as a simple voltage source behind transient reactance. Therefore. There is high risk for voltage instability following a grid fault when the closed loop terminal voltage control is used.

38     .

More accurate model can help the power system operators to study the dynamic performance of the gird under disturbances and predict its behavior in a more realistic way. The under/over voltage protection can be considered. In this study the model has been evaluated by simulating the model subjected to grid disturbances. The PSS/E built-in model investigated in this study emphasizes a detailed representation of reactive power and voltage control. the wind farm must be disconnected from the grid when disturbance is permanent otherwise it can be damaged by high amount of short circuit current and also the under voltages. This parameter has a significant effect on voltage stability of the wind turbine after fault clearing.   39     . the model can be used to study dynamic simulation when it is connected to bulky power system.5 5. Another subject that could be of interest is evaluation of the model for islanding operation. The model does not differentiate between short and longterm/permanent faults. It was found that the under voltage and over current protection is not implemented in the model. Accordingly the model is applicable with care. One of the most important parameters is the closed loop terminal voltage control limits. It is observed that there no limitation in fault time duration and severity.2   Future  works   The results of this work can be basic point for future studies. 5. Some basic parts of DFIG same as rotor crowbar can be added to the dynamic model. however some considerations related with the model must be take into account. The simulation results show that the behavior of the DFIG wind turbine model in PSS/E is dependent on the parameters used in the model. In general.1   Conclusion and F uture Wor ks Conclusions   The modeling of wind turbine generators for power system studies is an important aspect and it is in a rapid evolution state. The dynamic model for the full converter and fixed-speed wind turbines also can be introduced. While in reality. On the other hand the model the model has number of advantages such as the possibility to control the reactive power and voltage using different mode of operation. while the generator model of the wind turbine is simply modeled as a voltage source behind transient impedance.

  40     .

se/Tekniska-krav/Foreskrifter/ [3] and Environment. Finland [7] 4. Helsinki University of Technology. dissertation. Chalmers University of Technology.php?id=178 [2] driftsäk http://www. http://www. Laboratory of Electro mechanics. [6] Ph. [8] [9] PSS/E Program Application Guide -Speed Wind Turbines with Doubly-Fed Induction Generator in Short Term Stabi Int.2002. Göteborg.D. Sweden. [4] Stockholm.2008. issues and recommendations. 2008. Department of Electrical and Communications Engineering. Göteborg.Stoclhol m. Department of Energy and Environment.Workshop on Transmission Networks for Offshore Wind F arms. Chalmers University of Technology.svk. 2009. Brussels: EWEA.ewea.1 Jan. Helsinki 2006. Sweden [5] ion. Large scale integration of wind energy in the European power supply: analysis.References [1] EWEA (European Wind Energy Association) (2005).org/index.Sweden -   41     . Sweden.

436 1.1 Doubly-fed induction generator (WT3G1) Symbol Xeq Kpll Kipll Description Equivalent reactance for current injection(pu) PLL first integrator gain PLL second integrator gain V alue 0.05 3 0.10 0.45 42     .296 -0.A ppendices A .10 0.1 1.05 0.8 30 0 0.W ind T urbine and Networ k Parameters A.05 0.15 18 5 0.12 0.5 PLLMX PLL maximum limit Prated Turbine MW rating A.2 Doubly-fed induction generator electrical control (WT3E1) Symbol Tfv Kpv Kiv Xc Tfp Kpp Kip PMX PMN QMX QMN IPmax Trv RPMX Description Filter time constant in voltage regulator(sec) Proportional gain in voltage regulator(pu) Integrator gain in voltage regulator(pu) Line drop compensation reactance(pu) Filter time constant in torque regulator(sec) Proportional gain in torque regulator(pu) Integrator gain in torque regulator(pu) Max limit in torque regulator(pu) Min limit in torque regulator(pu) Max limit in voltage regulator (pu) Min limit in voltage regulator(pu) Max reactive current limit(pu) Voltage sensor time constant(sec) Max power order derivative(pu) V alue 0.6 1.

12 0.90 V alue VMAXCL Max voltage limit Kqv XIQmin XIQmax Tv Tp Fn WPMIN Wp20 Wp40 Wp60 Pwp Wp100 Voltgae/MVAR gain Min limit of diff.05 1.3 Turbine (WT3T1) Symbol VW H DAMP Kaero Theta2 Description Initial wind.78 0. between Vterm and Eq cmd Max limit of diff.05 0.0070 21. pu P/pu speed Aerodynamic gain factor Blade pitch at twice rated wind speed (deg) V alue 1.RPMN T_Power Kqi VMINCL Symbol Min power order derivative(pu) Power filter time constant(sec) MVAR/Voltage gain Min voltage limit Description -0.0 0.95 0.20 40 -0. MW*sec/MVA Machine damping factor . pu of rated wind speed Total inertia constant.0 0.74 1.0 0.20 A.50 0. between Vtrem and Eq cmd Lag time constant in WindVar controller(sec) Pelec filter in fast PF controller A portion of online wind turbines Shaft speed at Pmin(pu) Shaft speed at 20% rated power(pu) Shaft speed at 40% rated power(pu) Shaft speed at 60% rated power(pu) Minimum power for operation at Wp100 speed (pu) Shaft speed at 100% rated power(pu) 1.00 0.25 4.40 0.69 0.45 5.98 1.98 43     .

50 DSHAFT Shaft damping factor (pu) A.875 1.30 150 25 3.Htfrac Frec1 Turbine inertia fraction (Hturb/H) First shaft torsional resonant frequency(Hz) 0.0 10.0 TetaMax Upper pitch angle limit(deg) RTeta PMX Upper pitch angle rate limit(deg/sec) Power reference(pu on MBASE) 44     .0 30 0.4 Pitch control (WT3P1) Symbol Tp Kpp Kip Kpc Kic TetaMin Description Blade response time constant(sec) Proportional gain of PI regulator(pu) Integrator gain of PI regulator(pu) Proportional gain of the compensator(pu) Integrator gain of the compensator(pu) Lower pitch angle limit(deg) V alue 0.80 1.0 1.0 27.