1

Direct Drive Synchronous Machine Models for

Stability Assessment of Wind Farms
Sebastian Achilles and Markus Poller

Abstract The increasing size of wind farms requires power

system stability analysis including dynamic wind generator models. For turbines above 1MW, doubly-fed induction machines
are the most widely used concept. However, especially in Germany, direct-drive wind generators based on converter-driven synchronous generator concepts have reached considerable market
penetration. This paper presents converter driven synchronous
generator models of various order that can be used for simulating
transients and dynamics in a very wide time range.
Index TermsTransmission networks, variable speed wind
turbines, direct drive synchronous machines, off-shore wind
power generation, power system stability

I. I NTRODUCTION

II. C ONVERTER D RIVEN S YNCHRONOUS G ENERATORS

Figure 1 shows the principal arrangement of a direct drive synchronous generator. Rotor and generator shafts are mounted to the
Sebastian Achilles and Markus Poller are with DIgSILENT GmbH, HeinrichHertz-Str. 9, 72810 Gomaringen, Germany (email: [email protected],
[email protected])

e n

c y

e r t e r

G
G

r i d

Fig. 1. Direct-drive synchronous generator

same shaft without gear-box. The generator is a high-pole synchronous

generator designed for low speed. It can either be an electrically excited synchronous generator or a permanent magnet machine.

ORLDWIDE , the total installed wind power capacity as well as

the average rated power per wind mill is constantly increasing.
In many European countries, particularly in Denmark, Germany, and
Spain, wind-power experienced a rapid growth during the past years
and contributes nowadays considerably to the overall electricity production.
As a result, there is a strong need for transmission system operators to carry out power system stability analysis of the whole system
including dynamic models of on- and off-shore wind farms, so that all
interactions between conventional power plants and wind power generators can be assessed.
For this purpose, standard models are needed that can be used for
simulating complete power systems with hundreds of generators. For
these applications the model complexity must be chosen in a way that
model accuracy is sufficient with regard to the response to wind fluctuations and faults in the electrical network but the calculation time for
simulating large systems remains in an acceptable range.
Nowadays, for units above 1MW, variable speed concepts are usually applied that are either based on doubly-fed induction machines
(e.g. GE Wind-Power, Vestas, RE Power, Nordex, NEG-Micon) or
converter-driven synchronous machines (e.g. ENERCON, Pfleiderer).
Some of the largest units currently available and therefore especially suited for off-shore applications are built on the direct drive
synchronous machine concept. This paper presents an approach for
standard variable speed wind generator models based on this concept.
It includes models of all components, generator and converter (diode
or PWM), typical approaches for the control circuits and aerodynamics
of the turbine.
The focus of the presented models is on power system stability analysis. However, some of the presented models can also be used for
flicker analysis, harmonics studies and electromagnetic transients simulations.
All models have been implemented and tested in the power system
analysis program DIgSILENT PowerFactory [11].

r e q
C

r e q
C

S
G

e a r

u
v

e n

c y

e r t e r

r i d

Fig. 2. Converter-driven synchronous generator with gear-box

Synchronous machines of direct-drive wind generators tend to be

very large due to the high number of poles. Figure 2 shows a compromise between a direct drive concept and a design with gear box.
Here, a robust, single-stage gear box with low ratio is used allowing
for a lower number of poles and hence a smaller generator than in a
pure direct drive wind generator. Electrically, this concept behaves
analogously to the direct drive concept of figure 1.
Permanent magnet machines have usually a higher efficiency and
are more compact than electrically excited machines. However, they
are still considerably more expensive and require more advanced rectifiers because they dont allow for reactive power or voltage control.
The most common concept for direct drive wind-generators is
the Enercon-concept using a salient pole, electrically excited synchronous generator [3].
For allowing variable speed operation, the synchronous generator
must be connected to the grid through a frequency converter. There
is an almost infinite number of possibilities realizing frequency converters and most of them have already been used for wind power applications in the past. The concept having nowadays the widest spread
is shown in figure 3. The generator is connected to an intermediate
DC-circuit by a diode rectifier. The grid-side connection is realized by
a self commutated pulse-width modulated (PWM)-converter that imposes a pulse-width modulated voltage to the AC-terminal. The PWMconverter is connected to the network through a filter, symbolized by
the L-C circuit in figure 3. The level of harmonics in the voltage at the
connection point is extremely low.
The control concept for the inverter of figure 3 consists of fast current controllers that regulate AC currents in an AC-voltage oriented

2
P
g

e n

r e f

- I

r e f

e n

r e f

- I

i d

r e f

i q

r e f

r i d

c
g

e n

- I

- I

i q

e n

e r a t o

r e c t i f i e r

r - s i d

e
b

s t e r

g
i n

/ A

- s i d

e r t e r

i d

r e f

r e f

r e f

r e f

- I

r i d

D
e

Fig. 3. Frequency-converter with generator-side diode-rectifier and grid-side

PWM-converter

III. E LECTRICAL C OMPONENT M ODELING

A. Synchronous Generator
Synchronous machine models for power system analysis are usually
based on the assumption that the magnetic flux distribution in the rotor
is sinusoidal (e.g. [10]). With this assumption, the flux can entirely be
described by a vector.
The resulting synchronous machine model, also named the Parkmodel can best be described in a rotor-oriented d-q-reference frame.
Figures 5 and 6 show the d-axis and q-axis equivalent circuit diagrams of the synchronous generator model of seventh order. The
model considers stator windings, the excitation winding in the d-axis
and damper windings in the d- and q-axis. The inductances associated
to damper windings can either represent real windings or just model
eddy current losses in the rotor. Hence, the described model can be
used for synchronous generators with or without damper windings.

i d

r e f

i q

r e f

e n
c o

/ A

e r a t o
n

- I

reference frame (d-axis: active current, q-axis: reactive current). This

fast current control loops can be found in almost all grid connected
PWM-inverters.
In a slower control loop, active and reactive power are controlled.
The power reference is usually defined by a speed-power characteristic
that drives the wind-generator automatically in the operating point of
highest efficiency (Maximum Power Tracking strategy).
But controlling active power is just one of several possibilities for
realizing the outer, d-axis control loop. Controlling torque instead of
power is another alternative. In many designs, generator speed instead
of power or torque defines the active (d-axis) current reference.
The DC-booster stabilizes the voltage at the DC-terminal of the inverter. The controller is realized by a fast current controller set by an
outer voltage control loop.
Figure 4 shows another frequency converter concept that is very
common, especially for permanent magnet generators. Here, both converters, the generator-side and the grid-side converter are realized by
self-commutated PWM converters.
The grid-side control can be realized in exactly the same way as in
the concept according to figure 3, including all options with regard to
torque or speed control.
The generator-side rectifier regulates DC-voltage through the d-axis
current and AC-voltage through the q-axis current controller.
In case of electrically excited generators having AVRs, reactive
power could be controlled instead of AC-voltage.

i q

r e f

i d

r i d

- I

i d
i q

m
m

r e f

r e f

i d
D

D
C

r - s i d

- I

e r t e r

c o

i q

/ A

r i d

- s i d

e r t e r

Fig. 4. Frequency-converter with two PWM-converters and intermediate DCvoltage circuit

 

 








 



 


Fig. 5. d-axis synchronous machine model

According to the equivalent circuits of figures 5 and 6 the statorvoltage equations can be expressed as follows:

 "
 !$#%'( & *)+ ( ,
,  "
 !$# , %-(& , %.+ ( 

(1)

The rotor-voltage equations are:

0/. 1
 /#2/3%4( & /
5  1
 67#268%9( & 6
5  1
 :#2:%4( & :

(2)

The synchronous machine model is completed by the mechanical

equation:

; + & <>=.?@%A=.B
C & <1A+<

(3)

There is the following relationship between the mechanical variables

+ED and F D and the corresponding electrical values that are used in (1):

+GIHJ+<
C IH C <









 

 




Vdc

UVACac

Fig. 6. q-axis synchronous machine model

Fig. 7. PWM-converter circuit

with H being the number of pole-pairs.

The electrical torque can be calculated by:

=.B H (  # , ) (, #

(4)
G

For completing the synchronous machine model the flux linkage equations are required.
Stator flux linkage:

( 
  "!%#%$&2#%#"!#2/3%#"! #26
(,  "! , %'%$&2# , %#"! , #2:

C
v

a l v

a l v

(5)
Fig. 8. Valve representation by equivalent switches

Rotor flux linkage:

( /.("!
( G
6 ("!
( :.("! ,

#%) "!%#%$+* %#%$ /%2#2/3%, "!%#%$+*-2#26

#)
%  "!#
% %$+*-2#2/3%) "!%#%$+* %#%$ 62#26
#, ,
%  "! , #
% %$ .: 2#2:

(6)

Automatic voltage regulators (AVR) of electrically excited machines can be considered by connecting the AVR-model to  / .
Permanent magnet generators can be modeled with the above equations by keeping the excitation current # / to a constant value.
This synchronous machine model of seventh order is able to represent rotor and stator transients correctly. In stability studies however,
transient phenomena of the electrical network are usually not considered [10]. Applying the principle of neglecting stator transients to
the synchronous machine model reduces stator equations to arithmetic
equations:

1 "
 !$#*) + ( ,
,  "
 !$# , %.+ ( 

(7)

Equation (7) allows expressing the electrical torque (4) by voltages and
currents only:

H , ,
,
,
=.B  
+ /  # %I# )I"!0#21 %I#21 43*6+5 <  7B7)879

However, since switching frequencies are usually in a kHz-range,

such a detailed model requires very small simulation time steps why
calculation time gets very large.
For applications, in which the control behaviour of PWMconverters is in the center of interest (e.g. power system stability),
a fundamental frequency model should be used.
Assuming an ideal DC-voltage and an ideal PWM modulation (infinite modulation frequency), the fundamental frequency line to line AC
voltage (RMS value) and the DC voltage can be related to each other
as follows:

:; :
;
!=< ?
  >  A @  <
(9)
>
The AC-voltage phase angle is defined by the PWM converter.
The pulse-width modulation index @ is5C
the
of the
B  control
@ B variable
PWM converter. Equation (9) is valid for
5 . For values
larger than 1 the converter starts saturating and the level of low order
harmonics starts increasing. The complete characteristic of the PWM
converter, including the saturated range is shown in Fig. 9. Because of
the mentioned increase of low-order harmonics, the saturated are of a
PWM-converter is usually not used or just up to 
.

5D 5D D D 5D

(8)
PWM-Converter Characteristic

Hence, in the fifth-order-model, the electrical torque is equal to the

electrical power (minus stator losses) devided by the mechanical
speed.

0,9
0,8
0,7

Grid-side converters are usually realized by self commutated pulsewidth modulated circuits (see Figure 7). The circuit is built from six
valves with turn-off capability and six antiparallel diodes. Valves with
turn-off capability are typically realized with bipolar transistors or IGBTs (insulated gate bi-polar transistors) because they allow for higher
switching frequencies than classical GTOs.
In older wind generators, line commutated, thyristor-converters
were sometimes used, but due to their high level of harmonics and the
lack of reactive power control capabilities there are not used in modern
designs.
Depending on the application, the circuit according to Figure 7 is
actually modeled using switches according to Figure 8 for representing
the valves.

|VAC |/VDC

0,6

B. PWM Converter

0,5
0,4
0,3
0,2
0,1
0
0

0,5

1,5

2,5

Pulse-Width Modulation Index m

Fig. 9. PWM Converter fundamental frequency characteristic

The converter model is completed by the power conservation equa-

tion:

 <
 < % > 



;
 =! < != < E 5

For the AC-side, the following approximate equations can be used:

This equation assumes a loss-less converter. Because the switching

frequency of PWM converters is usually very high (typically several
hundreds Hz), switching losses are the predominant type of losses.
Since the average switching losses are basically proportional to 6 ,
switching losses can be considered by a resistance between the two
DC-poles in a fundamental frequency model.

V dc

7!=<,7 <
% 
8 5

(10)

D. DC-booster
The average values of voltage and current pulse streams generated
by the dc-booster (see Figure 3) are as follows:

;
;
; 1  @;
1 1  

V ac

Rl

(14)

(15)


B @ B  @
5

The control variable is the pulse-width modulation index

cuit shown in Figure 3 can realize values of
.

. The cir-

Fig. 10. PWM-converter model including losses

IV. P OWER E LECTRONICS C ONTROLLERS

A. Grid-Side Converter

The model according to figure 10 represents switching losses by

the equivalent resistance   . Load losses can be considered by copper
losses of the coupling-reactance.

The grid-side converter (see Figures 3 and 4) operates in a statorvoltage oriented reference frame. Hence, d-axis represents the active
and q-axis the reactive current component.
E 9

C. Diode Rectifier
I
d

243 5 6

ACB=D

2
243 5 6
V
d

V
c

78

;:

@ 356

Fig. 11. Diode Rectifier



The diode-rectifier-circuit is shown in Figure 11. As in case of

the PWM-converter, a fundamental frequency model is used for stability applications. The fundamental frequency model represents ACfundamental frequency and the DC-average values of voltages and currents. At fundamental frequency, a diode rectifier can be represented
Z

V
d

I
c

V
0

according to the equivalent circuit of Figure 12. DC-voltage drops due

to commutation, which are proportional to the dc-current, are represented by the equivalent resistance
.
The source voltage  and the equivalent resistance
are:

<

+>

;

<  +
;

% 
A5 

7 356


!$#%/( )1& 01,

8 !$#% &
( )1>?,

249 8

2 9

Fig. 13. Grid-side current controller

A very fast inner control loop regulates the d- and q-axis current
components of the grid-side PWM-converter. Current references are
defined by a slower outer control-loop regulating active- and reactive
power.
The objective of the MPT-characteristic defining the active power
reference is to drive the generator into the optimum speed-power operation point (optimum GF +H ) at any time.
As mentioned before, there are possible options for the active power
regulation shown in figure 13. In many designs, the active-current
reference is given by a speed controller, instead of the power controller.
In this case, a speed reference is calculated from the actual electrical
power using the inverse of the MPT-characteristic.

B. Generator-Side Converter

(11)
(12)

C. Booster Controller

(13)

The DC-booster of Figure 3 regulates the DC-voltage at the DCside of the grid-side PWM-controller. The concept is analogous to the
PWM-converter control concept. An inner loop regulates DC-current
and an outer voltage-controller loop defines the current reference.

The commutation angle can be calculated using

< 

.-

At the generator-side converter AC-voltage (optional: reactive

power) and DC-voltage of the intermediate DC-circuit is regulated.
As before, the controller is equipped with fast current controllers.
The current reference values are here defined by voltage regulators as
indicated in figure 14.

<

!=<

!$#% ( )+& <=,

78356 

0 

Fig. 12. Equivalent circuit of the fundamental frequency diode rectifier model

"!$#%'( )+& *,

a c

 
    

)    

  




%$  
& 



`#bc d

 


!#" 

'  
( 

!" 

`#a

]
_

Fig. 16. Generic model of the pitch-control system

A. Wind Fluctuations

Fig. 14. Generator-side PWM converter controller

D. Reactive Power Control

Reactive power control is possible through the q-axis component of
the grid-side converter. Variable speed wind power generators can be
operated at a constant power factor, which is the standard operation
mode today. Alternatively, an AC-voltage controller defining the daxis current reference can be used in Fig. 13 instead of the reactive
power controller, or secondary voltage control can be supported by
adjusting the Q-reference (e.g. [4]).

V. T URBINE
Equation (16) shows the aerodynamic equation of a wind turbine
that relates mechanical power to wind speed and mechanical speed of
the turbine (e.g. [5]):

7?E GF0+H*,+ -   1/.

(16)

with:
? : Mechanical power of the wind turbine
- : Air density
: Rotor radius
H
: Tip speed ratio
+ : Blade pitch angle
GF
coefficient as a function H and +
 / :: Power
Wind speed

Wind fluctuations can either be modeled by deterministic models,

where the wave-form of a gust or a wind-speed ramp is defined and
injected into the input  / of the turbine model.
Alternatively, stochastic wind-speed models can be used that generate stochastic signals representing wind-turbulence. A turbulence
model including rotational sampling- and tower-shadow effects was
presented in [6] and is briefly described in [8].

B. Torsional Oscillations
When the simulated applications are limited to the impact of wind
fluctuations, it is usually sufficient to consider just a single-mass shaft
model because shaft oscillations of variable speed wind generators are
not reflected to the electrical grid due to the fast active power control
[5].
In stability analysis however, when the system response to heavy
disturbances is analyzed, the shaft must be approximated by at least
a two mass model. One mass represents the turbine inertia, the other
mass is equivalent to the generator inertia.
The equations describing the mechanical coupling of turbine and
generator through the gear box by a two-mass model can be expressed
as follows (see e.g. [9]):

; ?ik+Ej ?  .
= ?) =.D
i+l <
; <Si  =.G
D %A= B 
i?l <
 +Ej ? ) +<
i'm
i=.
l D>o
 n ? < m ? <7%qp ? <  +Ej ?) +<

The tip speed ratio H is defined by:

H

 E+  ?2
/

(17)

with +E? being the rotational speed of the wind turbine.

According to (16) the aerodynamic behaviour is modeled by a
steady state equation and therefore, dynamic stall effects cannot be reflected by this model. An approximate method for including dynamic
stall effects is described in [6].

CE FG
0

ef g h

) 

CD

QSRUTVNWX
Y Z[\

39

0 678 7,:

;=< 7>@?8@A,B

24

LNM

C M
Fig. 15. Generic wind turbine model

In case of rotor frequencies below + D !#O , active power is regulated

according to the maximum power tracking (MPT) characteristic that
defines the maximum power depending on the shaft speed as power
reference of the power controller. When the maximum shaft speed
is exceeded, the active power setpoint remains constant and the pitch
angle control system (see Fig. 16) starts acting driving the shaft speed
back to the maximum permitted value.

(19)
(20)
(21)

equations turbine inertia, turbine torque and turbine frequency

E+Inj ? these
are related to the generator nominal frequency. The turbine torque
is related to the turbine power (see Fig. 15) by:

=.?E +E7 ??
j

IKJ
1 2 3 45

(18)

The electrical torque is defined by (4) and (8).

C. Protection
In case of heavy disturbances, direct drive synchronous machines
quickly disconnect from the system by blocking the PWM converter.
Usually, the following protection criteria a supervised:
r Under- and overvoltage
r Overcurrent
r Overspeed
With regard to under-voltage tripping, the tendency is more and more
to enable wind generators to stay on the network as long as possible,
even for deep voltage dips (voltage ride-through). With the grid-side
converter, reactive current support can be provided during voltage dips
for lifting the voltage at the generator terminals in case of faults near
to the connection point.

VI. S IMPLIFIED M ODEL FOR T RANSIENT

A NALYSIS

STABILITY

DIgSILENT

2.00

For investigating transient stability phenomena, large system models have to be studied considering the interaction of all generators in
the system. For reducing calculation time, simplified models of direct
drive synchronous generators should be used.
In stability analysis, the model should represent correctly:
r Electrical output at the connection point (P and Q depending on
voltage and angle).
r Generator and turbine acceleration.

-0.030 s
0.965 p.u.

1.00

-0.020 s
0.967 p.u.

0.00

-1.00
P

P
g

r e f

e n

- I

i d

r e f

r e f

Q
g

r i d

-2.00
-0.04
-0.02
-0.00
[s]
PWM Grid-Side Converter: Current, phase A in p.u., switched model
PWM Grid-Side Converter: Current, phase A in p.u., fundamental frequency EMT model
PWM Grid-Side Converter: Current, phase A in p.u., stability model

i q

r e f

0.02

1.25
i d

- I

i q

1.00

V
d

m
d

50.000 Hz
0.966 p.u.

0.75

D
c

/ A

C
1100.000 Hz
0.028 p.u.

0.50
g
c o

r i d

- s i d

900.000 Hz
0.036 p.u.
e

e r t e r

0.25

Fig. 17. Simplified model for converter driven synchronous generator

0.00
0.00

A simplified model can be derived under the assumption that the DCvoltage is constant. This assumption can be justified by the large capacitance in the DC-circuit and by the DC-voltage controllers, which
can be found in both concepts, the concept according to figure 3 and
figure 4.
With this assumption, only the grid-side converter needs to be represented fed by a DC-voltage source (see Figure 17).
Considering converter losses by a constant value (assumption of
constant dc-voltage), the electrical torque can be calculated by (8) using:
B 
 ) 
(22)

< < 87

The model can further be simplified by neglecting the time constants

of the fast current-controllers. Generally, this is a valid assumption for
stability analysis because the current controllers act in sub-cycle timeframes. However, the current-source representation can cause problems in case of faults close to the converter:
Assuming an ideal, three phase fault at the connection point, active and
reactive currents cannot be defined independent from each other, but
they result from the impedance seen from the converter. In reality, the
pulse-width modulation indices would limit and the converter behaves
basically like a voltage source. With the current source approach, numerical problems have to be expected because no solution exists to this
problem.
Except for faults near to the converter, the current source model
provides results with sufficient accuracy. But due to the mentioned
problems, the current source approach is not recommended. However,
for being able to select reasonable step sizes, an A-stable integration
algorithm with variable step size should be used when working with
the detailed fundamental frequency PWM-coverter model.

VII. C ASE S TUDIES

All models presented in the previous sections were implemented
and tested using the commercially available power system analysis
package DIgSILENT PowerFactory [11].

2000.00
4000.00
6000.00
PWM Grid-Side Converter: Current, phase A in p.u., frequency domain

[Hz]

8000.00

Fig. 18. Benchmark of different PWM-converter models

In order to present results of the wind generator models relevant

to power system stability analysis, close fault simulations were performed.
No real data were used for parameterizing the models because of
confidentiality agreements with various manufacturers. Instead, typical parameters have been used for all components. In all cases, 5MW
turbines were assumed, which are not yet available but should be available for off-shore applications in the near future.

A. PWM converter model

In a first example, the different models of the grid-side PWMconverter and the AC-network are benchmarked. The three different
models are:
r Switched PWM-converter model according to the circuit of figure 7 and an EMT1 -like representation of the AC network
r Fundamental frequency PWM-converter model and an EMT-like
representation of the AC network
r Fundamental frequency PWM-converter model and a steady state
AC-network model (stability model).
All models are directly supported by PowerFactory. The network
model (steady state or EMT-representation) can be selected when initializing a simulation run. The PWM-converter model (switched or
fundamental frequency) can be specified with one parameter.
Figure 18 compares the AC-current of one phase calculated by the
three different models. The depicted currents are in p.u. Instantaneous


Electromagnetic Transients, a network model using differential equations

for all AC-network elements

currents are rated to the nominal peak-value, currents of the steady

state network model are rated to the nominal RMS-value.
The labels depicted in figure 18 show that the current of both fundamental frequency models match very well. The spectrum obtained by
an FFT of the current allows comparing the fundamental frequency
component of the switched model with the fundamental frequency
models, which match extremely well.

DIgSILENT

1.20
1.00
0.80
0.60
0.40
0.20

DIgSILENT

B. Wind generator with grid-side PWM converter and

generator-side diode rectifier.

Breaker

0.00
-0.00

0.25
0.50
0.75
Point of Interconnection: Voltage, Magnitude in p.u.
Point of Interconnection: Voltage Phasor Magnitude in p.u., EMT-model

[s]

1.00

-0.00

0.25
0.50
PWM Grid-Side Converter: Active Power in MW
PWM Grid-Side Converter: Active Power in MW, EMT-model

0.75

[s]

1.00

-0.00

0.25
0.50
PWM Grid-Side Converter: Reactive Power in MVAr
PWM Grid-Side Converter: Reactive Power in MVAr, EMT-model

0.75

[s]

1.00

6.00
5.00
4.00

Point of Interconnection

3.00
2.00
1.00

AC Filter

0.00

PWM Grid-Side Converter

2.50

PWM DC Capacitor

2.00

DC/DC Converter
1.50

1.00

DC Bus 1

0.50

DC Inductance
0.00

DC Bus 2
DC Capacitor

Generator Side Rectifier

Fig. 20.
concept

Voltage sag simulation of the Diode-Rectifier, PWM-Inverter-

switched model. However, the speed-results from both models match

very well.
Generator

G
~

Fig. 19. Electrical representation of a wind generator with a grid-side PWM

converter and a generator-side diode rectifier.

The electrical representation of the wind generator according to the

concept of figure 3 is presented in figure 19. The PWM Grid-Side
Converter model includes a series inductance on the AC side. The
controllers applied in the model have the structure depicted in figure 3.
For validating the models under most severe disturbances, voltagedip ride-through capability was assumed even if this is not (yet) provided by most commercial wind-generators of this design. The response of the system to a voltage sag of 80% at the connection point
is presented in figures 20 and 21. The active and reactive power of the
grid side converter recovers quickly. Speed and power of the generator
are also presented (see figure 21). The small dips in the converter and
generator power are associated to the changes in the active power reference generated by the Maximum Power Tracking as a result of the
generator speed.
The results of both models match well. The fast generator power
fluctuations that can be observed in the results of the EMT model are
due to the fact that the connected diode rectifier is represented by a

C. Wind generator with two PWM converters and intermediate

DC circuit
The same voltage sag of 80% was simulated for a wind-generator
according to the concept of figure 4. All controllers were tuned in a
similar way as in the previous cases.
The results of a voltage sag of 80% are presented in figures 22 and
23. Because both rectifier and inverter is modeled by fundamental
frequency models, also in the EMT-representation, the active power of
the generator looks much smoother and is much closer to the stabilityresult than in the previous section.

D. Simplified model of converter driven synchronous wind generators for transient stability simulations
According to section VI, a simplified model was represented. The
electrical model of the generator and the DC circuit dynamics are neglected. However, the mechanical behavior of the generator (shaft,
aerodynamics and blade angle control) is represented and allows for
consideration of wind turbulences [8].
Figure 24 presents the results of the simulation of a voltage sag of
80% in the connection point with the simplified model. The resulting curves are compared against the results obtained with the detailed
model of the concept according to figure 3. The results match very
well. The behavior of the active and reactive power at the point of

1.275

DIgSILENT

DIgSILENT

1.20
1.00

1.250

0.80
0.60

1.225

0.40
0.20

1.200

0.00

1.175

-0.00

0.25
0.50
Point of Interconnection: Voltage, Magnitude in p.u.
Point of Interconnection: Voltage, Magnitude in p.u., EMT-model

0.75

[s]

1.00

-0.00

0.25
0.50
PWM Grid-Side: Active Power in MW
PWM Grid-Side: Active Power in MW, EMT-model

0.75

[s]

1.00

-0.00

0.25
0.50
PWM Grid-Side: Reactive Power in MVAr
PWM Grid-Side: Reactive Power in MVAr, EMT-model

0.75

[s]

1.00

6.00
5.00

1.150
0.00

1.25
2.50
Pitch Controller: Generator Speed
Pitch Controller: Generator Speed, EMT-model

3.75

[s]

5.00

4.00
3.00

8.00

2.00
1.00

6.00

0.00

4.00

1.00

0.70
2.00

0.40

0.10

-0.00

-0.20
-2.00
0.00

Fig. 21.
concept

1.25
2.50
Generator: Active Power in MW
Generator: Active Power in MW, EMT-model

3.75

[s]

5.00

Voltage sag simulation of the Diode-Rectifier, PWM-Inverter-

connection are mainly driven by the PWM control characteristics. The

simplified calculation of the generator speed is perfectly adequate.

VIII. C ONCLUSIONS
The modeling of converter driven synchronous machines commonly
applied to wind generation applications was treated in this paper. The
required representation of the different components particularly for
power system stability assessment was carefully described. Detailed
representation of these wind generator concepts was implemented and
simulation results were presented. A simplified model is also proposed
for large-scale power system analysis. The simulation results of this
model are compared to those obtained with the detailed models for
severe system disturbances. The proposed simplified model provides
adequate accuracy for transient and dynamic stability analysis and is
efficient with regard to calculation time.

R EFERENCES
[1] C. Ender, Wind Energy Use in Germany - Status 31.12.2002, DEWI
Magazin Nr. 22, February 2003
[2] Verband der Elektriztatswirtschaft (VDEW), Stromerzeugung in
Deutschland 2002,17.02.2003, http://www.strom.de
[3] S. Heier Grid Integration of Wind Energy Conversion Systems John
Wiley and Sons, 1999
[4] J. L. Rodriguez-Amenedo, Automatic Generation Control of a Wind
Farm With Variable Speed Wind Turbines, IEEE Transactions on Energy Conversion, Vol 17, No. 2, June 2002

-0.50

Fig. 22. Voltage sag simulation of the wind generator concept with two PWM
converters

[5] J. G. Slootweg, S. W. H. de Haan, H. Polinder, W. L. Kling, Aggregated

Modelling of Wind Parks with Variable Speed Wind Turbines in Power
System Dynamics Simulations Proceedings of the 14th Power Systems
Computation Conference, Sevilla, 2002
[6] P. Soerensen, A.D. Hansen, L. Janosi, J. Bech and B. BakJensen Simulation of interaction between wind farm and
power
system,
Technical-Report,
Risoe-R-128(EN),
2001,
http://www.risoe.dk/rispubl/VEA/veapdf/ris-r-1281.pdf
[7] M. Poller Doubly-Fed Induction Machine Models for Stability Assessment of Wind Farms Proceedings of the 2003 IEEE PowerTech Conference, Bologna, 2003
[8] M. Poller, S. Achilles Aggregated Wind Park Models for Analyzing
Power System Dynamics accepted for: 4th International Workshop on
Large-Scale Integration of Wind Power and Transmission Networks for
Offshore Wind Farms, Billund, Denmark 2003
[9] P. M. Anderson, B. L. Van Agrawal and J. E. Ness, Subsynchronous
Resonance in Power Systems, IEEE Press, New York, 1989
[10] P. Kundur Power System Stability and Control, McGraw-Hill, Inc.,
1994
[11] DIgSILENT GmbH DIgSILENT PowerFactory V13 - User Manual,
DIgSILENT GmbH, 2003

DIgSILENT

1.275

1.250

1.225

1.200

1.175

0.00

1.25
2.50
Speed measurement: speed
Speed measurement: speed, EMT-model

3.75

[s]

5.00

DIgSILENT

1.150
1.20
1.00
0.80

6.00

0.60
0.40

5.00
0.20
0.00

4.00

-0.00

0.50
1.00
Point of Interconnection: Voltage, Magnitude in p.u.

1.50

[s]

2.00

-0.00

0.50
1.00
PWM Grid Side: Active Power in MW (Simplified Model)
PWM Grid-Side Converter: Active Power in MW (Detailed Model)

1.50

[s]

2.00

-0.00

0.50
1.00
1.50
PWM Grid Side: Reactive Power in MVar (Simplified Model)
PWM Grid-Side Converter: Reactive Power in MVar (Detailed Model)

[s]

2.00

-0.00

0.50
Simplified Generator: speed in p.u.
Generator: speed in p.u.

[s]

2.00

6.00

3.00

5.00
4.00
3.00

2.00
2.00
1.00

1.00

0.00

0.00
0.00

1.25
2.50
Permanent Magnet Generator: Active Power in MW
Permanent Magnet Generator: Active Power in MW, EMT-model

3.75

[s]

5.00

3.50
3.00
2.50
2.00

Fig. 23. Voltage sag simulation of the wind generator concept with two PWM
converters

1.50
1.00
0.50

B IOGRAPHIES
Sebastian Achilles received the Electrical Engineer degree from
the University of La Plata, Argentina and the MS degree in Electric
Power Eng. from the Rensselaer Polytechnic Institute in Troy, NY in
1996 and 2000 respectively. He worked on power systems consulting
activities for ABB T&D Systems (Spain), GE Power Systems Energy
Consulting (NY, USA) and for the Instituto de Investigaciones
Tecnologicas de Redes y Equipos Electricos (Argentina). He currently
works in the Consulting Department of DIGSILENT GmbH carrying
out training and consulting activities on several topics including
integration of wind generation to power systems, power system
stability, network planning, test and modeling of generator controls
and insulation coordination
Markus Poller received Dipl.-Ing. degrees from the University
of Stuttgart and Ecole Nationale Superieure des Telecommunications
Paris in 1995. In 2000, he received the Dr.-Ing. degree from the
University of Hannover. Since 1995 he works with DIgSILENT
GmbH, Germany, where he holds the position of Manager Research
& Development. His activities involve the development of models
and algorithms for the power system analysis program DIgSILENT
PowerFactory, power system studies and the presentation of softwareand power system analysis courses. His current research interests
include wind power systems, optimal power flow dispatch and
probabilistic load flow analysis.

1.275
1.250
1.225
1.200
1.175
1.150
1.00

1.50

Fig. 24. Comparison of detailed and simplified stability models