Abstract— Wind farms are characterized with some unique features during their normal and faulty operating condi-tions. Different factors participate usually into these conditions such as the distributed generation concept, the own behavior of the induction generator, varying wind speed, … etc. This, conse-quently, arises different challenges regarding the behavior of their protection and control schemes. In this paper, the perform-ance of the conventional protection schemes (currently in use with wind farms) is thoroughly investigated. The aimed investiga-tion study is carried out on a 225 MW wind farm in AL-Zaafarana-Egypt as a simulation example using MATLAB. The package represents an ideal candidate for dynamic simulation purposes with its own highly performance graphical interface and built-in libraries. The major effective factors that may influ-ence the fault characteristics are considered such as fault posi-tion, fault type, fault resistance and pre-fault loading. This ex-plores the performance of wind farms during faulty conditions honestly. It also visualizes the performance of their protection elements and assists enhancing their performance in the future.

Index Wind farms Protection, Dynamic modeling, MATLAB-

Simulink, Fuse.

I. INTRODUCTION

Owing to the rapid increase of the global population and their energy needs, traditional means to satisfy the burgeoning energy demands need careful reevaluation. Coupled with the uneven distribution of resources around the world, economic impacts of large-scale importation and the environmental im-pacts of continued dependence on nonrenewable fossil fuels, there is an imminent need to transfer, at least partly, the de-pendence on to renewable energy resources. Among these re-sources, wind electric conversion has emerged as the leader at the present time. The impressive growth in the utilization of wind energy has consequently spawned active research activi-ties in a wide variety of technical fields. Moreover, the in-creasingly penetration of wind energy into conventional power systems highlights several important issues such as reliability, security, stability, power quality, … etc. Among these issues, providing wind farms with the proper protection is quite es-sential.

The essential benefits from the dedicated protection func-tions are to avoid the possible local damage resulting from in-cident faults and minimize the impact of these abnormal con-

Tamer.A. Kawady, is with the Electrical Engineering Dept., Faculty of

Engineering, Menoufiya University (e-mail: t_kawady@ieee.org). Naema M. Mansour is a Post Graduate Student, Electrical Engineering

Dept., Faculty of Engineering, Menoufiya University (e-mail: n_m_mansour @yahoo.com). A.I. Taalab is with Electrical Engineering Dept., Faculty of Engineering, Menoufiya University (e-mail: taalab@yahoo.com).

ditions on the other sound parts of the network. This reduces the associated negative impacts of the faults on the service continuity and the system stability. Consequently, it enhances the reliability and dependability of the overall grid perform-ance. These terms continuity, reliability …etc. have recently received much attention due to the new deregulation policies and marketing liberalization. Thus, the need for further efforts to improve the existing relaying devices as well as to develop new ones is obvious. On the contrary, it surprisingly has not garnered a sufficient attention tell present. The economic per-spective plays a major role, in which the enormous cost pres-sures usually coerce the wind farm designers for economic causes to remarkably reduce the utilized protection schemes. As a result, wind farms still utilize simple and none-integrated protection methodologies [1]. Also, research efforts regarding wind farm protection are still limited in the literatures [2-5]. As reported by Bauscke et al. in [6], different levels of damage were recorded resulting occasionally from the drawbacks of the associated protection system. Thus, the need for revaluat-ing and improving the existing known protection philosophies is obvious. Hence, the investigation of the wind farm behavior under abnormal operating conditions in addition to exploring its interrelation with other adjacent parts of the overall net-work is a real demand. Based on these investigations, expres-sive evaluation of the existing relaying schemes of the wind farm and possibility of improving their characteristics can be achieved.

The primary objective of this paper is to examine in detail the performance of the relaying and protection systems cur-rently in use with existing wind farms. These investigations are carried out based on well prepared simulation examples. Among the known packages for dynamic simulation purposes MATLAB [7] was employed for developing a successful dy-namic simulation of a wind farm that is used to carry out this study due to its modeling capabilities and superior develop-ment facilities. The major effective factors that may influence the wind farm behavior during faults are considered such as fault type, fault resistance, fault location, wind speed, …. etc. A 225 MW in Al-Zaafarana- Egypt is considered as a simula-tion example for this study.

The paper was organized as follows. The description of the basic wind farm protection systems is outlined in section II. The description of the Al-Zaafarana wind farm, that is used as a simulation example, as well as its dynamic model prepara-tion are explored in section III. Section V highlights the simu-lation results and its corresponding fault analysis study. Fi-nally, the paper recommendations and conclusions are summa-rized in section VI.

Performance Evaluation of Conventional Protection Systems for Wind Farms

Tamer A. kawady, MIEEE, Naema Mansour and Abd El-Maksoud I. Taalab, Senior MIEEE

©2008 IEEE.

Aerodynamicpart Generator

Tower Cable

Local Cotroller Local

TransormerCollector

feedr

Other coll. feeders

Point of Common Coupling

Main Step-up

Transormer

Grid-Connection

Generator area

Feeders collecting area

Grid Interconnection

area

C.B C.B

Other wind turbine units

Fig. 1 Principle layout of a typical wind farm system.

II. CONVENTIONAL PROTECTION SYSTEMS FOR WIND FARMS

Fig. 1 shows a schematic of a typical wind farm consisting of (n) units of wind turbines. Nowadays, modern wind farms include 20 to 150 units with typical size from 0.5 MW to 3 MW wind turbine generators. Larger sizes up to 5 MW are re-cently available in the market, in which they were successfully installed in some European countries. The use of induction generator in wind farm installations is today a standard prac-tice, due to its suitable characteristics for the wind turbines. The typical generator’s terminal voltage may range from 575 to 690 V with frequency of 50 (or 60) Hz. The generator ter-minal voltage is stepped up to the Collector Bus system with typical voltage of 22 to 34.5 kV. The step up transformer is an oil cooled, pad mounted located at the base of the wind turbine unit. Sometimes, the step up transformer is mounted in the turbine nacelle. Certain considerations should be applied for avoiding the harmonic effects. The typical wind farm collector system consists of a distribution substation collecting the out-put of the distributed wind turbine generators through the in-coming feeders. Usually some reactive power compensation units are provided by a collection of switched capacitors. Fi-nally, the collected power is transferred to the utility side via an interconnection step up transformer.

The wind farm protection system is usually divided into dif-ferent protection zones including the wind farm area, wind farm collection system, wind farm interconnection system and the utility area. First, the induction generator protection is typically accomplished via the generator controlling system covering some certain protection functions such as under/over voltage, under/over frequency, and generator winding tem-perature (RTDs). The generator control system does not con-tribute for the interconnecting system or the utility zone. The generator is protected against short circuits with its circuit breaker, which is practically dimensioned to 2-3 times the generator rated current. The generator step up transformer is usually protected with fuses dimensioned to 2-3 times its rated current. The collector feeder protection is simplified consider-ing it as a radial distribution feeder using overcurrent protec-tion (50/51). A basic challenge arises due to the distributed generators connected together to the radial feeder in determin-ing the minimum faulty zone. That is in order to keep the re-

maining sound parts of the farm supplying the power. On the other hand, the protection of the wind farm substation collec-tor bus and main power transformer consists of multi-function numerical relay system including main transformer differential relay, transformer backup overcurrent relay, collector bus dif-ferential relay and breaker failure relay. Further details are available in the literatures [2]-[4]. It should be considered that, the wind farm interconnection would be applied to MV distri-bution network, HV system ... etc. Therefore, the coordination of utility relays and the wind farm will be quite different. Communication system with dedicated SCADA is quite im-portant for wind farm operation. Nowadays, the data from each wind generator control is transmitted via optic cables and spread to the main substation for general control and monitor-ing purposes. This provides an ideal situation for providing them with an integrated monitoring and control system.

III. MODELING OF AL-ZAAFARANA WIND FARM

A. Al-Zaafarana wind farm description

A 225 MW wind farm was recently established in Al- Zaafa-rana (220 south east of Cairo, Egypt) and connected to the 220 kV Egyptian grid. This promising area is distinctive with dif-ferent superior features such as an average annual wind speed of 9.5 m/s, and its excellent geographical and environmental features. The farm was structured through five stages of 30, 33, 30, 47 and 85 MW respectively as illustrated in Fig. 2. Ex-cept the latter one, other stages are with fixed speed and vari-able pitch operation. Currently, two further stages are being constructed adding extra 205 MW to the farm.

Fig. 2 Geographical outline of Al-Zaafarana wind farm

The fourth stage of the farm was selected as a simulation ex-ample in the paper. It consists of 71 wind turbines (with a 660KW squirrel cage induction generator for each turbine) providing a total power of 47MW. Fig. 3 illustrates its distri-bution schematic, in which the associated wind turbines are

compounded at feeders 1,2,3 and 4 of the farm collecting feeders with 21, 17, 18 and 15 turbines for each feeder. Each wind turbine is connected to a 690V:22- KV local step-up transformer. The collected power are then fed to the 220 kV network through two 75 MVA, 22/220 kV step-up transform-ers. The rest step-up transformers are shared with other feed-ers of other wind farm stages.

Feeder 3

Feeder 2

Feeder 1

Main Step up transformers

Feeder 4

Fig. 3 Schematic of the fourth stage of Al-Zafarana wind farm.

B. Dynamic model Development

Fig. 4(a) illustrates the developed model of the fourth stage of Al-Zaafarana wind Farm in MATLAB-Simulink using the SimPower toolbox. Three masked blocks were utilized for each collecting bus including its own connected turbine units. These buses were assigned as B22_18, B22_15, Bus22 respec-tively. Each of them was connected to the 220 kV bus system “B220” via its related 22/220 kV step-up transformer. The 220 kV grid was modeled with its equivalent constant voltage be-hind its equivalent impedance.

Fig. 4(b) shows the detailed schematic of each wind unit constructed with the built-in wind turbine model in MATLAB. The turbine operation was characterized with the wind speed, the generator speed and its individual pitch control via fixed speed – variable pitch mode of operation. The nominal wind speed was assigned to 9.5 m/sec “the annual average wind speed in its corresponding location”, whereas the “cut-in” wind speed was assigned to be 4.5 m/sec. Each wind turbine was equipped with its induction generator model based on the asynchronous machine built-in model in MATLAB.

The relatively large number of wind turbine units, in which each of them was constructed with different individual items “Turbine, generator, local transformer, feeding cable, …” in-creased remarkably the corresponding source of code. This is characterized with a huge operation time (around 105 min. for each single running on a 3.2 GHz, 2GB-RAM machine). This resulted in impractical testing profile for those simulation pur-poses that are characterized with huge amounts of simulation cases. Moreover, the aforementioned problem is significantly exaggerated for larger systems. Therefore, the need for reduc-ing the overall wind farm model is obvious.

On the other hand, reduced model should be conditioned with the following restrictions:

• Model Accuracy for each individual power sys-tem element should be kept in its higher level

Wind Generation12 feeder

21WTG1 13.86MW17WTG 11.22MW

Wind Generation118WTG10.8MW

Wind Generation115WTG9MW1

GroundingTransformer

X0=4.7 Ohms2

ABCN

abc

GroundingTransformer

X0=4.7 Ohms1

ABCN

abc

GroundingTransformer

X0=4.7 Ohms

ABCN

abc

B22_18(22 kV)2

A

BC

ab

c

B22_15(22 kV)1

A

BC

abc

B220(220 kV)

ABC

abc

B22(22 kV)1

ABC

ab

c

5.5 km line1

ABC

ABC

4km line2

ABC

ABC

4.8 km line

ABC

ABC

3.3ohms 2

3.3ohms 1

3.3ohms

2500 MVAX0/X1=3

ABC

ABC

220 kV/22 kV85 MVA2

ABC

abc

220 kV/22 kV85 MVA1

ABC

abc

220 kV/22 kV85 MVA

ABC

abc

220 kV

NABC

(a)

C3

B2

A1

Wind Turbine

Generator speed (pu)

Pitch angle (deg)

Wind speed (m/s)

Tm (pu) PI

Rate Limiter

pitch

[Vabc_B1][Iabc_B1]

wr

[Iabc_B1]

[wr]

[Vabc_B1]

Data acquisition

Vabc_B1

Iabc_B1

Pmes

Pmec/Pnom

B1

cA

B

C

ab

c

Asynchronous Machine

Tm

mA

B

C

Wind (m/s) 1

<Rotor speed (wm)>

(b) Fig. 4 Simulink-based diagram of Al-Zaafarana wind farm

(a) Overall wind farm diagram (b) Single unit diagram

• The essential concepts for distributed generation must be satisfied.

• Equivalence of currents for each individual unit as well as overall farm currents for both detailed and re-duced model should be realized

• Equivalence of the generated power for each in-dividual unit as well as for the overall farm for both de-tailed and reduced model should be realized.

• Total power losses (due to connecting cables) should be considered.

Fig. 5 illustrates the proposed reduced model for the fourth stage of Al-Zafaranna wind farm. The first three collecting feeders were lumped with their power equivalency with total lumped equivalent generators of 21, 18, and 15 units for each one respectively. For the latter collecting feeder, among its wind turbine generators, 14 generators were represented with their equivalent lumped generator, whereas the rest ones (the first, second and last units) were represented individually for keeping the distributed generation concept. For those lumped units, cable lengths were considered for keeping the total power losses equal to those resulted with the corresponding detailed model. The response of the reduced model was vali-dated compared with the corresponding detailed one via dif-ferent simulation examples for both faulty and non-faulty op-erating conditions. Details for the proposed modeling method-ology were fully addressed in [8].

WT1 15*660KW

WT2 18*660KW

WT3 21*660KW

WT4 17 units

WT1

1*660KW

WT2

1*660KW

WT3

14*660KW

WT17

1*660KW Fig. 5 Schematic of the reduced wind farm model.

IV. SIMULATION RESULTS

Depending on the developed reduced model in the preceding section, the behavior of the simulated stage of Al-Zaafarana wind farm was thoroughly investigated under various faulty and non-faulty operating conditions. The prepared simulation cases covered a wide variety of operating conditions including fault type, fault location, fault resistance and wind speed varia-tions. These cases were applied on the forth collecting feeder as describe in Fig. 6. Four different fault locations were con-sidered: at the generator terminals(position A), before the fus-ing element (position B), after the fusing element (position C) and at the collecting bus (position D). Voltage and current quantities were recorded for each running case at different lo-cations including the generator terminals (Bus690), the high voltage side of the local transformer (BusB1, BusB2, BusB17), the collector bus (Bus22) and the grid connection bus (Bus220). This facilitated to explore the overall performance of the wind farm properly.

Summation 14 unit

Fuse Bus1690

Busb2

Unit_17

.69/22Kv

Fuse

Fuse

Fuse

C B

Busb1

A

C.B1

Collector bus 22Kv

.69/22Kv

Bus2690

Busb17 Bus17690

D

Network

22 /220 Kv

B220

Bus22

Unit_1

Unit_2 C.B2

C.B17

IG

IG

Collector feeder

IG

IG

Fig. 6 Selected fault locations for applied tests.

A. Normal operation with varying wind speed

First, the response of a single wind induction generator was examined for a constant wind speed of 8 m/sec. for one sec-ond. The applied wind speed was then increased to its nominal value of 9.5 m/sec. The measured electrical quantities includ-

ing the generator currents, voltages and power were demon-strated in Fig. 7(a), whereas the corresponding pitch angle variation was illustrated in Fig. 7(b). As remarked, no pitch control reaction was recognized since the wind was still with the nominal wind speed range. The measured electrical quanti-ties including the currents, voltages and power at the 220KV grid bus were demonstrated in Fig. 7(c) to indicate the total power from the 71*660Kw units.

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-1000

0

1000

I b690

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-800

0

800

Vb6

900 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

-1-0.5

00.5

1x 106

time(sec)

P&

Qb6

90

PQ

(a)

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.1

0.2

Pitc

h an

gle

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 28

9

10

Win

d sp

eed

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 21

1.01

1.02

1.03

Rot

or s

peed

(b)

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-200

0

200

I b220

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-2

0

2x 105

V b220

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-5

0

5x 107

P&Q

b220

PQ

(c)

Fig. 7 Simulation response to a change in wind speed from 8m/sec to 9.5m/sec.

(a) Electrical quantities at the generator terminals (b) Pitch angle variation (c) Total wind farm generation at Bus220.

B. Simulating Fault Cases 1) Single line to ground faults

Ground fault is generally the most common fault type in electrical networks, whereas its behavior depends mainly on the fault position, soil resistivity, fault resistance and the ap-plied grounding methodology. For a solid A-G fault at the ge-nerator terminals (position A), the currents and the voltages at the generator terminals (Bus690) are illustrated in Fig. 8. No sensible fault current was remarked as a result of the un-grounded stator winding. The resulting overvoltage permitted the local controller to open the local C.B. within 100m.sec.

0.4 0.45 0.5 0.55 0.6 0.65-1000

0

1000

I b690

0.4 0.45 0.5 0.55 0.6 0.65-1000

0

1000

time(sec)

Vb6

90

Fig. 8. Response to A-G fault at pos. A.

On the other hand, repeating the solid A-G fault before the fuse (position B) yielded the shown fault currents in Fig. 9(a) and (b) fed from the associated local generator and other gen-erating units (in addition to the main grid) respectively. Sur-prisingly, the fault current fed from the local generator was not sufficient to permit tripping of its local breaker (CB1) as re-marked from Fig. 9(a). On the other hand, the accumulated fault current from both the other generating units and the grid network is sufficient to melt the local fusing element as shown remarked from Fig. 9(b). As soon as the aforementioned fuse was melted, the fault was fed only from the local generator characterized with voltage reduction at the generator terminals (Bus690) as shown in Fig. 10. This undervoltage condition, for-tunately, permitted the local control to trip the generator ter-minals. Similarly, a solid A_G fault after the fusing element (position C) resulted in a relatively lower fault current fed from the local generator. The larger counterpart fault current fed from other generating units (in addition to the main grid), on the contrary, did not permit to trip the associated fuse. For-tunately, the collector bus breaker was successfully tripped. Accordingly, the overall power generated with the correspond-ing collecting feeder was totally inhibited pinpointing a typical distributed generation problem.

More complex situations were visualized with non-solid ground faults resulting from the occurred lower fault currents even with small fault resistance values. Also, repeating the fault before the local generator breaker (along the tower cable) is a challenge as well. Then, the need for more advanced pro-tecting schemes for detecting such faults as well as for mini-mizing the tripped generation units is obvious.

0.4 0.45 0.5 0.55 0.6 0.65-2000

0

2000

I b690

0.4 0.45 0.5 0.55 0.6 0.65-800

0

800

time(sec)

V b690

(a)

0.4 0.45 0.5 0.55 0.6 0.65-2500

0

2500

I b1

0.4 0.45 0.5 0.55 0.6 0.65-4

-2

0

2

4x 104

time(sec)

V b1

(b)

Fig. 9. Response to A-G fault at pos. B. (a) Voltages and currents at generator terminals BusB690. (b) Voltages and currents at BusB1 (fed from other wind turbines)

0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8

-2000

-1000

0

1000

2000

I b690

0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8

-500

0

500

time(sec)

V b690

Fig. 10. Response to A-G fault (at pos. B) after clearing the fault.

2) Double phase faults

For A-B fault at the generator terminals (Position A), the re-sulting fault current exceeded the pre-determined current set-ting for the associated generator breaker as well as the fuse of the faulted unit as described in Fig. 11(a) and (b) for both fault feeding currents. The fault was accordingly tripped from both sides. Although the individual fault current fed from the sec-ond wind turbine was demonstrated in Fig. 11(c), its local fuse was not permitted to trip its branch. This was resulted from the obviously larger fault current passing through the fuse associ-ate with the faulty feeder, which accelerating its tripping ac-tion.

Similar behavior was obtained with repeating the same fault condition (A-B fault) before the fuse (position B). It resulted from exceeding both counterparts of the fault current the set-ting boundaries of the associated breaker and the fuse. With repeating the same fault condition after the fusing element (position C), each participated wind generators fed almost the same fault current through its corresponding fuse. This re-sulted in completely losing the overall collecting feeder rather than tripping the faulty branch only. This typical distributed generation figure still represents a challenge for the utilized conventional protection elements.

0.4 0.45 0.5 0.55 0.6 0.65-1

-0.5

0

0.5

1x 10

4

I b690

0.4 0.45 0.5 0.55 0.6 0.65

-500

0

500

time(sec)

Vb6

90

(a)

0.4 0.45 0.5 0.55 0.6 0.65

-500

0

500

I b1

0.4 0.45 0.5 0.55 0.6 0.65

-2

-1

0

1

2

x 104

time(sec)

V b1

(b)

0.4 0.45 0.5 0.55 0.6 0.65-200

-100

0

100

200

I b2

0.4 0.45 0.5 0.55 0.6 0.65-2

-1

0

1

2x 104

time(sec)

Vb2

(c)

Fig. 11 Response to A-B fault at the generator terminals (position A).

(a) Voltages and currents at generator terminal BusB690. (b) Voltages and currents at BusB1. (c) Voltages and currents for wind turbine 2

3) Three phase faults

Fig. 12 illustrates the response for a three phase fault on the generator terminals (position A). Fortunately, the three phase voltage and current quantities at the generator terminals were rapidly decreased to zero. The local controller of the associ-ated generator disconnected its local breaker successfully due to the occurred undervoltage condition.

0.4 0.45 0.5 0.55 0.6 0.65-1

0

1x 104

I b690

0.4 0.45 0.5 0.55 0.6 0.65-1000

0

1000

Vb6

90

0.4 0.45 0.5 0.55 0.6 0.65-1

0

1x 106

time(sec)P

&Qb6

90

PQ

Fig. 12. Three phase fault at the generator terminals

Repeating the same fault before the fusing element (position B) yielded similar voltage and current profiles for the corre-sponding generator. Fortunately, the large fault current feeding from other generation units in addition to the grid network ex-ceeded the fuse setting as seen in Fig. 13(a). On the other hand, other generating units sharing the same step up trans-former had similar voltage and current profiles as shown in Fig. 13(b). This resulted in disconnecting these units by their undervoltage control, if the fuse associated with the faulty unit failed to operate.

0.4 0.45 0.5 0.55 0.6 0.65-1

-0.5

0

0.5

1x 10

4

I b1

0.4 0.45 0.5 0.55 0.6 0.65-2

-1

0

1

2x 104

time(sec)

Vb1

(a)

0.4 0.45 0.5 0.55 0.6 0.65-1

-0.5

0

0.5

1x 104

I b690

0.4 0.45 0.5 0.55 0.6 0.65-1000

-500

0

500

1000

time(sec)

V b690

(b)

Fig. 13 Response to a three phase fault before the fuse (position B). (a) Voltages and currents at the collector feeder terminals (b) Response of an unfaulted unit on the same collector bus

Unfortunately, repeating the same fault condition after the fuse element (position C) resulted in disconnecting the whole collecting feeder. Also, other collecting feeders sharing the same step-up transformer had similar situation.

In order to investigate the impact of network faults, a three phase fault was applied 5 km apart from the grid connection bus. The voltages and currents of all wind generation units were rapidly decreased to zero as shown in Fig. 14.

0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8-4000

-2000

0

2000

4000

I b690

1

0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8-1000

-500

0

500

1000

time(sec)

V b690

1

Fig. 14 Unit response to a three phase fault 5 km apart from the wind farm

V. CONCLUSIONS

There is no doubt that the impressive growth in the utiliza-tion of wind energy systems has consequently exaggerates their remarkable effect on the power system performance. As a result, minimizing the outages of considerable large wind farms is quite essential to keep the security of the whole power system network as much as possible. Hence, providing these wind farms with the proper, selective and integrated protection system has an obvious importance. Toward this goal, under-standing the behavior of these wind farms and their associated protection systems during faulty and non-faulty operating conditions plays a basic role. The paper aimed to visualize a deep investigation of the performance of conventional protec-tion systems that are commonly used with wind farms. For this target, a 225 MW wind farm in Al-Zaafarana-Egypt was dy-namically modeled using the MATLAB package. The wind farm construction represents the collector feeder as a typical radial feeder with multiple distributed generation units. Ac-cordingly, different problems arise with the simple and non-integrated protection schemes that are usually utilized with wind farms. Among these problems, unwanted disconnection of wind generation units, rather than disconnecting the faulty unit only, is not acceptable. Other problems were highlighted depending on the associated fault type as well as its position. The study emphasized the need for enhancing the existed pro-tection schemes for wind farms to realize better power system performance as well as minimize the possible damages result-ing from the fault occurrence. Intelligent techniques may play a role toward this aim. Further research efforts are being car-ried out for fulfilling this target.

VI. REFERENCES

[1] R. Ramakumar, N. Butler, A. Rodriguez and S. Venkata, “Eco-nomic aspects of advanced energy technologies”, Proceedings of the IEEE, Volume 81, Issue 3, March 1993, pp. 318 – 332.

[2] D. Hornak, N. Chau, “Green power - wind generated protection and control considerations”, Protective Relay Engineers, 2004 57th Annual Conference for 30 Mar-1 Apr 2004, pp. 110 – 131.

[3] S. Haslam, P. Crossley and N. Jenkins, “Design and evaluation of a wind farm protection relay”, Generation, Transmission and Distri-bution, IEE Proceedings, Volume 146, Issue 1, Jan. 1999, pp. 37 – 44.

[4] R. Fuchs, “Protection schemes for decentralized power genera-tion”, Developments in Power System Protection, 2004. Eighth IEE International, 5-8 April 2004, Vol. 1, pp. 323 – 326.

[5] K. Maki, S. Repo, P. Jarventausta, “Effect of wind power based distributed generation on protection of distribution network,” De-velopments in Power System Protection, 2004. Eighth IEE Interna-tional Conference, 5-8 April 2004, Vol.1, pp. 327 – 330.

[6] Stefan Bauschke1, Clemens Obkircher2, Georg Achleitner2, Lo-thar Fickert2, Manfred Sakulin2 , PSP 2006, Effect of distributed generation on power system protection “Improved Protection sys-tem for electrical components in wind energy plants”.

[7] The MathWorks Inc., MATLAB, Ver. 7.2, 2006, " http://www.mathworks.com/"

[8] Tamer A. Kawady, Naema Mansour, Aly Osheiba, Abdel-Maksoud Taalab and Rama Ramakumar, “Modeling and Simula-tion Aspects of Wind Farms for Protection Applications”, 40th ANNIVERSARY FRONTIERS OF POWER CONFERENCE Oc-tober 29-30, 2007, Stillwater, Oklahoma, USA

Acknowledgement The authors are expressing their gratitude to the US-Joint for funding this project. Deep gratitude to the Egyptian New and Renewable Energy Authority (NREA) an all members in AL-Zaafarana wind farm for their constructive cooperation and support.

VII. BIOGRAPHIES Tamer A. Kawady (M’02) was born in Shebin El-kom, Egypt on Sept. 30, 1972. He received his B.Sc. (honors) and M.Sc. degrees in Electrical Engineer-ing, Menoufiya University, Egypt, Ph.D. degree (excellent) from Technical University Darmstadt, Germany in 1995, 1999 and 2005 respectively. Dr. Ka-wady is currently an assistant professor at Menoufiya University, Egypt since April 2005. His interests are in digital protection, Power system simulation us-ing the Electromagnetic Transient Program (EMTP) and Artificial Intelligence applications to power system protection. N. M. Mansour is born in El-shohadaa, Egypt on 1978. She received her B.Sc. & M.Sc in Electrical Engineering from Menoufiya University, Egypt in 2000 and 2006, respectively. She has been appointed as a research engineer at the same Department for getting the Ph.D. She conducted her M.Sc thesis on the application of the DWT to line protection. Abdel-Maksoud I. Taalab (M’99–SM’03) received his B.Sc degrees in 1969, in Electrical–Engineering from Menoufiya University, Egypt, M.Sc. and Ph.D degrees from Manchester University (UMIST), U.K., in 1978, and 1982, re-spectively. In the same year of his graduation, he was appointed as an assis-tant professor at the Menoufiya University. He joined GEC Company in 1982. He is now a full Professor at the department of Electrical Engineering, Faculty of Engineering and vice dean of the Desert Environment Institute, Menoufiya University. His interests are in hvdc transmission systems, power system pro-tection, and power electronics applications.