Why Wind Power Works for Denmark CIEN.158.2.66

7/31/2019 Why Wind Power Works for Denmark CIEN.158.2.66

1/7

Hugh SharmanBSc, ACGI

Proceedings of ICE

Civil Engineering 158 May 2005Pages 6672 Paper 13663

Keywordselectrical engineering & distribution;

power stations (non-fossil fuel);weather

is principal of international energyconsulting and broking company

Incoteco (Denmark) ApS

The global oil price rise in the 1970sprompted the Danish Government toswitch to imported coal for its thermalpower stations and to start a wind energyprogramme targeted at generating 10% of

electricity by 2000. The target wasachieved and there are now 5500 windturbinesincluding the worlds twolargest offshore wind farms at Nysted(Fig. 1) and Horns Revproducing the

energy equivalent of 16% of nationaldemand.

This paper reports on performance dataof the west Denmark power grid, towhich 80% of the countrys wind power

is connected. The east Denmark powergrid is entirely separate but both grids areheavily interconnected to the nationalgrids of neighbouring countries to thenorth and south.

Denmark generates more wind power per head of population

than any other country in the world. Its 5500 wind turbines,

including the worlds two largest offshore wind farms, generate

16% of national demand.With increasing concerns over fossil

fuels, the country is now being closely monitored by energy

planners and funders worldwide. However, as this paper

reveals, Denmark is exporting most of its wildly fluctuating

wind power to larger neighbours while finding other solutions

for supply and demand at home.As an island grid based on

slow-reacting thermal power stations, Britain may find its

comparable wind-power aspirations more difficult to achieve.

Why wind powerworks for Denmark

C I V I L E N G I N E E R I N G

Fig. 1. Completed in December 2003, Nysted offshore wind farm in Denmark has 72 turbines with atotal capacity of 166 MW, making it the worlds largest (copyright: Elsam A/S)


2/7

A 15-year programme

In 1990 Denmark had six large powerstations, all of them designed as combinedheat and power plants providing districtheating to west Denmarks largest towns(Fig. 2). Five are coal-fired power stationsand the sixth is designed for ultra-super-

critical steam operation on gas. The coal-fired station near Aalborg has since beenextended and this and the gas-fired sta-tion are now the most efficient Rankine-cycle plants in the world. The installed

capacity of the main thermal power sta-tions at the end of 2003 was 3.5 GW.

During the past 15 years there was alsoan intensive construction programme toupgrade the district heating plants in mostDanish towns and villages to combinedheat and power. The total capacity of thesede-centralised power units in 2004 was

1450 MW. They are mostly based on natur-al gas engines, though a significant numberare bio-fuelled. Many industrial combinedheat and power plants and factories alsosupply heat to the district heating system.

During the 1990s, there was a crash-building programme for wind power,totalling 2374 MW1 at the end of 2003(Fig. 3). Although more wind power overallhas been installed in the US and Germany,the wind intensity of western Denmark isstill unmatched. It is equivalent to 0.88 kWof installed capacity per person in western

Denmark compared with 0.18 kW per per-son in Germany and 0.01 kW per person inthe UK (at the end of 2003).

Furthermore, Denmark boasts one of theworlds most developed wind energy man-

WHY WIND POWER WORKS FOR DENMARK

67C I V I L E N G I N E E R I N G

0 km 30

Local combined heat and power plants

Key:

Thermal power plant

Aalborg

Odense

Aarhus

Kolding

North

Sea

DENMARK

GERMANY

Skive

Fig. 2. West Denmarks six thermal power stations of 3516 MW are sup-plemented by a further 1450 MW from combined heat and power plantsin most towns and villages

0 km 30

Aalborg

Odense

Aarhus

NorthSea

DENMARK

GERMANY

Skive

Wind turbine(s)

Key:

Kolding'Horns Rev'offshore farm

Fig. 3. West Danish wind carpet has an installed capacity of 2374 MW,equivalent to 0.88 kW per person


3/7

ufacturing industries and, as mentionedabove, the worlds two largest offshorewind farms (Fig. 4). West Denmarks pro-duction from wind generators during 2003was 4.36 TWh out of a total of nearly27 TWh, with local consumption at justover 20 TWha record amount.

Highly relevant to the UK

On the face of it, west Denmark hasalready nearly achieved the 20% renew-able energy goal to which the UK aspiresfor 2020. Its proximity and shared lati-tudes with the UKas well as the large

contribution its wind turbine manufactur-ers are already making to the UKs windenergy portfoliomake its experiencewith wind energy highly relevant.

The term wind carpet was coined inDenmark and is frequently used todescribe the way many wind farms jointogether to form what is effectively alarge, virtual power station. Individualwind farms, in different parts of the coun-try, will be observed giving very different

outputs at the same time from winds thatmay be strong in one area but weak, andfrom a different direction, in another.However, from time to time, a cyclonecovers the whole of west Denmark.

West Denmark covers the same lati-tudes as Sunderland to the Moray Firthand is comparable in size to Scotland.Critically, it would appear that the windregime in Denmark is broadly similar tothat in the UKthough it is unarguablethat there are better wind locations in theBritain, particularly in the north and west(Fig. 5). It is therefore worth evaluating

the experience of the Danish experimentwith wind energy in some detail. Indeed,Danish experience is highly relevant to theUK at this very early stage in the develop-ment of a large national wind carpet.

Load factor is just 20%

Load factor (or capacity factor) hasbeen averaging about 20% in westDenmark (Fig. 6). For those not familiarwith the term load factor, if a 1 MWwind generator is working at 100% load

factor, it would produce 1 MW per h,24 h per day, 365 days per year, giving atotal of 8760 MWh per year. If it pro-duces half this annual amount4380 MWhits load factor would be50%, and so on.

In the UK the onshore load factor hasdropped to around 25%. As time goesby and the development of wind powerat the best wind sites gives way to lessfavourable ones, the specific output ateach new site in the UK is likely todecline further.

Furthermore, wind generators are not

exempt from breakdown and, like anyother rotating plant, need regular servic-ing. Breakdown outages and even rou-tine service outages can be extended,especially if these are offshore. In

Fig. 4. Horns Rev wind farm has 80 turbines generating 160 MW (copyright: Elsam A/S)

SHARMAN

68 C I V I L E N G I N E E R I N G

UK onshore

West Denmark

01999 2000 20022001 2003

5

10

15

20

25

30

35

Loadfactor:%

Fig. 6. Load factors are now relatively steadyin west Denmark but declining in the UK(sources: ELTRA andwww.dti.gov.uk/energy/inform/dukes)

North Sea

Atlantic Ocean

0 km 300

>7.5 m/s

Key

6.5 7.5 m/s

5.5 6.5 m/s

Fig. 5. Britain and Denmark have similar wind regimes (source: European Wind Atlas2)


4/7

Denmark there is a very large andhealthy service sector devoted solely tothe maintenance and repair of wind tur-bines. Maintenance and repair of an age-ing wind-carpet probably accounts forsome of the long-term decline in Danishload factor.

It is axiomatic that a high load factorwill result in high output. A lower loadfactor necessitates a larger installedcapacity just to deliver the volume of

energy required. If the Danish load fac-tor is a reasonable guide to that whichwill be reached in the UK, the projec-tions of the capacity needed to deliverthe 60 to 70 TWh in 2020 foreseen bythe UKs energy planners will be large.

In Fig. 7, it can be seen that if 70 TWhare needed to fulfil the UK Governmentsrenewable energy targets, this can be sup-plied from 23 GW if the load factor is35%but 41 GW would be needed tosupply the same amount if the load factorwere to be as low as the wind farms ofwest Denmark. Each GW of wind gener-

ation costs roughly 1 billionthis is theaverage cost of equal amounts of offshoreand onshore wind capacityso the finan-cial penalty for not getting the load factorright is enormous. To these costs must beadded the infrastructural costs of deliver-ing the power to load centres and theaccompanying costs of standby powerand load balancing.

A highly variable supply

From an installed capacity in west

Denmark of 2374 MW there were nineoccasions during 2003 when the windcarpet output exceeded 2000 MW(Fig. 8). Furthermore, there were manyevents during 2003 when the whole car-pet was delivering an output of over 75%of its installed capacity. When this hap-pens, it behaves as a single power station,responsive only to nationwide weatherconditions, not demand.

Sometimes the Danish wind carpetproduces maximum output when thereis little demand. On other occasions itdelivers no energy when demand is high.

There were 54 days in 2002, for exam-ple, when wind supplied less than 1% ofdemand (Fig. 9). On one of those days(16 August 2002) the wind power sys-tem steering requirements exceeded

wind output and the wind carpet con-sumed more power than it could pro-duce. In other words, the wind carpetbecame a net energy consumer. Therewas also a whole week in February 2003when virtually no wind power was gen-erated in west Denmark.

In view of the Danish experience, it isprobably imprudent to believe the UKcan expect firm capacity from its windcarpet, despite its larger size and more

westerly geography. If the wind does notblow, no power can be generatedandhigh-pressure systems can be very largeand persistent. Paradoxically, if the windblows too much, there is also no powergenerated. A modern wind turbine needsa wind speed of around 4 m/s before itcan begin generating but it must featherits blades at around 25 m/s to preventmechanical damage (Fig. 10).

Figure 11 illustrates how the country-wide wind carpet, consisting of the modernwind machines on which the UKs renew-

ables hopes are founded, actually behaves.Output increases more or less linearlybetween a wind speed of 5 m/s and13 m/s. Above this speed, output levels offand, at 20 m/s (45 mph, force 8), stallsaltogether for many machines in the sys-tem. Most modern turbines are designed tooperate above this speed with a 20% lower



020 25 30 35

5

10

15

20

25

40

30

45

35

Capacity:GW

Load factor: %

Fig. 7. Installed wind farm capacity needed todeliver the UK Governments 2020 target of70 TWh increases dramatically as load factordecreases

0

500

1000

1500

2000

2500

Power:MW

Fig. 8. Wind power production in west Denmark exceeded 2000 MW nine times during 2003

0

500

-500

1000

1500

2000

2500

3000

3500

Power:MW

Demand

Wind output

Fig. 9. Wind power during the week of 1117 August 2002 fell a long way short of overall demand

January February March April May June

Month (2003)

Date (August 2002)

July August September October November December

11 12 13 14 15 16 17


5/7

SHARMAN

output and even these are turned off com-pletely when the wind speed exceeds25 m/s (56 mph, force 910).

How the Danish grid is balanced

In every power grid, load and demandare balanced dynamically as there iseffectively no storage in the transmissionand distribution system. This is not to beconfused with energy storage systems

(e.g. pumped or compressed air storage),which buy low-cost power in times ofexcess generation and deliver it verymuch as any other generator in times ofhigher demand.

As mentioned above, the west and eastDanish grids have strong interconnectionswith neighbouring countries but are notlinked to each other. West Denmark is tiedinto the much bigger grids of its neigh-bours Sweden, Norway and Germany witha total interconnector capacity of2400 MW (Fig. 12). This is equivalent totwo-thirds of the regions peak winter

demand and, interestingly, about the sameas its current wind capacity.

As shown in Fig. 13, west Denmarkmakes full use of its interconnections forbalancing wind power as there is a strongcorrelation between wind output and netpower outflows. However, the intercon-nectors were built primarily to linkNorway and Sweden to Germany and,without their prior existence, it may nothave been viable for west Denmark tobuild wind capacity on the scale it has.

Furthermore, the success of the inter-

connections has much to do with theextent to which both Sweden andNorway generate hydropowerwhichcan supply 50% and nearly 100% of


0

20

40

60

80

100

0 5 10 15 20 25 30 35

Outputasaproportion

ofcapacity:%

Wind speed: m/s

If installed capacity =2315 MWthen +/ 1 m/s => +/320 MW

% of installed capacity

Fig. 11. How power output in west Denmark increases, decreases and finally shuts down with increasing wind speed (source: ELTRA)

Fig. 10. The Vestas V80 2 MW turbines at Horns Rev need 4 m/s wind speed before they can begin gener-ating but must feather their blades at around 25 m/s to prevent mechanical damage (copyright: Elsam A/S)


6/7

their respective needs from water tur-bines. Hydropower output can be adjust-ed very rapidly as the highly variablewind power flows through the intercon-nectors.

It is also relevant that all three neigh-bouring systems, including Germanys,are many times bigger than westDenmarks and can act as a power sink tostabilise west Denmarks much smallergrid. Even in 2000, when wind capacity

was 1000 MW less, the Danish balancingsolution was mainly handled across theinterconnectors.

The extent of the balancing challengecan be seen in Fig. 14. This shows thechanges in wind carpet output each hourduring 2002, expressed as a percentageof the total carpet capacity, which wasthen 2310 MW. In a conventional grid,thermal power stations must balancethese flows by ramping up and down orby some other means. There is littledoubt that this chart will be representa-tive of the manner in which a UK wind

carpet will perform.Norway and Sweden, with their mainly

hydro-supplied grids, are generally ableto accommodate power surges from westDenmark. Hydro is a good match forwind power and both these countries arewell placed to accommodate a large pene-

tration of wind capacity in future. In con-trast, an island system depending mostlyon thermal generating capacity will haveto accommodate power swings by ramp-ing thermal capacity up and down, some-times very quickly.

Problems with forecasting

Although the climate of the British Islesis notoriously fickle, Denmark also experi-ences difficulties with weather forecasting.Fig. 15 shows a bad but unfortunately



-2000

-1500

-1000

-500

0

500

1000

1500

Power:MW

h/h

Time: h

120

39

58

77

96

115

134

153

172

191

210

229

248

267

286

305

324

343

362

381

400

419

438

457

476

495

514

533

552

571

590

609

628

647

666

685

704

723

742

Wind power output

Net exchange

Fig. 13. West Denmarks wind power output compared to net power outflow to neighbouring coun-tries in December 2003the amounts are closely in balance

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

30%

Outputchangeperhourasproportio

n

ofcapacity:%

Fig. 14. Hourly changes in west Danish wind output power in 2002 as a percentage of overall wind

capacity. In the absence of interconnectors, thermal power stations would need to be ramped upand down to match this

0 km 100

Sweden

Norway

Denmark

Germany

NorthSea

Skagerrak

BalticSea

950MW

1000M

W

580MW

610MW

800MW

1200MW

Fig. 12. West Denmark has a total intercon-nector capacity of 2400 MW with its neigh-bours Germany, Norway and Swedenaboutthe same as its installed wind power capacity

January February March April May June

Month (2002)

July August September October November December

0

150

300

450

600

750

900

1050

1200

1350

1500

Power:MW

Time: h

Forecast wind power outputActual wind power output

2 4 6 8 10 12 14 16 18 20 22 24

Fig. 15. Forecast versus actual wind power output in west Denmark on 27 October 2002 (source:Jyllands Posten)


7/7

SHARMAN

common day for Danish wind forecasts.Wind forecasting to the requirements of

the transmission system operator is a majorchallenge as even the best mean averagewind speed can only be predicted to within1 m/s. As indicated in Fig. 11, for westDenmark wind power this equates to +/-320 MW or 12% of wind capacity.

West Denmarks transmission systemoperator, ELTRA, is making significantinvestments to improve the accuracy of

wind forecasting and it is possible thatduring the next 10 to 15 years moreaccurate weather predictions will makesurprise events somewhat rarer. However,it is unlikely they can be ruled out alto-gether or that their probability can bereduced to an insignificant level.

Conclusions

Denmark has the most intense windcarpet in the world, with a total of

3000 MW installed by the end of 2003equivalent to 0.88 kW of wind energyper person in west Denmark. The aver-age annual load factor for the wind tur-bine carpet in west Denmark is measuredat approximately 20%. There are consid-erable and often rapid output variationsthroughout the day and throughout theyear. Accurate forecasting of wind speedsis still difficult and output rarely matchesdemand, sometimes dropping below zero

as stalled wind turbines still requirepower for their steering systems.The variations, which are inherent in

any wind energy system, can be readilyaccommodated in west Denmark becausethere are very strong electrical connec-tions to the much larger grid systems ofNorway, Sweden and Germany that canabsorb these variations, particularly dueto their reliance on rapid-reactinghydropower. Countries such as the UK,which operate an island grid, will find it

difficult to do this with slower-reactingthermal power stations and may thushave to limit their reliance on windpower.


References1. ELTRA. Oversigt 2003 (Overview 2003).

ELTRA,Fredericia,Denmark, 2003.

2. TROEN I and PETERSEN E.L. European Wind

Atlas. Ris National Laboratory, Roskilde,Denmark,1989.

What do you think?If you would like to comment on this paper,

please email up to 200 words to the editor at

[email protected].

If you would like to write a paper of 2000 to 3500

words about your own experience in this or any

related area of civil engineering, the editor will be

happy to provide any help or advice you need.

Documents

Why Wind Power Works for Denmark CIEN.158.2.66