Burton T. (et. al.) Wind energy Handbook

Подождите немного. Документ загружается.

9.6.3 Support mechanisms for wind energy

Historically, electrical energy from wind turbines was not competitive in commer-

cial markets with other forms of generation, particularly the use of a combined cycle

gas turbine (CCGT) plant burning natural gas. Hence, in order to take account of

external costs, and to meet commitments to reduce CO

emissions, various supp ort

mechanisms have been used by gover nments to encourage the development of

wind power as well as other forms of renewable energy. These support me chan-

isms, together with the markets for electrical energy, are subject to very rapid

change but the main principles are described.

Perhaps the most obvious approach to support wind power is to require ﬁxed

premium tariffs to be paid for all power generated by renew able sources. This was

the basis for the ‘Public Utilities Requirement to Purchase Act (PURPA) introduced

in the USA in 1978 but abandoned in the late 1980s and the German ‘Electricity

Feed-in-Law’. A similar approach was adopted in Spain and Denmark for a period.

The PURPA legislation was based on ‘avoided cost’, i.e., the marginal co st of

generation from an alternative plant was paid. This varied from state to state but in

California in the early 1980s the avoided cost was calculated to be US$0.06/kWh.

This rate coupled with an inﬂation factor and tax incentives led to some 2000 MW

of wind turbine capacity being installed between 1981 and 1989. However, develop-

ment slowed dramatically as gas ﬁred CCGT plant reduced the avoided cost. The

German ‘Electricity Feed-in-Law’ requ ired the utility to purchase wind generated

power at 90 percent of the average price paid by all customers. As the bulk

wholesale price of electrical energy is typically a third of the retail price it may be

seen that this was a most attractive arrangement for wind farm operators. The cost

of the ‘Feed-in-Law’ was borne by the utilities who made efforts to have the law

rescinded. Denmark also operated a similar system with the utilities paying 70–

85% of the retail price for wind-generated power. This rate, together with tax

incentives, encouraged individuals and co-operatives to buy wind turbines and

some 75% of turbines in Denmark are owned in this way.

The general effect of this type of premium tariffs is to provide very considerable

stimulus to the development of wind power. However, this form of support may

not apply strong downward pressure on wind farm costs. Hence, the price paid per

kWh in Germany in 2000 was approximately double that available under the last

round of the Non Fossil Fuel Obligation in the UK which was based on an auction

of re newable contracts. However, it should be noted that the UK sites generally had

higher wind speeds than many of those in Germany. It is difﬁcult to reconcile the

concept of ﬁxed premium tariffs with that of a deregulated market for electrical

energy and so this form of support may not be acceptable to European Union

competition authorities in the long term.

Both the UK and Ireland have used a process based on competitive auctions of

ﬁxed-term, ﬁxed-price power purchase contracts. This type of support mechanism

has the effect of providing very considerable downward pressure on costs and also

offers the devel oper stability as the price is guaranteed for the long term (in the UK

for 15–20 years with an additional allowance for general inﬂation). The major

disadvantages are that: (1) the auctions are held intermittently and so there is no

continuous market for the turbine manufacturers; (2) very considerable costs are

FINANCE 555

incurred in bidding which may be wasted if the bid is unsuccessful; and (3) there is

little penalty on not constructing the projects and so some bids may be unrealisti-

cally low. The UK Non Fossil Fuel Obligation (NFFO) was remarkable in reducing

the price for wind energy by two-thirds from 1989 (9p/kWh) to 1999 (2.9p/kWh)

although the 1989 contracts were for only 9 years while the 1999 contracts were for

20 years. However, only limited capacity was actually built. The NFFO has now

been abandoned and has been replaced by an obligation on electricity suppliers to

provide a percentage of their energy from renewable sources and a system of

tradable ‘Green Certiﬁcates’.

The use of some form of ‘Green Certiﬁcates’ together with an obli gation on users

or suppliers to ensure that a fraction of their electrical energy comes from renewable

sources is presently seen as the most likely future development for support

mechanisms. There need be no physical conn ection between the ‘Green Certiﬁcate’

and the electrical ene rgy and the ‘Certiﬁcates’ may be traded in one market and the

energy (kWhs) in another. The advantages of this type of arrangement are that: (1)

it is compatible with a competitive market for electrical energy; (2) it is likely to

exert a downward pressure on costs; (3) it may develop a large market attractive to

major companies; and (4) the full range of market instruments (e.g., futures

markets) may be developed.

If Government support mechanisms are to be based on som e form of ‘Green

Certiﬁcates’ then the wind farm operator is still left with the problem of selling the

energy. In markets for electrical energy, the wholesale price varies considerably

throughout the day and year and so the wind farm operator is likely to be exposed

to volatile prices. After privatization of the electricity supply industry in the UK a

pool-type trading arrangement was developed whereby any generator might sell

into the pool at the system marginal price of generation. Additional contracts made

outside the pool, the so-called ‘contracts for differences’ were used by many

conventional generators to manage the uncertainty of the variation of the pool price

but the availability of the pool marginal price always provided a reference price for

power. However, the pool is now being superseded by a system of bilateral energy

contracts which will place a very high premium on controllability and predictability

of generating plant. The New Electricity Trading Arrangements (NETA) will

comprise three elements: (1) long-term bilateral contracts between generators, large

customers and those supply companies serving smaller customers; (2) a short-term

market, 24–3.5 hours ahead of delivery, between generators, customers and suppl i-

ers; (3) a balancing mecha nism, 3.5 h ahead of delivery, operated by the National

Grid Company (the Transmission Operator). Failure to match contracted demand

with supply will probably result in higher costs in the short-term market or

balancing market. This is likely to provide considerable difﬁculties for wind farm

operators due to the intermittent and unpredictable nature their output.

Some consumers are prepared to pay a premium for electrical energy provided

from renewable sources. This ‘Green Pricing’ is being developed in Germany and the

UK but the extent of the market demand it is not yet clear. There are also difﬁculties

with accrediting which generating plants are ‘green’ and suitable for such a scheme.

In parts of the USA ‘net metering’ is allowed for some small wind turbines. This

is the use of a single electricity meter which operates backwards to reduce the

measured kWh at times of export from the wind turbine and its local load, typically

556 WIND TURBINE INSTALLATIONS AND WIND FARMS

a house or farm. In economic terms this approach can be criticized as it ignores the

variation with time in the value (and price) of electrical energy and also that a large

element of the cost of electrical energy delivered to the customer is associated with

the transmission and distribution facilities and their management.

References

Addison, J. F. D. et al., (2000). ‘A neural network version of the Measure–Correlate–Predict

algorithm for estimating wind energy yield’. International Congress and Exhibition on

Condition Monitoring and Diagnostic Management, Texas, USA.

British Standard, (1999). Wind turbine generator systems – Part 11: Acoustic noise measure-

ment techniques’. BS EN61400–11:1999. BSI.

British Standard, (1997). ‘Method for rating industrial noise affecting mixed residential and

industrial areas’. BS4142: 1997. BSI.

British Wind Energy Association, (1994). ‘Best practice guidelines for wind-energy develop-

ment’. British Wind Energy Association, www.bwea.com.

Colson, E. W., (1995). ‘Avian interactions with wind energy facilities: a summary’. Proceedings

of the American Wind Energy Association Conference, Windpower ’95, pp 77–86.

Danish Statutory Order on Noise from Windmills, (1991). ‘Bekendtgørelse om stoj fra wind-

moller’ (in Danish).

Department of the Environment, (1993). ‘Planning Policy guidance note 22: Renewable

energy, annex on wind energy’. PPG22, Department of the Environment, Welsh Ofﬁce.

Derrick, A., (1993). ‘Development of the Measure–Correlate–Predict strategy for site assess-

ment’. Proceedings of the European Wind Energy Association Conference, pp 681–685.

English Nature, (1994). ‘Nature conservation guidelines for renewable energy projects’.

English Nature, Peterborough, UK.

ETSU, (1993). ‘Attitudes to wind power – a survey of opinion in Cornwall and Devon’. ETSU

Report W/13/00354/038/REP. ETSU, Harwell, UK.

ETSU, (1994). ‘Cemmaes wind farm sociological impact study – ﬁnal report’. ETSU Report

W/13/00300/REP. ETSU, Harwell, UK.

ETSU, (1995). ‘Wind farm radar study’. ETSU Report W/32/00228/49/REP. ETSU, Harwell, UK.

ETSU, (1996). ‘Birds and wind turbines: can they co-exist’. ETSU-N-113. ETSU, Harwell, UK.

ETSU, (1997a). ‘The assessment and rating of noise from wind farms’. Final Report of the

Working Group on Noise from Wind Turbines, ETSU-R-97. ETSU, Harwell, UK.

ETSU, (1997b). ‘Investigation of the interactions between wind turbines and radio systems

aimed at establishing co-siting guidelines, Phase 1: Introduction and modelling of wind

turbine scatter’. ETSU Report W/13/00477/REP. ETSU, Harwell, UK.

ETSU, (1999). ‘UK Onshore wind energy resource’, ETSU Report R-99, ETSU, Harwell, UK.

ETSU, (2000). ‘A critical appraisal of wind-farm noise propagation’, ETSU Report W/13/00385/

REP. ETSU, Harwell, UK.

EUREC Agency, (1996). ‘The future for renewable energy, prospects and directions ’. James and

James, London, UK.

European Wind Atlas, (1989). Riso National Laboratory, Roskilde, Denmark, and www.wind-

power.dk.

EWEA. ‘European best practice guidelines for wind-energy development’. European Wind

Energy Association, www.ewea.org.

Gipe, P., (1995). Wind energy comes of age. John Wiley and Sons, New York, US.

Hall, H. H. (1992). ‘The assessment and avoidance of electromagnetic interference due to

windfarms’. Wind Engng 16, 6, 326–337.

REFERENCES 557

International Energy Agency, (1994). Recommended practice for wind turbine testing 4. Acoustics

– Measurements of noise emission from wind turbines, Third Edition.

International Energy Agency, (1997). Recommended practice for wind turbine testing 10. Measure-

ments of noise emission from wind turbines at noise receptor locations, First Edition IEC61400-11

1998.

International Energy Agency, (1998). Wind turbine generator systems Part II: Acoustic noise

measurement techniques. IEC 61400-11: 1998.

Khatib, H., (1997). Financial and economic evaluation of projects in the electricity supply industry.

IEE Power Series, London, UK.

Landberg, L. and Mortensen, N. G., (1993). ‘A comparison of physical and statistical methods

for estimating the wind resource at a site’. Proceedings of the 15th British Wind Energy

Association Conference (ed. K. F. Pitcher). Mechanical Engineering Publications, Bury St

Edmunds, UK pp 119–125.

Madders, M. and Walker, D. G., (1999). ‘Solutions to raptor–wind farm interactions’.

Proceedings of the 21st British Wind Energy Association Conference (Ed. P. Hinson). Mechani-

cal Engineering Publications, Bury St Edmunds, UK, pp 191–195.

Moglia, A., Trusszi, G. and Orsenigo, L., (1996). ‘Evaluation methods for the electromagnetic

interferences due to wind farms’. Proceedings of the Conference on Integration of Wind Power

Plants in the Environment and Electric Systems, Paper 4.6.

Mortimer, A. A., (1994). ‘A new correlation/prediction method for potential wind farm sites’.

Proceedings of the 16th British Wind Energy Association Conference (Ed. G. Elliot). Mechanical

Engineering Publications, Bury St Edmunds, UK, pp 349–352.

Pinder, J. N., (1992). ‘Mechanical noise from wind turbines’. Wind Engng, 16, 3, 158–168.

Spera, D. A., (1994). Wind turbine technology: fundamental concepts of wind turbine engineering.

ASME Press, New York, US.

Stanton, C., ‘The visual impact and design of wind farms in the landscape’. Proceedings of the

16th British Wind Energy Association Conference. Mechanical Engineering Publications, Bury

St Edmunds, UK, pp 249–255.

Stiesdal, H. and Kristensen, E., (1993). ‘Noise control on the BONUS 300 kW wind turbine’.

Proceedings of the 15th British Wind Energy Association Conference. Mechanical Engineering

Publications, Bury St Edmunds, UK, pp 335–340.

Still et al., (1994). ‘The birds of Blyth Harbour’. Proceedings of the British Wind Energy

Association Conference. Mechanical Engineering Publications, Bury St Edmunds, UK,

pp 241–248.

Taylor, D. and Rand, M., (1991). ‘Planning for wind energy in Dyfed’. EERU 065. Energy and

Environment Research Unit, Open University, UK.

Van Kats, P. J. and van Rees, J., (1989). ‘Large wind turbines: a source of interference for TV

broadcast reception’. Wind energy and the environment (Ed. Swift-Hook, D. T.). IEE Energy

Series, London, UK, pp 95–104.

Verkuijlen, E. and Westra, C. A., (1984). ‘Shadow hindrance by wind turbines’. Proceedings of

the European Wind Energy Conference. pp 356–361.

Wagner, S., Bareis, R. and Guidati, G., (1996). Wind turbine noise. Springer-Verlag, Berlin,

Germany.

Walmsley, J. L., Salmon, J. R. and Taylor, P. A., (1982). ‘On the application of a model of

boundary-layer ﬂow over low hills to real terrain’. Boundary-layer Meteorology, 23, 17–46.

Walmsley, J. L., Taylor, P. A. and Keith, T., (1986). ‘A simple model of neutrally stratiﬁed

boundary-layer ﬂow over complex terrain with surface roughness modulations –

MS3DJH/3R’. Boundary-layer Meteorology, 36, 157–186.

P–www.wasp.dk and Chapter 8 of European Wind Atlas. Riso National Laboratory,

Roskilde, Denmark, www.windpower.dk.

558

WIND TURBINE INSTALLATIONS AND WIND FARMS

Electrical Systems

10.1 Power-collection Systems

Figure 10.1 shows a schematic of a typical ﬁxed-speed wind turbine electrical

system. The main power circuit is from the generator, via three ﬂexible pendant

cables to a mo ulded case circuit breaker (MCCB). The MCCB is ﬁtted with an

overcurrent protection trip with an instantaneous setting against faults and a

delayed (thermal) function which operates at a lower current level and is intended

to detect overload of the generator. This arrangement follows normal practice for

the control of large motors. An anti- parallel thyristor soft-start unit, normally with a

bypass contactor, is used to reduce the current inrush when the generator is

energized. There are a number of additional circuits including those for the power

factor correc tion capacitors (PFC) and the auxiliary supplies. The power factor

correction capacitors are switched in stages to provide a greater degree of control of

reactive power and also to limit capacitive switching currents. Small reactors may

be connected in series with the capacitors to reduce the inrush current.

Most of the auxiliaries use alternating current but there may also be a require-

ment to provide direct current for the wind turbine controller. The auxiliary circuits

are protected by a smaller MCCB and fuses rated at lower currents. Surge diverters

are ﬁtted to protect the internal electrical system from overvoltages being trans-

mitted from the site electrical network and another set of surge diverters may be

connected at the generator terminals in the nacelle. The normal scope of supply of

Turbine transformer

MCCB and protection trip

Surge diverter

Pendant cable

Soft-start

PFC Capacitors

Generator

Surge diverter

110 V auxiliaries

690V auxiliaries

Lights and sockets

Supply to controller

Figure 10.1 Electrical Schematic of a Fixed-speed Wind Turbine

the wind turbine manufacturer will end at the low voltage connection to the turbine

transformer.

The voltage level chosen for the main power circuit from the gen erator to the

tower base is usually less than 1000 V and is selected to be one of the internationally

standard voltages, in Europe typically 690 V between phases. This voltage is

surprisingly low for a large electrical machine and leads to rather high currents. For

example a 600 kW wind turbine opera ting at 690 V requires a current of over 500 A.

However, it is found to be convenient and cost-effective to restrict all voltages

within the wind turbine to less than 1000 V as, in many countries, the safety

regulations become very much more severe for voltages above this value and

special precautions are required including the provision of dedicated equipment to

earth the circuits before work on them is permitted. Further reasons for this low

voltage include a wider choice of switchgear and ﬂexible cables for the pendant and

higher production volumes and hence lower costs of the generators.

However, this low generator voltage leads to a requirement for a transformer

located in the tower or immediately adjacent to it (e.g., Figure 10.2). In some early

wind farms a number of small wind turbines were connected to one transformer

(Figure 10.3) but as turbine ratings have increased with consequently higher cur-

rents this approach would lead to excessive voltage variations and electrical losses

in the low-voltage cables. It is only with rather small wind turbines located close

together that the connection of multiple turbines to one transformer is likely to be

cost-effective.

Figure 10.2 Transformer and Switchgear of 1.5 MW Wind Turbine, MV Switchgear on right,

LV Switchgear on left (Reproduced by permission of NEG MICON; www.neg-micon.dk)

560

ELECTRICAL SYSTEMS

The choice of medium-voltage (MV) level for the wind-farm power-collection

system is usually determined by the practice of the local distribution utility. In this

way cable and switchgear is readily available. Thus, in the UK the choice is

generally between 11 kV and 33 kV while in continental Europe wind farm collec-

tion circuits are likely to be either 10 kV, 15 kV or 20 kV.

Local utility practice also inﬂuences the choice of the neutral earthing arrange-

ment on the MV circuits. 11 kV circuits in the UK tend to have their neutral point

either solidly earthed or connected to ground through a low-value resistor (typi-

cally 6.35  to allow 1000 A to ﬂow into a phase/ground fault). Solid neutral

earthing is cheaper as it requires no extra equipment but can lead to high earth fault

currents which may cause damage and high step or touch voltages (ANSI/IEEE,

1986). In contrast, on some continental European wind farms the neutral point of

the medium voltage system is left unearthed. This allows operation of the wind

farm to continue with a single phase-earth fault but with the voltages of the healthy

phases raised by a factor of

ﬃﬃﬃ

The design of the power collection is very similar to that of any MV power

network but with one or two particular aspects. It is generally not necessary to

provide any redundancy in the circuits to take account of MV equipment failures.

Both operating experience and reliability calculations have shown that the power

collection system of a wind farm is so much more reliable than the individual wind

turbines that it is generally not cost-effective to provide any duplicate circuits. This

also applies to the single circuit or transformer linking the wind farm to the utility

network where duplication cannot usually be justiﬁed. If part of the wind power-

collection circuit does fail then the only lost revenue is that of the energy output of

that part of the wind farm. This is easily estimated and is generally mode st. In

contrast, if a public distribution utility circuit fails then the consequence is that the

customers do not have access to electrical power. The commercial value of this loss

of supply is more difﬁcult to quantify but is generally considered to be several

orders of magnitude higher than the price of the energy. Thus, public distribution

circuits often have duplicate supplies while wind farm power-collection networks

tend to consist of simple radial circuits with limited switchgear for isolation and

switching. The wind turbine transformers are connected directly to the radial

circuits.

In Europe almost all power-collection circuits within wind farms are based on

underground cable networks. This is both for reasons of visual amenity and for

Multiple turbines per transformer Individual turbines per transformer

Figure 10.3 Grouping of Turbines to Transformers

POWER-COLLECTION SYSTEMS 561

safety as large cranes are required for the erection of the turbines. In other parts of

the world (e.g., India and the USA) overhead MV lines are sometimes used within

the wind farm to reduce costs. If cables are used then their low series inductive-

reactance leads to small variations of voltage with current and so the circuits are

dimensioned on considerations of electrical losses. Any wind farm project which

reaches the stage of requiring a detailed design of the electrical system should also

have a good estimate of the wind farm output. Thus it is straightforward to use

these data to calculate the losses in the electrical equipment at various wind turbine

output powers using a load-ﬂow program, sum these over the life of the project and

using discounted cash ﬂow techniques choose the optimum cable size and transfor-

mer rating. In countries where wind-generated electricity attracts a high price then

it is likely to be cost-effective to install cables and transformers with a thermal

rating considerably in excess of that required at full wind-farm output in order to

reduce losses. It may also be worthwhile to consider ‘low-loss’ transformer designs.

10.2 Earthing (Grounding) of Wind Farms

All electrical plant require a connection to the general mass of earth in order to:

• minimize shock hazards to personnel and animals,

• establish a low-impedance path for earth-fault currents and hence satisfactory

operation of protection,

• improve protection from lightning and retain voltages within reasonable limits,

and

• prevent large potential differences being established wh ich are potentially hazar-

dous to both personnel and equipment.

In the UK this subject is referred to as ‘earthing’ while in the US it is called

‘grounding’. The terms may be considered to be synonymous.

Because earthing has obvious safety implications there is considerable guidance

available in both national and international standards for conventional electrical

plant. For earthing of AC substations, the US standard (ANSI/IEEE, 1986) is widely

applied and Charlton has written an informative guide (Copper Development

Association, 1997) which is mainly focused on UK practice.

Wind farms, however , have rather unusual requirements for earthing. They are

often very extensive stretching over several kilometres, subject to frequent lightning

strikes because of the height of modern wind turbines, and are often on high-

resistivity ground being located on the tops of hills. Thus, normal earthing practice

tends not to be easily applicable and special consideration is required. The IEEE

Recommended Practice (1991), which is no longer current, recommended ‘that the

entire wind farm installation have a continuous metallic ground system connecting

all equipment. Thi s should include, but not be limited to, the substation, transfor-

562 ELECTRICAL SYSTEMS

mers, towers, wind-turbine generators and electronic equipment.’ This practice is

generally followed with bare conductor being laid in the power-collectio n cable

trenches to provide both bonding of all parts of the wind farm as well as a long

horizontal electrode to reduce the impedance of the earthing system.

A wind farm earthing system is required to operate effectively for both power

frequency 50/60 Hz currents and lightning surges which have rise-times typically

of less than 10 ìs. Although it is conventional to use the same physical earthing

network for power frequency currents and lightning surges the response of the

earthing system to the high-frequency components of the light ing current is

completely dif ferent to that at 50 Hz.

The performance of a wind-farm earthing system may be understood qualita-

tively by considering Figure 10.4.

At each wind turbine a local earth is provided by placing a ring of conductor

around the foundation at a depth of about 1 m (sometimes known as a counterpoise

earth) and by driving vertical rods into the ground. It is common to bond the steel

reinforcing of the wind-turbine foundation into this local earthing network. The

purpose of this local earth is to provide equipotential bonding against the effects of

both lightning and power frequency fault currents and to provide one element of

the overall wind-farm earthing system. Regulations differ for the value of this local

earth but in the UK it is necessary to achieve a resistance to earth of less than 10 

(British Standard, 1999) before the connection of any power or SCADA cables.

These local turbines’ earths are shown as R

turbine

in Figure 10.4. As the turbine-

earthing network consists of a ring of only, say, 15 m in diameter and local driven

rods it may be considered as purely resistive.

KEY

turbine

series

shunt

Figure 10.4 Schematic of Wind Farm Earthing System

EARTHING (GROUNDING) OF WIND FARMS 563

However, the long horizontal electrodes linking one turbine to the next have a

more complex behavio ur (similar to a transmission line) and are represented in

Figure 10.4 as a ð equivalent circuit. Thus the resistance to ground is indicated by

shunt

while the series impedance is the combination of R

series

and L

series

. R

series

comes about simply from the resistance of the earth wire while L

series

is the self-

inductance of the earth circuit. On the long earthing networks found on large wind

farms this series impedance cannot be ignored. It may be seen immediately that for

the high-frequency components of a lightning strike on a wind turbine the series

inductance acts effectively to reduce the earthing network to only the local turbine

earth. Ev en with 50 Hz fault current the series impedance leads to signiﬁcantly

higher earthing impedances than would be expected with geographically small

earthing system s where series impedance may be neglected. In this discussion

shunt capacitance has been ignored although this may become signiﬁcant at very

high frequencies.

This behaviour of wind-farm earthing systems has been conﬁrmed by site meas-

urements and Table 10.1 shows the measured values for the earth impedance at the

main substations of two UK wind farms.

It may be seen that even at 50 Hz the impedance to earth consists of almost equal

resistive and inductive parts (i.e., the X=R ratio of the circuit to earth is almost

unity). This has important implications both for the design and testing of wind-

farm earthing systems. It is not possible to use conventional calculation techniques

which were designed for small, fully resistive, earth networks and it is necessary to

take into account the effect of the long conductors. Sophisticated computer pro-

grams to do this are available and these may be used to consider both 50 Hz and

lightning surge behaviour. Similarly, simple testing methods such as those devel-

oped by Tagg (1964) are only intended to measure resistance to earth and will give

optimistic results on large wind farms. The only effective measurement method to

determine the 50 Hz earth impedance of a large wind farm is the so-called current

injection test. In this test a current (typical ly 10–20 A) is injected into the earth

electrodes at the wind farm and the potential rise measured with respect to a ‘true’

earth. However, the return path of the injected current must be remote from the

wind farm (usually some 5–10 km away) to ensure that it does not effect measure-

ments at the wind farm. Hence it is conventional to use the de-energized connection

circuit of the wind farm to the main power network as the path of the test current.

The rise in potential of the wind-farm earthing systems is measured against a

remote earth which classically is transferred to the site over a metallic telephone

circuit. Ideally the route of the telephone circuit should be orthogon al to the power

circuit to avoid induced effects. It may be seen that tests of this type are extremely

Table 10.1 Measured Earth Impedance at Substations of Two Wind Farms

Capacity of wind farm

(MW)

Number of turbines Route length of

horizontal earth

conductor (km)

Earth impedance at

main substation

( at 50 Hz)

7.2 24 6.7 0:89 þ j0:92

33.6 56 17.7 0:46 þ j0:51

564

ELECTRICAL SYSTEMS