1

Wind Turbine Modelling Approaches for Dynamic Power System Simulations

J. Soens, J. Driesen and R. Belmans

AbstractThe level of detail for modelling wind turbines depends on the application. In the first part, a detailed model is described for assessing the turbine and grid behaviour in a distribution grid. In the second part, the development of a simplified turbine model is described for constructing generic wind farm models. This allows estimating the potential in a given point of the transmission grid to absorb an amount of wind power. Simulation examples are given for both the detailed and the simplified model, applied for respectively a Belgian distribution grid and the Belgian transmission grid. Index Terms-- Power system simulation, Wind power generation, Wind power modelling,

(fast) output power or frequency control; voltage control; black start capability; economic dispatch and financial trade reinforcements

I. INTRODUCTION The steadily increasing amount of wind power throughout various UCTE-countries puts new challenges to the power system operators, who have to ensure a reliable, safe and economically manageable grid operation. Therefore, the modelling of wind turbines for power system simulations is a matter of high interest. The development of these models has been the subject of many discussions: it requires a compromise between making substantial simplifications to reduce computational efforts on the one hand, and maintaining the necessary adequacy to be able to predict the farms influence on the systems dynamic behaviour on the other hand. Apart from the static impact on the power grid, also the dynamic behaviour of a wind farm must be investigated. This way, more insight is obtained about the ability of a wind farm to provide grid support. Grid support, also known as ancillary services, represents a number of services that the power system operator requires from power generators, in order to secure a safe, reliable, stable and economically manageable grid operation. These ancillary services include support for [1]:
This research is part of the IWT-GBOU research project Embedded Generation: A Global Approach To Energy Balance And Grid Power Quality And Security, and of the Belgian Federal Science Office project Optimal Offshore Wind Energy Developments in Belgium. The authors are grateful to the Belgian Fonds voor Wetenschappelijk Onderzoek (F.W.O.) - Vlaanderen for their financial support of this work. J. Soens is a doctoral research assistant of the F.W.O.-Vlaanderen. J. Driesen holds a postdoctoral research fellowship of the F.W.O.- Vlaanderen. The authors are with the Department of Electrical Engineering, ESATELECTA, Kasteelpark Arenberg 10, B-3001 Heverlee Belgium (corresponding authors email: joris.soens@esat.kuleuven.ac.be)

The relation between wind farms and grid support has been extensively discussed over the past years, especially in Denmark and Germany, where the relative amount of wind power in the power grid is the highest of Europe. Specific grid connection requirements for wind turbines were issued first by the Danish and German grid operators, and are used as a reference by most European grid operators who have to take a large amount of wind power in their power system into account. The actual existing grid connection requirements are mainly focussed on the first two mentioned ancillary services: (fast) output power control and voltage control. With advanced technology for wind turbines generators, the performance of those generators can be considered as high as with conventional generators, regarding these issues. On the other hand, even the most advanced turbine technology does hardly improve the capability of wind power to facilitate the economic dispatch of the power market and financial trade reinforcements. These issues can not be enforced by technical grid connection requirements, but must be part of the economical risk that a wind farm operator is willing to take. The criterion for success in these issues is mainly the accuracy of wind speed predictions on a (mid)long term, rather than the turbine technology. This will not be further discussed in this paper. This paper gives an overview of the required properties for a detailed wind turbine model. A simulation example is performed to demonstrate the impact of a wind turbine in the distribution grid of Leuven (Belgium). Regarding ancillary services, the connection requirements for wind turbines in distribution grids are at this moment not yet focused on dynamic voltage control. In most cases, the power factor (cos) is required to be as close as possible to 1. A simulation example shows the impact of various types of wind turbine generator types on the voltage at the neighbouring nodes. In the second part of this paper, an introduction is given to the modelling of a wind turbine as an equivalent transfer function. This is more deeply discussed in the referred paper [6]. The purpose of this less-detailed model is to make an

2 estimate of the impact of large wind farms on the transmission grid behaviour. A simulation example with a hypothetical offshore wind farm connected to the Belgian transmission grid is given. II. DETAILED SINGLE TURBINE MODEL A. General References [1], [4], [5] describe detailed models of wind turbines for power system simulations, equipped with either squirrel cage induction generators, doubly fed induction generators or synchronous generators. Those models consist generally of: a wind speed model; an aerodynamic model for the turbine; a model for the shaft coupling and gearbox; a generator model, containing the voltage differential equations and flux equations, mostly in a rotor- or statorflux oriented (d,q) reference frame, as well as the generator motion equation; - models for the power electronic devices (if any); - controller models: pitch control, generator active and reactive power and current control, maximum power tracker; - protective relays; The wind speed model is mostly a time series of measured or well-chosen wind speed values. However, wind speeds can also be generated as stochastic signals, based on a power spectral density function [3]. The aerodynamic turbine model consists mostly of an approximate formula for the coefficient of performance Cp, as a function of wind speed, turbine speed and turbine design. For more detailed models, the Blade Element Method (BEM) can be used, as suggested in [4] but is by many authors assumed to require too much modelling and calculation effort. Shaft and gearbox are mostly modelled as an equivalent torsional spring [1], [4], [5]. The spring stiffness is relatively low. This may result in large torsional vibrations between turbine and generator, affecting considerably the electrical behaviour towards the power system. The shaft model must therefore always be included in models for turbines with fixed speed induction generators, as the shaft stiffness has a considerable impact on the torque pulsations and thus generator current. For variable speed turbines, the torque pulsations are mainly damped by the turbine speed variation, and the modelling of a soft shaft is only necessary when very fast transients are to be investigated. The model for a doubly fed or induction generator contains the stator and rotor voltage differential equations, the flux equations and the mechanical motion equation, as described a.o. in [5]. This is the so-called fifth order model, named after the number of differential equations in the model. In the seventh order model, the voltage equations for the damper winding in the d- and q-frame are also included. This level of detail is however rarely required for power system simulations. In the third order model, the stator transient flux terms are neglected, which is a reasonable assumption in many cases. In the first order model, also the rotor flux transients are neglected, resulting in a set of algebraic voltage equations. The protective relays include over- and undervoltage tripping relays and an overspeed relay. For a doubly fed induction generator, special precautions must be taken in order to protect the rotor frequency converter from overcurrent, as the converter is the most sensitive part of the system to be damaged by overcurrents. In case of rotor overcurrent, the basic action can be: - opening the rotor circuit; - bypassing the converter by shorting the rotor circuit with a crowbar; In [4] is claimed that the 3rd order generator model is sufficiently accurate for small signal stability investigations (e.g. flicker). However, the 3rd order model may give inadequate results for the calculated rotor and stator current. When simulating the transient behaviour of a doubly fed induction generator, these inadequate results may lead to a wrong assessment of the rotor current protection actions. The 5th order model must be used to obtain a correctly simulated transient behaviour. The controllers include pitch control, and, for variable speed turbines, also speed control and active and reactive power control. The reference speed is calculated by a maximum power tracker. The speed of the active and reactive current control loops depends on the generator type. With a synchronous generator with frequency converter, the reference current can be immediately obtained. For a doubly fed induction generator, the stator current is controlled through magnetic interaction with the controlled rotor current. Because of this, the stator current control speed is lower than for the synchronous generator with frequency converter. This is further discussed in paragraph III.B. B. Simulation Example 1) Model Description A detailed model of a wind turbine with fifth-order generator model was developed, following the guidelines of the previous paragraph. The model is fully described in [5]. A summary of the simulations performed in [5] is given here. The distribution grid of Haasrode, an industrial site near Leuven (Belgium), was modelled (Fig. 1). It consists of four radial 10kV-lines connected to a 70kV substation, at which the short circuit power is 430MVA. The total load is 10MW, equally distributed among the different nodes. A 2MW wind turbine is assumed to be connected at node 408. 2) Wind Speed Fluctuation The simulated wind speed is shown in Fig. 2. Two generator types are investigated: the squirrel cage (fixed speed) induction generator and the doubly fed induction generator (variable speed) of which the reference value for the

3 produced reactive power is permanently zero. The generator speed and its active and reactive power are shown for the two cases in Fig. 3, Fig. 4 and Fig. 5. The voltages at node 408 and node 2 are shown in Fig. 6, the pitch angle of the squirrel cage generator turbine is shown in Fig. 7. The pitch behaviour of the doubly fed induction generator turbine was found to be the same as for the squirrel cage generator.
101 SUB SUB 1 2 102 103 104 105 106 107 108 109 110 111

201

202

203

204

205

206

207

208

301

302

303

304

305

401

402

403

404

405

406

407

408

409

WINDT

Fig. 1. Distribution grid of an industrial site near Leuven, Belgium

3) Voltage Dip With the same wind speed input as above, a voltage dip is simulated by applying a three-phase grid fault (fault resistance 0,1 p.u.) at node 403, at t = 120 s. The fault is cleared after 0.5 s. The simulated cases are described in TABLE I. The resulting voltage at node 408, and the turbine and generator speeds for each of the cases are shown in Fig. 8 and Fig. 9. It is concluded that the doubly fed generator with high ridethrough capability (case doubly fed B) does least hamper the voltage restoration after the fault clearance, compared to the base case, as could be expected. The case doubly fed C hampers the voltage restoration considerably. This is because the generator consumes a lot of reactive power when accelerating towards its supersynchronous speed. Therefore, after the rotor converter is shorted because of rotor overcurrent, the converter should be immediately connected again or either the entire generator should be tripped. Continuing to operate as an induction generator causes too high voltage disturbances.
TABLE I SIMULATED CASES TO ASSESS IMPACT OF A VOLTAGE DIP BASE CASE A No wind turbine installed in the grid Squirrel cage generator, modeled with infinite shaft stiffness (turbine torque directly applied to generator) Squirrel cage generator, modeled with finite shaft stiffness Rotor converter shorted as a result of rotor overcurrent (induced by voltage dip). Generator entirely tripped after 100 ms. Rotor converter rides through voltage dip without being shorted; Rotor converter is shorted, and generator continues to operate as a singly fed induction generator without being tripped

Fig. 2. Wind speed, as input for the simulation

Fig. 3. Generator speed

SQUIRREL CAGE

A DOUBLY FED

Fig. 4. Generated active power

Fig. 5. Consumed reactive power

Fig. 6. Voltages at node 408 and node 2

Fig. 7. Wind speed and pitch angle

Fig. 8. Voltage at node 408, at moment of fault (t = 120 s) and restoration (t = 120.5 s)

The voltage fluctuations due to wind speed fluctuations are lower with the doubly fed induction generator, because its reactive power consumption remains constantly at zero. However the voltage fluctuations caused by a squirrel cage induction generator are also far within the normal voltage fluctuation range in a distribution grid. Only at the start-up of a squirrel cage generator, a large voltage dip is noted, due to the high reactive power consumption at that moment. This must be counteracted, using a softstarter for the squirrel cage generator, to limit the inrush current.

Fig. 9. Generator and turbine speed during voltage dip

4 III. SIMPLIFIED TURBINE AND FARM MODEL A. Active Power Model Starting from a detailed model, a more simplified model active and reactive power is constructed [6]. The detailed model of a variable speed turbine was taken from literature, published by the manufacturer [7]. It was used to perform simulations with sinusoidal wind speeds with a varying average value and varying fluctuation frequency (Fig. 10).
vw ind/vw ind,r , Pmech/P mech,r
a v a ila b le w in d speed tra n s fe r fu n c tio n fo r lo w w in d s p e e d

lo w -p a s s filte r
1

p o w e r c u rv e

v w in d ,a v a il
0 g ra d u a l tra n s itio n lo w w in d s p e e d -> h ig h w in d s p e e d

a d d itio n a l tra n s fe r fu n c tio n fo r h ig h w in d s p e e d

Fig. 12. Equivalent Transfer Function for Active Power

0.6 0.5 0.4 0.3 0.2 0.1 0

freq = 0.05Hz

0.6 0.5 0.4 0.3 0.2

freq = 0.5Hz

0.6 0.5 0.4 0.3 0.2

freq = 2Hz

20

a)

40

0.1

10

b)

20

0.1

c)

10

vw ind/vw ind,r , Pmech/Pmech,r

1.4 1.2 1 0.8 0 20 time [s] d) 40

1.4 1.2 1 0.8 0 10 e) time [s] 20

1.4 1.2 1 0.8 0 5 time [s] vwind Pmech

Thus, an equivalent active power model for a wind turbine is constructed, which is well suitable to estimate the turbine power fluctuations during continuous operation. The full explanation of the model derivation, as well as the aggregation into a model of an entire wind farm, is given in [6]. Also the simplified simulation of turbine yawing and active power control is given in [6]. B. Reactive Power Model The modelling of the reactive power generation and the behaviour during grid disturbances does not start from a predefined detailed model from literature. It is believed that future large wind farms will always be able to control the reactive power output, either by control action on the generator itself or by additional devices (such as SVCs or STATCOMs) connected at the point of common coupling. The supplied reactive power is calculated by either a Pcontroller or PI-controller with anti-windup, making sure that the reactive power that the wind farm must supply never exceeds a limit value. The implementation of a PI-controller is supported by most power system simulation software packages, and does not contain any particularities in its use for this model. The speed of the reactive current control depends on the generator type. The two most common generator types for variable speed turbines are - doubly fed induction generators - synchronous generators with frequency converter, (mostly combined with a direct drive turbine configuration) The current control loop of the synchronous generator can be much faster than with the doubly fed induction generator, as the entire machine active and reactive power is processed by power electronic converters. Suggested values for the time constant TICTL (the time constant of the current control loop) are: - TICTL = 20ms for synchronous generator - TICTL = 200ms for doubly fed induction generator
A detailed description about the modelling of the current controller, as well as the impact of the reactive current control on the active power control (in case of overcurrents) can be found in [6].

f)

10

Fig. 10. Wind speed and mechanical power for two values of average wind speed and three different wind speed fluctuation frequencies

All simulation results are summarized in the plot of Fig. 11, in which the amplitude of the power oscillations is set out against the fluctuation frequency of the wind speed, using the average wind speed as parameter. Two sets of curves can be distinguished: curves for high average wind speed (slope 20dB/decade in low frequency region) and for low average wind speed (horizontal in low frequency region). The reason for this is the different effect of speed and pitch control of the turbine, below and above rated wind speed. The full explanation of the results is given in [6]. The curves of Fig. 11 suggest to simplify the complicated active power model by two equivalent transfer functions (Fig. 12), one for high and one for low wind speeds. The time constants are calculated to have an optimal match between the curves of Fig. 11 and the frequency response of the transfer functions. Suggested values are: Tlow = 7 s d = 0.3 T0 = 0.52 s Khigh = 0.06
P
0 -10
vwind,avg = 9 m/s vwind,avg = 8 m/s vwind,avg = 7 m/s vwind,avg = 6 m/s
mech

amplitude-frequency characteristic
vwind,avg = 10 m/s vwind,avg = 11 .. 20 m/s

, am plitude [dB] fluctuating P

-20 -30 -40 -50 -60 -70 -80 -3 10

m ech

10

-2

10

-1

10

10

10

wind

fluctuation frequency [rad/s]

Fig. 11. Frequency Characteristic of Power Fluctuation Amplitude

5 C. Simulation example 1) Assumptions As an illustration, a simulation for a hypothetical 500MWwind farm in the Belgian North Sea is performed. The wind farm is connected at the 150kV-substation of Slijkens, one of the three coastal 150kV substations in Belgium. The connection is shown in Fig. 13. The wind farm power is assumed to be collected at 30kV, and transformed by an offshore transformer towards 150kV. The grid connection is made by a submarine 150kV cable. The cable characteristics have impact on the simulation results. Especially its capacitance, which is proportional to the cable length, is not neglectible. An active and reactive power model for the wind farm is made as described above. The wind farm is assumed to consist of five turbine rows, orthogonally oriented to the wind speed. The Belgian power grid model contains: - all 400 kV, 220kV 150kV and 70kV substations and high voltage lines of Belgium, including the planned 150kV cable between the coastal nodes Koksijde and Slijkens; - all generation and load data for each substation, as they have been recorded on a representative winter day (19/01/1994); - dynamic models of the governors and voltage controllers of most generators in the Belgian Grid, including the power plant of Herdersbrug, which is the power plant nearest to the coast. The active power production is shown in Fig. 15. The moments at which the wind gust at t = 1000 s reaches each of the five turbine rows can be clearly distinguished. A sudden wind speed increase results in a power increase towards rated power in approximately 150 s. The farm rated power is not fully achieved because of the farm losses, causing a reduction of wind speed for the turbines behind the first row. The rated power is achieved when the wind speed increases further to 25 m/s. The turbines then have to pitch the blades out of the wind in order to avoid excessive mechanical loads and excessive power production. The pitching action goes rather fast, and the farm is able to maintain its output power within a narrow range around its rated power. The moments at which the wind speed gust reaches each of the five turbine rows is again clearly seen. The change in wind speed direction also causes a short drop in power production, which is quickly restored by the yawing action of the turbines. For the active power production, no differences were noted between the four scenarios. The produced reactive power for each of the four scenarios is shown in Fig. 16. In the cases with voltage control, the reactive power production is negative: the farm behaves as an inductor. The resulting voltage in Slijkens is shown in Fig. 17. It is seen that, without voltage control, the voltage at Slijkens fluctuates if the wind speed and farm active power production changes. In the cases with voltage control, the voltage can well be maintained at a fixed value. The cable length has an impact on how the reactive power must be controlled in order to control the voltage at Slijkens. It is seen that the voltage at Slijkens either increases or decreases at the moment of increased active power production. This is because the cable capacitance, which has a large influence on the systems voltage behaviour, is proportional to the cable length, and thus much difference in the behaviour can occur with different cable lengths. The voltage fluctuations at the 150kV substation of Slijkens for the cases a) and b) are far less than 1% (Fig. 17), and thus well within the normal voltage fluctuations that appear on a power system. A wind farm operation strategy at which the farm reactive power is controlled at a fixed value does not result in a gravely decreased grid power quality. It is concluded that the impact assessment of wind speed fluctuations on the grid voltage does not provide an incentive for installing highly advanced generator types or highly advanced voltage control algorithms.
35
30
25
20
15
10
5
0
-5 0
b c c a d

Fig. 13. Assumed Grid Connection of Wind Farm to Belgian Power Grid

A wind speed sequence as in Fig. 14 is assumed. The wind speed direction undergoes a sudden change of 28 degrees (0,5 radians) at t = 5000 s. The turbines must yaw towards the new wind direction. The mismatch angle between the wind direction and the turbines orientation, calculated according to the description in paragraph 1, is also shown in Fig. 14.

w ds e d[m ] a dd c n[d g in p e /s n ire tio e ]

2) Wind Gust Simulation Four scenarios are considered: a) the wind farm produces nor consumes reactive power at the offshore 150kV-node; the transmission cable length is 10 km b) same as a), but with a cable length of 50km; c) the wind farm reactive power is dynamically controlled in such a way that the voltage at Slijkens remains at a fixed value. The transmission cable length is 10km; d) same as c), but with a cable length of 50km.

abcd-

w ind speed w ind direction farm angle mismatch rated w ind speed

1000

2000

3000 time [s]

4000

5000

6000

Fig. 14. Assumed Wind Speed and Wind direction for Wind Gust Simulation

6
600
600
600

500

500

500

f r p wr[ W am o e M ]

400

400

400

controller, such as a static var compensator, is installed onshore, near the point of common coupling (Slijkens). The conclusions from the figures are: - The voltage at the initial moment of the dip is the same for all cases. However, the voltage can be better maintained if voltage support is delivered by the wind farm generators. - The duration of typical voltages dips is some hundreds of milliseconds, and thus the dynamic voltage support by the wind farm must be fast enough. There is nearly no difference between the voltages at Slijkens for the case where the wind farm does not provide voltage support (Fig. 19) and where it provides voltage support very slowly (Fig. 22). - The cable length limits the voltage support that a wind farm can deliver. In each of the cases of Fig. 20, the wind farm supplies the maximum available reactive power (this was set in the simulation model to 1 p.u., i.e. 500MVAr). The effect on the voltage restoration is much less for a 50km-cable than for a 10km-cable. This effect was not yet visible on the curves of Fig. 21 and Fig. 22, because the maximum reactive power was not yet obtained due to the slower control systems.
170

300

300

300

200

200

200

100

100

100

1000

1100 time [s]

1200

0 2950

3000 3050 time [s]

3100

0 4950

5000 time [s]

5050

Fig. 15. Active Power Production by a 500MW-Wind Farm

100
80
60

100
80
60
40
20
a, b

100
80
60
40
20
a, b d

a,b- Q-neutr 10&50km c- DVA R 10km d- DVA R 50km

f r r a tv p wr[ V r ame ci e o e MA]

40
20
0
-20
-40
-60
-80
-100
1000
d c

0
-20
-40

0
-20
-40

a, b d

-60
-80
1100 time [s]

-60
-80
3050 3000 time [s]

-100 2950 1200

-100 4950 3100

5000 time [s]

5050

Fig. 16. Reactive Power Production by a 500MW-Wind Farm

158
157.5
157
156.5
a b

158
157.5
157
156.5
156
155.5
155
c, d c, d b a

158
157.5
157
156.5
156
155.5
b a
a- Q-neutr 10km b- Q-neutr 50km c- DVAR 10km d- DVAR 50km

[V k]

156
Sj e s l kn i

155.5
155
154.5
154
153.5
153
1000

155
c, d

154.5
154
153.5
1100 time [s]

154.5
154
153.5

1200

153 2950

3050 3000 time [s]

3100

153 4950

5000 time [s]

5050

V lta ea S e s[k ] o g t lijk n V

Fig. 17. Voltage at Slijkens 150kV-substation

165
164
163
162

160

165
164
163
162
b d a b d

165
164
163
162
b
a- Q-neutr 10km b- Q-neutr 50km c- DVAR 10km d- DVAR 50km

150

140

130

[V k]

161
of h r fs oe

161
160
159

161
160
159
d

160
159
158
157
156
155
154
1000

120

158
157
c

158
157
c

110

1km 10km 20km 30km 40km 50km

156
155

156
155
3050 3000 time [s]

100 0
5000 time [s]

0.5

1 time [s]

1.5

1100 time [s]

1200

154 2950

3100

154 4950

5050

Fig. 18. Voltage at offshore 150kV-node

Fig. 19. Voltage at Slijkens, wind farm keeps reactive power output at zero
170

3) Voltage Disturbance Simulation due to Grid Fault A grid fault is simulated at t = 1 s, by applying a short circuit in the substation of Brugge, which is located further inland and connected by a 150kV line to Slijkens. The fault is cleared after 300 ms. This results in a 300 ms voltage dip at Slijkens. The depth of the voltage dip depends on the wind farm reaction. For the following simulations, next assumptions were made: - the rated wind farm power is 500 MW, - the wind speed is constant and equal to 12m/s (below rated wind speed); - calculations were made with transmission cable lengths of 1, 10, 20, 30, 40 and 50 km; - in one scenario the wind farm keeps its reactive power output at zero (Fig. 19); - in the other scenario, the voltage at Slijkens is monitored and the wind farm provides dynamic support to control this voltage. The time constant of the farm current controller TICTL is either 20 ms (Fig. 20), 200 ms (Fig. 21) or 2 s (Fig. 22) The voltage at Slijkens for each of the scenarios is shown in Fig. 19, Fig. 20, Fig. 21 and Fig. 22. In terms of voltage control, the scenario with a cable length of 1 km is also representative for the case in which a dynamic voltage

V lt g a S e s[k ] o a e t lijk n V

160

150

140

130

120

110

1km 10km 20km 30km 40km 50km

100 0

0.5

1 time [s]

1.5

Fig. 20. Voltage at Slijkens, wind farm provides dynamic voltage support, TICTL = 20 ms
170

V lta ea S e s[k ] o g t lijk n V

160
150
140
130
120
110
100 0

1km 10km 20km 30km 40km 50km

0.5

1 time [s]

1.5

Fig. 21. Voltage at Slijkens, wind farm provides dynamic voltage support, TICTL = 200 ms
170

V lta ea S e s[k ] o g t lijk n V

160

150

140

130

120

110

1km 10km 20km 30km 40km 50km

100 0

0.5

1 time [s]

1.5

Fig. 22. Voltage at Slijkens, wind farm provides dynamic voltage support, TICTL = 2 s

7 IV. REFERENCES
[1] [2] [3] [4] S. Stoft, Power System Economics, John Wiley & Sons; 1st edition (May 17, 2002) Dynamic Modelling of Doubly-Fed Induction Machine WindGenerators, DigSilent GmbH Technical Documentation, 2003, available at http://www.digsilent.de M. Pller, S. Achilles, Aggregated Wind Park Models for Analyzing Power System Dynamics; V. Akhmatov, Modelling of Variable-Speed Wind Turbines with Doubly-Fed Induction Generators in Short-Term Stability Analysis, Proceedings of the 3rd International Workshop on Transmission Networks for Off-shore Wind Farms, Stockholm, April 11-12, 2002; J. Soens, T. Vu Van, J. Driesen, R. Belmans, Modelling wind turbine generators for power system simulations, European wind energy conference EWEC, Madrid, Spain, June 16-19, 2003; J. Soens, J. Driesen, R. Belmans, Generic Dynamic Wind Farm Model for Power System Simulations, Nordic Wind Power Conference NWPC04, Chalmers University of Technology, Gteborg, Sweden, 1-2 March 2004; R. W. Delmerico, N. Miller, W. W. Price, J. J. Sanchez-Gasca, Dynamic Modelling of GE 1.5 and 3.6 MW Wind Turbine-Generators for Stability Simulations, IEEE Power Engineering Society PES General Meeting, 13-17 July, Toronto, Canada; J. Soens, J. Driesen, R. Belmans, Generic Aggregated Wind Farm Model for Power System Simulations Impact of Grid Connection Requirements, International Conference of Renewable Energy and Power Quality, Barcelona, 31 March, 1-2 April 2004 Accepted for publication Johan Driesen (S93M97) graduated as an Electrotechnical Engineer and received the Ph.D. degree in electrical engineering from the Katholieke Universiteit Leuven (KULeuven), Leuven, Belgium, in 1996 and 2000, respectively. In 1996, he became a Research Assistant of the Fonds voor Wetenschappelijk OnderzoekVlaanderen (Fund for Scientific Research of Flanders F.W.O.-Vl.). From 2000 to 2001, he was a Visiting Lecturer with Imperial College, London, U.K. In 2002, he was a Visiting Scholar with the Electrical Engineering Department, University of California at Berkeley. He is currently a Postdoctoral Research Fellow of the F.W.O.-Vl. at KULeuven. Dr. Driesen received the 1996 R&D Award of the Belgian Royal Society of Electrotechnical Engineers (KBVE) for his Masters thesis on power quality problems. In 2002, he received the KBVE R. Sinave Award for his Ph.D. dissertation on coupled problems in electrical energy transducers. Ronnie Belmans (S77-M84-SM89) received the M.S. degree in electrical engineering in 1979, the Ph.D. in 1984, and the Special Doctorate in 1989 from the K.U.Leuven, Belgium and the Habilitierung from the RWTH, Aachen, Germany, in 1993. Currently, he is full professor with K.U.Leuven, teaching electrical machines and variable speed drives. He is appointed visiting professor at Imperial College in London. He is also President of UIE. He was with the Laboratory for Electrical Machines of the RWTH, Aachen, Germany (Von Humboldt Fellow, Oct.88-Sept.89). Oct.89Sept.90, he was visiting associate professor at Mc Master University, Hamilton, Ont., Canada. During the academic year 1995-1996 he occupied the Chair at the London University, offered by the Anglo-Belgian Society. Dr.Belmans is a fellow of the IEE (United Kingdom). He is the chairman of the board of Elia, the Belgian transmission grid operator.

[5] [6]

[7]

[8]

V. BIOGRAPHIES
Joris Soens was born in 1978 in Belgium. He received the M.S. degree in 2001 as Electrotechnical Engineer from the K.U. Leuven, Belgium. He received the Sidmar Award for his Masters thesis on the power quality of a cycloconverter-driven rolling machine. Since 2001, he has been working as a doctoral research assistant of the Belgian 'Fonds voor Wetenschappelijk Onderzoek Vlaanderen'. His research interests lie in the impact of large wind farms and small distributed generation units, on the transmission system and distribution system level.